# 404 Memory Found: full text of every article > 404 Memory Found is a nostalgia blog about the technology, games, websites and companies of the 90s and 2000s: what they cost, why they won or lost, and what happened to them. Source: https://404memoryfound.com. Each article below carries its URL, author, publication date, last update, a direct-answer summary, key facts, the article text and its sources. # Is SkyMall Still Around in 2026? Yes, Online Only URL: https://404memoryfound.com/posts/skymall-still-around-who-owns-it.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-19 Topics: Business Blunders, Internet Culture **Summary:** SkyMall is still around in 2026, but only as an online store at skymall.com. C&A Marketing bought the brand out of Chapter 11 bankruptcy in 2015 for $1.9 million, after phones and tablets emptied out the audience that made an in-flight catalog work. The seat-pocket edition that ran from 1990 to 2015 is gone for good. **Key facts:** - Launched: 1990, on Eastern Airlines, from Phoenix, Arizona - Revenue collapse: $33.7 million in 2013 to $15.8 million in the first nine months of 2014 - Sold for: $1.9 million at auction in 2015, against $50.1 million paid in 2001 - Status today: Online store at skymall.com; no standalone seat-pocket catalog - Owner today: C&A Marketing, since April 2015 ## Who put the first catalog in a seat pocket? Robert Worsley was an accountant at Price Waterhouse when the idea arrived on a flight in 1989. He was paging through the Giftmaster catalog in the seat pocket in front of him and decided the merchandise was terrible. Worsley incorporated SkyMall in Arizona and mailed a 100-page business plan to 50 venture capitalists and individual investors. In February 1990, Alan Ashton, the co-creator of WordPerfect (https://404memoryfound.com/posts/what-happened-to-wordperfect-word-processor.html), agreed to finance it with $250,000. Getting started cost nearly $3 million. That money bought a 21,000-square-foot warehouse in Phoenix and a customer service center running NeXT computers and Sun Microsystems servers, which in 1990 was serious equipment for a mail-order startup. The first SkyMall catalog went out on Eastern Airlines in 1990. It carried goods from Hammacher Schlemmer, Spiegel, The Nature Company, The Wine Enthusiast and Bits and Pieces, which SkyMall bought at a 35 percent discount and resold at list. Revenue in that first year was $200,000. The reach was real and the economics were not, which is a sentence that also fits Pets.com (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html) a decade later. ## How did SkyMall get into almost every airline? Once the first carrier signed, the rest came quickly. TWA began carrying the catalog in February 1991 and Continental Airlines added it that April. July 1991 was the flood: Alaska, Atlantic Southeast, Delta, Horizon, SkyWest and USAir all started carrying SkyMall in the same month. The deal was easy for an airline to say yes to. Carriers were paid a percentage of what SkyMall sold, so the catalog cost them nothing up front. The complication was weight. Early editions ran 60 to 80 pages, and paper burns jet fuel. When the sales commissions did not cover the difference, SkyMall paid the airlines for the extra fuel their catalogs consumed. Revenue reached $5.4 million in 1991. The company was also losing about half a million dollars a month at the time. That pairing of enormous distribution and thin margins defined SkyMall for the next 25 years. It sat in front of more American shoppers than most department store chains and still had to argue about postage and fuel surcharges. ## What did people actually buy from it? SkyMall's reputation came from the strange end of the catalog: the life-size garden yeti, the dog staircases, the gadgets that solved problems nobody had. Those pages are the reason anyone remembers the brand at all. The everyday business was duller. SkyMall resold merchandise from established catalog houses such as Hammacher Schlemmer and Spiegel, buying at a discount and keeping the spread. The gap between reading and buying is the whole story, and the man who eventually bought the brand said so out loud. Chaim Pikarski of C&A Marketing told CNN Money in 2015: "Even in the past when people did read SkyMall in-flight, they read it as entertainment. It was not a call to action to buy." That is an unusual thing for an owner to say about his own asset. It also explains why a catalog with that much reach never converted like one that arrives at a house. Catalog retailers that owned their customer lists learned the same lesson from the other side. Montgomery Ward (https://404memoryfound.com/posts/montgomery-ward-failed-who-owns-it.html) built a century of sales on a book a household asked for, not one a traveler met by accident. ## Who owned SkyMall before the bankruptcy? SkyMall did not stay independent. On July 18, 2001, Gemstar-TV Guide International acquired all of SkyMall's outstanding common stock. Gemstar's later annual report to the Securities and Exchange Commission put the aggregate purchase price at $50.1 million, made up of $22.2 million in cash and roughly 741,000 Gemstar shares valued at $36.58 each. SkyMall shareholders received $1.50 and 0.03759 of a Gemstar share for every share they held. That $50.1 million is the figure to remember. Fourteen years later the entire business, brand included, changed hands for under four percent of it. Ownership kept moving after 2001, and every owner inherited the same structural weakness. SkyMall's audience was rented from the airlines. There were no subscribers, only passengers, and they stayed only as long as the carriers kept signing contracts. A catalog business that controls its own mailing list can survive a bad year. Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html) mailed to names it owned for most of a century. SkyMall never owned a name until a customer placed an order, and most passengers never did. ## Why did SkyMall go bankrupt in 2015? Phones finished it. SkyMall's revenue fell from $33.7 million in 2013 to $15.8 million for the nine months ending in September 2014. Acting chief executive Scott Wiley put the cause into the company's 2015 court filings without decoration: "With the increased use of electronic devices on planes, fewer people browsed the SkyMall in-flight catalog." The carriers moved before the creditors did. Delta Air Lines ended its SkyMall contract on November 30, 2014. Southwest Airlines told the company it would stop carrying the catalog after April 1, 2015. Losing two airlines that size gutted the distribution SkyMall's merchants were paying for. SkyMall filed for Chapter 11 protection on January 23, 2015. The collapse looks abrupt in the headlines and was not. Seat-back screens, in-flight wifi and a phone loaded with downloaded video had spent years eating the catalog's only real asset, which was a bored passenger with nothing else to look at. Once regulators let travelers keep their devices switched on from gate to gate, even the taxi and takeoff minutes belonged to the phone. ## Who bought the brand, and what is it now? The assets went to auction in Phoenix in March 2015. Seven bidders prequalified and only two put in an offer. FSG Distributors bid $2.5 million. C&A Marketing bid $1.9 million and won anyway, because its offer carried far fewer contingencies. An Arizona bankruptcy judge approved the sale on March 27, 2015. C&A Marketing, a New Jersey consumer products company, still owns SkyMall and runs it as an online store at skymall.com. It sells gadgets, home goods and travel gear to people who go looking for them, which is the opposite of the original model. Print did come back, briefly. In October 2015 SkyMall returned as an insert inside United Airlines' in-flight magazine, Hemispheres, and the company followed with a 36-page holiday edition mailed to its best customers. Pikarski described the new plan to CNN Money in 2015: "The way you run a catalog business is by mailing them to people or by engaging them in e-commerce." SkyMall in 2026 is the e-commerce half of that sentence. The seat pocket version is finished, and so are most of the airline magazines that used to sit beside it. ### Where to find a SkyMall catalog today Old SkyMall issues are ephemera rather than collectibles, so they trade for about what a used paperback costs rather than what a rarity costs. Sellers usually list them by year and by airline edition, and condition is the only thing that really moves the price. For a copy to keep, browse the vintage SkyMall catalog listings on Etsy (https://www.etsy.com/search?q=vintage+skymall+catalog). For a copy to read, the Internet Archive has scanned full editions from the catalog's final years. ## Frequently Asked Questions ### Is SkyMall still in business in 2026? Yes. SkyMall exists in 2026 as an online store at skymall.com, owned by the New Jersey consumer products company C&A Marketing, which bought the brand out of bankruptcy in 2015. What no longer exists is the standalone SkyMall catalog that sat in airline seat pockets from 1990 until the Chapter 11 filing in January 2015. ### Why did SkyMall go bankrupt? SkyMall filed for Chapter 11 bankruptcy protection on January 23, 2015, after revenue fell from $33.7 million in 2013 to $15.8 million in the nine months ending September 2014. Passengers spent flights on phones and tablets instead of the seat-pocket catalog, and Delta Air Lines ended its SkyMall contract on November 30, 2014, with Southwest Airlines dropping it after April 1, 2015. ### How much did SkyMall sell for? C&A Marketing bought the SkyMall assets for $1.9 million at a bankruptcy auction, a sale an Arizona judge approved on March 27, 2015. That is a steep markdown from 2001, when Gemstar-TV Guide International paid an aggregate $50.1 million for SkyMall's outstanding stock. **Sources:** - History of SkyMall, Inc. (International Directory of Company Histories): https://www.fundinguniverse.com/company-histories/skymall-inc-history/ - Gemstar-TV Guide International, Form 10-K/A, FY2001 (SEC EDGAR): https://www.sec.gov/Archives/edgar/data/0000923282/000095016803001148/d10ka.htm - SkyMall Magazine Files For Bankruptcy, BuzzFeed News, January 2015: https://www.buzzfeednews.com/article/mollyhensleyclancy/skymall-magazine-files-for-bankruptcy - SkyMall Clinches $1.9M Asset Sale To C&A, Law360, March 2015: https://www.law360.com/retail/articles/636620/skymall-clinches-1-9m-asset-sale-to-c-a - SkyMall is back, Yetis and all, CNN Money, December 2015: https://money.cnn.com/2015/12/07/media/skymall-returns-from-bankruptcy/index.html --- # Why Montgomery Ward Failed and Who Owns It Now URL: https://404memoryfound.com/posts/montgomery-ward-failed-who-owns-it.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-18 Topics: Business Blunders, Money & Tech **Summary:** Montgomery Ward failed in stages: it shut its own mail-order catalog in 1985 after 113 years, filed for Chapter 11 in July 1997, and announced on December 28, 2000 that all 250 stores would close. The chain was liquidated during 2001, ending 128 years in business. The Montgomery Ward name survives as an online and catalog retailer owned since 2008 by Colony Brands Inc. **Key facts:** - Launched: August 1872, Chicago, by Aaron Montgomery Ward with $1,600 in capital and a one-sheet catalog - Catalog closed: 1985, ending 113 years of mail order - What ended the stores: December 28, 2000 announcement: all 250 stores closed and about 37,000 jobs cut during 2001 - Owner today: Colony Brands Inc., a family-owned direct-mail company, since 2008 - Status today: No stores anywhere. Montgomery Ward sells online at wards.com on installment credit ## How one sheet of paper in 1872 became American shopping Montgomery Ward opened in Chicago in August 1872. Britannica records that Aaron Montgomery Ward started with capital of $1,600 and issued his first catalog as a single sheet listing about 150 items. The company's own history puts the count at 163 items, and adds the detail that makes the business make sense: from the start, Montgomery Ward sold on credit, buy now and pay later. The sheet turned into a book, and the book turned into a habit. By 1904 Montgomery Ward was mailing catalogs to three million customers, according to that company history. Farm families ordered stoves, sewing machines, coats and wedding rings from a merchant none of them had ever seen. That is the scale worth holding on to before the rest of the story. For most of the 19th century a rural American shopped at whatever general store was within a wagon ride. Montgomery Ward replaced that store with the mail, decades before a telephone was common and more than a century before a shopping cart appeared on a screen. ## The day two soldiers carried the chairman out of his office By the Second World War the company was run by Sewell Avery, and Avery spent more energy fighting Washington than fighting his competitors. In April 1944 Attorney General Francis Biddle traveled to Chicago to press Avery to extend a union contract, so that Montgomery Ward's warehouses would keep supplying the home front. Avery would not do it. The University of Wyoming's American Heritage Center, in its account of the seizure, records what Avery said to the Attorney General: "To hell with the government!" Biddle stopped arguing. Two soldiers picked Avery up by an arm each and carried him out of his office, still sitting upright, past the photographers waiting outside. The federal government took control of Montgomery Ward facilities, and the picture of a company chairman being removed from his own building ran on front pages nationwide. It is a strange thing for a catalog company to be famous for. It also set the tone for the next sixty years: Montgomery Ward was rarely the retailer that moved first, and it was often the one arguing with whoever wanted it to. ## Montgomery Ward shut down its own catalog in 1985 Ownership moved out of retail entirely in the 1970s. Britannica records that Mobil Oil Corporation bought 54 percent of the voting shares of Marcor, Ward's parent company, in 1974, and that Marcor merged into the new Mobil Corporation two years later. An oil major now owned the business that invented mail order. The catalog did not survive the arrangement. In 1985, Britannica records, Montgomery Ward ended its 113-year-old mail-order operation to concentrate on its retail stores. Read that twice. The company that taught the country to buy sight unseen from a printed list closed that business to bet everything on mall anchors, roughly a decade before anyone typed a credit card number into a browser. The stores then walked into a fight they could not win. Britannica notes that in the 1990s Montgomery Ward faced stiff competition from discount retailers such as Target and Wal-Mart, and filed for bankruptcy in 1997. Ward was not the only casualty of that decade. Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html) and Kmart (https://404memoryfound.com/posts/is-kmart-still-open-stores-left.html) were losing the same customers to the same discounters. Both outlasted Montgomery Ward, and both ended up shrunken to a fraction of what they were. ## What the July 1997 bankruptcy filing actually said Montgomery Ward filed for Chapter 11 protection on Monday, July 7, 1997. The Spokesman-Review reported that day that Ward was the largest privately owned retailer in the country, 125 years old, operating 400 stores in 43 states. The figure behind the filing was a loss of $237 million in 1996. A chain that size does not lose that much on merchandising alone; it loses it on stores in the wrong places selling the wrong things at prices a discounter could beat. GE Capital Services, which held the majority stake, backed the filing and promised a $1 billion line of credit so that the shelves would stay stocked while Ward reorganized. That money bought time, not a turnaround. Montgomery Ward spent the rest of the decade inside a restructuring, still owned by a finance company rather than a retailer, and came out of it with the same problem it went in with: hundreds of large stores that shoppers no longer had a reason to drive to. ## December 28, 2000: the day 128 years ended Three and a half years after the first filing, Montgomery Ward gave up. WWD reported that the company was shutting all 250 stores, cutting about 37,000 jobs, and ending a business that had traded for 128 years. Chairman and chief executive Roger Goddu blamed the Christmas that had just finished. "Overall weak holiday sales and a very difficult retail environment simply did not permit us to complete the turnaround that might have been possible in an otherwise thriving economy," he said in the statement announcing the filing. He was blunter in the next line: "Sadly, today's action is unavoidable." The decision was not entirely his. GE Capital, the owner, was withdrawing its financial support after the weak holiday season, and without that backing there was no version of Montgomery Ward that opened in January. Britannica ends the story in a sentence: the company announced in December 2000 that it was going out of business, and closed its remaining stores in 2001. The liquidation ran through the first half of that year, store by store, with the fixtures sold off after the merchandise. ## Is Montgomery Ward still in business in 2026? Yes, but only as a name on a website. There has been no Montgomery Ward department store anywhere in the United States since 2001, and the company that ran them no longer exists. The trademark was bought and restarted as a catalog and internet business. Montgomery Ward's own site states that in 2008 the brand was acquired by Colony Brands Inc., a family-owned direct-mail business, which has run it ever since. What that company sells at wards.com is furniture, appliances, electronics and home goods, on installment credit, under the phrase "Buy Now, Pay Later." The mechanism is the one Aaron Montgomery Ward used in 1872, pointed at a browser instead of a mailbox. So the honest answer to who owns Montgomery Ward now is Colony Brands, a Wisconsin direct marketer, and the honest answer to whether the old company survived is no. A dead retail name outliving its stores is common enough: Woolworth (https://404memoryfound.com/posts/is-woolworth-still-in-business.html) did the same thing in a different way. Where to find one today: the paper is the collectible. Christmas Wish Books, fall and winter catalogs, store signage, appliance manuals and Wards-branded tools turn up in the vintage listings on Etsy (https://www.etsy.com/search?q=vintage+montgomery+ward+catalog). Year, page count and condition drive the asking price far more than anything else, so check whether a catalog is complete before paying for it. ## Frequently Asked Questions ### Is Montgomery Ward still in business? Montgomery Ward exists today only as an online and catalog retailer at wards.com, with no physical stores. The original Montgomery Ward chain announced in December 2000 that it was going out of business and closed its remaining stores in 2001. The brand was later acquired by Colony Brands Inc. in 2008 and has been run as a direct-marketing business since. ### When did Montgomery Ward close its stores? Montgomery Ward announced on December 28, 2000 that it was closing all 250 of its stores, and WWD reported the shutdown cost about 37,000 jobs after 128 years in business. The stores were liquidated during 2001. Montgomery Ward had already closed its mail-order catalog in 1985, after 113 years. ### Who owns Montgomery Ward now? Colony Brands Inc., a family-owned direct-mail company, has owned the Montgomery Ward brand since 2008, according to Montgomery Ward's own company history. Before that, the original retailer was majority-owned by GE Capital Services, which backed its 1997 Chapter 11 filing and withdrew support in December 2000. No department store company owns the name today. **Sources:** - Britannica Money: Montgomery Ward & Co., American Retailer and Mail-Order Pioneer: https://www.britannica.com/money/Montgomery-Ward-and-Co - American Heritage Center, University of Wyoming: Defiance and Consequences, The Montgomery Ward Seizure: https://ahcwyo.org/2018/12/27/the-montgomery-ward-seizure/ - The Spokesman-Review: Ward Files For Bankruptcy (July 8, 1997): https://www.spokesman.com/stories/1997/jul/08/ward-files-for-bankruptcy/ - WWD: Long-Struggling Wards Calls It Quits, Shutting All 250 Stores (December 2000): https://wwd.com/fashion-news/fashion-features/feature/long-struggling-wards-calls-it-quits-shutting-all-250-stores-1188758-1718094/ - Montgomery Ward: About Us and company history: https://www.wards.com/cm/history-about-us.html --- # Is Homestar Runner Still Updating? The Site in 2026 URL: https://404memoryfound.com/posts/homestar-runner-still-updating.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-18 Topics: Internet Culture, Software & Apps **Summary:** Homestar Runner is still online and still updating in 2026, more than 26 years after Mike Chapman and Matt Chapman launched homestarrunner.com on January 1, 2000. New cartoons arrive irregularly rather than weekly, and the whole Flash archive has run on the Ruffle emulator since December 31, 2020. The brothers still own the site, and it has never carried advertising. **Key facts:** - Launched: January 1, 2000, by Mike Chapman and Matt Chapman - Started as: A 1996 photocopied children's book by Mike Chapman and Craig Zobel - Strong Bad Emails: 200 by September 23, 2008, then years apart - Owner today: The Brothers Chaps, independent and ad-free - Status today: Still updating in 2026, on no fixed schedule ## Who made Homestar Runner, and where did it start? Homestar Runner started on paper. In 1996 Mike Chapman and Craig Zobel wrote and drew a short children's book called The Homestar Runner Enters the Strongest Man in the World Contest, a parody of the cheaply made kids' books they had seen at a mall near Atlanta. They photocopied about 15 copies and handed them to friends. The website came four years later. Mike Chapman and his brother Matt Chapman taught themselves Macromedia Flash, rebuilt the characters as animation, and put homestarrunner.com online on January 1, 2000. The two worked under the name The Brothers Chaps. They wrote, drew, animated and voiced nearly everything themselves, which is why every character sounds like one of two guys in a room in Georgia. Flash made that possible. A cartoon that would have needed a broadcast budget in 1995 could be drawn by two people and downloaded over a 56k modem in 2000, which is the same reason Macromedia Flash (https://404memoryfound.com/posts/definitive-history-macromedia-flash.html) carried almost every web cartoon of that decade. ## How did Strong Bad take the site over? Homestar Runner was the title character. Strong Bad, a masked heavy in boxing gloves, was supposed to be the villain, and he ended up running the place. The reason was a format. In Strong Bad Email, Strong Bad reads a message from a real fan on an ancient computer and answers it badly. A new one went up most Mondays from 2002 through 2008, and the email was the thing people checked for. Email number 58, called dragon, went online on January 13, 2003. In it Strong Bad draws a dragon out of "consummate V's" and names it Trogdor the Burninator. Trogdor became the most reproduced thing the site ever made, printed on shirts, worked into a heavy metal theme song, and eventually turned into a board game. The 200th Strong Bad Email, email thunder, arrived on September 23, 2008. The Brothers Chaps treated 200 as a round number to stop at, and the weekly rhythm never came back. New emails now show up years apart, usually attached to a holiday or a joke about the gap itself. ## How did two brothers make a living with no ads? Homestar Runner never carried advertising. No banners, no pre-roll, no sponsor cards, in an era when that was the only model anyone in web video used. The money came from merchandise. The first shirt run was tiny and paid for by checks mailed to the Chapman family house. Sales grew into shirts, hoodies, plush toys, DVD compilations and a CD, and by late 2002 both brothers had quit their other work to do Homestar Runner full time. The traffic justified it. Matt Chapman told Wired in 2003 that the site drew "a few million" unique visitors a month, at a point when the entire American internet population was roughly 66 million people. The brothers never thought it would last. "We're on borrowed time here, there's just no way to make a living off of this, because it's unsustainable," Mike Chapman told Gizmodo in its 2017 oral history of the site. There was no Kickstarter yet, no Patreon, and they did not want ads. In 2008 Telltale Games licensed the characters for Strong Bad's Cool Game for Attractive People, five episodes released for Windows and WiiWare, which was the closest the site came to an outside paycheck. ## Why did the site go quiet after 2010? The updates thinned out through 2010 and then stopped. The last real update of that era was a Halloween costume toon posted on December 22, 2010, and nothing new appeared for 1,197 days. Both brothers had families and television work by then. Matt Chapman moved into writing and voice work on animated shows, and the site had no deadline structure left once the Monday email habit broke. The return came on April 1, 2014, with a new April Fools' Day cartoon, and the internet reaction was large enough that the brothers kept going. They were careful about what they were promising. "We're not expecting this to be our full-time gig, we just want this to be fun," Matt Chapman told Rolling Stone in 2014, adding that the original run had been a great job and also a stressful one. A year later they took a paid gig with the same sensibility. Two More Eggs ran on Disney XD from June 2015 to a final season in 2017, which kept The Brothers Chaps animating while homestarrunner.com updated on its own clock. ## How often do new cartoons appear now? Homestar Runner is still updating in 2026, without a schedule. New toons arrive in clusters rather than weekly, often months apart, and Halloween is still the most reliable date on the calendar: the annual costume cartoon has been a fixture since the site's first year. The 25th anniversary passed on January 1, 2025, and the brothers marked it with a song called Back to a Website, which is about missing the version of the internet the site was built for. The site itself has not changed shape. It is free, it has no account, no feed, no recommendation engine and still no advertising. You open homestarrunner.com and click a cartoon, which in 2026 makes it an unusual thing to find rather than a normal one. That puts it in a very short club. Most of its neighbors from the Flash cartoon era either shut down or became something else entirely, and Newgrounds (https://404memoryfound.com/posts/what-happened-to-newgrounds.html) is one of the few other survivors still run by the people who started it. ## What happened when Flash died? Adobe ended Flash Player on December 31, 2020, and every browser dropped support. For a site whose entire archive, hundreds of cartoons and games, was built in Flash, that was an extinction event. Homestar Runner beat the deadline by one day. On December 31, 2020, the site relaunched using Ruffle, an open-source Flash emulator written in Rust that runs the original files inside the browser with no plugin to install. The rebuilt site pairs HTML5 and JavaScript around embedded Ruffle players, and it kept the old look on purpose. The switch fixed something else. The archive became usable on phones for the first time since 2012, because Flash had never worked on iOS at all. Games made after about 2007 can still be rough under emulation, and the original Flash version is kept at old.homestarrunner.com for reference. Plenty of that era's writing simply vanished instead. Text blogs had it easier than Flash did, which is why LiveJournal (https://404memoryfound.com/posts/what-happened-to-livejournal-blogging-platform.html) archives outlasted most of the animation of the same years. ## Where to find one today Everything on homestarrunner.com is free to watch, so there is nothing to buy in order to see it. New official merchandise, including Trogdor shirts and plush, is sold through the site's own store and through Fangamer. The old physical items are a different matter. The Everything Else DVD volumes and the 2003 CD Strong Bad Sings (and Other Type Hits) are long out of print, and copies circulate second-hand along with the 2018 board game. Prices swing with condition, so compare a few listings before buying on a marketplace such as Etsy (https://www.etsy.com/search?q=homestar+runner). ## Frequently Asked Questions ### Is Homestar Runner still updating in 2026? Yes. Homestar Runner still posts new cartoons at homestarrunner.com in 2026, but on no fixed schedule, often with months between updates. The site returned from a three-year silence on April 1, 2014, and has been running irregularly ever since, with the Halloween cartoon as its most dependable annual entry. ### Do the old Homestar Runner cartoons still work without Flash? They do. Adobe killed Flash Player on December 31, 2020, and on that same day homestarrunner.com relaunched using the Ruffle emulator, which plays the original Flash files in a normal browser with nothing to install. The archive also works on phones now, which it never did while the site depended on Flash. ### Who owns Homestar Runner? Mike Chapman and Matt Chapman, the pair known as The Brothers Chaps, created Homestar Runner in 2000 and still own and run it in 2026. The site was never sold to a network or a studio, has never carried advertising, and was funded by merchandise sales from the early 2000s onward. **Sources:** - An Oral History of Homestar Runner, the Internet's Favorite Cartoon (Gizmodo, 2017): https://gizmodo.com/an-oral-history-of-homestar-runner-the-internets-favor-1791519879 - Homestar Runner Returns! The Brothers Chaps on Bringing the Site Back (Rolling Stone, 2014): https://www.rollingstone.com/culture/culture-news/homestar-runner-returns-inside-a-cult-classics-comeback-195674/ - Post-Flash Site Update (Homestar Runner Wiki): http://hrwiki.org/wiki/Post-Flash_Site_Update - Homestar Runner (Wikipedia): https://en.wikipedia.org/wiki/Homestar_Runner - Trogdor!! The Board Game (Kickstarter, 2018): https://www.kickstarter.com/projects/1999933720/trogdor-the-board-game --- # Who Owns Packard Bell Now? Acer Revived the Brand URL: https://404memoryfound.com/posts/packard-bell-computer-brand-today.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-18 Topics: Hardware, Business Blunders **Summary:** Packard Bell is owned by Acer, which took over the brand in 2008 and then left it dormant for about 13 years. In September 2026 Acer relaunched Packard Bell at IFA in Berlin as a budget lifestyle brand, led by a 14.5-inch DotBook laptop starting at 499 euros. The range is sold across Europe, the Middle East and Africa, and Acer has announced nothing for the United States. **Key facts:** - Launched: 1986, Packard Bell Electronics, Inc., on a brand name dating to a 1933 Los Angeles radio maker - Peak: Best-selling PC brand in the United States in 1995, above 15 percent of US units shipped - What ended it: NEC closed the US business on November 3, 1999 and the name left US shelves by 2000 - Owner today: Acer Inc., which took over Packard Bell in 2008 after buying Gateway - Status today: Relaunched September 2026 in Europe, the Middle East and Africa; the DotBook laptop starts at 499 euros ## How a Los Angeles radio maker ended up in every US electronics store Packard Bell began in 1933, when Herb Bell and Leon Packard started building radios in Los Angeles. Teledyne bought the company in 1968 and eventually let the name go quiet. In 1985 a group led by Beny Alagem, with Jason Barzilay and Alex Sandel, bought the dormant Packard Bell trademark from Teledyne for less than $100,000. They founded Packard Bell Electronics, Inc. in 1986 and aimed it at a customer nobody else wanted: the person buying a first computer from a store that had never sold one. That meant Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html), Walmart, Best Buy and Circuit City (https://404memoryfound.com/posts/what-happened-to-circuit-city-electronics.html) rather than computer dealers. The machines shipped preloaded and cabled, ready to plug in, which mattered to buyers who had no idea what a sound card was. It worked quickly. Fortune ran a feature on June 12, 1995 built around the fact that Packard Bell sold more PCs in the United States than anyone, and the brand's share of US units shipped passed 15 percent that year. Larry Mondry, then an executive at the retailer CompUSA (https://404memoryfound.com/posts/is-compusa-still-around.html), told the magazine: "In a business in which some companies can't put out a new line in six months, they did it in two weeks." ## Why Packard Bell computers earned such a bad reputation The complaints started early. In 1992 the company disclosed that 17 percent of the computers it sold came back, against an industry norm of 5 to 8 percent, and it shelved a plan to go public that year. Reviewers noticed. PC World, PC Magazine and Consumer Reports rated the machines poorly through the decade, even as they outsold everything else on the shelf. The bundled software did not help. Packard Bell shipped its own front end, Packard Bell Navigator, aimed at the same nervous first-time buyer that Microsoft Bob (https://404memoryfound.com/posts/what-happened-to-microsoft-bob-friendly-interface.html) was chasing, and it drew the same complaint about standing between people and Windows. Then came the parts. Attorneys general in 22 states found that Packard Bell stripped returned machines, tested the components and built them into computers sold as new. The case settled in 1998. The company paid $3.5 million to the federal government, $70,000 in fees and investigative costs to each of the 22 states, and more than $1 million in refunds and repair bills, without admitting wrongdoing. It also agreed that boxes for products sold after December 31, 1996 would disclose that the contents "may include components from previously sold computers." ## What NEC actually bought when it merged its PC business in NEC had been buying into Packard Bell for years. In April 1996 it added $283 million, lifting a 19 percent holding to roughly 40 percent once preference shares were counted. Earlier that year Packard Bell had bought Zenith Data Systems from Groupe Bull of France, which handed it a business customer base and a European footprint it had never had. In June 1996 the two sides announced they would combine their PC operations, with about $300 million of NEC assets transferred across. The merger closed that July and the business became Packard Bell NEC. The combination briefly returned the company to the top of US unit shipments. It did not hold. Compaq (https://404memoryfound.com/posts/compaq-lost-pc-business-hp-merger.html) took the number one spot in early 1997, and the squeeze underneath was brutal: Compaq from above, the discount maker eMachines from below, and a brand that could not charge a premium because nobody believed it had earned one. Packard Bell NEC lost more than $1 billion in 1997. ## When did Packard Bell stop selling computers in the United States? On November 3, 1999, NEC called it. It would close the Sacramento assembly plant, move the headquarters to Mountain View, and cut 80 percent of the 2,600 people still working there. Forbes, writing the same day, put the bill at more than $2 billion poured in by NEC "only to have it continue to bleed red ink like a stuffed pig." By 2000 the Packard Bell name was off the US market. The retail shelf space it had effectively invented for home computers did not vanish with it. That space went to Compaq, Hewlett-Packard and Gateway (https://404memoryfound.com/posts/what-happened-to-gateway-2000-cow-box-computer.html), and later to cheaper laptops from Acer and Dell. No single decision killed it. A model built on volume at thin margins, a return rate that ate the margin, and a name that had become a joke in the same stores that made it are enough on their own. ## How Acer ended up with the name NEC kept the European business running for another six years, then sold that too. In 2006 it agreed to hand Packard Bell BV, run out of the Netherlands, to Lap Shun "John" Hui, co-founder of eMachines. Neither side disclosed a price. Acer had bought Gateway in 2007, and Gateway held a right of first refusal over any sale of Packard Bell. When Hui opened talks with Lenovo, Acer used it. The takeover completed in 2008 and put Packard Bell, Gateway and eMachines under one Taiwanese owner. Acer kept Packard Bell selling in Europe for a few more years, retired eMachines in 2013, and then let Packard Bell go quiet. The website stayed up. The products stopped. For roughly 13 years the brand existed as a trademark and little else, which is the usual fate of a name that outlives the company that earned it. ## Can you buy a new Packard Bell in 2026? Yes, if you live in Europe, the Middle East or Africa. At IFA in Berlin in September 2026, Acer relaunched Packard Bell as a budget lifestyle brand under the line "Tech it easy," timed to what it calls the brand's 100th anniversary. That anniversary takes some liberties. Acer dates the founding to Los Angeles in 1926, while the Packard Bell partnership between Herb Bell and Leon Packard was formed in 1933. Bell had worked in radio since 1926, which appears to be where the round number comes from. The relaunch covers nine products. The lead item is the DotBook, a 14.5-inch laptop starting at 499 euros and reaching European retail in the fourth quarter of 2026. Around it sit the DotLoop turntable, the DotTune speaker, the DotLine 4G feature phone, the DotLook audio glasses, the DotShine ring light, the DotCombo keyboard and mouse set and the DotBass headphones. Valerie Piau, EMEA vice president at Acer Europe, framed the pitch this way in 2026: "People need technology that helps them study, work, and stay in touch, all at a fair price." Acer has announced nothing for the United States. Where to find one today: the beige machines are a separate market. Complete 1990s towers, keyboards, mice, restore discs and the occasional boxed system turn up through Etsy's vintage Packard Bell listings (https://www.etsy.com/search?q=vintage+packard+bell+computer), mostly out of house clearances. Condition and completeness drive the asking price far more than the model does, so compare several listings and assume the hard drive is dead. ## Frequently Asked Questions ### Is Packard Bell still in business in 2026? Packard Bell exists in 2026 as a brand owned by Acer, not as an independent company. Acer relaunched it at IFA in September 2026 with nine budget products led by the DotBook laptop, sold across Europe, the Middle East and Africa. The company founded as Packard Bell Electronics, Inc. in 1986 stopped selling computers in the United States in 1999. ### Why did Packard Bell fail in the United States? Packard Bell sold on price through general retail, and that model stopped working once Compaq attacked from above and eMachines from below in the mid 1990s. Quality was the other half of it: the company disclosed a 17 percent return rate as early as 1992, and in 1998 it settled with 22 state attorneys general over building parts from returned machines into computers sold as new. NEC closed the US business on November 3, 1999. ### Can you buy a Packard Bell laptop in the United States? No. The Packard Bell range Acer announced in September 2026, including the 499 euro DotBook, is sold only in Europe, the Middle East and Africa, and Acer has not announced a United States release. Anyone in the US who wants the name has to buy a 1990s machine second hand, because no new Packard Bell computer has been sold there since 2000. **Sources:** - Fortune: Packard Bell sells more PCs in the U.S. than anyone. So just who are these guys? (June 12, 1995): https://money.cnn.com/magazines/fortune/fortune_archive/1995/06/12/203837/index.htm - Wikipedia: Packard Bell: https://en.wikipedia.org/wiki/Packard_Bell - San Francisco Chronicle: Packard Bell Settles Suit Over Used Parts (1998): https://www.sfgate.com/business/article/packard-bell-settles-suit-over-used-parts-2967237.php - Forbes: NEC finally pulls plug on Packard Bell (November 3, 1999): https://www.forbes.com/1999/11/03/mu3.html - Acer newsroom: Acer Relaunches Packard Bell with Full New Lifestyle Collection to 'Tech It Easy' (September 2026): https://news.acer.com/acer-relaunches-packard-bell-with-full-new-lifestyle-collection-to-tech-it-easy --- # How Many Payphones Are Left? The FCC Stopped Counting URL: https://404memoryfound.com/posts/payphones-left-united-states.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-17 Topics: Then vs Now, Hardware **Summary:** Nobody can give an exact number of payphones left in the United States, because the Federal Communications Commission stopped requiring payphone audits in March 2018. Its last full count put the total under 100,000 at the end of 2016, down from more than 2.1 million at the 1999 peak. Thousands still work in 2026, run mostly by private operators on rural highways, in parks, prisons, hospitals and transit stations. **Key facts:** - Peak: More than 2.1 million payphones in 1999 - Last official count: Under 100,000 at the end of 2016 (FCC) - Status today: Still operating, but uncounted since the FCC dropped payphone audits in 2018 - Largest operator: Pacific Telemanagement Services, which told CNN in 2018 it ran 20,000 phones - Cost of a call: 50 cents on Verizon payphones from September 2001, up from 35 cents ## What the last official payphone count showed The Federal Communications Commission counted payphones for years, and the numbers it published are still the only reliable trend line anyone has. At the 1999 peak there were more than 2.1 million payphones in service in the United States. By the end of 2016 the count was under 100,000, a fall of about 95 percent from 2000. The collapse was close to uniform. Most states lost more than 90 percent of their payphones between 2000 and 2016. Hawaii was the outlier, down only about 60 percent, because remote roads and visitors kept the phones earning. New York was the other exception, in the opposite direction. In 2016 the state still held roughly one in five of every payphone left in the country, according to FCC figures reported by CNN in 2018. Those two facts explain most of the map. Payphones survived where cell coverage was bad, where travelers arrived without a working plan, and where a sidewalk carried enough traffic to justify the maintenance visit. ## Why nobody has an official number anymore Congress opened the payphone business to competition in the Telecommunications Act of 1996. The law stopped the Bell companies from favoring their own phones and set up a system to pay independent operators for calls that never touched a coin slot, such as toll-free and calling-card calls. That compensation system needed policing, so the FCC made carriers run annual audits of the call tracking systems that counted those calls. By 2018 the audits cost more than the calls were worth. On March 15, 2018 the agency adopted an order scrapping them, writing that "Audit requirements are no longer needed as safeguards to ensure that [pay phone service providers] receive the compensation they are due." The rule changes took effect on April 16, 2018. The side effect is that the federal government stopped generating payphone data. Anyone who states an exact number of payphones for 2026 is estimating from the 2016 baseline, from what operators disclose, or from walking around with a camera. ## Are there still working payphones in 2026? Yes. They are just no longer where most Americans walk. The largest operator is Pacific Telemanagement Services of San Ramon, California, which absorbed AT&T's payphone estate in 2008 and most of Verizon's in 2011. Its president, Tom Keane, told CNN in 2018 that the company ran 20,000 phones and that payphones can still turn a profit where there is no cell or landline coverage. Keane named the two conditions that keep a coin phone alive. The first is dead signal: "We have phones in Yosemite Valley that are extremely busy when there's not snow on the ground," he said. The second is everything else failing at once: "Every time there's a disaster our phone use goes through the roof." That is where the survivors sit. Rural highways, national parks, jails and prisons where inmates cannot hold a handset of their own, hospitals, bus and rail stations, and university campuses. A payphone stopped being street furniture somewhere around 2010 and became emergency infrastructure instead. ## How New York lost 6,000 street phones and kept four booths New York City had more than 6,000 active public payphones on its sidewalks in 2014, when the franchise that became LinkNYC began. Removal started in 2015 and ran for seven years. On May 23, 2022 a crane lifted the last free-standing public payphone off Seventh Avenue at West 50th Street, a block south of Times Square. It went to the Museum of the City of New York for an exhibit about the city before computers. Matthew Fraser, then commissioner of the city's Office of Technology and Innovation, marked the moment: "As a native New Yorker, saying goodbye to the last street payphone is bittersweet because of the prominent place they've held in the city's physical landscape for decades." The title was disputed the same week. Four full-length glass booths still stand on West End Avenue on the Upper West Side, kept there in large part by a neighborhood resident named Alan Flacks, who spent years lobbying to save them. Payphones inside private buildings were never part of the city count either. ## What a payphone call cost before the coins stopped Coin rates were deregulated in 1997, and the price climbed as the customer base shrank. On September 8, 2001 CNN reported that Verizon, then the largest payphone operator in the country, would raise the price of a local call on its 430,000 payphones across 33 states from 35 cents to 50 cents. The company pointed at wireless competition: fewer calls meant each remaining call had to carry more of the cost. What emptied the coin boxes is no mystery. Cheap handsets did it, first with phones like the Nokia 3310 (https://404memoryfound.com/posts/what-happened-to-nokia-3310.html) from 2000 and then with the Motorola Razr V3 (https://404memoryfound.com/posts/motorola-razr-v3-coolest-phone-ever-made.html) from 2004, which put a dial tone in the pocket of people who had never owned a landline. For what the rest of that decade's hardware cost in current money, see our table of 90s tech prices in 2026 dollars (https://404memoryfound.com/posts/90s-tech-inflation-calculator.html). ### Where to find one today Payphones have a second life as decor and as restoration projects. Single-slot Western Electric units, booth signs, coin doors, receivers and instruction cards all turn up in the vintage telephone listings on Etsy (https://www.etsy.com/search?q=vintage+payphone). Condition drives the price, and the gap between a gutted shell and a unit with a working coin mechanism and its keys is wide. Read the listing carefully: reproductions built to look like 1950s phones sit in the same search results as genuine Bell System hardware, and a phone with no key is a phone you cannot open. ## Frequently Asked Questions ### How many payphones are left in the United States? There is no current official count of payphones in the United States. The Federal Communications Commission's last full figure put the number under 100,000 at the end of 2016, down from more than 2.1 million at the 1999 peak, and the agency eliminated its payphone audit and reporting rules in March 2018, which ended the data series. ### Are there still working payphones in 2026? Yes, working payphones still exist in the United States in 2026, mostly on rural highways, in national parks, inside jails and prisons, and in hospitals and transit stations. Pacific Telemanagement Services, the largest operator, told CNN in 2018 that it ran 20,000 of them, and its president said payphones remain profitable where cell coverage is poor. ### When did payphones start disappearing? Payphones peaked in 1999 at more than 2.1 million and fell by about 95 percent between 2000 and 2016 as cell phones spread. New York City, which held roughly one in five of the country's remaining payphones in 2016, started pulling its street phones in 2015 and removed the last public one on May 23, 2022. **Sources:** - CNN Money: There are still 100,000 pay phones in America (2018): https://money.cnn.com/2018/03/19/news/companies/pay-phones/index.html - Federal Register: Modernization of Payphone Compensation Rules (FCC, March 15, 2018): https://www.federalregister.gov/documents/2018/03/15/2018-05201/modernization-of-payphone-compensation-rules - Stacker and Spokeo: The decline of pay phones in every state: https://stacker.com/stories/retrospective/decline-pay-phones-every-state - NPR: New York City removes its last public pay phone, a disputed title (2022): https://www.npr.org/2022/05/24/1100931534/last-pay-phone-new-york-city-public-nyc - CNN: Verizon raises pay phone rates to 50 cents (2001): https://www.cnn.com/2001/US/09/08/pay.phone.calls/ --- # Are LaserDiscs Worth Anything? The 2026 Collector Market URL: https://404memoryfound.com/posts/laserdisc-value-collector-market.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-17 Topics: Hardware, Then vs Now, Money & Tech **Summary:** Most LaserDiscs are worth very little: common titles sell for a dollar or less, and the great majority of 1990s discs are not scarce. The money is in late pressings, Japanese imports and Criterion editions, where prices run from $20 to $50 up to a reported $5,700 for Tokyo Raiders. The format has been dead since 2001 and Pioneer stopped building players in 2009, which is the real limit on what any disc is worth. **Key facts:** - Launched: 1978 in the United States, as MCA DiscoVision, with Jaws as the first title - First player price: $749 for the Magnavox VH-8000 in 1978 - Peak US reach: About 2 percent of American households in 1998, roughly 2 million players - Last releases: Bringing Out the Dead, October 3, 2000 in the US; Tokyo Raiders, September 2001 in Japan - Status today: Dead format. Common discs under $5, scarce pressings $160 to $5,700 ## The honest answer: most discs are worth under $5 A box of LaserDiscs found in a garage is almost always a box of common titles, and common titles are close to worthless. Widescreen Review's 2025 survey of the market put many LaserDiscs at a dollar or less. The value sits in a thin top layer. The same survey placed most genuinely collectible discs in the $20 to $50 range, which means ten good titles might come to $200 or $500 rather than a windfall. That top layer is real, though. A small number of pressings have sold for hundreds of dollars, and one has reached a reported $5,700. So the answer depends entirely on which discs are in the box. LaserDisc never had the mass-market second life that vinyl got, and it is not going to get one. What it has instead is a small group of buyers chasing a small group of pressings, closer to the market around Sony's MiniDisc (https://404memoryfound.com/posts/what-happened-to-minidisc-sony.html) than to anything on a store shelf. ## Which titles carry real money, and why Three things push a LaserDisc into three-figure territory: a late release date, a Japanese pressing, or extras that never moved to any later format. Late releases are the easiest rule to apply. By 1999 the format was dying in the United States, so studios pressed small runs of big movies. Widescreen Review lists The Matrix at up to $160 and Star Wars: The Phantom Menace at up to $35, and puts Pulp Fiction near $999 at the top of its recorded sales. The most expensive disc is the one that ended the format. Paramount released Martin Scorsese's Bringing Out the Dead on October 3, 2000, the last LaserDisc issued by a major studio in the United States. Japan kept pressing for another year, and in September 2001 the Hong Kong action film Tokyo Raiders became the last LaserDisc released anywhere. Copies have sold for as much as $5,700. None of those are prices anyone should count on. They are ceilings set by a handful of collectors, and condition decides the rest. ## Why so few Americans ever owned one LaserDisc reached American stores in 1978 under the MCA DiscoVision name, with Jaws as the first movie released on the format in the United States. The player was the obstacle. The Magnavox VH-8000, the first consumer LaserDisc player sold in the United States, cost $749 in 1978. A VHS deck was expensive too, but VHS could record and LaserDisc could not, the same practical advantage that had already decided the Betamax and VHS format war (https://404memoryfound.com/posts/betamax-vs-vhs-format-war.html). The result was a format for enthusiasts. By 1998, LaserDisc players were in about 2 percent of American households, roughly two million homes. In Japan, where players and discs were priced much closer to tape, they reached 10 percent of households in 1999. That gap explains a lot about the market now. Japanese pressings are more numerous, they are often better mastered, and they carry titles that never got a release in the United States at all, which is why so many of the expensive discs are imports. ## Does laser rot destroy the value? Laser rot is the biggest risk in buying LaserDiscs unseen, and despite the name it has nothing to do with the laser. It is oxidation: the aluminum reflective layer loses its shine, the picture fills with speckles, and eventually the disc stops tracking. The cause was manufacturing, not use. Impurities in the adhesive that bonded the two halves of a disc worked through the lacquer seal and attacked the metal underneath. A disc that was never played can rot just as readily as one played hundreds of times. That gives collectors a pattern to work with. Double-sided discs rot and single-sided discs largely do not. Early MCA DiscoVision pressings from the late 1970s were badly affected, and Sony's DADC plant in Terre Haute, Indiana kept turning out rot-prone discs into the 1990s, after most other plants had fixed the problem. For a buyer that means the pressing plant matters as much as the title on the sleeve. A rotted disc is worth nothing no matter what a clean copy sells for, which is why serious buyers ask for photographs of the playing surface before money changes hands. ## The Criterion premium and the first commentary track The Criterion Collection started on LaserDisc in 1984, and its discs still carry the strongest premium of any American label. Criterion's King Kong release did something no home video edition had done before. For its 1984 LaserDisc of the 1933 film, the company recorded film historian Ronald Haver talking over the movie on a separate analog audio channel. "I'm going to take you on a lecture tour of King Kong as you watch the film," Haver says at the start of that 1984 commentary, which invented the bonus feature more than a decade before DVD existed. Criterion expected the track to interest film students and almost nobody else. It became a standard instead, and Haver kept recording commentaries for the label. That history is why Criterion LaserDiscs hold value even when the film itself is streaming. Some of those supplements were never carried over to later formats, so the disc is the only place the material exists. Owners who want the content rather than the object end up transferring it themselves, the same calculation people make with old VHS tapes (https://404memoryfound.com/posts/how-to-digitize-vhs-tapes-2026.html). ## Can you still play a LaserDisc in 2026? Yes, but only on hardware that has not been built since 2009. Pioneer announced on January 14, 2009 that it was ending LaserDisc player production after one final run of 3,000 units. The company's news release blamed supply rather than demand: "it has become difficult for Pioneer to procure the parts required to produce LD players." Pioneer had shipped more than 9.5 million players worldwide by then, out of roughly 16.8 million LaserDisc players sold by all manufacturers. That is the entire installed base, and it has been shrinking through failure ever since. Players are the real constraint on prices. Laser diodes dim, drive belts perish, and the loading mechanisms on older units are usually the first thing to fail. Every dead player removes a buyer from the disc market, which is part of why value concentrates in a few dozen titles instead of spreading across thousands of them. Where to find one today: common LaserDiscs turn up for a few dollars, most collectible titles sit in that $20 to $50 band, and sealed or late-run pressings go well above it. The LaserDisc listings on Etsy (https://www.etsy.com/market/laserdisc) are a reasonable place to start, and a working player is worth budgeting for separately. ## Frequently Asked Questions ### Are LaserDiscs worth anything in 2026? Most LaserDiscs are worth very little. Widescreen Review's 2025 market survey found that many sell for a dollar or less, while genuinely collectible LaserDisc titles sit in the $20 to $50 range and a small number of scarce pressings reach the hundreds. Condition and the pressing plant matter as much as the film itself. ### What is the most valuable LaserDisc? Tokyo Raiders, a Hong Kong action film issued on LaserDisc in Japan in September 2001, is the highest priced title on Widescreen Review's 2025 list, with copies selling for as much as $5,700. It was the last LaserDisc released anywhere, which is exactly why so few of them exist. ### Can you still buy a LaserDisc player? Not a new one. Pioneer, which built more than 9.5 million of the roughly 16.8 million LaserDisc players ever sold, stopped production on January 14, 2009 after a final run of 3,000 units. Every LaserDisc player on the market in 2026 is second-hand, and laser diodes and loading mechanisms are the usual failure points. **Sources:** - Pioneer Corporation, Pioneer Announces End of LaserDisc Player Products (news release, January 14, 2009): https://global.pioneer/en/corp/news/press/index/1380/ - Widescreen Review, Your LaserDiscs Could Be Worth $250 Or More, Here's Where To Sell Them: https://www.widescreenreview.com/news_detail.php?id=21576 - Digital Archivist, LaserDisc Laser Rot Explained: Why Only Some Discs Rot: https://digital-archivist.com/laserdisc-laser-rot/ - Internet Archive, King Kong (1933) Criterion Collection LaserDisc audio commentary with Ronald Haver: https://archive.org/details/king-kong-1933-criterion-collection-laserdisc-audio-commentary-with-historian-ronald-haver - Wikipedia, LaserDisc: https://en.wikipedia.org/wiki/LaserDisc --- # Is Tower Records Still Around? Online and in Japan URL: https://404memoryfound.com/posts/tower-records-still-around.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-17 Topics: Business Blunders, Music & Entertainment **Summary:** Tower Records is still around, but not as an American record store. The US chain was sold to a liquidation firm at a bankruptcy auction in October 2006 and all 89 remaining stores closed. The brand returned as an online store on November 13, 2020, and Tower Records Japan, a separate company, still runs stores in Japan. **Key facts:** - Founded: 1960 in Sacramento, California, by Russ Solomon - Peak: Nearly 200 stores in more than a dozen countries, over $1 billion a year in sales - Sold for: $134.3 million to Great American Group at the October 2006 liquidation auction - Status today: No US stores. Online only since November 13, 2020 - Owner today: US brand run by chief executive Danny Zeijdel. Tower Records Japan is a separate company ## How a Sacramento drugstore turned into a billion-dollar chain Russ Solomon made his first record sale at 16, selling used jukebox discs out of his father's drugstore in Sacramento, California. The shop was called Tower, after the theater next door, and the name stayed on everything he built afterward. Solomon opened the first Tower Records store in Sacramento in 1960. The idea was simple and, for the time, close to radical: put every genre under one roof, keep deep catalog on the shelves instead of only the hits, and stay open late enough that working musicians could shop after a gig. It scaled further than anyone expected. At its peak the company ran nearly 200 stores in more than a dozen countries and took in more than $1 billion a year, according to the Washington Post's 2018 account of Solomon's career. The stores in Tokyo, London and on the Sunset Strip in Los Angeles became destinations rather than errands. Record retail was crowded in those years. Sam Goody (https://404memoryfound.com/posts/sam-goody-stores-left.html) had the malls and Columbia House (https://404memoryfound.com/posts/is-columbia-house-still-around.html) had the mailbox, but neither of them had Tower Records' selection, and neither of them had its opening hours. ## What the stores felt like when the money was good The clearest measure of the chain's pull is who shopped there. Elton John was a regular at the Sunset Strip store, and in the 2015 documentary "All Things Must Pass" he put it plainly: "I can say, without exaggeration, that I spent more money at Tower than any other human being." The staff was part of the product. Dave Grohl worked at a Tower Records store before Nirvana, one of many musicians who passed through the payroll, and Billboard's 2015 report on the documentary treated the hiring as a subculture of its own: clerks who knew the catalog and argued about it out loud, in front of customers. Tower Records also published its own music magazine, Pulse!, and handed it out free in the stores. A retailer reviewing the records it was trying to sell should not have worked, and it did. Colin Hanks directed the documentary, and by the time it reached theaters in 2015 the subject was already history. Every American Tower Records had been closed for nine years. ## Why Tower Records collapsed in 2006 Two things went wrong at the same time. Compact discs, which carried the chain's margins, stopped being the only way to own music, and discount chains went after the discs people still bought. The Washington Post's obituary of Solomon named both causes, describing a company "felled by the rise of digital downloads and growing competition from discount chains." File sharing finished the job that the discounters started, and Napster (https://404memoryfound.com/posts/napster-destroyed-music.html) taught a generation of buyers that the album they wanted did not have to cost anything at all. The finances gave way first. The company filed for bankruptcy in 2004 and, in the Washington Post's phrasing, after "some financial maneuverings by outsiders" it was forced into bankruptcy a second time in 2006 and liquidated. By August 2006, when Tower Records filed for Chapter 11 protection for the second time, the American company was down to 89 stores in 20 states and owed creditors about $200 million. Those are the numbers that went into the courtroom that fall, and they are the reason nobody wanted to run the business. ## The 30-hour auction that ended the American stores The end came in a bankruptcy court rather than a boardroom. Tower Records was auctioned in October 2006, and the winning bidder wanted the inventory rather than the company. Great American Group, a liquidation firm, won with a bid of $134.3 million after almost 30 hours of bidding that attorneys called "robust" and "vigorous," according to the Associated Press report carried by the Deseret News. A federal bankruptcy judge approved the sale that Friday, and the going-out-of-business signs went up. The margin was $500,000. Trans World Entertainment, which ran record stores of its own and wanted to keep at least some Tower Records locations trading, lost by half a million dollars on a company carrying about $200 million of debt. A slightly higher bid from Trans World and some of the stores might have stayed open under new ownership. Instead the 89 stores emptied through the fall while shoppers picked over the racks at closeout prices. The American chain was 46 years old when it stopped trading. ## What the brand sells in 2026 and who runs it Tower Records came back, in a much smaller shape. On November 13, 2020, the brand relaunched as an online store selling vinyl, cassettes, compact discs and merchandise, with Pulse! revived alongside it. Rolling Stone reported the relaunch that month, including the company's stated plan to open physical locations later. The chief executive is Danny Zeijdel. He told Rolling Stone in 2020 that the return "has been met with tremendous success," and described customers photographing their Tower Records deliveries and posting them on Instagram, which is a fair description of what the brand now sells: a shipping box with a logo people recognize. What that means for a reader in 2026 is straightforward. There is no Tower Records store to walk into anywhere in the United States. The website takes orders and ships records. The browsing, the listening stations and the clerks with opinions, which is what people actually remember, is the part that did not come back. Where to find one today: original Tower Records store bags, staff shirts and back issues of Pulse! circulate steadily in the vintage market, and what they cost depends on condition and rarity rather than any set rate. The Tower Records listings on Etsy (https://www.etsy.com/market/tower_records) are a reasonable place to start looking. ## Why the Japanese stores outlived the American ones Tower Records stores are still open every day, just not in the country where the chain started. Tower Records Japan became a separate company from the American parent in the early 2000s, which is why it was never part of the 2006 liquidation, and it still runs stores across Japan, including the Shibuya flagship in Tokyo. Marketplace reported in 2022 on why those stores kept working. Japanese listeners went on buying physical discs long after American listeners stopped, compact discs stayed the dominant format there, and the streaming services that hollowed out American music retail arrived in Japan late. The gap is easy to underestimate. In the United States the question about Tower Records is which year it closed. In Japan it is which branch to visit, and the chain is treated as a normal part of buying music rather than a memory of it. That split is also the answer to a question people ask after seeing a photograph of a Tower Records sign taken last week in Tokyo. The stores in Japan were never sold to Great American Group, were never part of the American bankruptcy, and never had to come back, because they never went away. ## Frequently Asked Questions ### Is Tower Records still in business? Tower Records is still in business as an online store, which relaunched on November 13, 2020 and sells vinyl, compact discs, cassettes and merchandise. The American retail chain is gone: its 89 remaining stores were liquidated after the October 2006 bankruptcy auction, and no Tower Records store operates in the United States in 2026. Tower Records Japan, a separate company, still runs stores in Japan. ### Why did Tower Records go out of business? Tower Records went out of business because compact disc sales fell away while its debts did not. The Washington Post's 2018 obituary of founder Russ Solomon said the company was felled by digital downloads and competition from discount chains. Tower Records filed for bankruptcy in 2004, filed again in August 2006 with 89 stores in 20 states and about $200 million owed to creditors, and was sold to a liquidation firm at auction that October. ### Who owns Tower Records now? The Tower Records name outlived the chain that made it. The American brand changed hands after the 2006 liquidation and has been run since its November 13, 2020 online relaunch by a management team under chief executive Danny Zeijdel, with no stores and no connection to the original retail operation. Tower Records Japan is owned and run separately, which is why its stores stayed open through and after 2006. **Sources:** - Deseret News (Associated Press): Tower Records bankruptcy filing to result in liquidation, October 2006: https://www.deseret.com/2006/10/9/19978382/tower-records-bankruptcy-filing-to-result-in-liquidation/ - The Washington Post: Russ Solomon, Tower Records founder who created a mecca for music lovers, dies at 92 (2018): https://www.washingtonpost.com/local/obituaries/russ-solomon-tower-records-founder-who-created-a-mecca-for-music-lovers-dies-at-92/2018/03/05/fa0267be-20b4-11e8-86f6-54bfff693d2b_story.html - Billboard: Elton John Shopped There and Dave Grohl Worked There, Behind Colin Hanks' New Tower Records Doc (2015): https://www.billboard.com/music/music-news/colin-hanks-all-things-must-pass-tower-records-doc-elton-john-dave-grohl-david-geffen-6502450/ - Rolling Stone: Tower Records Relaunches as Online Music Store (November 2020): https://www.rollingstone.com/music/music-news/tower-records-online-music-store-1090626/ - Marketplace: Tower Records stores have actually been thriving all this time, in Japan (2022): https://www.marketplace.org/story/2022/11/01/tower-records-stores-have-actually-been-thriving-all-this-time-in-japan --- # Who Owns eToys.com Now? Toys R Us Bought the Name URL: https://404memoryfound.com/posts/etoys-com-who-owns-it-now.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-17 Topics: Business Blunders, Internet Culture **Summary:** eToys.com was the online toy store that briefly passed Toys R Us on the stock market in 1999, then filed for Chapter 11 on March 7, 2001. Toys R Us has owned the eToys.com address since February 2009, when it bought the name out of a second bankruptcy for a reported $2.15 million. In 2026 there is no eToys store behind it. **Key facts:** - Launched: 1997, founded by Toby Lenk in Santa Monica, California - IPO: Priced at $20 a share on Nasdaq, May 20, 1999; closed the first day near $76 - Fiscal 2000 results: $151.0 million in revenue against a $189.6 million net loss, per its SEC annual report - Status today: Bankrupt since March 7, 2001; no eToys store operating in 2026 - Owner today: Toys R Us, which bought the eToys.com name in February 2009 ## Who started eToys, and what did it promise shoppers? eToys, Inc. was founded in 1997 by Toby Lenk, who had been a corporate vice president at the Walt Disney Company before he left to sell toys over a modem. The site opened later that year from offices in Santa Monica, California. The pitch was aimed at the parts of toy shopping people hated. No parking lot in December, no line at the register, no aisle where the one thing a kid asked for was sold out. Shoppers could sort by age, by price or by brand, which no physical store could do. It worked well enough that eToys became the name investors reached for whenever they wanted an example of retail moving onto the web. By the 1999 holidays the company was spending heavily on television advertising and on warehouses it had to build from nothing, because unlike Toys R Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html) it owned no stores to ship from. That was the trade eToys made. It carried no rent, no cashiers and no shelves. It also had no existing distribution network, and toys are bulky, seasonal and cheap per box, which is close to the worst possible cargo for a company learning to ship. ## How did a toy site pass Toys R Us on the stock market? eToys sold shares to the public at $20 each and started trading on Nasdaq on May 20, 1999. The offering raised about $166 million. The stock never traded anywhere near $20. It opened well above the offering price, ran into the mid 80s during the session and closed its first day near $76, roughly 280 percent above where the underwriters had priced it. At that close the market valued eToys at about $7.8 billion. Toys R Us, the largest toy retailer in the country, was worth about $5.6 billion the same week. The gap between the two valuations and the two businesses was not subtle. In the fiscal year that ended March 31, 2000, eToys reported revenue of $151.0 million. Toys R Us was selling billions of dollars of toys a year out of real buildings. Investors were paying for the second company on the assumption that it would turn into the first. The loss line explained the risk they were taking. eToys lost $189.6 million in that same fiscal year, more than it collected in sales. ## What was Toywar, and why did it embarrass eToys? A European art group called etoy had registered etoy.com in 1995, two years before eToys opened its store. In late 1999 eToys sued, arguing that shoppers would confuse the two names. A Los Angeles court granted a preliminary injunction in November 1999 and ordered etoy off the address, with penalties of up to $10,000 a day if the artists kept using it. The timing could not have been worse for the retailer, because the order landed in the middle of the holiday shopping season. etoy answered by turning the lawsuit into an artwork. Toywar was a browser-based campaign that signed up thousands of players to flood eToys with email, tie up its website and argue the case in public while the company was trying to sell Christmas presents. etoy described the project as "not about hunting stupid little pixel monsters: it is about the destruction of Internet brands and valuable domain names," in its own account of the campaign in 1999. eToys dropped the suit in January 2000 and agreed to cover up to $40,000 of etoy's legal costs. The artists kept the address, and etoy.com is still theirs. ## What happened during the Christmas of 2000? The 2000 holiday season was the one that finished the company. eToys had built and staffed for a Christmas far bigger than the one it got, and by January 2001 it was telling investors it was running out of money. The cuts came fast. eToys closed its European operations in January 2001 and eliminated about 700 jobs in the United States, roughly 70 percent of its workforce. The stock went with it. Shares that had closed near $76 on their first day in 1999 were trading at 9 cents by early 2001, and Nasdaq halted them before removing the company from the exchange. Two things had changed underneath eToys while it was raising and spending. Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html) had widened its catalog into toys, and Toys R Us had given up on building a rival site by itself and teamed with Amazon instead. eToys was suddenly the third choice in a two-horse category. The dot-com crash did the rest. Money that had been free in 1999 was gone by 2001, and a retailer losing nine figures a year had no way to raise more of it. Pets.com (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html) hit the same wall in the same stretch of months. ## What did the bankruptcy leave behind? eToys filed for Chapter 11 in the United States Bankruptcy Court in Wilmington, Delaware on March 7, 2001, and shut the website the next day. The filing listed $416.9 million in assets against $285 million in debts as of December 31, 2000. The site left a short notice for anyone who arrived after the doors closed. It was no longer taking orders but would fill the ones already placed, and it signed off with a line that read like a store clerk waving from an empty aisle: "It's been our sincere pleasure serving you." Operations ended entirely in April 2001. What was left got sold off in pieces, which is how a company the market had priced at $7.8 billion came apart into line items on a bankruptcy docket. The brand and the address went to KB Holdings, parent of the mall chain KB Toys (https://404memoryfound.com/posts/kb-toys-stores-left.html), at a bankruptcy auction in May 2001 for about $3.35 million. The market had valued the same company at $7.8 billion two years earlier. ## Where did the eToys name end up? KB Toys reopened eToys.com as an online store, and the address kept changing hands as its operators did. By the mid 2000s the site was run by eToys Direct, which became part of a company called The Parent Company. The Parent Company filed for Chapter 11 on December 22, 2008, which put eToys.com on the auction block for the second time in seven years. Toys R Us bought it in February 2009, together with BabyUniverse.com and ePregnancy.com, for a reported $2.15 million. The chain eToys had briefly passed on the stock market ended up owning the eToys name for less than KB Holdings had paid in 2001. That is where it sits in 2026. Toys R Us still holds the eToys.com address and there is no eToys store behind it. The name is an asset on a shelf rather than a shop, kept for the same reason big retailers keep dead brands: it costs almost nothing to hold, and something to lose. ## Frequently Asked Questions ### Is eToys still in business? No. eToys, Inc. filed for Chapter 11 bankruptcy on March 7, 2001, closed eToys.com the following day and ended operations that April. A later company reopened the eToys.com address as an online toy store, but that business went through bankruptcy too, in December 2008, and no eToys store has operated since. ### Who owns eToys.com now? Toys R Us has owned the eToys.com address since February 2009, when it bought the name out of The Parent Company's bankruptcy for a reported $2.15 million. Before that the name belonged to KB Holdings, which took it at the original eToys bankruptcy auction in May 2001 for about $3.35 million. ### How much was eToys worth at its peak? eToys was worth about $7.8 billion at the close of its first day of trading on May 20, 1999, when the stock finished near $76 after being priced at $20. That was more than Toys R Us, the largest toy retailer in the country, which was valued at about $5.6 billion at the time. **Sources:** - eToys Inc., Form 10-K for the fiscal year ended March 31, 2000, SEC EDGAR: https://www.sec.gov/Archives/edgar/data/0001052245/000091205700027125/a10-k.txt - eToys IPO plays well on Wall Street, CNNfn, May 20, 1999: https://money.cnn.com/1999/05/20/technology/etoys/ - Internet toy retailer eToys files for bankruptcy, closes Web site, Associated Press via Deseret News, March 8, 2001: https://www.deseret.com/2001/3/8/19573862/internet-toy-retailer-etoys-files-for-bankruptcy-closes-web-site/ - TOYWAR 1999, etoy.CORPORATION project archive: https://etoy.com/projects/toywar/ - EToys.com, Wikipedia: https://en.wikipedia.org/wiki/EToys.com --- # Broadcast.com: The $5.7 Billion Site Yahoo Shut Down URL: https://404memoryfound.com/posts/broadcast-com-yahoo-billion-dollar-deal.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-17 Topics: Business Blunders, Internet Culture **Summary:** Broadcast.com was the Dallas streaming company that Mark Cuban and Todd Wagner sold to Yahoo on July 20, 1999 for $5.7 billion in stock, after 1998 revenue of $22.4 million. Yahoo shut the last of the service down in 2002. In 2026 the name survives only as a domain Yahoo still owns, and Cuban says he has tried to buy it back. **Key facts:** - Launched: 1995 in Dallas as AudioNet; renamed Broadcast.com in 1998 - 1998 results: $22.4 million revenue, $16.4 million net loss, per its SEC annual report - Sold to Yahoo: $5.7 billion in stock, completed July 20, 1999 - Status today: Service closed in 2002; no site at the address - Owner today: Yahoo, owned by Apollo Global Management funds since 2021 ## Who built Broadcast.com, and what did it actually stream? The company began in Dallas in 1995 as AudioNet, a service that put live radio signals onto the internet. Mark Cuban and Todd Wagner, both Indiana University graduates living in Texas, took it over because they wanted to hear Hoosiers basketball games they could not pick up on a radio. What it sold was simple and, in 1995, close to unheard of: a link in a browser that played live audio. AudioNet carried radio stations, college and professional sports, press conferences and political speeches, then added video. A second line of business turned out to pay better than advertising. Companies wanted their earnings calls, product launches and training sessions streamed, and almost none of them knew how. The company renamed itself Broadcast.com in 1998 and sold that service to corporate clients. The quality was poor by any later standard. Most listeners were on 28.8k or 56k modems, and video, when a stream carried it, ran in a window the size of a postage stamp. Nobody minded much, because the alternative was nothing. Most of the decoding was handled by RealPlayer (https://404memoryfound.com/posts/what-happened-to-realplayer-streaming.html), the software almost every streaming site depended on at the time. ## Why was the 1998 IPO a record? Broadcast.com priced 2.5 million shares at $18 in July 1998, above the $11 to $13 range in its early filings with the Securities and Exchange Commission. The stock closed its first day of trading at $62.75. That was a gain of 249 percent, the largest first-day jump any US IPO had posted to that point, edging out Secure Computing's 247 percent from 1995, as CBS News reported that week. The record did not last the year: theglobe.com (https://404memoryfound.com/posts/what-happened-to-theglobe-com-dot-com-ipo.html) took it four months later. The business underneath the stock price was much smaller than the trading suggested. Broadcast.com finished 1998 with $22.4 million in revenue and a net loss of $16.4 million, according to the annual report it filed with the SEC for that year. Investors were not buying the revenue. They were buying the idea that live audio and video would move to the internet, and that the company carrying the most of it in 1998 would still be carrying the most of it in 2005. ## Why did Yahoo pay $5.7 billion for a company earning $22 million? Yahoo completed the purchase on July 20, 1999, in an all-stock deal valued at $5.7 billion. "Broadcast.com's tremendous first-to-market advantage has made it the leading destination on the Web for audio and video broadcasts," Yahoo chairman and chief executive Tim Koogle said in the announcement that day. The reasoning was about inventory. Yahoo was the front door of the web in 1999, with directory, mail and news traffic it could not sell video advertising against, because it had no video. Broadcast.com had the streams, the encoding staff and the contracts with radio stations and sports teams. The price still worked out to roughly 250 times the previous year's revenue for a company that had never turned a profit. Yahoo paid it in stock, which in 1999 felt free: the shares were going up faster than the cash would have been spent. The scale of it is easier to judge with hindsight. Google bought YouTube in October 2006 for $1.65 billion, less than a third of what Yahoo paid seven years earlier for a service that no longer existed by then. ## What killed the service after the deal closed? Bandwidth did most of the damage. Every listener cost money in server capacity and transit, and a free stream with a banner ad beside it did not cover the bill. The advertising market that was supposed to grow into the cost collapsed instead when the dot-com crash hit in 2000 and 2001. Yahoo folded the operation into Yahoo Broadcast Services and began cutting. The corporate streaming business was sold off, the consumer programming was thinned out, and in 2002, three years after the purchase, Yahoo closed the last of the broadcast operations it had bought, including Yahoo Radio and its FinanceVision video channel. The technology was not the problem. Live streaming became the biggest thing on the internet within a decade, on YouTube from 2005 and on services built for broadband rather than dial-up. Broadcast.com was early enough to prove the demand and too early to serve it profitably. It is the purest example of what the 1999 market rewarded. The price Yahoo paid has been on every list of the worst acquisitions of the era since the crash, alongside its own purchase of GeoCities (https://404memoryfound.com/posts/history-of-geocities-websites.html) the same year. ## Who owns Broadcast.com now, and does the address still work? Yahoo has kept the domain since 1999 and never sold it. There is no Broadcast.com site to visit: the address forwards to Yahoo's own pages, and has done so for most of the past two decades. Cuban has tried to get it back at least twice. Fortune reported in August 2017 that he wanted the name for a new venture, and in May 2025 he posted a list of patents his lawyers had drafted for Broadcast.com in January 1999 and never filed, adding: "I tried to buy it back. Still willing to if anyone there cares!" Yahoo is no longer the company that bought Broadcast.com either. It was sold to Verizon, then sold again to funds managed by Apollo Global Management, which has run it as a standalone business since 2021. The details of that chain are in our piece on who owns Yahoo now (https://404memoryfound.com/posts/who-owns-yahoo-now.html). So the answer people search for is unglamorous. The company is gone, the service is gone, and the name sits unused in the domain portfolio of a private-equity-owned internet company that has no plans for it. ## What did the $5.7 billion actually leave behind? It left several hundred people in Dallas with life-changing money. Cuban has said publicly that the great majority of Broadcast.com's roughly 330 employees came out of the sale as millionaires, because the staff held stock and the stock was sold at the top of the market. It left Cuban himself with Yahoo shares he was smart enough not to hold. He hedged the position with a collar that set a floor under the price, which is why his fortune survived Yahoo's fall after 2000, and in January 2000 he bought the Dallas Mavericks. It left Wagner running 2929 Entertainment and Landmark Theatres with him. And it left Yahoo with a reputation for overpaying that followed the company into every deal it did afterward. ## Frequently Asked Questions ### What happened to Broadcast.com? Broadcast.com, the Dallas streaming company founded as AudioNet in 1995, was bought by Yahoo in an all-stock deal completed on July 20, 1999 and valued at $5.7 billion. Yahoo folded it into Yahoo Broadcast Services and shut the last of those operations down in 2002. ### How much did Yahoo pay for Broadcast.com? Yahoo paid $5.7 billion in stock for Broadcast.com in 1999, against revenue of $22.4 million and a net loss of $16.4 million that Broadcast.com had reported to the SEC for 1998. It remains one of the most expensive acquisitions of the dot-com era. ### Does Broadcast.com still work in 2026? No. There has been no Broadcast.com service since Yahoo closed it in 2002, and the domain forwards to Yahoo's own pages. Mark Cuban, who ran Broadcast.com before the 1999 sale, said in May 2025 that he had tried to buy the name back and was still willing to. **Sources:** - Broadcast.com, Encyclopaedia Britannica: https://www.britannica.com/topic/Broadcast-com - Broadcast.com Inc., Form 10-K405 for fiscal 1998, SEC EDGAR: https://www.sec.gov/Archives/edgar/data/0001061236/000095013499002266/0000950134-99-002266.txt - Broadcast.com IPO Soars 249%, CBS News, July 1998: https://www.cbsnews.com/news/broadcastcom-ipo-soars-249/ - Yahoo! Completes Broadcast.com Acquisition, July 20, 1999: https://www.newmediawire.com/news/yahoo-completes-broadcast-com-acquisition-3064719 - Mark Cuban Reveals Lost Broadcast.com Patents, Benzinga, May 2025: https://www.benzinga.com/tech/25/05/45340957/mark-cuban-reveals-lost-broadcast-com-patents-that-couldve-shaped-techs-future-still-willing-to-buy-it-back-if-anyone-there-cares --- # EverQuest in 2026: The 1999 MMO That Still Runs URL: https://404memoryfound.com/posts/everquest-still-active-2026.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-17 Topics: Gaming, Internet Culture **Summary:** EverQuest is still active in 2026. Darkpaw Games, part of Daybreak Game Company, still runs the 1999 MMORPG and opened a new time-locked progression server called Lethar on August 26, 2026. A separate solo-friendly rebuild, EverQuest Legends, launched on July 28, 2026 for $19.99 plus $9.99 a month. **Key facts:** - Launched: March 16, 1999, by Verant Interactive for Sony Online Entertainment - Early sales: 10,000 copies on day one, 225,000 within six months - Owner today: Darkpaw Games, part of Daybreak Game Company, owned by Enad Global 7 - Status today: Live in 2026. Newest server, Lethar, opened August 26, 2026 - Cost now: Client is a free download. EverQuest Legends is $19.99 plus $9.99 a month ## How a 1999 fantasy game outgrew its own servers EverQuest went on sale on March 16, 1999, built by Sony's 989 Studios and its spinoff Verant Interactive and published by Sony Online Entertainment. Nobody inside the building expected what followed. Ultima Online, the 1997 game that first proved players would pay every month for a shared world, had sold about 120,000 copies, and by PC Gamer's account of the launch the EverQuest team would have been pleased to sell a quarter of that. EverQuest sold 10,000 copies on its first day. By April 1999 it had sold 60,000. Six months after launch the figure was 225,000, roughly double what Ultima Online had managed in twice the time. Servers filled faster than Sony could add them, and the company had to buy more internet capacity to keep the world standing up. The design is what made it stick. Norrath had no quest markers, no fast travel and a death penalty that sent you running unarmed across a continent to recover your own corpse. It asked for whole evenings rather than sessions, and several hundred thousand people handed them over. ## Who owns EverQuest now? Four owners later, the game has outlasted every one of them. Sony Online Entertainment ran EverQuest from launch. That studio was sold off and renamed Daybreak Game Company, the same organization under a different flag. At the start of 2020, Daybreak split itself into three studios, one for each of its biggest properties. Darkpaw Games took EverQuest and EverQuest II, Rogue Planet Games took PlanetSide 2, and Dimensional Ink Games took DC Universe Online. Darkpaw has run EverQuest ever since. In December 2020 the Swedish publisher Enad Global 7 agreed to buy Daybreak for $300 million, with the sale set to close by the end of that year. EG7 had bought MechWarrior developer Piranha Games the week before. So the 2026 answer needs three names. Darkpaw Games operates EverQuest day to day, Daybreak Game Company publishes it, and Enad Global 7 owns Daybreak. ## Which EverQuest servers opened in 2026? Darkpaw Games still ships expansions and still opens new servers, which is the clearest evidence the game is alive rather than merely switched on. The newest is Lethar, a time-locked progression server that went live on August 26, 2026 at noon Pacific time. Progression servers are EverQuest's answer to its own nostalgia. They start the world at an older expansion and unlock the rest on a published schedule, so a returning player can replay the game's history in order instead of landing in a world that moved on without them. Lethar skips the first ten expansions and opens at The Serpent's Spine, then unlocks The Buried Sea twelve weeks later and everything after that every eight weeks. That schedule runs a long way out. Shattering of Ro, the current expansion, is not due to unlock on Lethar until October 2029, which says something about how long Darkpaw expects the server to be there. Lethar also brought in a personal loot ruleset, where bonus loot drops to individual players at the end of a group fight instead of being rolled for. Where to find one today: the client itself is a free download, so there is nothing to buy in order to play. The collecting is all physical. The 1999 big box, the jewel-case reissues and the cloth maps of Norrath that shipped with the early expansions still change hands, and condition decides the price, since the disc is scenery without a live account. Sellers keep them listed among Etsy's EverQuest listings (https://www.etsy.com/search?q=everquest). ## Why Daybreak sued an EverQuest fan server EverQuest has a shadow version. Emulated servers have rebuilt old eras of the game for two decades, and Daybreak left most of them alone for most of that time. The Heroes Journey was the exception. It launched in November 2024, pitched itself as an EverQuest for solo and duo players, and grew quickly. Daybreak took it to court. The 41-page complaint named Kristopher Takahashi and Alexander Taylor, and accused the server of copying Daybreak's game content, circumventing its technical protections and diluting the EverQuest trademark. It said the operation "generates revenue through a thinly-disguised 'donation' system." Scale is what made the case unusual for a fan project. US District Judge Cynthia Bashant ruled in Daybreak's favor, and the ruling put The Heroes Journey's income at roughly $100,000 a month in donations. The server shut down. The client came off every download page and the site came off the web. Its operators had described the project as a "three-year labor of love aimed at creating an incredible soloduo EQ experience." In 2026 that description is most of what survives it. ## What EverQuest Legends does differently On July 28, 2026, Daybreak launched a second EverQuest. EverQuest Legends is a rebuilt version of the 1999 game, developed by the independent studio Game Jawn rather than by Darkpaw Games, and published by Daybreak. It keeps the look, the world and the music, and strips out the thing that defined the original: the requirement to find other people. The whole game can be finished alone. Groups cap at four players and raids at eight, and one character can carry up to three classes at once, so a single avatar can be a rogue, a paladin and a wizard. The pricing is a deliberate throwback. The base game costs $19.99 and includes one month of subscription time, and after that it runs $9.99 a month. David Youssefi, executive producer at Daybreak, made the pitch personal when the game was announced in 2026. "I've dreamt of a solo/casual version of EQ for over 20 years," he said. Legends runs on its own servers and does not replace anything. Darkpaw's EverQuest is still taking new characters on the same day. ## Why a 27-year-old MMO still has players EverQuest's survival has less to do with the game than with the sunk cost of a character somebody built in 2001 and a guild that still logs in on Tuesday nights. The same force keeps other old worlds open. RuneScape still draws a crowd (https://404memoryfound.com/posts/is-runescape-still-popular-2026.html) in 2026, and Second Life is still running (https://404memoryfound.com/posts/what-happened-to-second-life-virtual-world.html) more than twenty years on, for one reason: a virtual place holding a decade of somebody's life is very hard to switch off. The business math cooperates. A game with a small, loyal, paying base and a team measured in dozens does not need millions of players to justify itself, which is why the expansions have kept arriving every December. What actually ended was the culture around it, the era when playing together meant hauling a tower across town for a LAN party (https://404memoryfound.com/posts/golden-age-of-lan-parties.html) and splitting a phone line. The servers are the part that never stopped. ## Frequently Asked Questions ### Is EverQuest still active in 2026? Yes. EverQuest, released on March 16, 1999, is still live in 2026 under Darkpaw Games, which is part of Daybreak Game Company. Darkpaw opened a new time-locked progression server named Lethar on August 26, 2026, and Shattering of Ro is the game's current expansion. ### Who owns EverQuest now? EverQuest is operated by Darkpaw Games, a studio Daybreak Game Company created at the start of 2020 to run EverQuest and EverQuest II. Daybreak, the former Sony Online Entertainment, was itself bought by the Swedish publisher Enad Global 7 in a $300 million deal agreed in December 2020. ### Is EverQuest Legends the same game as EverQuest? No. EverQuest Legends is a separate, solo-friendly rebuild of the 1999 game, developed by Game Jawn and published by Daybreak on July 28, 2026. It costs $19.99 with one month of subscription included and $9.99 a month after that, and it runs on its own servers alongside the original EverQuest. **Sources:** - PC Gamer: Breaking the internet, the story of EverQuest: https://www.pcgamer.com/breaking-the-internet-the-story-of-everquest-the-mmo-that-changed-everything/ - Gematsu: Enad Global 7 to acquire Daybreak Game Company: https://www.gematsu.com/2020/12/enad-global-7-to-acquire-daybreak-game-company - MMOHuts: EverQuest's Lethar TLP server starts at The Serpent's Spine on August 26: https://mmohuts.com/news/everquests-lethar-tlp-server-starts-at-the-serpents-spine-on-august-26 - TweakTown: EverQuest emulator The Heroes' Journey loses copyright suit, shuts down: https://www.tweaktown.com/news/107868/highly-profitable-everquest-emulator-the-heroes-journey-thj-loses-copyright-suit-shuts-down/index.html - Gematsu: EverQuest Legends launches July 28: https://www.gematsu.com/2026/06/everquest-legends-launches-july-28 --- # Is Angelfire Still Online? Lycos Closed It in 2026 URL: https://404memoryfound.com/posts/is-angelfire-still-online.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-16 Topics: Internet Culture, Business Blunders **Summary:** Angelfire is gone. Lycos shut the free web host down on April 24, 2026, about three months after a server outage in January 2026 had already made every hosted page unreachable. The brand still belongs to Lycos, Inc., now a subsidiary of Brightcom Group, but nothing of the service runs and the user pages were purged. **Key facts:** - Launched: 1996, as a paid website builder and medical transcription service - Owner today: Lycos, Inc., a subsidiary of Brightcom Group - Status today: Shut down April 24, 2026; hosted pages purged - Free signups ended: 2010, when Lycos stopped allowing new free Angelfire sites - What replaced it: Neocities and GitHub Pages for new sites, the Wayback Machine for old ones ## What did Lycos say before it pulled the plug? On March 6, 2026, a short notice appeared on the Lycos homepage. It read: "We apologize for the service interruptions. Unfortunately we will be shutting down in the next 30 days." Thirty days put the deadline at April 5, 2026. The notice named no backup tool, no download link and no export option. It told people to move their hosting somewhere else. Lycos then took the message down. Archivists found it still sitting in the homepage source code after it vanished from the visible page, and on April 3, 2026 it reappeared for anyone who looked. April 5 came and went without a formal goodbye. Angelfire's homepage stayed reachable for another three weeks. On April 24, 2026 it stopped answering, and Tripod, the other free host Lycos had owned since 1998, went dark the same day. That is the entire public record. No press release, no blog post, no named executive attached to the decision. A service that had been online for 30 years closed with two sentences of homepage copy. ## Why did Angelfire pages break in January 2026? The March notice was not the first sign. User sites on Angelfire stopped loading in early January 2026, roughly two months before Lycos said anything at all. Visitors got HTTP 502 and HTTP 403 errors instead of pages. The Angelfire homepage itself flickered in and out, which made the whole thing look like a bad week rather than an ending. Nobody outside the company could tell which it was. There was no status page and no support answer, so people who ran Angelfire sites spent the first quarter of 2026 guessing. The timing matters for anyone hoping to recover a page. Most Angelfire URLs were last crawled before January 2026, because after that the server was handing out error codes rather than HTML. Whatever archives captured in 2025 or earlier is what survives. It also means the 30-day warning was closer to zero days of real access. By March 6, 2026 the files had already been unreachable for two months. ## Who owns Angelfire now, and is any of it running? Nothing of Angelfire runs in 2026. The builder is gone, the hosted pages were purged, and the address serves no site. The brand still has an owner on paper. Angelfire belongs to Lycos (https://404memoryfound.com/posts/what-happened-to-lycos-search-engine.html), Inc., which is a subsidiary of Brightcom Group, the digital advertising company that was called Ybrant Digital when it bought Lycos in 2010 for $36 million. That $36 million is the punchline of the whole chain. On May 16, 2000, near the top of the dot-com bubble, Lycos announced it would be bought by Terra Networks for $12.5 billion. Terra sold Lycos to South Korea's Daum Communications in 2004, and Daum sold it to Ybrant Digital in 2010. Angelfire came along for every one of those rides. It had been a Lycos property since 1998 and was still one 28 years later, which is how a free homepage service outlived three of its own parent companies before the fourth closed it. Lycos itself is still trading in 2026. Angelfire is not, and Lycos has published no plan to bring it back. ## How a transcription startup became a homepage factory Angelfire launched in 1996 as an odd hybrid: part website builder, part medical transcription service. The transcription half was dropped, and what was left was a paid tool for putting a page on the web. Free came later, and it came from Lycos money. WhoWhere, a company based in Mountain View, California, picked up Angelfire in 1997. A Lycos filing with the Securities and Exchange Commission records the deal as the purchase of "certain net assets of Angelfire, Inc. for approximately $2,720,000 in cash and notes payable" in October 1997. Lycos bought WhoWhere in August 1998 for $133 million, which is how Angelfire, the MailCity email service and WhoWhere's white pages all landed in one portfolio. Lycos had already bought Tripod in February 1998 for $58 million in stock. Then Lycos made the builder free and paid for it by putting a banner ad on every hosted page. If you wanted the ad gone, you paid. That trade is the reason a generation of American teenagers had a homepage at all. ## What did people actually put on Angelfire? The most-read surviving Angelfire page is probably one its author never meant to be read. In April 2013, The Register and several other outlets reported on a 1999 Angelfire site built under the handle "mez51" whose source code named Mark Zuckerberg as the author. Zuckerberg has never confirmed it, and the reporting has always been careful to call it unverified. The page was the work of someone who said he was 15, lived in a small town near New York City, and had just finished his freshman year of high school. It carried a set of Java applets: a drawing tool, a base converter, a pong game, and one called The Web that connected visitors to each other. "As of now, the web is pretty small. Hopefully, it will grow into a larger web," the page said of that applet, five years before Facebook launched. The rest of Angelfire ran in the same register. Hobby code, band pages, fan shrines, wrestling stats, tiled backgrounds and under-construction signs. GeoCities (https://404memoryfound.com/posts/history-of-geocities-websites.html) is the name people reach for when they remember that web, but Angelfire and Tripod held a large share of it, and their 2026 closure removed it in one move. ## Can you still find your old Angelfire page? Sometimes. The Internet Archive's Wayback Machine is the only realistic route, and angelfire.com is not excluded from it, so archived copies of user sites do exist. Start with the full original URL, including the neighborhood folder that early Angelfire accounts used. If you only remember the account name, the archive's index will list the captured paths underneath it. Two things work against you. Angelfire deleted accounts it judged inactive, long before 2026, and no mirror of the pages removed that way is known to exist. The January 2026 outage then meant almost nothing new was captured in the service's final months. Archive Team ran a dedicated Angelfire grab during the shutdown, with volunteer scripts pointed at as many accounts as its tracker could enumerate. Anything it caught goes to the Internet Archive rather than back to the original address. If you kept your own copy on an old hard drive or a Zip disk (https://404memoryfound.com/posts/what-happened-to-iomega-zip-drives.html), that file is worth more than anything still online. For a new page, free static hosts such as Neocities and GitHub Pages do the job Angelfire did, without the banner ad. ## Frequently Asked Questions ### Is Angelfire still online in 2026? No. Angelfire, the free web host Lycos had owned since 1998, went offline for good on April 24, 2026, and its hosted user sites had already been unreachable since a server outage in early January 2026. Lycos posted 30 days of notice on March 6, 2026 and the hosted pages were purged. ### Did Tripod shut down at the same time as Angelfire? Yes. Tripod, the other free hosting service Lycos owned, stopped answering on April 24, 2026, the same day Angelfire's homepage went dark. Lycos had bought Tripod in February 1998 for $58 million in stock, and the single shutdown notice posted on March 6, 2026 covered both services. ### How do I find my old Angelfire website? Search the full original angelfire.com URL, neighborhood folder included, in the Internet Archive's Wayback Machine. Angelfire pages are not excluded from the archive, but Angelfire deleted accounts it considered inactive years before the April 24, 2026 shutdown, and no mirror of those deleted pages is known to exist. **Sources:** - Angelfire: https://en.wikipedia.org/wiki/Angelfire - Angelfire (shutdown record and archiving): https://wiki.archiveteam.org/index.php/Angelfire - Lycos, Inc. Form 8-K/A (1998), SEC EDGAR: https://www.sec.gov/Archives/edgar/data/0001007992/000092701698002099/0000927016-98-002099.txt - Lycos: https://en.wikipedia.org/wiki/Lycos - Ancient website from 1999: By Mark Zuckerberg aged 15 3/4, The Register (2013): https://www.theregister.com/2013/04/04/teenage_zuckerberg_angelfire_website/ --- # Are Heelys Still Sold in 2026? Who Owns Them Now URL: https://404memoryfound.com/posts/heelys-still-sold-2026.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-16 Topics: Then vs Now, Money & Tech **Summary:** Heelys are still sold new in 2026. BBC International has owned the wheeled-shoe brand since April 26, 2021, when it bought Heelys from Sequential Brands Group for $11 million in cash, and current models run about $60 to $75 a pair in the United States. The boom that made them famous peaked in 2006 at 6.2 million pairs and $188.2 million in net sales, then fell to $70.7 million by 2008. **Key facts:** - Launched: 2000, by Roger Adams through Heeling Sports Limited of Carrollton, Texas - Price then: $49.99 to $89.99 in 2007, about $80 to $145 in 2026 dollars - Peak year: 6.2 million pairs and $188.2 million in net sales in 2006 - The crash: Net sales fell to $70.7 million in 2008 and $43.8 million in 2009 - Status today: Still made and sold new in 2026, roughly $60 to $75 a pair - Owner today: BBC International, which bought the brand for $11 million on April 26, 2021 ## Who put the wheel in the heel, and when? Roger Adams was a psychologist in his forties, on a 1998 vacation in California, when he watched skaters and cyclists roll past and decided a shoe ought to be able to do that on command. He went home, cut the heel out of a sneaker and fitted a wheel into the hole. Skating was the family trade. Adams grew up in Tacoma, Washington, where his parents ran a roller rink, and at nine months old he was entered in the Guinness Book of World Records as the youngest person to roller skate. He launched Heelys in 2000 through a company he set up as Heeling Sports Limited, based in Carrollton, Texas, just north of Dallas. The design was one wheel seated in a removable cassette in each heel, so a kid could walk normally, shift weight back and roll, then pop the wheel out with a plug when an adult objected. Adams held the patents and made a fortune from them. He died of pancreatic cancer on March 24, 2026, at his home in Glenbrook, Nevada, at the age of 71, and the New York Times obituary that ran in April 2026 credited him with the invention outright. ## How big did the wheeled-shoe boom actually get? Enormous, and almost overnight. Heelys sold 1.4 million pairs in 2005 and 6.2 million pairs in 2006, an increase of 338.1 percent in a single year, according to the company's annual report for fiscal 2007. Net sales followed the same curve, from $44.0 million in 2005 to $188.2 million in 2006. The shoes carried domestic retail price points of $49.99 to $89.99, which is roughly $80 to $145 in 2026 dollars (https://404memoryfound.com/posts/90s-tech-inflation-calculator.html). The company took that momentum to the Nasdaq, pricing its initial public offering at $21.00 a share in December 2006 under the ticker HLYS. The stock rose sharply on its first day, and for about nine months Heelys looked like a footwear company rather than a craze. It was a craze. Toy booms of that decade ran on the same fuel, a cheap object that other kids could see you had, which is exactly what drove the Beanie Babies bubble (https://404memoryfound.com/posts/beanie-babies-bubble-what-happened.html) a decade earlier, and they tended to break the same way. ## Why did schools start banning them? Because children were rolling down hallways into other children. By August 21, 2006, the Deseret News reported that schools across Utah's Wasatch Front had banned athletic shoes with wheels in the heels, and that teachers at one elementary school had been arguing about a ban since 2003. What ended the argument there was an injury. "We had an incident at school where a student was riding their Heelys and fell back and broke their arm," Reed Hodson, principal at Highland Elementary, told the Deseret News in 2006. The bans spread through 2006 and 2007 across American school districts and shopping malls, usually on two grounds: collisions in crowded corridors, and wheel marks on gym floors. There was never a national rule, which is why the question still gets typed into search engines twenty years later. The picture in 2026 is the same patchwork. Some districts prohibit all wheeled footwear, others allow the shoes as long as the wheels stay out and the heel plugs stay in, and the policy lives in each school's handbook rather than in state law. ## How fast did the money disappear after 2007? Faster than it arrived. Net sales were $183.5 million in 2007, essentially flat against 2006, then collapsed to $70.7 million in 2008 as retailers worked through inventory nobody wanted. Sales fell again in 2009, to $43.8 million. That is a 62 percent drop in one year and a 77 percent drop in two, on a product whose tooling and distribution had been built for the 2006 number. The stock never recovered, and the company spent the next several years as a licensing operation with a large pile of cash and a shrinking business. Sequential Brands Group bought it in a deal announced in December 2012. Because Heelys held so much cash, Sequential later put the net purchase price of the brand itself at about $5.5 million in 2013, which is what a name worth $188.2 million in annual sales six years earlier was finally traded for. It is the standard shape of a novelty product that gets mistaken for a category, the same arc that turned the Segway (https://404memoryfound.com/posts/what-happened-to-segway-personal-transporter.html) from a world-changing invention into a mall-security vehicle. ## Who makes Heelys now, and what do they cost? BBC International owns the brand. Sequential Brands Group closed the sale of Heelys to BBC International on April 26, 2021 for $11 million in cash, and BBC International had already been the brand's core licensee since 2013, so the shoes never stopped shipping during the handover. "We believe the brand is in great hands under BBC's leadership going forward," said William Sweedler, executive chairman of Sequential Brands Group, in the April 2021 announcement. Sequential itself filed for bankruptcy later that year, so the sale outlasted the seller. Heelys are still made and still sold new in the United States in 2026, in one-wheel and two-wheel versions, through the brand's own site and mainstream chains such as Famous Footwear, Journeys and Walmart. Most current models list between about $60 and $75, which in real terms is close to what the mid-range pairs cost in 2007. Where to find a pair today: new Heelys are the cheap route, at roughly $60 to $75 for a current model from a US retailer, and they come with the wheels and plugs in the box. Discontinued 2000s colorways, old wheel cassettes and Heelys-era merchandise turn up on Etsy's marketplace (https://www.etsy.com/search?q=heelys), which is the practical place to look for a pair in the original styling rather than a single listing that disappears. They also outlasted most of what shared a shelf with them. Tamagotchi (https://404memoryfound.com/posts/tamagotchi-digital-pet-that-made-us-feel.html) and Heelys are two of the very few 1990s and 2000s crazes you can still buy new, from the original brand, in a US store in 2026. ## Frequently Asked Questions ### Are Heelys still sold in 2026? Yes. Heelys, the sneakers with a removable wheel in the heel that Roger Adams launched in 2000, are still manufactured and sold new in the United States in 2026, in both one-wheel and two-wheel models, with most current styles listing between about $60 and $75. ### Why were Heelys banned in schools? American schools began banning Heelys in 2006 and 2007 over hallway collisions, falls and marks left on gym floors. The Deseret News reported on August 21, 2006 that schools across Utah's Wasatch Front had prohibited them. There was never a national ban, and in 2026 the rule still varies by district, with many schools allowing the shoes only with the wheels removed. ### Who owns Heelys now? BBC International owns Heelys. It bought the brand from Sequential Brands Group for $11 million in cash in a sale that closed on April 26, 2021, after acting as the brand's core licensee since 2013. Sequential Brands Group had acquired Heelys, Inc. in a deal announced in December 2012. **Sources:** - Heelys, Inc., Form 10-K for fiscal year 2007, U.S. Securities and Exchange Commission: https://www.sec.gov/Archives/edgar/data/0001373980/000104746908002921/a2183631z10-k.htm - Heelys, Inc., fourth quarter and full year 2008 results, Form 8-K (2009), SEC: https://www.sec.gov/Archives/edgar/data/0001373980/000110465909021749/a09-9070_1ex99d1.htm - Schools put brakes on Heelys, Deseret News (August 21, 2006): https://www.deseret.com/2006/8/21/19969720/schools-put-brakes-on-heelys/ - Sequential Brands Group Announces Sale of Heelys Brand, GlobeNewswire (April 26, 2021): https://www.globenewswire.com/news-release/2021/04/26/2216988/25123/en/Sequential-Brands-Group-Announces-Sale-of-Heelys-Brand.html - Roger Adams dies at 71; he invented the rolling sneakers known as Heelys, New York Times obituary (April 2026): https://www.texarkanagazette.com/news/2026/apr/17/roger-adamsdies-at-71-he-invented-the-rolling/ --- # Why Discovery Zone Failed and What's Left in 2026 URL: https://404memoryfound.com/posts/discovery-zone-failed-indoor-playground.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-16 Topics: Business Blunders, Then vs Now **Summary:** Discovery Zone failed because it borrowed heavily to build about 350 indoor playgrounds and then could not fill them more than once or twice a year. The chain filed for bankruptcy in March 1996, filed again in April 1999, and closed its last locations by the end of 2001. In 2026 a single unaffiliated playground still uses the Discovery Zone name, at Florence Mall in Kentucky. **Key facts:** - Founded: 1989; first FunCenter opened January 1990 - Peak size: About 350 outlets in March 1996, per the San Francisco Chronicle - 1996 bankruptcy: $366.2 million in liabilities against $164.4 million in assets - Status today: Original chain liquidated; last locations closed by the end of 2001 - Owner today: An unaffiliated operator runs one Discovery Zone at Florence Mall, Kentucky ## Who built Discovery Zone, and what was inside one? Discovery Zone was founded in 1989 by Ronald Matsch, Jim Jorgensen and David Schoenstadt. The first center opened in January 1990, and the company called its locations FunCenters. The layout barely changed over the next eleven years. A FunCenter was a warehouse of padded tubes bolted to the ceiling, a ball pit deep enough to lose a shoe in, rope bridges, foam obstacle courses and a bank of arcade games that ran on plastic tokens. Parents were parked behind glass in a waiting area the company called the Quiet Zone. The product was a birthday party with no weather, no grass stains and no setup, sold by the session rather than by the slice. Tennis champion Billie Jean King was an early investor and a public supporter of the chain, which marketed physical play to girls and boys on identical terms. The pitch landed in the right decade. American malls were still full, two-income households were normal, and an indoor playground that charged admission looked like a business with almost no cost of goods. ## How did a 1990 startup reach 350 locations? It bought them. In April 1993, Blockbuster (https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html) Entertainment invested $10.3 million for 20 percent of Discovery Zone, with an option to raise that stake to 50.1 percent in June 1994. Blockbuster's money turned a regional chain into a national one in about eighteen months. In July 1994, Discovery Zone bought the Leaps and Bounds chain from McDonald's for $111 million in stock and cash, absorbing its only serious national rival in a single move. It also bought back the FunCenters its own franchisees had built. When Viacom acquired Blockbuster in September 1994, Discovery Zone came along inside the package, which is how a chain of ball pits ended up part of a Hollywood conglomerate's balance sheet. By March 1996, the San Francisco Chronicle counted about 350 outlets. Every one of them had cost real money to open. A FunCenter was a purpose-fitted big-box space, not a storefront, and the buildout was paid for long before the first token went into a machine. ## What did the 1996 bankruptcy petition actually say? Discovery Zone filed for Chapter 11 protection in Wilmington, Delaware on March 26, 1996. The petition listed $366.2 million in liabilities against $164.4 million in assets, including roughly $230 million in bank debt and subordinated debentures run up during the buying spree. The trajectory behind those numbers was ugly. Discovery Zone last made money in 1993, when it reported a profit of $3.3 million. It lost $24.9 million in 1994, lost $114.4 million through the first three quarters of 1995, and fell into technical default on loan covenants that September. The company was unusually blunt about the cause. "The bottom line is we have too many stores," spokesman Robert Mead told the San Francisco Chronicle in March 1996, adding that Discovery Zone would renegotiate unfavorable leases and close the centers it could not fix. Chief executive Donna Moore struck a brighter note the same week. "Discovery Zone is a viable concept, and we have an opportunity to revitalize our operations with new marketing programs and entertainment concepts," she said as the filing was announced in March 1996. ## Why did the second attempt fail faster than the first? Discovery Zone emerged from Chapter 11 on July 29, 1997 under new ownership led by the investment firm Wellspring Associates, with Blockbuster no longer in control. Wellspring spent money refurbishing FunCenters and brought in a new management team. It did not help, because the fixtures were never the problem. Discovery Zone sold novelty, and novelty does not produce repeat visits the way food does. A family might book one party a year and stay away in between, while the chain still had to light, staff, clean and insure every one of those buildings seven days a week. Discovery Zone filed for Chapter 11 a second time on April 20, 1999. Two months later it stopped pretending. On June 25, 1999, it closed more than 100 FunCenters at once, with no warning to staff or to the families holding reservations. The Washington Post described the result the next morning: a seven-year-old, his mother and sixteen friends arrived at the Falls Church, Virginia center for a party booked two months earlier and found the windows painted white and a typed notice taped up announcing that the store had closed. ## Is there a Discovery Zone open anywhere in 2026? One is, and it has nothing to do with the original company. A single indoor playground trades under the Discovery Zone name in 2026, inside Florence Mall in Florence, Kentucky, run by an unaffiliated operator that revived the name. The company that built the tubes is long dead. In July 1999, CEC Entertainment, the owner of Chuck E. Cheese (https://404memoryfound.com/posts/chuck-e-cheese-animatronics-still-there.html), bought the twenty Discovery Zone locations still open, thirteen owned and seven leased, together with the trademarks and trade names. A bankruptcy judge converted the case to liquidation in June 2000, and the last Discovery Zone locations closed by the end of 2001. The name reappeared on February 7, 2020, when a new operator opened a Discovery Zone at Eastgate Mall in Cincinnati. A second followed at Florence Mall on July 23, 2021. The Cincinnati location closed in 2025, which leaves Florence as the only place in the United States where the sign is still lit. So the honest answer to the search query is no. The chain people remember from 1996 does not exist, nobody is rebuilding it at national scale, and the surviving location is a tribute rather than a continuation. Discovery Zone died the way a lot of 1990s mall businesses died, a few years ahead of Toys R Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html), and for the same reason: fixed costs sized for a level of demand that never arrived. ## Frequently Asked Questions ### Is Discovery Zone still around in 2026? The original Discovery Zone chain is not. It closed its last locations by the end of 2001 after two bankruptcies. One indoor playground does trade under the Discovery Zone name in 2026, at Florence Mall in Florence, Kentucky, but it is run by an unaffiliated operator that revived the name in 2020, not by the company founded in 1989. ### When did Discovery Zone close? Discovery Zone closed in stages. It filed for Chapter 11 on March 26, 1996, emerged in July 1997, filed a second time on April 20, 1999, and shut more than 100 FunCenters without warning on June 25, 1999. A bankruptcy judge converted the case to liquidation in June 2000, and the last Discovery Zone locations closed by the end of 2001. ### What happened to Leaps and Bounds, the McDonald's indoor playground? Leaps and Bounds was McDonald's own chain of indoor play centers, and Discovery Zone bought it in July 1994 for $111 million in stock and cash. The Leaps and Bounds sites were rebranded as Discovery Zone FunCenters, so they went down inside the same collapse and were gone along with the rest of the chain by the end of 2001. **Sources:** - Discovery Zone seeks Chapter 11 protection, San Francisco Chronicle (1996): https://www.sfgate.com/business/article/discovery-zone-seeks-chapter-11-protection-3156186.php - Discovery Zone will close some outlets, Las Vegas Sun (March 25, 1996): https://lasvegassun.com/news/1996/mar/25/discovery-zone-will-close-some-outlets/ - Discovery Zone Inc. files for bankruptcy, Baltimore Sun (March 26, 1996): https://www.baltimoresun.com/1996/03/26/discovery-zone-inc-files-for-bankruptcy-acquisition-binge-caused-debt-company-plans-reorganization-recreation/ - Discovery Zone Abruptly Closes 100 Centers, The Washington Post (June 26, 1999): https://www.washingtonpost.com/archive/business/1999/06/26/discovery-zone-abruptly-closes-100-centers/937bb97e-6807-49c6-95bd-f2390edd0263/ - Discovery Zone, Wikipedia: https://en.wikipedia.org/wiki/Discovery_Zone --- # How Many Rainforest Cafes Are Left? 16 in the US URL: https://404memoryfound.com/posts/rainforest-cafe-locations-left.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-16 Topics: Business Blunders, Then vs Now **Summary:** Rainforest Cafe is still open in 2026, but only 16 restaurants are left in the United States, spread across 11 states, plus a handful abroad for about 22 worldwide. That is down from a peak of 59. The chain has belonged to Landry's, Tilman Fertitta's Houston restaurant group, since Landry's bought it for roughly $75 million in October 2000. **Key facts:** - Launched: October 1994, Mall of America, Bloomington, Minnesota - Peak size: 59 restaurants, late 1990s to early 2000s - Sold: October 2000 to Landry's, $3.25 a share, about $75 million - Owner today: Landry's, owned by Tilman Fertitta - Stores left: 16 in the United States across 11 states, about 22 worldwide ## Where did the Rainforest Cafe idea come from? Steven Schussler spent about three years and roughly $400,000 turning his own house in suburban Minneapolis into a working prototype of a jungle restaurant. He started in 1989, and he was an advertising salesman, not a restaurateur. Fortune magazine described the result in 1996: "Artificial waterfalls tumbled down custom-made rock formations, animatronic crocodiles bobbed their heads, and speakers piped in the roar of a tropical thunderstorm." The animals in the house were real ones. In his memoir It's a Jungle In There, Schussler listed "Forty tropical birds, two 150-pound tortoises, a baboon, an iguana and a bevy of tropical fish housed in ten 300-gallon fish tanks." Powering the demonstration took 3,700 extension cords running 20 separate sound, lighting and fog systems. The house did its job. Lyle Berman, the Minneapolis executive who founded Grand Casinos, put in $1.2 million and became the company's first chairman and chief executive. Rainforest Cafe, Inc. was formed in February 1994, with Schussler as executive vice president for development. ## How fast did the chain grow after the 1995 IPO? The first restaurant opened at the Mall of America in Bloomington, Minnesota, in October 1994. Some reference sites give February 1994, but that date belongs to the company's formation; Minnesota newspapers put the restaurant's debut in the fall, and the second location, at Woodfield Mall outside Chicago, opened about a year later in October 1995. Rainforest Cafe went public on the Nasdaq in April 1995 under the ticker RAIN. Shares were offered at $6, each with a warrant attached. The offering raised $9.5 million and the warrants brought in another $14.2 million. Revenue went from $13.45 million in 1995 to $262.69 million in 1999, close to twentyfold in four years. The company opened 34 restaurants in that stretch and topped out at 59. The rooms were not cheap to build. The Mall of America restaurant cost around $1,000 per square foot, because almost nothing in it came off a shelf. Every waterfall, every animatronic gorilla and every fiberglass banyan trunk was made to order. ## Why did the jungle stop making money? The first crack showed in early 1998. Rainforest Cafe warned that fourth-quarter earnings would fall short, and the stock dropped 40 percent. Restaurants open longer than 18 months were running 11 percent behind the previous year. Shareholders sued, alleging that management knew sales were sliding while selling their own shares at higher prices. Net income of about $14.65 million in 1998 fell to $5.69 million in 1999, in the same year revenue hit its record. The math underneath was simple. A themed restaurant sells a first visit brilliantly and a fifth visit badly, and Rainforest Cafe had put most of its rooms in regional malls where the same families walked past every weekend. When those malls started losing anchors, chains like Sam Goody (https://404memoryfound.com/posts/sam-goody-stores-left.html) and Toys R Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html) went with them. By 2000 the company was planning to shut as many as 20 restaurants in the first quarter of 2001. ## Who bought Rainforest Cafe, and for how much? Landry's, the Houston restaurant group run by Tilman Fertitta, had offered roughly $125 million earlier in 2000 and been rejected by Rainforest Cafe shareholders. It came back in September with a smaller number. The second offer was $3.25 a share in cash, made through a subsidiary called LSR Acquisition Corp. and worth about $75 million. It expired on October 27, 2000. Around 13,360,000 shares were tendered, and together with the 1,030,800 shares Landry's already owned, that came to nearly 65 percent of the company. Landry's changed the plan straight away. Instead of chasing shopping-mall traffic, it concentrated the surviving restaurants where tourists go and where a two-hour wait is part of the day out: Disney property, Las Vegas, Niagara Falls, Galveston, Times Square. ## Which states still have a Rainforest Cafe? Eleven of them. In 2026 Rainforest Cafe runs 16 restaurants in the United States across 11 states, plus locations in Canada, Japan, France and Malta, for roughly 22 worldwide. Texas and Florida have more than one. Most of the other states have exactly one. California is down to a single restaurant, at the Ontario Mills mall east of Los Angeles. The state had four, and the other three closed between 2013 and 2018. New Jersey is also down to one, after the Menlo Park Mall restaurant in Edison closed on February 19, 2025, when its landlord declined to renew the lease. The tourist closures stung more. The Downtown Disney location at the Disneyland Resort in Anaheim closed on June 30, 2018, for a hotel that was later cancelled, and the building is now a Star Wars Trading Post. Woodfield Mall closed when its lease expired on January 1, 2020. The Rainforest Cafe at Disney Village beside Disneyland Paris, open since 1999, is now expected to close in November 2026, moved back from a reported September date. Disney has not confirmed either date publicly. ## Who owns Rainforest Cafe now? Landry's does, and Landry's belongs to Tilman Fertitta. He bought in the remaining public shares in 2010 and took the company private, and the restaurant group now sits inside Fertitta Entertainment next to Bubba Gump Shrimp Co., Morton's The Steakhouse, Del Frisco's and the Houston Rockets. Private ownership is part of why the chain has neither vanished nor grown. There is no share price to defend and no quarterly pressure to open units, so a Rainforest Cafe closes when a lease ends rather than in a restructuring announcement. The one recent addition was a pop-up at the Empire State Building in New York in 2024, not a new mall restaurant. The contrast with its old rival is stark. Planet Hollywood, which had more than 60 locations worldwide at its 1990s peak, was down to about three restaurants and five hotels as of May 2025. Rainforest Cafe lost two thirds of its restaurants and kept the format intact, including the thunderstorm that still rolls through the dining room on a timer every 20 to 25 minutes depending on the location. ## Why is Rainforest Cafe popular online again? Because the children who were taken there in 1998 are adults with phones. Fast Company reported in 2024 on a wave of Rainforest Cafe videos posted by people too young to remember the chain at full size, running the line "You're not just there for the food" as its headline. The appeal is the same one Chuck E. Cheese animatronics (https://404memoryfound.com/posts/chuck-e-cheese-animatronics-still-there.html) still have: a room built before screens did the work, where a mechanical elephant raises its trunk on a cycle and a fiberglass storm breaks over your appetizers. Very little has been built that way since. Where to find one today. The restaurants are the only place to see the animatronics, but the merchandise has its own market: souvenir glasses, 1990s graphic tees, mugs, safari hats and the Cha! Cha! tree frog plush that every kid wanted. Sellers list them by the hundred on Etsy's vintage Rainforest Cafe listings (https://www.etsy.com/search?q=vintage+rainforest+cafe), where condition and whether the tag survived do most of the pricing. ## Frequently Asked Questions ### Is Rainforest Cafe still open in 2026? Yes. Rainforest Cafe still operates 16 restaurants in the United States in 2026, across 11 states, along with locations in Canada, Japan, France and Malta for about 22 worldwide. That is down from a peak of 59 restaurants around the turn of the 2000s, and the Disney Village restaurant near Disneyland Paris is expected to close in November 2026. ### Who owns Rainforest Cafe now? Landry's owns Rainforest Cafe, and has since a tender offer at $3.25 a share, worth about $75 million, that expired on October 27, 2000. Landry's itself is owned by Tilman Fertitta, who took it private in 2010 and holds it through Fertitta Entertainment along with Bubba Gump Shrimp Co. and Morton's The Steakhouse. ### Why did so many Rainforest Cafe locations close? Rainforest Cafe built most of its restaurants in American shopping malls in the late 1990s, and a themed restaurant depends on first visits rather than repeat ones. Sales at older locations were already falling 11 percent year over year by early 1998, the company planned to close up to 20 restaurants in early 2001, and mall closures plus the 2020 pandemic took out most of the rest. **Sources:** - Atlas Obscura: The True Story of the Rainforest Cafe Is Even Wilder Than You Thought: https://www.atlasobscura.com/articles/the-true-story-of-the-rainforest-cafe-is-even-wilder-than-you-thought - Encyclopedia.com: Rainforest Cafe Inc. company history: https://www.encyclopedia.com/social-sciences-and-law/economics-business-and-labor/businesses-and-occupations/rainforest-cafe-inc - SEC EDGAR: Rainforest Cafe Inc., Schedule TO-T tender offer by Landry's, 2000: https://www.sec.gov/Archives/edgar/data/0000924919/000095013000005261/0002.txt - Chowhound: Where Can You Still Find Rainforest Cafes in the US?: https://www.chowhound.com/1832653/how-many-rainforest-cafes/ - Fast Company: Social media is falling in love with Rainforest Cafe all over again: https://www.fastcompany.com/91239649/social-media-falling-in-love-with-rainforest-cafe-all-over-again --- # Who Owns Kinko's Now? FedEx Retired the Brand URL: https://404memoryfound.com/posts/is-kinkos-still-around.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-16 Topics: Business Blunders, Then vs Now **Summary:** Kinko's no longer exists as a store name in the United States. FedEx bought the 1,200-store copy chain for $2.4 billion in a deal announced on December 30, 2003, renamed it FedEx Office on June 2, 2008, and booked a charge of approximately $891 million largely to write off the Kinko's trade name. The only business still trading under the name is Kinko's Japan, which Konica Minolta bought from FedEx in 2012. **Key facts:** - Founded: 1970 in Isla Vista, California, by Paul Orfalea - Peak size: About 1,200 locations worldwide and roughly $2 billion in annual revenue in 2003 - Sold to FedEx: $2.4 billion in cash, announced December 30, 2003 - Name retired: Announced June 2, 2008, with a charge of about $891 million - Status today: No US Kinko's; FedEx Office runs roughly 2,000 US stores. Owner today: FedEx, while Konica Minolta owns Kinko's Japan ## What was Kinko's before FedEx bought it? Paul Orfalea opened the first Kinko's in 1970 in Isla Vista, California, a few steps from the University of California, Santa Barbara. He had a $5,000 Bank of America loan cosigned by his father, a rented Xerox copier, and a former hamburger stand to put them in. Copies cost four cents a page. When the room filled up, he wheeled the copier out onto the sidewalk. Accounts of the space disagree: the FundingUniverse company history calls it 80 square feet, while other retellings of the story say 100. The name came from his hair. Friends called Orfalea "Kinko" because of his red curls, and that is what went on the sign. What grew out of that shop was strange for a national chain. Into the 1990s Kinko's was not one company but a federation of more than 120 separate partnerships, each with its own owners and its own way of doing things. The New York private equity firm Clayton, Dubilier & Rice bought into the business in 1996, and the partnerships were rolled up into a single corporation with centralized management. That tidying up is what made Kinko's sellable. By the time FedEx came calling, Clayton, Dubilier & Rice controlled the chain and the head office had moved from the California coast to Dallas. ## Why did FedEx pay $2.4 billion for a copy chain? FedEx announced the purchase on December 30, 2003: $2.4 billion in cash for roughly 1,200 Kinko's locations worldwide and estimated annual revenue of about $2 billion for the year ending December 31, 2003. The number that mattered was smaller. More than 400 of those 1,200 stores were open 24 hours a day, seven days a week. FedEx had drop boxes and a fleet. It did not have 400 staffed rooms with the lights on at 2 a.m., and it could not build them quickly. Retail shelf space was also the thing FedEx's rival already had. United Parcel Service had converted a franchise chain into The UPS Store, which put a manned counter in strip malls across the country. Kinko's was the equivalent answer sitting on the market. "The FedEx and Kinko's combination will substantially increase our retail presence worldwide and will enable both companies to take advantage of growth opportunities in the fast-moving digital economy," Frederick W. Smith, then chairman, president and chief executive officer of FedEx, said in the December 30, 2003 announcement. The deal closed in the first quarter of 2004, and the chain went in as an operating company of its own inside FedEx rather than as a bolt-on to the delivery business. ## When did the Kinko's name come off the stores? In stages, and then all at once. For the first four years FedEx put its own name first and kept the old one in second place, so the stores traded as FedEx Kinko's. On June 2, 2008, FedEx announced that FedEx Kinko's would become FedEx Office and that retail locations would be rebranded over the following several years. The same announcement carried the bill: a charge of approximately $891 million, or $696 million net of tax and $2.22 per diluted share, in the fiscal fourth quarter that ended May 31, 2008. Most of it was a one-time, non-cash impairment tied to the decision about using the Kinko's trade name and to the goodwill from the acquisition. Brian D. Philips, who had been named president and chief executive officer of FedEx Office the month before, gave the reasoning in the release: "Kinko's was primarily a copy and print-service provider when it was acquired in 2004." The new name, he said, "more accurately represents our broader role of providing superior information and services through our company-owned, digitally connected locations around the world." Put the two announcements side by side and the arithmetic is blunt. FedEx paid $2.4 billion in 2004 for a business whose best-known asset was a word, and in 2008 it wrote down close to a billion dollars for the right to stop saying it. ## Can you still walk into a store called Kinko's? Not in the United States. The rebranding that began in 2008 took the name off American storefronts, and FedEx Office is what stands there now: roughly 2,000 stores in the United States, run as an operating company of FedEx Corporation from a headquarters in Plano, Texas. The name survived somewhere else. Kinko's had a Japanese arm, and in May 2012 Konica Minolta agreed to buy FedEx Kinko's Japan and its 61 locations from FedEx. Konica Minolta renamed the business Kinko's Japan, and the Kinko's name stayed on the doors there. So the answer to who owns Kinko's now comes in two parts. The company that was Kinko's is FedEx Office, owned by FedEx and no longer using the word. The only business still trading as Kinko's is Kinko's Japan, owned by Konica Minolta. That is an unusual exit for a retail name. When Woolworth disappeared from American main streets (https://404memoryfound.com/posts/is-woolworth-still-in-business.html), the company survived by leaning on a different brand it already owned. Kinko's went the other way: the buyer kept every store and threw away the word on the sign. ## What replaced the all-night copy shop? Kinko's did not lose to a competitor. It lost to the job going away. In 1995, a 40-page proposal that had to be in a client's hands in another city on Tuesday meant a physical original, a copier and a counter. Three things removed all of that. Email attachments and the PDF made duplication unnecessary. Cheap home laser and inkjet printers put a machine on every desk. And when a file was too fat to email, people carried it, which is how the era's other dead technologies got involved: Zip disks (https://404memoryfound.com/posts/what-happened-to-iomega-zip-drives.html) and blank CD-Rs did the courier work for a few years before broadband made even that pointless. What is left at the counter is the errand FedEx wanted in the first place. FedEx Office still prints signs, banners and business documents, but the queue is mostly people handing over boxes and printing shipping labels. It is the same slow squeeze that emptied out RadioShack (https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html): a store built around a task that the internet quietly took over, still standing after the reason for the visit was gone. ## Frequently Asked Questions ### Is Kinko's still around in 2026? Not in the United States. Kinko's, founded in 1970 in Isla Vista, California, was bought by FedEx in a $2.4 billion deal announced on December 30, 2003, and FedEx announced on June 2, 2008 that the stores would be renamed FedEx Office. The only business still trading under the Kinko's name in 2026 is Kinko's Japan, which Konica Minolta agreed to buy from FedEx in May 2012. ### Did Kinko's become FedEx Office? Yes. The Kinko's chain FedEx acquired in 2004 traded as FedEx Kinko's until June 2, 2008, when FedEx announced it was renaming the business FedEx Office and taking a charge of approximately $891 million, most of it a non-cash write-down of the Kinko's trade name and the goodwill from the purchase. FedEx Office operates roughly 2,000 stores in the United States today. ### Are FedEx Office stores open 24 hours the way Kinko's was? Mostly not. When FedEx announced the Kinko's purchase in December 2003, more than 400 of the chain's roughly 1,200 locations worldwide were open 24 hours a day, seven days a week. FedEx Office does not run its network that way in 2026: most stores keep daytime and evening hours, and only a small number list round-the-clock service on FedEx's own store locator. **Sources:** - FedEx to Acquire Kinko's, press release of December 30, 2003 (SEC 8-K exhibit): https://www.sec.gov/Archives/edgar/data/0000230211/000110465903029453/a03-6596_1ex99d1.htm - FedEx Changes Name of FedEx Kinko's to FedEx Office, press release of June 2, 2008 (SEC 8-K exhibit): https://www.sec.gov/Archives/edgar/data/0001048911/000136231008003100/c73562exv99w1.htm - History of Kinko's Inc., FundingUniverse company history: https://www.fundinguniverse.com/company-histories/kinko-s-inc-history/ - Company Structure and Facts, FedEx: https://www.fedex.com/en-us/about/company-structure.html - Konica Minolta to Acquire FedEx Kinko's Japan, press release of May 10, 2012: https://www.konicaminolta.com/about/releases/2012/0510_01_01.html --- # Is Toonami Still On TV in 2026? Yes, on Adult Swim URL: https://404memoryfound.com/posts/is-toonami-still-on-tv.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-16 Topics: Music & Entertainment, Internet Culture **Summary:** Toonami is still on TV in 2026. The block airs on Adult Swim every Saturday night from 12:00 a.m. to 4:00 a.m. Eastern, hosted by TOM and extended by half an hour on January 3, 2026. Cartoon Network canceled the original Toonami on September 20, 2008, and an April Fools prank in 2012 brought it back. **Key facts:** - Launched: March 17, 1997, on Cartoon Network - Canceled: September 20, 2008, after 11 years - Revived: May 26, 2012, on Adult Swim - Status today: On the air, Saturdays 12:00 a.m. to 4:00 a.m. ET on Adult Swim - Owner today: Warner Bros. Discovery, pending sale to Paramount Skydance ## How a 4 p.m. cartoon block turned America on to anime Toonami went on the air on Cartoon Network on March 17, 1997, as a weekday afternoon block running from 4 p.m. to 6 p.m. Eastern. It was built by Sean Akins and Jason DeMarco under Cartoon Network programming chief Mike Lazzo, and it was assembled on almost nothing. "There was literally no money to do anything," Akins told Complex in 2017 for the block's 20th anniversary oral history. The first host was Moltar, the lava-headed character borrowed from Space Ghost Coast to Coast and voiced by C. Martin Croker. He introduced ThunderCats, Voltron and The Real Adventures of Jonny Quest alongside the shows that mattered more in hindsight: Sailor Moon, Dragon Ball Z and Mobile Suit Gundam Wing. For a lot of American kids, that 4 p.m. slot was the first place they saw Japanese animation with its serialized plots left intact. On July 11, 1999, Moltar was replaced by TOM, the Toonami Operations Module, a small robot voiced first by Sonny Strait and then, from September 2000 onward, by Steve Blum. The block moved to Saturdays, grew its own science-fiction storylines, and ran for 11 years. ## Why Cartoon Network pulled the plug in 2008 The final Cartoon Network broadcast aired on Saturday, September 20, 2008. The network said the block was no longer pulling the ratings it needed, and named Adult Swim its main home for anime that same night. The decision came from Stuart Snyder, who ran Cartoon Network from 2007 to 2014. In a radio interview surfaced by fans and reported by CBR in 2026, Snyder said "the Toonami block on Cartoon Network just was not drawing a large enough audience," and described a network trying to rebuild itself around new shows for children. By 2008 the schedule leaned heavily on Naruto reruns, which made the Toonami branding hard to justify against airtime the network wanted for originals. Snyder has said it was a hard call precisely because the fanbase was so loud. TOM's last words on Cartoon Network ended with "stay gold," a nod to Cowboy Bebop's Spike Spiegel, a role Blum also voiced. Viewers treated the sign-off as a funeral. It turned out not to be one. ## The April Fools joke that brought Toonami back On April 1, 2012, Adult Swim started its usual April Fools stunt by playing The Room. Twenty seconds in, the feed cut to Toonami, with TOM noting the date before introducing that week's episode of Bleach. The rest of the night ran Dragon Ball Z, Mobile Suit Gundam Wing, Outlaw Star, The Big O, Yu Yu Hakusho, Tenchi Muyo!, Trigun, Blue Submarine No. 6, Astro Boy and Gigantor. The idea came from Lazzo. DeMarco told Anime News Network in 2022 that Lazzo asked, "what if we just brought Toonami back for a night," during a brainstorm about April Fools, and that he and Gill Austin went off to work out how to do it with no time and no budget. Adult Swim then asked viewers on Twitter whether they wanted it back, under the hashtag #BringBackToonami. They did. On May 16, 2012, Adult Swim announced the block was returning, and on May 26, 2012, Toonami came back as a weekly Saturday late-night block, opening with Bleach at midnight and the United States premiere of Deadman Wonderland at 12:30 a.m. ## What time does Toonami air in 2026? Toonami airs on Adult Swim on Saturday nights, from 12:00 a.m. to 4:00 a.m. Eastern. In practice that is the early hours of Sunday morning, which is part of why the question keeps getting asked: almost nobody runs into it by accident. The block got longer at the start of 2026. On January 3, 2026, Toonami extended its end time to 4:00 a.m. and reshuffled the lineup, dropping Naruto and opening with Bleach: Thousand-Year Blood War, then Blue Exorcist, Mashle: Magic and Muscles, One Piece, Dan Da Dan, Dragon Ball DAIMA, Zom 100 and Rick and Morty: The Anime. It has kept moving since. My Adventures with Superman season three premiered on the block on June 13, 2026, after a full-season marathon the week before, and Scavengers Reign joined the schedule in September 2026. One thing that trips people up: Toonami is not part of Cartoon Network's own schedule anymore. Adult Swim takes over the same cable channel overnight, so the dial position has not changed, but the block belongs to Adult Swim. Without cable, Adult Swim programming streams on HBO Max. ## Who owns Toonami now, and what the Paramount deal changes Toonami belongs to Warner Bros. Discovery, through Adult Swim and Cartoon Network. That ownership is about to change hands. Paramount Skydance agreed to buy Warner Bros. Discovery at $31.00 per share in cash, valuing the company at roughly $110.9 billion, and Warner Bros. Discovery shareholders approved the transaction on April 23, 2026. The companies expect it to close in the third quarter of 2026, subject to regulatory clearance. Under the terms, Paramount takes the Turner entertainment channels, Cartoon Network and Adult Swim among them. The linear business those channels sit in has shrunk badly. Combined annual advertising revenue for Cartoon Network and Adult Swim fell from $668.3 million in 2014 to $133.7 million in 2024, a drop of about 83 percent, according to S&P Global Market Intelligence figures reported by Bloomberg Businessweek. Cable built around young men has been dying for a decade, which is what finished G4 and its two attempts at a gamer channel (https://404memoryfound.com/posts/g4-tv-failed-twice.html), and it is the same arithmetic that turned MTV into a network that mostly does not play music (https://404memoryfound.com/posts/when-mtv-actually-played-music-rise-fall.html). ## Where to find the old Toonami broadcasts today The 1997 to 2008 block was never released commercially. The shows it carried have been licensed and relicensed many times over, but the parts people actually miss, the bumpers, the TOM segments and the Total Immersion Events, survive mostly as off-air recordings. A lot of those recordings sit on VHS tapes in American basements, taped by kids who left the VCR running on a weekday afternoon. Magnetic tape loses quality every year it sits there. The cheap route is to do it yourself with a working VCR and a USB capture dongle, which is the approach laid out in our guide to digitizing VHS tapes in 2026 (https://404memoryfound.com/posts/how-to-digitize-vhs-tapes-2026.html). The mail-in route costs more and takes weeks: Legacybox (https://legacybox.com/pages/pricing) lists tape digitizing from about $9 a tape in 2026, with kits priced by how many items you send in one box. Fan uploads of the April 1, 2012 broadcast are archived on the Internet Archive, which is the closest thing to a clean copy of the night that restarted everything. ## Frequently Asked Questions ### Is Toonami coming back to Cartoon Network? There is no announced plan to move Toonami back to Cartoon Network. The block left Cartoon Network's schedule on September 20, 2008, and has aired on Adult Swim since May 26, 2012. A short revival called Toonami Rewind ran Friday afternoons from May 31, 2024 until December 27, 2024, when Adult Swim ended it to make room for Family Guy. ### Is Cartoon Network shutting down in 2026? Cartoon Network has not shut down as of September 2026, but it is a far smaller business than it was. Advertising revenue for Cartoon Network and Adult Swim combined fell from $668.3 million in 2014 to $133.7 million in 2024. Warner Bros. Discovery shareholders approved a sale to Paramount Skydance on April 23, 2026, which would put Cartoon Network and Adult Swim under a new owner. ### Does TOM still host Toonami? Yes. TOM, the Toonami Operations Module, still hosts the block on Adult Swim in 2026, voiced by Steve Blum, who has played the character since September 2000. TOM first replaced the original Toonami host Moltar on July 11, 1999, and the robot has been redesigned several times since. **Sources:** - Complex: The Oral History of Cartoon Network's Toonami: https://www.complex.com/pop-culture/a/elijah-watson/oral-history-of-toonami - CBR: Why Cartoon Network Canceled Toonami Is Finally Revealed: https://www.cbr.com/naruto-reruns-cartoon-network-cancels-toonami/ - Anime News Network: How an April Fools Day Joke Revived Toonami: https://www.animenewsnetwork.com/feature/2022-04-01/how-an-april-fools-day-joke-revived-toonami/.184015 - Cartoon Brew: The Future of Cartoon Network Looks Very Bleak: https://www.cartoonbrew.com/business/the-future-of-cartoon-network-looks-very-bleak-247237.html - Warner Bros. Discovery: Shareholder Meeting of April 23, 2026 on the Paramount Skydance Transaction: https://www.wbd.com/news/warner-bros-discovery-sets-shareholder-meeting-date-april-23-2026-approve-transaction --- # Why G4 TV Failed Twice: The 1,000-Viewer Collapse URL: https://404memoryfound.com/posts/g4-tv-failed-twice.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-15 Topics: Gaming, Business Blunders, Music & Entertainment **Summary:** G4 TV failed twice. The gaming channel launched on April 24, 2002, went dark on December 31, 2014, returned to cable on November 16, 2021, and was shut down again on October 16, 2022 after Nielsen measured its average audience at 1,000 viewers. G4 has been off the air ever since, and the brand still belongs to Comcast Spectacor. **Key facts:** - Launched: April 24, 2002, as a Comcast cable venture - Merged with TechTV: May 28, 2004, after Comcast paid about $300 million for it - First shutdown: December 31, 2014 - Second shutdown: October 16, 2022, with 45 jobs cut - Status today: Off the air; the G4 brand is owned by Comcast Spectacor ## How a week of Pong put a gaming channel on cable G4 went on the air on April 24, 2002, as a Comcast venture built on a simple bet: that video games were big enough to carry a cable channel of their own. The soft launch was a week of uninterrupted Pong. Comcast aimed the channel squarely at viewers aged 12 to 34, the group advertisers chased hardest and basic cable reached worst. G4 was not the first to try it. TechTV (https://404memoryfound.com/posts/what-happened-to-techtv-screen-savers-channel.html), the San Francisco channel owned by Paul Allen's Vulcan, had been covering computers and games since 1998, when it launched as ZDTV. Two channels chasing the same small audience was one channel too many, and both of them knew it. TechTV had the credibility and the hosts. G4 had a cable operator with 20 million subscribers behind it and no particular reason to lose. What G4 did not have was a second source of revenue. Cable networks live on carriage fees paid per subscriber per month, and a niche channel only collects them if operators agree to carry it. Everything that happened to G4 over the next twenty years traces back to that arithmetic. ## What Comcast bought when it paid $300 million for TechTV Comcast settled the standoff by buying the competition. The deal closed on May 10, 2004, when Comcast acquired the outstanding stock of TechTV "from Vulcan Programming, Inc. for approximately $300 million in cash," according to the company's own announcement that day. On May 28, 2004, the two channels began broadcasting as one network under the name G4techTV. The merger was not a merger in any meaningful sense. TechTV's San Francisco operation was shut down and production moved to Los Angeles, and most of the TechTV staff did not make the trip. The hybrid name lasted less than a year. On February 15, 2005, Comcast dropped TechTV from the title and the channel went back to being G4. What survived the purchase was the programming block that defined the network for a decade: X-Play, the game review show, and Attack of the Show!, the daily internet-culture hour. Both were TechTV inheritances. Neither was something Comcast had invented, and both would outlive the channel's original purpose. ## What killed G4 the first time, in 2014 By 2012 the channel had stopped being about games. Comcast planned to rebrand G4 as the Esquire Network under a licensing deal with Hearst, a pivot from controllers to cocktails aimed at a richer audience. G4's studio programming was cancelled to make room for it. X-Play and Attack of the Show! aired their final original episodes in January 2013, and the network filled its schedule with reruns of Cops and Cheaters. Then the rebrand moved. Esquire Network launched on September 23, 2013, on the Style Network's slot instead of G4's, because G4's ratings were too low to be worth converting. That left a channel with no original shows, no plan, and no new name. Comcast dropped G4 from its own cable systems during 2014. The last carriage agreements, with providers including AT&T U-verse and Google Fiber, expired at midnight on December 31, 2014. The final program was an episode of X-Play, and the screen faded out on a game of Pong shrinking to a dot. It was the same ending MTV (https://404memoryfound.com/posts/when-mtv-actually-played-music-rise-fall.html) avoided by abandoning its founding format. G4 tried the same move and the slot went to somebody else. ## Why the 2021 relaunch lasted eleven months The comeback was announced on July 24, 2020, at Comic-Con@Home, with the brand moved from NBCUniversal to Comcast's sports and esports arm, Comcast Spectacor. G4 spent a year as a YouTube and Twitch operation before returning to television. The linear relaunch came on November 16, 2021, on Xfinity TV, Verizon Fios, Cox and the streaming service Philo, alongside a multiyear partnership with Twitch. X-Play and Attack of the Show! came back with them. Russell Arons, a former Mattel and Electronic Arts executive, had been named president that September. The audience did not follow. Nielsen measured G4 at an average of 1,000 viewers in 2022, in primetime and across the full day alike, which made it the least-watched network Nielsen rated that year. Arons left at the end of August 2022. Layoffs followed in September, falling heavily on the X-Play crew, and hosts started leaving within days of each other. On October 16, 2022, Comcast Spectacor chairman and chief executive Dave Scott sent a memo ending the whole thing. "Over the past several months, we worked hard to generate that interest in G4, but viewership is low and the network has not achieved sustainable financial results," Scott wrote, in a memo reported by Variety. Forty-five people lost their jobs, and the shutdown was effective immediately. ## Is G4 coming back, and who owns the name in 2026? G4 has been off the air since October 16, 2022. There is no channel, no streaming feed, and no announced plan to bring either back. The brand sits with Comcast Spectacor, which has said nothing about it since the memo. The reason it is unlikely to return a third time is that the job G4 invented is now done for free. Twitch and YouTube carry live game coverage, reviews and personality-driven shows around the clock, without carriage negotiations or a Nielsen panel deciding whether the audience counts. That is also what makes the 2022 number so brutal. A thousand viewers is a small Twitch channel on a slow afternoon. G4 was paying for studios, crew, distribution and on-air talent to reach it. The rest of the gaming press consolidated the same way. Game Informer (https://404memoryfound.com/posts/who-owns-game-informer-now.html) was shut down by GameStop (https://404memoryfound.com/posts/is-gamestop-still-in-business.html) in 2024 and revived under new ownership, and the outlets that survived did it by getting smaller and cheaper, not by buying airtime. What is left of G4 is an archive. Fan-run collections on the Internet Archive hold recordings from both eras of the network, which is more than the company itself has kept online. ## Frequently Asked Questions ### Is G4 TV still on the air in 2026? No. G4 TV has been off the air since October 16, 2022, when Comcast Spectacor shut the gaming network down for the second time. There is no G4 cable channel or streaming feed in 2026, and no announced plan to relaunch it. ### Why did G4 TV shut down in 2022? G4 TV was shut down on October 16, 2022 because almost nobody watched it. Nielsen put the network's average audience at 1,000 viewers for 2022, the lowest of any network it measured, and Comcast Spectacor cut 45 jobs when it closed the channel. ### When did G4 TV go off the air the first time? G4 TV went dark the first time at the end of December 31, 2014, after Comcast dropped the channel from its own systems and the remaining carriage deals expired. The network had launched on April 24, 2002 and stopped making original shows in January 2013. **Sources:** - Variety: G4 TV Shuts Down, Resulting in Layoff of 45 Employees: https://variety.com/2022/digital/news/g4-tv-shuts-down-layoffs-comcast-1235405384/ - The Hollywood Reporter: Comcast Shutters Gaming Network G4 TV: https://www.hollywoodreporter.com/business/business-news/comcast-shutters-gaming-network-g4-tv-1235242764/ - IndieWire: Why Did the G4 Channel Close? Well, It Averaged 1,000 Viewers: https://www.indiewire.com/features/general/g4-canceled-ratings-1234773398/ - Comcast Corporation: Comcast Completes Acquisition of TechTV: https://corporate.comcast.com/news-information/news-feed/comcast-completes-acquisition-of-techtv - Wikipedia: G4 (American TV network): https://en.wikipedia.org/wiki/G4_(American_TV_network) --- # Is Photobucket Still Around? Why Your Photos Vanished URL: https://404memoryfound.com/posts/is-photobucket-still-around.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-15 Topics: Internet Culture, Business Blunders **Summary:** Photobucket is still around in 2026, but not as the free image host that served a large share of the MySpace-era web. It sells subscription storage starting at $5 a month for one terabyte, and it has had no free tier since June 2023. The billions of broken images on forums date to June 20, 2017, when Photobucket began charging $399.99 a year for third-party hosting. **Key facts:** - Launched: 2003, by Alex Welch and Darren Crystal - Peak sale: July 2007, to Fox Interactive Media for $250 million plus up to $50 million in performance payments - What broke the links: June 20, 2017 terms change: third-party hosting moved to a $399.99-a-year plan - Status today: Alive, subscription only. No free accounts since June 2023 - Owner today: Photobucket Inc., privately held, Denver, Colorado ## What broke every Photobucket link in 2017? On June 20, 2017, Photobucket quietly updated its terms of service and stopped serving images to other websites. Anyone who had pasted a Photobucket link into a forum thread, a fan page or a marketplace listing watched the photo turn into a gray box asking for money. The company had folded third-party hosting, the feature that let an image live on Photobucket but appear somewhere else, into a single expensive plan. It was called Plus 500, and it cost $399.99 a year. Almost nobody was warned. Millions of posts across hobbyist boards, motorcycle forums, quilting communities and shopping listings broke in the same afternoon, and the graphic that replaced each picture told readers the account holder needed to upgrade. Photobucket chief executive John Corpus defended the decision in a 2017 statement, blaming "the rise of ad blockers and the Company's explosion of 3rd party hosting that generates zero revenue." He said 75 percent of the company's costs came from non-paying users whose images were being loaded elsewhere. "This model is no longer sustainable," Corpus wrote, in remarks 9News reported that July. ## How Photobucket became the photo album of the MySpace era Photobucket launched in 2003, built by Alex Welch and Darren Crystal, and it answered a question the early web had no good response to: where does a picture live when the site you want to post it on will not store it for you? That question got much louder once MySpace (https://404memoryfound.com/posts/rise-and-fall-of-myspace.html) arrived. MySpace profiles were endlessly customizable and came with almost no storage, so the glitter graphics, layout backgrounds and profile banners of 2005 and 2006 were served from somewhere else. Photobucket became that somewhere else for a very large share of them. By 2007 the site had roughly 40 million registered users. Fox Interactive Media, the News Corp division that also owned MySpace, bought Photobucket in July 2007 for $250 million in cash plus performance payments of up to $50 million. The fit lasted about two years. News Corp sold Photobucket in 2009 to Ontela, a Seattle mobile imaging startup, in a deal that valued the site at about $60 million, roughly a fifth of what News Corp had paid. Ontela took the Photobucket name and has run it independently ever since. ## Why free accounts disappeared in 2023 The 2017 price did not hold. In May 2018 a new management team cut the hosting plan to $1.99 a month, or $19.99 a year, and switched the blocked images back on. The reprieve did not fix the underlying math. Storing billions of files costs money every month, and an image loaded inside somebody else's forum thread shows none of Photobucket's advertising. In June 2023 Photobucket ended free accounts altogether. Free users received a message telling them their account had been deactivated because the company no longer supported free subscriptions. The photos were not deleted. The door was simply locked. The effect on the open web was permanent. Every thread whose author had stopped logging in years earlier stayed broken, and nothing is coming along to fix it. The habit had moved on anyway: free rivals such as Imgur absorbed the forum-posting crowd, Flickr (https://404memoryfound.com/posts/what-happened-to-flickr-yahoo-photo-sharing.html) kept the photographers, and phone backup services took everyone else. ## What does Photobucket cost in 2026, and who owns it? Photobucket is still running in 2026. It is a subscription storage company now rather than a free image host, and there is no free tier to sign up for. The entry plan, My Bucket, is $5 a month for one terabyte of photo and video storage. The plan that includes third-party hosting, the feature that made Photobucket essential and then notorious, runs $13 a month. Ownership has not changed in more than fifteen years. Photobucket Inc. is the former Ontela, privately held and based in Denver, Colorado. There is no parent conglomerate behind it and no larger company covering the storage bill. So the short answer to the question people type is that the company exists and the service you remember does not. The free image host that carried a chunk of the 2005 web is gone, and the pictures from lapsed accounts sit on Photobucket's servers behind a subscription. ## Can you get your old Photobucket photos back? Usually yes, if you can still get into the account. Deactivated accounts were locked rather than erased, so the files are generally still there, and starting a paid plan restores access long enough to download the albums. The obstacles are practical ones. Plenty of accounts were opened with a college address that stopped working in 2008, or under a username nobody wrote down, and password recovery through a dead mailbox goes nowhere. When the account itself is unreachable, the Internet Archive's Wayback Machine is the next stop. Public forum threads were often crawled while the images still loaded, so an archived copy of the page can hold a picture the original poster lost. If the originals only ever existed as prints or negatives in a shoebox, that is a different job. Mail-in services such as ScanMyPhotos (https://www.scanmyphotos.com/) handle a box at a time, and scanning them yourself (https://404memoryfound.com/posts/scan-old-photos-negatives-2026.html) costs less if you have the hours to spare. ## Why Photobucket wants to sell 13 billion photos to AI companies In April 2024 Photobucket chief executive Ted Leonard told Reuters the company was in talks with multiple technology companies about licensing its archive of roughly 13 billion photos and videos to train image-generating models. He put the rates under discussion at 5 cents to $1 per photo, and more than $1 per video. "We need to pay our bills, and this could give us the ability to continue to support free accounts," Leonard told Reuters in 2024, in comments PetaPixel reported that April. Users had not clearly agreed to any of it. Photobucket revised its terms and gave account holders a 45-day window to opt out, after which consent was treated as given. In December 2024 a proposed class action, Pierce v. Photobucket, was filed in federal court in Colorado over biometric privacy and the licensing plan. On March 11, 2026, the court allowed most of the case to go forward, holding that leaving an old account sitting untouched is not the same as agreeing to new terms. ## Frequently Asked Questions ### Is Photobucket still around in 2026? Yes. Photobucket still operates in 2026 as a paid photo and video storage service run by Photobucket Inc. of Denver, Colorado, with plans starting at $5 a month for one terabyte. What no longer exists is the free image host people used between 2003 and 2023, and free accounts have not been available since June 2023. ### Why are my old Photobucket images broken on forums? Photobucket stopped serving images to other websites on June 20, 2017 unless the account paid $399.99 a year for third-party hosting, and it deactivated all remaining free accounts in June 2023. Any Photobucket picture embedded in a forum post by an account that has since lapsed will stay broken, because only an active paid subscription makes those links load again. ### Can I get my old Photobucket photos back? Probably, if you can still sign in. Photobucket locked deactivated free accounts in 2023 rather than deleting them, so subscribing to a paid plan usually restores access and lets you download the albums. If the account email is dead, the Internet Archive's Wayback Machine sometimes holds copies of forum pages captured before the images stopped loading in 2017. **Sources:** - Photobucket (company history): https://en.wikipedia.org/wiki/Photobucket - 9News: Free service is unsustainable in an age of online ad blockers, Photobucket CEO says (2017): https://www.9news.com/mobile/article/mobile/money/business/free-service-is-unsustainable-in-an-age-of-online-ad-blockers-photobucket-ceo-says/454952304 - PetaPixel: Photobucket is in Negotiations With AI Companies to License 13 Billion Images (April 2024): https://petapixel.com/2024/04/08/photobucket-is-in-negotiations-with-ai-companies-to-licence-13-billion-images/ - Olshan: Colorado Federal Court Allows Privacy and AI Training Claims to Proceed Against Photobucket (2026): https://www.olshanlaw.com/newsroom/alerts/client-alert-colorado-federal-court-allows-privacy-and-ai-training-claims-to-proceed-against-photobucket - Photobucket storage plans (official): https://my.photobucket.com/storageplan-ppc-nb --- # Does Yahoo Messenger Still Exist? Why It Shut Down URL: https://404memoryfound.com/posts/does-yahoo-messenger-still-exist.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-15 Topics: Software & Apps, Internet Culture **Summary:** Yahoo Messenger no longer exists. Oath, the Verizon division that then owned Yahoo, shut the service down on July 17, 2018, a little over 20 years after it launched as Yahoo Pager in March 1998. Yahoo has run no instant messaging product since its replacement, Yahoo Together, closed on April 4, 2019, and the company itself has belonged to Apollo Global Management since September 2021. **Key facts:** - Launched: March 9, 1998, by Yahoo, as Yahoo Pager - Renamed: Yahoo Messenger in 1999, when voice chat arrived - Shut down: July 17, 2018, announced by Oath on June 8, 2018 - What replaced it: Squirrel, renamed Yahoo Together, closed April 4, 2019 - Status today: Discontinued; owner today is Apollo Global Management ## The chat app that started life as a pager Yahoo Messenger did not launch under that name. It arrived on March 9, 1998 as Yahoo Pager, a small Windows program whose entire job was to tell you when a friend signed on and when new Yahoo Mail landed. The first release shipped with three status settings: available, busy and on vacation. That was the product. Yahoo renamed it Yahoo Messenger in 1999, the year it added voice chat, once it became obvious that naming a piece of internet software after pager hardware would age badly. The rename stuck for the next 19 years. The timing helped. AOL Instant Messenger (https://404memoryfound.com/posts/what-happened-to-aim-aol-instant-messenger.html) had launched in May 1997 and owned the American teenage buddy list, but Yahoo already ran mail, search, finance, games and fantasy sports, and it wired the chat client into all of them. Yahoo Messenger was also the front door to Yahoo Chat, the public topic rooms that later caused the company most of its trouble. ## What Yahoo Messenger had that AIM did not Yahoo competed on personality rather than reach. Version 6.0, released in May 2004, pulled Yahoo Games, music, photos and Yahoo Search into the messaging window and introduced Yahoo Avatars, the cartoon stand-ins users dressed up and kept on their profile. The same era brought Audibles, short voice animations you fired at someone mid-conversation, and IMVironments, themed chat backgrounds that could play clips behind a conversation. None of it made the service faster or more reliable. All of it made it feel personal. Then there was Buzz. Pressing it shook the recipient's window and played a sound, and it existed for exactly one purpose: getting the attention of someone who had stopped answering. MSN Messenger (https://404memoryfound.com/posts/what-happened-to-msn-messenger.html) had its own nudge, but Yahoo's is the one people still imitate out loud. Version 6.0 also added a stealth mode, which let a user appear offline to specific people on the list rather than to everyone. Twenty years before read receipts became a social problem, Yahoo Messenger had already shipped the tools for avoiding one. ## The 2005 chat rooms scandal that pushed advertisers out In May 2005 the Houston television station KPRC reported that Yahoo was carrying user-created chat rooms with names such as "Girls 13 And Under For Older Guys", some of them filed under the education category. The rooms opened straight from Yahoo Messenger, and national brand advertising was running against them. The advertisers did not wait. Pepsi, Georgia-Pacific, State Farm, Countrywide and T-Mobile all pulled advertising tied to the user-created rooms. "As soon as we found out we pulled our ads," Pepsi spokesman Dave DeCecco told the Associated Press in 2005, adding that the company kept advertising elsewhere on Yahoo. Georgia-Pacific spokeswoman Robin Keegan was blunter in the same report: "We were horrified to find out we were on those sites." In June 2005 Yahoo closed every user-created chat room and blocked users from making new ones. The rooms that remained ran under Yahoo's own categories and never regained the traffic. It was the first time Yahoo treated a piece of Yahoo Messenger as a liability instead of an asset, and it would not be the last. ## The 275 million user truce with Microsoft On October 12, 2005 Yahoo and Microsoft announced the first interoperability agreement between two major consumer instant messaging networks, letting Yahoo Messenger and MSN Messenger users add each other as contacts. Their joint announcement put the combined community at "more than 275 million strong", which would have made it the largest consumer instant messaging community in the world. The engineering took longer than the press release suggested. The two networks planned the link for the second quarter of 2006 and turned it on in July 2006. What the deal did not include was AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html). The two companies chasing AOL Instant Messenger connected to each other and left the market leader outside, which is a reasonable summary of how instant messaging worked before phones ate the category. The alliance outlived neither partner's product. Microsoft folded its messenger into Skype (https://404memoryfound.com/posts/skype-shut-down-what-replaced-it.html), and Yahoo spent the following decade rebuilding Yahoo Messenger instead of defending it. ## How Yahoo shut down Yahoo Messenger twice The first shutdown came on August 5, 2016, when Yahoo retired the legacy desktop client and the interface that third-party apps used to reach it. Anyone still running the old Windows program, or connecting through Pidgin or Trillian, was locked out that day. Yahoo had announced the replacement in December 2015: a rebuilt mobile and web app with unsend, likes and animated images, aimed at people who had already moved to WhatsApp and group texts. It did not bring them back. On June 8, 2018, Oath, the Verizon division that owned Yahoo at the time, announced the real ending. Yahoo Messenger would stop working on July 17, 2018, a little over 20 years after Yahoo Pager shipped. "We know we have many loyal fans who have used Yahoo Messenger since its beginning as one of the first chat apps of its kind", the company said in its announcement, adding that it was focusing on communication tools that were a better fit for users. The accompanying help pages were more honest: there was no replacement product, and users had six months to download their chat history before it went away. ## Is there any way to use Yahoo Messenger in 2026? Not through Yahoo. The company has not operated an instant messaging service since July 17, 2018, and nothing in the current Yahoo lineup replaces it. Yahoo did try, briefly. Squirrel, an invite-only group chat app, went into testing in May 2018 and was the only product Yahoo pointed departing Messenger users toward. It was renamed Yahoo Together in October 2018 and shut down on April 4, 2019, roughly six months after it launched publicly. Yahoo itself changed owners after that. Apollo Global Management completed its purchase of the company on September 1, 2021 in a deal valued at about $5 billion, with Verizon keeping a 10 percent stake, and Yahoo has run as a standalone business (https://404memoryfound.com/posts/who-owns-yahoo-now.html) under Apollo since. No version of Yahoo Messenger has returned. What still works is unofficial. Hobby projects run replacement servers that accept logins from a few of the old Yahoo Messenger builds, with PhoenixIM and Escargot the names that come up most in that community. You supply a decades-old installer, point it at a community server, and end up with a buddy list containing only other people doing the same thing. The export window closed in early 2019. Conversations that were never downloaded during those six months are gone, which is the part of this story that still annoys people who used Yahoo Messenger for a decade. ## Frequently Asked Questions ### When did Yahoo Messenger shut down? Yahoo Messenger shut down on July 17, 2018, a little over 20 years after it launched as Yahoo Pager on March 9, 1998. Oath, the Verizon division that owned Yahoo at the time, announced the closure on June 8, 2018 and gave users six months to download their chat history before the service went dark. ### What replaced Yahoo Messenger? Nothing from Yahoo. When Yahoo Messenger closed on July 17, 2018, Yahoo pointed users at Squirrel, an invite-only group chat app still in testing, which was renamed Yahoo Together in October 2018 and shut down on April 4, 2019. Yahoo has had no instant messaging product since. ### Can you still download Yahoo Messenger? No official Yahoo Messenger download has worked since July 17, 2018, and the legacy desktop client stopped connecting on August 5, 2016. Fan-run servers such as PhoenixIM and Escargot accept logins from a handful of old Yahoo Messenger builds, but Yahoo itself operates nothing. **Sources:** - What Ever Happened to Yahoo! Messenger? (TechSpot): https://www.techspot.com/article/2484-yahoo-messenger/ - Yahoo Messenger will shut down on July 17th (Engadget, June 8, 2018): https://www.engadget.com/2018-06-08-yahoo-messenger-discontinued-july-17th.html - Microsoft and Yahoo! Announce Landmark Interoperability Agreement (Microsoft, October 12, 2005): https://news.microsoft.com/source/2005/10/12/microsoft-and-yahoo-announce-landmark-interoperability-agreement-to-connect-consumer-instant-messaging-communities-globally/ - Yahoo shuts chat rooms amid child-sex fears (Associated Press via NBC News, 2005): https://www.nbcnews.com/id/wbna8334047 - Apollo Funds Complete Acquisition of Yahoo (Yahoo Inc., September 1, 2021): https://www.yahooinc.com/press/apollo-funds-complete-acquisition-of-yahoo --- # Who Owns Game Informer Now? Gunzilla Games Does URL: https://404memoryfound.com/posts/who-owns-game-informer-now.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-15 Topics: Gaming, Business Blunders **Summary:** Game Informer is owned by Gunzilla Games, the studio behind Off the Grid, which bought the magazine in March 2025 after GameStop shut it down and deleted its website in August 2024. The full staff came back, the archive was restored and expanded, and Game Informer is printing 10 issues a year in 2026. **Key facts:** - Launched: 1991, as a FuncoLand store handout - Shut down: August 2, 2024, by GameStop, after 33 years - Owner today: Gunzilla Games, since March 25, 2025 - Peak circulation: Near 8 million in 2011, third largest US magazine - Status today: Publishing, 10 print issues a year; issue 381 is the 35th anniversary ## How a FuncoLand giveaway became the biggest magazine in gaming Game Informer started in 1991 as an in-house handout from FuncoLand, a Minnesota chain that sold used cartridges. The first issue was thin, free and written to move product off the shelf behind the counter. The retail connection never went away, and it is the reason the numbers eventually got strange. GameStop inherited the magazine in 2000 when it absorbed FuncoLand's business, then folded a subscription into its paid loyalty tier. Every shopper who bought the upgraded membership became a subscriber. That turned a store newsletter into a mass-circulation title. By 2011 Game Informer's circulation sat near 8 million, which The Washington Post described as the third largest magazine in the nation. Its rivals never got that lifeline. Electronic Gaming Monthly and Nintendo Power both stopped printing in the early 2010s, and cable's games coverage had already thinned out after TechTV (https://404memoryfound.com/posts/what-happened-to-techtv-screen-savers-channel.html) disappeared. By 2024 Game Informer was the last monthly multi-platform games magazine still printing in the United States. ## Why did GameStop shut down Game Informer in 2024? On Friday, August 2, 2024, the staff were called into a meeting with a GameStop human resources executive and told the publication was closing, effective immediately. Everyone was laid off the same day. No real reason was given. The notice went up on Game Informer's own social accounts and opened with a line that has been quoted ever since: "it is with a heavy heart that we announce the closure of Game Informer," GameStop wrote in the August 2024 statement, after 33 years of publishing. The staff made clear the goodbye was not theirs. The wording read like something generated rather than written, and the people who had actually made the magazine said they had no hand in it. The business logic was a cost line. GameStop (https://404memoryfound.com/posts/is-gamestop-still-in-business.html) spent 2024 cutting everything that did not directly sell games, and a print magazine carrying paper, postage and a full editorial payroll was an easy target for a retailer shrinking its store count. The last issue published under GameStop put The Casting of Frank Stone on the cover. Its cover reveal had gone out days before the shutdown, so the magazine died mid-cycle rather than at a natural stopping point. ## What happened to 33 years of Game Informer articles? The website was not simply frozen. Every URL on gameinformer.com was redirected to the closure statement, so a link to a 2007 review, a studio profile or a cover story all returned the same farewell page. Forbes reported the deletion on August 3, 2024, one day after the announcement. Three decades of reporting on a secretive industry, including previews and developer interviews that exist nowhere else, went dark in an afternoon. That is the part of the story that reversed most completely. After the 2025 sale, the new owner restored the digital archive and then went past where it had been, putting every issue back to the first one online, rebuilt with help from the Video Game History Foundation, the scanning community at Retromags and individual collectors. The whole run from 1991 to 2024 is now readable free with a Game Informer account, and the Video Game History Foundation's digital library carries the magazine as well. The archive is in better shape than it was on the day GameStop owned it. ## What is Gunzilla Games, the studio that bought the magazine? Gunzilla Games is a European development studio best known for Off the Grid, a cyberpunk battle royale built around its own blockchain. Neill Blomkamp, who directed District 9, is a co-founder and the studio's chief creative officer. On March 25, 2025, close to eight months after the shutdown, Gunzilla Games announced it had acquired Game Informer and rehired the staff, from editorial through production, with the publication set up as its own company. Gunzilla Games chief executive Vlad Korolov framed the purchase as a rescue. The studio was proud to "join their fight to preserve the heart of video game journalism," he said in the March 2025 announcement. The obvious problem is that a game developer now owns a review outlet. Both sides answered that the same way in 2025: the editorial team decides what it covers and how, with no input from the owner. That is a promise rather than a structure, and readers have been checking it ever since. The parent company has had its own turbulence. Aftermath reported in 2026 that some Gunzilla Games developers said they had gone months without pay, while Game Informer said its own work and pay had continued without interruption. ## Is Game Informer still publishing in 2026? Yes, in print and online. The magazine returned to mailboxes in 2025 and now runs 10 print issues a year, with each one released digitally to subscribers on the day it launches. On August 24, 2026, Game Informer opened print subscriptions outside the United States for the first time since the relaunch, adding Canada, the United Kingdom and about 40 other countries. New international subscribers start with the Exodus issue. That issue, number 381, doubles as the 35th anniversary edition. It carries a 12-page retrospective of the magazine's covers with comments from former editors and from developers including Ed Boon, Ken Levine and Tony Hawk (https://404memoryfound.com/posts/what-happened-to-tony-hawks-pro-skater.html), plus a pack-in poster showing almost every cover in the run. So the status is unusual for anything in this archive. A print product from 1991 was killed, deleted from the web, and came back with a larger archive and wider distribution than it had on the day it stopped. ### Where to find one today Reading the old material costs nothing: every issue from 1991 through 2024 is free with a Game Informer account, and the Video Game History Foundation's digital library holds the same run. A current subscription is the only route to the new print issues, which arrive 10 times a year. For a physical back issue with the original ads, fold-outs and cover stock, collectors sell them singly and by the lot, usually for less than the price of a year's subscription, and Etsy's vintage Game Informer listings (https://www.etsy.com/search?q=game+informer+magazine+vintage) are the practical place to browse. ## Frequently Asked Questions ### Is Game Informer still in print in 2026? Yes. Game Informer returned to print in 2025 under Gunzilla Games and publishes 10 print issues a year alongside daily coverage on its website. In August 2026 it extended print subscriptions to Canada, the United Kingdom and about 40 other countries, and issue 381 marks the magazine's 35th anniversary. ### Why did GameStop shut down Game Informer? GameStop closed Game Informer on August 2, 2024 without naming a reason, laying off the entire staff in one meeting and ending a 33-year run that began in 1991. The retailer was cutting costs across its business that year, and the magazine's paper, postage and editorial payroll were expenses it chose to stop paying. ### Can you still read old Game Informer issues? Yes, and for free. After Gunzilla Games bought Game Informer in March 2025, the magazine restored and expanded its digital archive so every issue from 1991 to 2024 can be read with a free account. The Video Game History Foundation's digital library carries the same run of the magazine. **Sources:** - The Washington Post: Game Informer's shutdown ends an era of gaming's glossy prestige: https://www.washingtonpost.com/entertainment/video-games/2024/08/02/game-informer-shut-down-gamestop/ - Forbes: A Deleted Game Informer Is Now Redirecting To GameStop's Closure Statement: https://www.forbes.com/sites/paultassi/2024/08/03/a-deleted-game-informer-is-now-redirecting-to-gamestops-ai-written-statement/ - Gunzilla Games: Gunzilla Games Acquires Game Informer: https://gunzillagames.com/en/news/63-Gunzilla-Games-Acquires-Game-Informer - Aftermath: Gunzilla, The Company That Owns Game Informer, Has Reportedly Not Paid Some Developers For Months: https://aftermath.site/gunzilla-gunz-gun-off-the-grid-game-informer/ - Game Informer: International print subscriptions are now available: https://gameinformer.com/2026/08/24/game-informer-international-print-subscriptions-are-now-available-in-canada-united --- # How Many KB Toys Stores Are Left? Zero Since 2009 URL: https://404memoryfound.com/posts/kb-toys-stores-left.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-15 Topics: Business Blunders, Then vs Now **Summary:** No KB Toys stores are left. The mall toy chain shut its last 461 locations on February 9, 2009 after a second bankruptcy, and nothing has reopened under the name since. In 2026 KB Toys exists as a licensed brand selling collectibles online and at conventions, not as a retailer. **Key facts:** - Founded: 1922 in Pittsfield, Massachusetts, as Kaufman Brothers - Peak size: 1,324 stores in 1999, second only to Toys R Us - Stores left: Zero. The last 461 closed on February 9, 2009 - Owner today: Strategic Marks, LLC; licensing run by Firefly Brand Management since 2024 - Status today: Brand only: online collectibles and convention pop-ups, no stores ## Why a mall toy store could not beat the discounters KB Toys was built for a floor plan that stopped mattering. The stores were small, bright and stacked to the ceiling, wedged between a shoe shop and the food court, and they worked because the enclosed mall was where American families spent Saturday afternoon. The chain peaked in 1999 at 1,324 stores, second only to Toys R Us among American toy retailers. It had grown out of a Pittsfield, Massachusetts candy wholesaler founded by brothers Harry and Joseph Kaufman in 1922, which moved into toys after World War II and started selling at retail as Kay-Bee Toy and Hobby. The trouble was that the same box of toys sold for less somewhere else. Walmart and Target used toys to pull families through the door and priced them to do it, and a small mall shop paying mall rent could not answer that on a boxed action figure. So KB Toys multiplied formats instead: KB Toy Works in strip centers, KB Toy Outlet in outlet malls, and short-term KB Toys Holiday stores that opened for Christmas and vanished in January. Those four formats were still the whole company in its last week. ## What pushed KB Toys into Chapter 11 twice? Consolidated Stores, the parent of Big Lots, sold KB Toys in December 2000 to an investment group led by Bain Capital and the chain's own management for $305 million. Bain put up roughly $18 million of its own money, and the rest of the price was borrowed against the toy company. In 2002 KB Toys took on more debt to pay its owners an $85 million dividend. Creditors later sued, arguing that the payout helped tip an already weak retailer into insolvency, and the case eventually settled. The first bankruptcy came in January 2004, weeks after the parent of FAO Schwarz (https://404memoryfound.com/posts/fao-schwarz-store-rockefeller-center.html) filed its own. KB Toys closed hundreds of stores, came out of Chapter 11 in August 2005 under Prentice Capital Management, and went into the 2008 holiday season smaller and still carrying debt. It filed again on December 11, 2008, two weeks before Christmas, with sales down 20 percent in the month that was supposed to carry the year. At that point the Pittsfield company ran 277 mall stores, 114 outlets, 40 KB Toy Works and 30 holiday stores, 461 in all, and it was the largest mall-based toy retailer in the country. ## Who bought the KB Toys name after the liquidation? The 2008 filing turned into a wind-down almost immediately. Going-out-of-business sales ran through the winter, and the last KB Toys stores closed on February 9, 2009, the same day the company's website went dark. The trademark outlived the leases. Later in 2009 the KB Toys logos and web addresses were sold to CE Stores, a unit of Toys R Us, so the name ended up owned by the competitor that had outlasted it. CNBC covered that sale in September 2009 as part of a pattern rather than a one-off, grouping KB Toys with Sharper Image (https://404memoryfound.com/posts/who-owns-sharper-image-now.html) and Linens 'n Things as brands that kept trading as intellectual property after the stores were gone. Toys R Us never rebuilt KB Toys as a chain and eventually let the registration lapse. In 2016 Strategic Marks, a California company that buys and revives dead American brands, filed for the KB Toys trademark. Its founder, Ellia Kassoff, has been behind every KB Toys announcement since. ## What happened to the 1,000 pop-up stores promised in 2018? In March 2018, Toys R Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html) announced that it would liquidate its American stores, and within days Kassoff said KB Toys would move into the gap. The plan was as many as 1,000 temporary stores open for Black Friday, with the best performers converted into permanent leases. "I think we could build the concept out pretty quick, with help from the toy industry and manufacturers. We don't have that much time," Kassoff told The Boston Globe in March 2018. Not one of them opened. The relaunch slipped to 2019, was scaled back to a few hundred proposed stores, and then stopped, because the money never arrived. Kassoff later explained the failure to the Licensing Industry Merchandisers Association, saying that "toy companies had lots of conflicts of interest that prevented them from investing in KB given that they sell to other retailers," and that mall operators do not usually fund tenants who do not exist yet. Those 2018 headlines are still indexed, which is why Americans keep searching to find out whether the stores ever opened. ## Is there a KB Toys store you can walk into in 2026? No. There is no KB Toys store in any American mall, strip center or outlet, and there has not been one since February 2009. Nothing in the current plan is a chain. What exists instead is a licensing program. In November 2024, Firefly Brand Management announced a master toy deal for KB Toys with MH Enterprises, covering toys, apparel, paper goods, drinkware, comics and the right to use the KB Toys store design for live experiences. MH Enterprises is run by Marc and Marianna Heon. "Having met at KB Toys over 25 years ago, this brand has always held a special place in our hearts," the couple said in the 2024 announcement carried by License Global. In 2026 the brand is marking 80 years since the Kaufmans moved into toys, with collectible buckets, pins, trading cards and a first pop-up at San Diego Comic-Con rather than a lease anywhere. Strategic Marks still holds the brand and Firefly handles the licensing. ### Where to find one today Old KB Toys material is cheap and plentiful: store shopping bags, price stickers, shelf signage and employee shirts, mostly from the 1980s and 1990s. Vintage listings sit in the low tens of dollars, well under what the new licensed collectibles cost. Etsy's vintage KB Toys listings (https://www.etsy.com/market/vintage_kb_toys) are the practical place to browse them. ## Frequently Asked Questions ### Is KB Toys coming back? Not as a store chain. KB Toys has been a licensed brand since 2016, and the 2018 plan for up to 1,000 pop-up stores never opened because the funding did not come together. As of 2026 KB Toys sells collectibles online and appears at conventions, with no announced retail locations. ### When did KB Toys close? KB Toys filed for Chapter 11 on December 11, 2008, and its last stores closed on February 9, 2009. The chain had 461 locations at the end: 277 in enclosed malls, 114 outlets, 40 KB Toy Works and 30 seasonal holiday stores. ### Who owns KB Toys now? The KB Toys brand belongs to Strategic Marks, a California brand-revival company run by Ellia Kassoff, which filed for the trademark in 2016 after Toys R Us let the earlier registration lapse. Since November 2024, Firefly Brand Management has run KB Toys licensing, including a master toy deal with MH Enterprises. **Sources:** - KB Toys files for bankruptcy protection (NBC News, 2008): https://www.nbcnews.com/id/wbna28179348 - Will KB Toys Live On After Liquidation? (CNBC, 2009): https://www.cnbc.com/2009/09/08/will-kb-toys-live-on-after-liquidation.html - Entrepreneur plots comeback for KB Toys chain (The Boston Globe, 2018): https://www.bostonglobe.com/business/2018/03/21/entrepreneur-plots-comeback-for-toys-chain/kxifv4G3QLN1alxZImPveP/story.html - Strategic Marks Seeks Funding for KB Toys Relaunch (Licensing International, 2019): https://licensinginternational.org/news/strategic-marks-seeks-funding-for-kb-toys-relaunch/ - Firefly Brand Management Announces Master Toy Deal for KB Toys Brand (License Global, 2024): https://www.licenseglobal.com/toys-games/firefly-brand-management-announces-master-toy-deal-for-kb-toys-brand --- # Where Is FAO Schwarz Now? Inside Rockefeller Center URL: https://404memoryfound.com/posts/fao-schwarz-store-rockefeller-center.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-15 Topics: Then vs Now, Business Blunders **Summary:** FAO Schwarz is still open. The store left its Fifth Avenue home on July 15, 2015 and reopened three years later at 30 Rockefeller Plaza in New York, where it trades today under owner ThreeSixty Group. In June 2026 it added a second New York shop inside the Nordstrom flagship in Manhattan, along with eight Jewel Box shops in Nordstrom stores around the country. **Key facts:** - Founded: 1862, as a toy bazaar in Baltimore; New York store from 1870 - Fifth Avenue store closed: July 15, 2015, after giving up a 45,000 square foot lease - Owner today: ThreeSixty Group of Irvine, California, which bought the brand on October 4, 2016 - Stores left: One US flagship at 30 Rockefeller Plaza, plus a Nordstrom shop in Manhattan and eight Nordstrom Jewel Box shops - Status today: Open and expanding, with FAO Schwarz toys sold on Nordstrom.com since June 2026 ## How a Baltimore toy bazaar became a New York institution FAO Schwarz started in 1862, when Frederick August Otto Schwarz opened a toy bazaar in Baltimore. The New York half of the story began eight years later. On September 16, 1870, Schwarz ran an advertisement for his new toy and fancy store at 765 Broadway in Manhattan, asking people to come and shop. The Library of Congress files that advertisement as the start of the New York business that every later FAO Schwarz store descended from. The name moved uptown with the money over the next century and settled on Fifth Avenue, where the store stopped being a shop and became a stop on the tour. Reporting its closure in 2015, NBC News called FAO Schwarz the oldest toy store in the United States and "a retailer once considered accessible only to the rich." Both halves of that description were accurate. The Fifth Avenue store sold stuffed animals taller than the children looking at them, and the prices matched the ceiling height. It also collected screen time, appearing in the 1988 Tom Hanks film Big and in Woody Allen's 1995 film Mighty Aphrodite. That is why the 2015 closing ran on national news, and why the question people still type is whether anything survived it. ## Why did FAO Schwarz leave Fifth Avenue in 2015? It was a rent story rather than a toy story. FAO Schwarz said in May 2015 that it would give up its three-level, 45,000 square foot Fifth Avenue space, and it shut the doors on July 15, 2015, leaving its lease early. The company did not blame shoppers or the internet. The move was "due to the continuing rising costs of operating a retail location on Fifth Avenue in New York City," FAO Schwarz said in the 2015 statement NBC News reported. The arithmetic is unforgiving on that block. A store whose merchandise is priced in tens of dollars has to sell a great deal of it to cover rent on a stretch of Fifth Avenue shared with luxury houses that sell a single handbag for more than a cart of toys. The owner at the time was Toys R Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html), which had acquired FAO Schwarz in 2009. When the closure was announced, the company said it was hunting for a new Midtown Manhattan home for the flagship. It never opened one. ## Who owns FAO Schwarz in 2026? ThreeSixty Group owns it. On October 4, 2016, Toys R Us sold the FAO Schwarz brand to ThreeSixty Group, a company based in Irvine, California that designs and sells toys, home goods and other products. Terms were not disclosed. CBS News, covering the sale in 2016, noted that Toys R Us had bought FAO Schwarz in 2009 and then shuttered the store to save money. Selling the name was the last act of that decision. ThreeSixty Group already ran a shelf of recognizable American brands, including The Sharper Image, Animal Planet and Discovery Kids. If the pattern sounds familiar, it is the same one behind Sharper Image (https://404memoryfound.com/posts/who-owns-sharper-image-now.html): buy a name people trust, keep the name, drop the expensive stores. The plan announced in 2016 was to design a new collection of toys under the FAO Schwarz name and to build in-store experiences for other retailers, recreating a piece of the famous store inside somebody else's shop. A decade later that is exactly what the company is doing, at scale. ## How many FAO Schwarz stores are left in the US? One full-size American store, plus a growing set of shops inside other retailers' buildings. The flagship is at 30 Rockefeller Plaza in New York, and it opened on November 16, 2018, three years after the Fifth Avenue store went dark. It sits about ten blocks south of the old location, on the plaza where the Christmas tree goes up. In June 2026, FAO Schwarz opened a second New York shop inside the Nordstrom flagship in Manhattan, its first new store in the city in more than a century. Nordstrom and FAO Schwarz also announced eight Jewel Box shops, a store-within-a-store format, in Nordstrom locations elsewhere in the country, and FAO Schwarz product went on sale at Nordstrom.com beginning in June 2026. So the answer to "how many FAO Schwarz stores are left" depends on what counts as a store. One if you mean a building with FAO Schwarz on the outside. Ten if you count the shops inside Nordstrom. That is the shape a lot of surviving mall names now take, including the Disney Store (https://404memoryfound.com/posts/is-disney-store-still-open.html), which moved most of its selling into Target and its own website. ## Is the dance-on piano from Big still in the store? Yes, in a rebuilt form. When the Rockefeller Plaza store opened in November 2018, the dance-on piano came with it, installed on the second floor rather than tucked into a corner of the sales floor. CNBC reported in 2018 that the new piano was given a mirrored ceiling above it, so people standing in the plaza could watch the keys light up under the dancers' feet, with the skating rink visible below. It is a piece of 1988 nostalgia re-engineered as a window display. What it is not is the exact instrument Tom Hanks used. Versions of the walking piano have been built, replaced and rebuilt over four decades, and the one on the second floor in 2026 is a current model, not a movie prop under glass. The Nordstrom shops carry the idea forward too. The New York Nordstrom location that opened in 2026 includes a dance-on piano of its own, alongside toy soldiers and plush animals. Where to find one today: the store's own merchandise is where the collecting happens. Old FAO Schwarz teddy bears, shopping bags, catalogs and the boxed dance-mat pianos sold under the brand turn up regularly through Etsy's vintage FAO Schwarz listings (https://www.etsy.com/search?q=vintage+fao+schwarz), where condition and the original box drive most of the price difference. ## What the Nordstrom deal says about the brand The 2026 partnership is the clearest statement yet of what FAO Schwarz is now: a name that fills a department store's toy aisle rather than a chain that pays its own rent. "Nordstrom is a perfect partner for FAO Schwarz, with its long tradition of beautiful stores, unique and quality products, and its commitment to the best in customer service," said David Niggli, the chief merchandising officer of FAO Schwarz, in the 2026 announcement. Nordstrom framed it as traffic. "FAO Schwarz represents imagination and joyful discovery, which aligns naturally with how we think about creating meaningful experiences for our customers," said Tacey Powers, the Nordstrom executive vice president who runs shoes, kids and home, in the same 2026 release. Translated, both companies get something the other cannot buy easily. FAO Schwarz gets floor space in stores that already have shoppers in them, without signing a Fifth Avenue lease. Nordstrom gets a reason for families to walk into a department store in a decade when fewer of them do. It is a smaller business than the one that closed in 2015, and a more durable one. The brand has outlived its own bankruptcy, two owners and the building most Americans picture when they hear the name. ## Frequently Asked Questions ### Is FAO Schwarz still open? Yes. FAO Schwarz is still open, with its American flagship at 30 Rockefeller Plaza in New York, which opened on November 16, 2018 after the Fifth Avenue store closed on July 15, 2015. Since June 2026 there is also an FAO Schwarz shop inside the Nordstrom flagship in Manhattan, plus eight Jewel Box shops in Nordstrom stores elsewhere in the country. ### Who owns FAO Schwarz now? FAO Schwarz is owned by ThreeSixty Group, a product design and marketing company based in Irvine, California, which bought the brand from Toys R Us on October 4, 2016 for an undisclosed sum. Toys R Us had owned FAO Schwarz since 2009 and closed the Fifth Avenue store in 2015. ThreeSixty Group also owns The Sharper Image, Animal Planet and Discovery Kids. ### Is the big piano still at FAO Schwarz? Yes. The dance-on piano made famous by Tom Hanks in the 1988 film Big is part of the FAO Schwarz store at 30 Rockefeller Plaza in New York, where it has been installed on the second floor since the store opened on November 16, 2018. CNBC reported in 2018 that the rebuilt piano was fitted with a mirrored ceiling so the keys can be seen lighting up from the plaza outside. **Sources:** - Toy Retailer FAO Schwarz Closing Flagship New York City Store - NBC News, 2015: https://www.nbcnews.com/business/business-news/toy-retailer-fao-schwarz-closing-flagship-new-york-city-store-n360086 - Toys R Us sells FAO Schwarz brand to ThreeSixty Group - CBS News, October 4, 2016: https://www.cbsnews.com/news/toys-r-us-sells-fao-schwarz-brand-to-threesixty-group/ - FAO Schwarz puts a new spin on its dance-on piano as part of Friday's NYC comeback - CNBC, November 13, 2018: https://www.cnbc.com/2018/11/13/fao-schwarz-comeback-puts-new-spin-on-old-favorites-like-the-big-piano.html - Nordstrom Partners With FAO Schwarz - Nordstrom press release, 2026: https://press.nordstrom.com/news-releases/news-release-details/nordstrom-partners-fao-schwarz-expand-iconic-toy-brand - F.A.O. Schwarz Opened His First Store in New York City - Library of Congress, This Month in Business History: https://guides.loc.gov/this-month-in-business-history/september/fao-schwarz-opened-his-first-store-in-new-york-city --- # Is Woolworth Still in Business? Only as Foot Locker URL: https://404memoryfound.com/posts/is-woolworth-still-in-business.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-14 Topics: Business Blunders, Then vs Now **Summary:** Woolworth is still in business in the narrowest sense. The corporation never went bankrupt: it closed all 400 of its American five-and-dime stores in 1997, renamed itself Venator Group in 1998 and then Foot Locker, Inc. in 2001, and Dick's Sporting Goods bought that company in 2025. No Woolworth store has traded in the United States since 1997, though a separate German company now runs about 1,000 Woolworth stores across Europe. **Key facts:** - Founded: 1879, Lancaster, Pennsylvania, by Frank Winfield Woolworth - US stores closed: All 400 remaining stores, announced July 17, 1997 - Renamed: Venator Group in 1998, then Foot Locker, Inc. on November 2, 2001 - Owner today: Dick's Sporting Goods, which completed its Foot Locker purchase on September 8, 2025 - Status today: No US Woolworth stores since 1997; about 1,000 Woolworth stores trading in Europe under a separate German company ## Why the five-and-dime ran out of customers Frank Winfield Woolworth opened his first lasting store in Lancaster, Pennsylvania, in 1879, selling nothing above ten cents and putting the goods out where shoppers could handle them before paying. That model carried the company for 117 years. It did not survive the move of American retail out of downtown. By the middle of the 1990s the variety store side of Woolworth Corporation was shrinking fast, and the count of general merchandise stores in the United States fell from 1,465 to 400 in five years, as The Washington Post reported in July 1997. The financial picture was worse than the store count suggested. In the quarter that ended April 26, 1997, the F. W. Woolworth division lost $24 million on $224 million of sales, according to the company's own filing with the Securities and Exchange Commission. A division losing that much on that little revenue has no route back. Woolworth was not alone in the squeeze. Kmart (https://404memoryfound.com/posts/is-kmart-still-open-stores-left.html) and Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html) spent the same decade learning that a store built for 1965 shoppers does not hold 1995 ones. ## What Roger Farah announced on July 17, 1997 On July 17, 1997, Woolworth Corporation told the Securities and Exchange Commission that it was leaving the domestic general merchandise business. All 400 remaining F. W. Woolworth stores in the United States would be closed, sold or converted to something else. Roger N. Farah, then chairman and chief executive, put it plainly in the filing: "We made the very difficult decision to close our domestic F.W. Woolworth general merchandise operations to help assure the continuing profitable growth of the Woolworth Corporation." The same filing said the business had kept losing money and could not be returned to profitability. About 100 of the better locations were converted to Foot Locker, to a larger Champs Sports format, or to other specialty stores. The rest were closed or sold, along with the distribution center in Denver, Pennsylvania, that had supplied them. The Washington Post counted 9,200 employees losing their jobs, 3,600 of them full time and 5,600 part time. The closings ran through the second half of 1997, ending a chain that had been open since the presidency of Rutherford B. Hayes. ## How a shoe chain ended up owning the company name Shutting the five-and-dimes left Woolworth Corporation as a holding company for specialty chains, and the healthiest of them sold sneakers. In 1998 the company renamed itself Venator Group, a deliberately blank label for a business that no longer knew what it was. The blankness did not last. Venator kept selling the non-athletic pieces, and by 2001 there was nothing left to be vague about. On November 1, 2001, Venator Group announced that it would begin doing business as Foot Locker, Inc. the following day. The company's own announcement said the divestitures had allowed it to be "totally focused in the athletic retail segment." At that point it ran roughly 3,600 athletic stores in 14 countries across North America, Europe and Australia, under the Foot Locker, Lady Foot Locker, Kids Foot Locker and Champs Sports banners. So the question of whether Woolworth went out of business turns on a corporate technicality. The stores died. The legal entity, share register and all, changed its name twice and kept trading. ## Who owns the company now that Foot Locker has a parent The business that used to be Woolworth is not independent any more either. Dick's Sporting Goods completed its purchase of Foot Locker, Inc. on September 8, 2025, in a deal valued at about $2.4 billion, which makes Dick's the corporate descendant of a five-and-dime chain founded in 1879. The store count is still moving. Dick's reported 2,478 stores in the Foot Locker business as of August 1, 2026, covering the Foot Locker, Kids Foot Locker, Champs Sports and WSS banners, after closing dozens of locations it judged unproductive. There is no Woolworth store anywhere in the United States, and there has not been one since 1997. The trademark did not disappear with the shelves: Foot Locker, Inc., now owned by Dick's, is the direct successor to the F. W. Woolworth Company and holds the American rights to the name. That is the pattern that also swallowed RadioShack (https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html). The signs come down, the trademark keeps changing hands, and the name outlives the business it described. ## Why Europe has about 1,000 Woolworth stores in 2026 American searches for Woolworth keep colliding with a chain that is very much alive. Woolworth Deutschland is a separate German company, and it has been opening stores at speed while the American original is a footnote. In March 2026 it opened its one thousandth store in Europe, at the Kronenberg Center in Essen. Roman Heini, who runs the business, said in a statement that "The 1000th store is an important step for us and confirms the viability of our business model." The chain runs around 850 stores in Germany and also trades in Austria, Poland, the Czech Republic and Slovakia, selling cheap household goods, clothing and stationery at fixed low prices. Management has talked publicly about reaching 5,000 European locations in the long run. The link back to Frank Winfield Woolworth is historical rather than corporate. The German business began as a prewar subsidiary of the American company and has been separately owned for decades, so buying a mug in a German Woolworth sends nothing to Dick's Sporting Goods. ## The lunch counter that outlived the chain One Woolworth interior still draws visitors every week. On February 1, 1960, four Black students from North Carolina Agricultural and Technical State University sat down at the whites-only lunch counter of the Woolworth store in Greensboro, North Carolina, and stayed in their seats after service was refused. The sit-in spread across the South within weeks and became one of the defining actions of the civil rights movement. The Greensboro building is now the International Civil Rights Center and Museum, which is why that address is the one Woolworth site most Americans can still name. Where to find one today: the chain left behind a great deal of small, cheap, well-made stuff. Lunch counter china, store signs, paper bags, employee badges and Christmas ornaments all circulate among collectors, and a search for five-and-dime memorabilia on Etsy (https://www.etsy.com/search?q=vintage+woolworth) is the practical way to browse them. Condition and whether a piece actually carries the store logo drive the asking price more than age does, so compare several listings before buying. ## Frequently Asked Questions ### Did Woolworth become Foot Locker? Yes. The F. W. Woolworth Company, trading as Woolworth Corporation, renamed itself Venator Group in 1998 after closing its American five-and-dime stores, then began doing business as Foot Locker, Inc. on November 2, 2001. It is the same legal entity under a new name, not a buyer of the old one. ### When did Woolworth close its stores in the United States? Woolworth Corporation announced on July 17, 1997 that it was shutting the 400 F. W. Woolworth variety stores still trading in the United States. About 100 were converted to Foot Locker and other specialty formats and the rest were closed or sold, which ended the American five-and-dime chain after 117 years. ### Who owns the Woolworth name today? In the United States the Woolworth trademark belongs to Foot Locker, Inc., the renamed Woolworth Corporation, which Dick's Sporting Goods acquired in a deal completed on September 8, 2025. In Europe the Woolworth name is run by a separate German company, Woolworth Deutschland, which opened its one thousandth store in March 2026. **Sources:** - Woolworth Corporation, Form 8-K, July 1997 (U.S. Securities and Exchange Commission): https://www.sec.gov/Archives/edgar/data/0000850209/000085020997000006/0000850209-97-000006.txt/seq-2 - Five-and-Dime Farewell, The Washington Post, July 18, 1997: https://www.washingtonpost.com/archive/politics/1997/07/18/five-and-dime-farewell/879889fe-6436-4898-8cc7-d211ee068aca/ - Venator Group, Inc. Announces Name Change to Foot Locker, Inc., Form 8-K exhibit, 2001 (SEC): https://www.sec.gov/Archives/edgar/data/0000850209/000095011701501503/ex-99.txt - DICK'S Sporting Goods, Inc., Form 10-Q for the quarter ended August 1, 2026 (SEC): https://www.sec.gov/Archives/edgar/data/0001089063/000108906326000036/dks-20260801.htm - Non-food discounter Woolworth reaches milestone of 1,000 stores in Europe, RetailDetail EU, 2026: https://www.retaildetail.eu/news/general/non-food-discounter-woolworth-reaches-milestone-of-1000-stores-in-europe/ --- # Who Owns Sharper Image Now? No Stores Since 2008 URL: https://404memoryfound.com/posts/who-owns-sharper-image-now.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-14 Topics: Hardware, Business Blunders **Summary:** Sharper Image is owned by ThreeSixty Group, a California consumer products company that bought the brand from Iconix Brand Group for $100 million in December 2016. The retail chain itself is gone: Sharper Image filed for Chapter 11 in February 2008 and its assets were auctioned for $49 million that May. What is left is a website, a catalog and a licensing program, not a store you can walk into. **Key facts:** - Peak revenue: $760.0 million in fiscal 2004, the year ended January 31, 2005 - Bankruptcy: Chapter 11 in February 2008; assets auctioned for $49 million in May 2008 - Owner today: ThreeSixty Group, which paid $100 million in December 2016 - Stores left: None in the United States as of 2026 - Status today: A licensed brand selling through sharperimage.com, a catalog and other retailers ## How big was Sharper Image before the collapse? Sharper Image spent the early 2000s as one of the most recognizable names in an American mall: a store built around massage chairs, air purifiers, chrome desk toys and whatever gadget the catalog was pushing that season. The business peaked in fiscal 2004. Sharper Image reported record revenues of $760.0 million for the year ended January 31, 2005, spread across its stores, its printed catalog and its website. That was the high point. Revenue slid in the years that followed, the company cycled through leadership, and the single product that had carried the growth turned into the liability that ended it. The timing makes the decline easy to file alongside the 2008 retail wipeout that also took Circuit City (https://404memoryfound.com/posts/what-happened-to-circuit-city-electronics.html). That is not what happened here. Sharper Image's trouble started years earlier, with one air purifier. ## What the Ionic Breeze air purifier cost the company The Ionic Breeze was a tall, silent, filterless air purifier, and for several years it was the product Sharper Image moved more of than anything else on the sales floor. It sold in enormous numbers. When the company settled a class action in January 2007, the settlement class covered roughly 3.2 million people who had bought an Ionic Breeze since May 6, 1999. The terms were expensive. Sharper Image agreed to more than $60 million in discounts on its merchandise, offered each of those buyers a $19 credit good for a year, and made an ozone-reducing attachment available to qualifying customers. The allegation at the center of the case was that the Ionic Breeze did not clean air the way the advertising said it did, and that it emitted ozone. A $19 credit does not repair that. The product that had built the company had become the reason shoppers stopped trusting it. ## Why suing Consumer Reports backfired Consumer Reports tested the Ionic Breeze and was blunt about the result. The magazine reported that the machine delivered "almost no measurable reduction in airborne particles," a finding it put in front of readers in its February 2002 and October 2003 issues. Sharper Image sued the magazine's publisher, Consumers Union, in September 2003, arguing the reviews were false. It lost. In November 2004 the US District Court for the Northern District of California threw the case out under California's anti-SLAPP statute. Judge Maxine Chesney wrote that Sharper Image "has not demonstrated a reasonable probability that any of the challenged statements were false." The lawsuit did what lawsuits against reviewers usually do. It pushed the review back into the news, welded the company's name to it permanently, and left Sharper Image covering the other side's legal bills. ## What happened in the 2008 bankruptcy Sharper Image filed for Chapter 11 bankruptcy protection in February 2008 from its San Francisco headquarters. No reorganization followed. The company's assets went to a bankruptcy auction in New York on May 28, 2008, and a group of private investment firms led by units of Hilco Consumer Capital Corp and Gordon Brothers Group won it with a bid of $49 million. That number is the entire story in one line. A retailer that had booked $760.0 million in revenue four years earlier changed hands for $49 million, and what the buyers wanted was not the shops. Hilco and Gordon Brothers said they had a global licensing strategy ready for wholesale, retail, direct-to-retail, ecommerce and catalog businesses. In practice that meant the name would be rented out and the stores would not reopen, the same path RadioShack (https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html) took a few years later. ## How the name traveled from a liquidator to ThreeSixty Group Trademarks that pass through liquidators rarely stay put, and this one moved twice more. The Sharper Image mark ended up with Iconix Brand Group, a company whose business is collecting names and licensing them out. ThreeSixty Group, a California consumer products manufacturer, had been licensing Sharper Image products since 2008 and had grown into the brand's largest licensee. In December 2016 the tenant bought the building. ThreeSixty Group acquired the Sharper Image brand and its intellectual property assets from Iconix Brand Group for $100 million in cash. Kirk McLean, co-founder of ThreeSixty Group, framed the purchase as the logical end of a long licensing relationship, saying the company had come to "understand the power of the Sharper Image name and see tremendous opportunity to leverage our platform." ThreeSixty also holds FAO Schwarz (https://404memoryfound.com/posts/fao-schwarz-store-rockefeller-center.html), which says a good deal about how it reads the category. ## Can you shop at a Sharper Image in 2026? Not in a store. There are no Sharper Image retail locations operating in the United States in 2026, and there have not been any since the 2008 liquidation emptied the chain. What survives is the name and the channels behind it. Sharper Image sells through sharperimage.com and a printed catalog, and ThreeSixty Group licenses the brand to manufacturers whose goods reach shoppers through general retailers rather than through a shop with the Sharper Image sign above the door. The product mix is recognizable enough: air purifiers, massagers, heated pillows, remote-control toys, roughly the gift-counter inventory the stores carried. What is missing is the demonstration, which was the whole reason the chain worked. The Segway (https://404memoryfound.com/posts/what-happened-to-segway-personal-transporter.html) hit the same wall, a product that needs a showroom does poorly without one. Where to find one today. Original Sharper Image gear lives in the vintage market now rather than at retail. In September 2026, listings ran from a few dollars for an early 1980s Sharper Image catalog up to about $75 for a boxed Sharper Image poker set, with desk gadgets and air purifier towers in between. Browsing the vintage Sharper Image listings on Etsy (https://www.etsy.com/search?q=vintage+sharper+image) is the practical way to see what is actually circulating. ## Frequently Asked Questions ### Is Sharper Image still in business? Sharper Image still exists as a brand in 2026, but not as a store chain. The retailer filed for Chapter 11 bankruptcy in February 2008 and its assets were auctioned that May, and since then Sharper Image has run as a website, a catalog and a licensing program rather than a company with shops of its own. ### Who owns Sharper Image now? Sharper Image is owned by ThreeSixty Group, a California consumer products company that bought the brand and its intellectual property from Iconix Brand Group for $100 million in December 2016. ThreeSixty had been licensing Sharper Image products since 2008, the year the original retailer collapsed. ### What happened to the Ionic Breeze air purifier? The Ionic Breeze was Sharper Image's best-selling product and the source of its worst publicity. Consumer Reports found in 2002 and 2003 that it barely reduced airborne particles, Sharper Image sued the magazine's publisher in September 2003 and lost in November 2004, and a January 2007 class action settlement gave roughly 3.2 million buyers a $19 merchandise credit. **Sources:** - Sharper Image Corporation, fiscal 2004 results (Form 8-K, SEC): https://www.sec.gov/Archives/edgar/data/0000811696/000119312505095689/dex991.htm - Libel suit over Sharper Image product review dismissed, Reporters Committee for Freedom of the Press: https://www.rcfp.org/libel-suit-over-sharper-image-product-review-dismissed/ - Sharper Image Settles Air Purifier Suit, CBS News: https://www.cbsnews.com/news/sharper-image-settles-air-purifier-suit/ - Sharper Image Reported Sold for $49 Million, Chief Marketer: https://www.chiefmarketer.com/sharper-image-reported-sold-for-49-million/ - Iconix Sells Sharper Image to Licensee ThreeSixty Group, Licensing International: https://licensinginternational.org/news/iconix-sells-sharper-image-to-licensee-threesixty-group/ --- # Is the Disney Store Still Open? 21 Left in 2026 URL: https://404memoryfound.com/posts/is-disney-store-still-open.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-14 Topics: Business Blunders, Then vs Now **Summary:** Yes, the Disney Store is still open, but barely: about 21 company-owned locations were left in North America as of April 2026, down from a peak of 747 stores in 1999. Disney closed at least 60 North American stores in 2021 and moved the selling to shopDisney and to Disney Store shop-in-shops inside Target. In 2026 it began testing mall shops again under the name Disney Store Limited Time. **Key facts:** - Launched: March 28, 1987, at the Glendale Galleria in Glendale, California - Peak: 747 stores in 11 countries in 1999 - Owner today: The Walt Disney Company, which took 231 North American stores back on May 1, 2008 - Stores left: About 21 company-owned locations in North America as of April 2026 - Status today: Open, with selling shifted to shopDisney and Target shop-in-shops, plus two Disney Store Limited Time pop-ups in 2026 ## How one mall shop in Glendale became 747 stores The first Disney Store opened on March 28, 1987 at the Glendale Galleria in Glendale, California. It was the first Disney shop built outside a theme park or a resort, and the interior was themed to Hollywood on the theory that every store would eventually reflect the town around it. The format was built to be loud. Video screens ran cartoon clips above the sales floor, staff were called cast members, and almost everything on the shelves was made for the chain rather than bought in from other suppliers. It expanded with the American mall. By 1997 there were more than 600 Disney Store locations in the United States and ten other countries. The chain peaked in 1999 at 747 stores across 11 countries, the largest it has ever been. That peak came the same year mall anchor traffic began to slide, and the chain rode the slide down with every other tenant on the concourse, from Toys R Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html) to the record shops at the far end of the food court. ## Why Disney handed its own chain to a kids' clothing company By the early 2000s the stores were a problem for Disney's consumer products division rather than a showcase for it. In November 2004 Disney sold the North American Disney Store business to The Children's Place, the mall-based children's clothing retailer, and licensed the brand to it instead of running the shops directly. The Children's Place operated the stores through a subsidiary named Hoop Holdings. The arrangement lasted less than four years. Hoop Holdings went into Chapter 11 bankruptcy protection in 2008, and on May 1, 2008 the transaction returning 231 North American Disney Stores to affiliates of The Walt Disney Company was completed. So Disney spent four years demonstrating that it did not want to be a mall landlord's tenant, then bought its way back in at the bottom of a recession. The stores it recovered were remodeled and reopened under Disney control, and for a decade the chain looked stable again. ## The 2021 decision that emptied the concourses On March 3, 2021 Disney said it would close at least 60 North American Disney Store locations before the end of that year, about 20 percent of a chain that then ran roughly 300 stores worldwide. The plan was expected to leave fewer than 250 open by the close of 2021. Stephanie Young, then president of Disney's consumer products, games and publishing unit, gave CNBC the reasoning in 2021. "While consumer behavior has shifted toward online shopping, the global pandemic has changed what consumers expect from a retailer," she said. The selling moved to shopDisney, the company's own e-commerce site, and to Disney Store shop-in-shops inside Target. Neither those Target shops nor Disney's 600-plus theme park stores were part of the 2021 closures, which is why the company could shut dozens of malls' worth of retail and still say its stores were growing. For shoppers the effect was the same one Kmart (https://404memoryfound.com/posts/is-kmart-still-open-stores-left.html) customers had already learned. The chain did not announce an ending. It just stopped being anywhere near you. ## The 2026 count, and what actually counts as a store As of April 2026 there were about 21 company-owned Disney Store locations left in North America. Against the 747 the chain ran in 1999, that is a reduction of roughly 97 percent in 27 years. The count is slippery because Disney now sells the same merchandise four different ways, and three of them get counted as Disney retail in company statements. There are the theme park shops, more than 600 of them. There are the Disney Store shop-in-shops inside Target. There is shopDisney online. And then there are the surviving standalone stores, which is what people mean when they type the question into a search box. Most of the survivors sit in outlet centers rather than the enclosed malls the chain was designed for, because outlet rents are cheaper and the traffic is already there to buy discounted brands. Disney has not published a store list, and fan trackers counting locations one by one in 2026 land in the same range, at about 20 to 21 in the United States. ## What the 2026 mall pop-ups are testing In April 2026 Disney announced Disney Store Limited Time, a run of temporary shops built with Go! Retail Group. The first opened on May 23, 2026 at Ross Park Mall in Pittsburgh, on the upper level. The second opened at Westfield Garden State Plaza in Paramus, New Jersey on September 3, 2026, and is scheduled to run through the 2026 holiday season. They are deliberately styled after the Disney Store of the 1990s and early 2000s, down to a daily opening ceremony, which tells you exactly which decade Disney thinks the nostalgia sits in. Patrick Sager, vice president of Disney Store, told NJBIZ in 2026 that the format lets Disney "bring the Disney Store experience into new communities, deepen our connections with fans and learn from how they engage with the brand." Nothing in that is a commitment. The leases are short, the inventory is limited, and Disney has said the shops are temporary. What it is, plainly, is a test of whether the chain still has a floor under it in 2026, run the way Disney has tested products since it bought Club Penguin (https://404memoryfound.com/posts/what-happened-to-club-penguin.html): quietly, in one market, with an exit already written. ## Where to find Disney Store merchandise today The chain's own products outlived most of its addresses. The 1990s bean bag plush lines, the exclusive VHS clamshells, the snow globes, the shopping bags and the cast member name badges all still circulate, mostly from people who worked in the stores or cleared one out when it closed. Nearly all of it reaches buyers through Etsy's vintage Disney Store listings (https://www.etsy.com/search?q=vintage+disney+store). Prices follow condition and character rather than scarcity, since none of it was manufactured to be rare, and anything still carrying its original hang tag sells for a premium over the same piece without one. ## Frequently Asked Questions ### Is the Disney Store still open in 2026? Yes. About 21 company-owned Disney Store locations were still trading in North America as of April 2026, most of them in outlet centers, alongside shopDisney online and Disney Store shop-in-shops inside Target. Disney also opened two temporary Disney Store Limited Time shops in 2026, in Pittsburgh in May and in Paramus, New Jersey in September. ### Why did Disney close most of its Disney Stores? Disney announced on March 3, 2021 that it would close at least 60 North American Disney Store locations, roughly 20 percent of about 300 stores worldwide, and shift spending to its shopDisney website. Stephanie Young, then president of Disney's consumer products, games and publishing unit, told CNBC in 2021 that the pandemic had changed what shoppers expected from a retailer. ### How many Disney Stores were there at the peak? The Disney Store chain peaked in 1999 at 747 stores in 11 countries, up from more than 600 locations across the United States and ten other countries in 1997. The first one had opened twelve years earlier, on March 28, 1987, at the Glendale Galleria in Glendale, California. **Sources:** - Disney Store, The - D23: https://d23.com/a-to-z/disney-store-the/ - The Children's Place Announces Completion of Transaction Between Hoop Holdings and Affiliates of The Walt Disney Company (May 1, 2008): https://www.globenewswire.com/news-release/2008/05/01/377457/141520/en/The-Children-s-Place-Retail-Stores-Inc-Announces-Completion-of-Transaction-Between-Its-Subsidiary-Hoop-Holdings-LLC-and-Affiliates-of-The-Walt-Disney-Company.html - Disney shuttering at least 20% of Disney Stores as it shifts focus to e-commerce - CNBC, March 3, 2021: https://www.cnbc.com/2021/03/03/disney-shuttering-at-least-20percent-of-disney-stores-as-it-shifts-focus-to-e-commerce.html - Disney announces limited-time, nostalgic retail experiences - TheStreet, April 2026: https://www.thestreet.com/retail/disney-announces-limited-time-nostalgic-retail-experiences - Disney pop-up stores coming to NJ, PA malls - NJBIZ, 2026: https://njbiz.com/disney-pop-up-stores-nj-pa-malls-go-retail/ --- # Where to Buy a Game Boy in 2026 and What to Pay URL: https://404memoryfound.com/posts/buy-working-game-boy-2026.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-14 Topics: Gaming, Hardware **Summary:** You can still buy a working Nintendo Game Boy in 2026, and a tested original handheld runs about $70, with boxed examples passing $300 and sealed ones reaching close to $2,000 at auction. Nintendo stopped production on March 31, 2003, so nothing carries a factory warranty and condition is the whole game. The faults that matter are missing LCD lines, corroded battery terminals and undisclosed reshells. **Key facts:** - Launched: July 31, 1989 in the United States, $89.95, Nintendo - Units sold: 118.69 million, Game Boy and Game Boy Color combined - Discontinued: March 31, 2003 - Status today: Sold used only; about $70 for a tested working original in 2026 - What replaced it: Game Boy Advance SP in 2003, and the $239.99 Analogue Pocket in 2026 ## Which Game Boy model fits what you actually want to play? Four machines share the name, and they are not interchangeable. The original DMG-01 Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html) reached the United States on July 31, 1989 at $89.95, ran on four AA batteries, and used an unlit green screen that needs a lamp or a window to read. The Game Boy Pocket followed in 1996 at $69.99, roughly half the bulk, with a true black and white display that is still unlit. The Game Boy Color arrived on November 18, 1998 at $79.95 and runs both its own cartridges and the older monochrome ones. The Game Boy Advance SP, released in February 2003 at $99.99, is the model most first-time buyers end up happy with. It folds shut, charges from a wall adapter instead of eating AAs, and plays Game Boy, Game Boy Color and Game Boy Advance cartridges. Nintendo cut it to $79.99 in September 2004. The later AGS-101 revision carries a backlit screen that is far brighter than the front-lit AGS-001, and buyers pay more for it. Gunpei Yokoi, who led the original design, never apologized for the gray brick. "Once you start playing the game, the colors aren't important," he said in a 1997 interview translated by Shmuplations. ## What does a working Game Boy cost in 2026? BGR put a bare original Game Boy in good working order at about $70 in 2026. That is the handheld on its own: no box, no manual, usually no game. Completeness is what moves the number. A Game Boy with its original box and documentation can pass $300, and a sealed unit that was never opened has gone for close to $2,000 at auction, according to the same 2026 report. Those upper figures are collector prices, and they are not what a player should pay. Nintendo shipped 118.69 million Game Boy and Game Boy Color units worldwide before production stopped, so the supply of ordinary, scuffed, perfectly playable handhelds is deep. Nintendo ended production of the Game Boy and the Game Boy Color on March 31, 2003. Every unit on the market in 2026 is at least 23 years old, none of them carry a warranty from Nintendo, and the price on the listing tells you nothing about whether the screen still shows all 144 rows. What the seller can prove is worth more than the number. ## What fails on a Game Boy that sat in a closet for 25 years? Three faults account for most dead units. The first is missing vertical lines on the LCD, where the connection to the screen ribbon has degraded and whole columns of pixels drop out. It is repairable, but it is a repair, not a scratch. The second is battery corrosion. Alkaline cells left in the compartment since the Clinton administration leak, and the blue-green crust on the terminals stops current before it reaches the board. Light corrosion cleans off. Deep corrosion eats the contacts and the traces under them. The third is the shell and everything bolted to it: a cracked screen lens, a scratchy speaker, buttons that need two presses, a missing battery cover. Battery covers went missing on most of these handhelds decades ago. Ask any seller for three things before money moves. A photo of the handheld powered on with a game running, not a title screen alone. A photo of the open battery bay. A plain answer on whether the sound works. A seller who will not power the unit on is selling you a repair project. ## Is a refurbished or modded Game Boy worth the premium? Refurbishers do the work most buyers cannot: a backlit IPS panel dropped into a 1989 shell, fresh capacitors, a new speaker, cleaned contacts, sometimes a new case. It costs more than a stock unit, and for anyone who plans to play indoors it solves the machine's single worst flaw, which is that the original screen is unlit. The tradeoff is that a modded handheld is no longer original, and the two markets price differently. A collector paying for a boxed 1989 unit wants nothing touched. A player wants the screen fixed. Watch for the reshell that is not disclosed. A Game Boy in a color Nintendo never sold, in condition too perfect for its age, with soft or shallow molded lettering and panel gaps that do not line up, has almost certainly been put in an aftermarket case. That is not fraud by itself, since the board inside is usually genuine Nintendo, but it should be stated in the listing and it should be reflected in the price. Ask who did the work and whether the mainboard is the original. ## Should you just buy a new handheld instead? Start with the cheap answer, because it is usually the right one: a used Game Boy Advance SP with the backlit AGS-101 screen plays the entire Game Boy, Game Boy Color and Game Boy Advance library on original cartridges and costs less than anything built this decade. Above that sits the Analogue Pocket, which takes real Game Boy cartridges and puts them on a modern screen. It launched at $199, moved to $219.99, and went to $239.99 in March 2026. Analogue was blunt about why: "Analogue Pocket is now $239.99 due to recent tariff announcements," the company said in the statement Engadget reported that month. The cheapest route runs through emulation boxes rather than cartridges. An Anbernic handheld (https://404memoryfound.com/posts/anbernic-handheld-worth-it-game-boy.html) costs a fraction of a Pocket and plays files instead of carts, which is a different hobby with a different legal footing. If what you miss is a backlit color screen from the 1990s, the Sega Game Gear (https://404memoryfound.com/posts/what-happened-to-sega-game-gear-handheld.html) got there first and drained six AAs doing it. ## Where to find one today Budget about $70 for a tested, working original Game Boy in 2026, more for a Game Boy Advance SP with the backlit screen, and well past $300 if you want the box and papers with it. Buy from a shop that tests hardware and says so in writing rather than from a listing that shows a powered-off screen, because a unit that arrives with dropped LCD lines is worth a fraction of what you paid. Specialist retro dealers list Game Boy hardware and accessories by system, including Stone Age Gamer's Game Boy console section (https://stoneagegamer.com/nintendo/game-boy/consoles/). Whatever you buy, take the batteries out before it goes back in a drawer. ## Frequently Asked Questions ### How much does a Game Boy cost in 2026? A bare original Nintendo Game Boy in good working condition runs about $70 in 2026, according to BGR. A complete example with its box and documentation can pass $300, and a sealed unit has gone for close to $2,000 at auction. Nintendo's own 1989 launch price in the United States was $89.95. ### What should you check before buying a used Game Boy? Before buying a used Nintendo Game Boy, check the LCD for missing vertical lines, check the battery compartment for blue-green corrosion, and confirm the speaker works. Ask for a photo of the handheld powered on with a game running. Nintendo stopped making the Game Boy and Game Boy Color on March 31, 2003, so nothing sold today carries a factory warranty. ### Do Game Boy Advance systems play original Game Boy games? Yes. The Game Boy Advance and the Game Boy Advance SP, released in February 2003 at $99.99, play Game Boy, Game Boy Color and Game Boy Advance cartridges. The Game Boy Advance SP is the practical choice for anyone who wants a lit screen, since its later AGS-101 revision is backlit while the earlier AGS-001 is only front-lit. **Sources:** - Game Boy (Wikipedia): launch date, price, sales and discontinuation: https://en.wikipedia.org/wiki/Game_Boy - Game Boy Advance SP (Wikipedia): 2003 launch price, AGS-001 and AGS-101: https://en.wikipedia.org/wiki/Game_Boy_Advance_SP - Gunpei Yokoi x Yukihito Morikawa, 1997 interview (Shmuplations translation): https://shmuplations.com/yokoi/ - BGR: Here's How Much Your Original Nintendo Game Boy Is Worth Today: https://www.bgr.com/2141069/how-much-original-nintendo-game-boy-is-worth-today/ - Engadget: The Analogue Pocket is back in stock with a tariff-related price increase: https://www.engadget.com/gaming/the-analogue-pocket-will-be-back-in-stock-this-week-but-theres-a-tariff-related-price-increase-182226016.html --- # Can You Still Get Dial-Up Internet? NetZero Does URL: https://404memoryfound.com/posts/dial-up-internet-still-available-2026.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-14 Topics: Internet Culture, Then vs Now **Summary:** You can still get dial-up internet in 2026. NetZero and Juno, both run by United Online, still sell it in the United States, and Microsoft still charges $21.95 a month for MSN Dial-Up Internet Access. What ended on September 30, 2025 was AOL's dial-up service, not dial-up itself. **Key facts:** - Status today: Still sold in the US in 2026 by NetZero, Juno and MSN Dial-Up - Owner today: NetZero and Juno are United Online brands, owned by B. Riley Financial since July 1, 2016 - Cheapest paid plan: MSN Dial-Up Internet Access, $21.95 a month or $179.95 a year - Free option: NetZero and Juno, ad-supported, capped at 10 hours a month - US homes on dial-up: 163,401 in 2023, about 0.13% of internet subscriptions (Census Bureau) ## Three companies will still sell you a dial-up account Dial-up internet is not gone. Three providers were taking new sign-ups in the United States in 2026, and two of them are the same brands that handed out free hours in the late 1990s. NetZero and Juno both still run national dial-up networks. Microsoft still sells MSN Dial-Up Internet Access through its own online store, a product that never came off the price list. PCWorld laid the prices side by side after AOL announced its exit in 2025. MSN Dial-Up costs $21.95 a month, or $179.95 paid a year at a time. NetZero charges $29.95 a month for its accelerated dial-up plan, the tier aimed at people who use the connection for more than email. Both prices were still current in 2026. The free tier survived too. NetZero and Juno each advertise an ad-supported plan capped at 10 hours a month, the same offer both companies built their names on. Ten hours works out to about 20 minutes a day, which felt generous in 1999 and is close to useless for a modern web page. The practical catch is geography. Dial-up access numbers are local, so the real question is not whether the service exists but whether a provider has a number in your area code that will not bill as a long-distance call. ## What actually ended on September 30, 2025 AOL stopped selling and supporting dial-up internet on September 30, 2025. The notice went up in August of that year, and the wording was as unsentimental as a discontinued printer driver. "AOL routinely evaluates its products and services and has decided to discontinue Dial-up Internet," the company wrote on its site, as CNN reported on August 11, 2025. Two pieces of software went with it. The AOL Dialer, which stored the access numbers and placed the call, and AOL Shield, a browser built for old operating systems and slow links, were both retired the same day. What did not end was dial-up as a technology. AOL was one seller among several, and the modems, the copper phone lines and the access numbers that make a dial-up call work were never AOL property. When the company switched off its dialer, nothing changed for a NetZero subscriber in a rural county. The confusion is understandable. AOL was dial-up to most Americans for a decade, so its exit read in headlines like a funeral for the entire category rather than the retirement of one product. Anyone curious about the mechanics can still hear what each part of the handshake screech meant (https://404memoryfound.com/posts/dial-up-modem-sound-explained.html). ## Who owns NetZero and Juno now? Both brands belong to United Online, and United Online belongs to B. Riley Financial. The firm agreed to buy the company in 2016 at $11.00 a share, a deal valued at about $170 million, and closed it on July 1, 2016. B. Riley was blunt about why it wanted a business in decline. The strategy was "purchasing companies with complex business dynamics in challenging or mature industries and then implementing operational changes to generate attractive returns," chairman and chief executive Bryant Riley said in the announcement of the completed deal in July 2016. That sentence explains dial-up in 2026 better than any nostalgia does. Nobody is winning new dial-up customers. The networks are already built, the billing runs itself, and a subscriber base that renews quietly is worth more to an owner like this than a growth story would be. The same arithmetic keeps other relics breathing. AOL still charges a monthly subscription (https://404memoryfound.com/posts/does-aol-still-exist-today.html) to customers who could get most of it free, and that is the business, not an accident. ## How many Americans still connect by modem? The Census Bureau counted 163,401 American homes on dial-up in 2023, roughly 0.13 percent of all internet subscriptions, a figure Smithsonian Magazine cited when AOL announced the shutdown in 2025. Nationally that is a rounding error. Locally it is a lifeline. Dial-up households cluster where cable and fiber never arrived, where the terrain blocks a clear view of the sky for satellite service, or where the monthly cost of anything faster does not fit the budget. For those homes the modem is not a nostalgia purchase. It is the only connection that runs over equipment already in the house, on a phone line that was paid for decades ago, and it still delivers email, a bank balance and a weather forecast. The count has been shrinking for twenty years and will keep shrinking, because the copper telephone network underneath it is being retired region by region. When a carrier turns off a copper exchange, dial-up in that area goes with it whether or not anyone was still selling the service. ## Why dial-up costs more today than it did in 1998 Here is the detail that stings. MSN Dial-Up Internet Access sells for $21.95 a month in 2026. That is exactly what AOL charged for unlimited dial-up after April 1998. AOL announced that increase on February 9, 1998, raising its unlimited plan by 10 percent from $19.95 to $21.95, and said it needed the money because advertising revenue alone could not cover how heavily subscribers were using the service. CNN's business desk covered the hike the day it was announced. So the price of a dial-up account has not moved in 28 years. The product has not moved either: the same modem tones, the same tied-up phone line, the same ceiling on speed that looks absurd next to a 2026 broadband plan (https://404memoryfound.com/posts/dialup-speed-then-vs-now.html). NetZero's $29.95 accelerated plan costs more than the 1998 AOL bill did, in dollars worth considerably less. Nobody buying dial-up in 2026 is buying it for value. They are buying it because the alternative on their road is nothing. ## Frequently Asked Questions ### Can you still get dial-up internet in 2026? Yes. Dial-up internet was still sold in the United States in 2026 by NetZero and Juno, both operated by United Online, and by Microsoft as MSN Dial-Up Internet Access. AOL left the business on September 30, 2025, but it was only one seller among several, and its exit did not shut down dial-up itself. ### Does AOL still have dial-up? No. AOL discontinued its dial-up internet service on September 30, 2025, after posting the notice in August of that year, and retired the AOL Dialer and the AOL Shield browser on the same day. AOL still sells other monthly subscription products, but dial-up internet access is no longer one of them. ### How much does dial-up internet cost now? MSN Dial-Up Internet Access cost $21.95 a month or $179.95 a year in 2026, and NetZero's accelerated dial-up plan cost $29.95 a month. NetZero and Juno also advertise ad-supported free plans capped at 10 hours a month, which is where most remaining dial-up users start. **Sources:** - CNN Business: AOL will stop offering dial-up internet service after more than 30 years in business (2025): https://www.cnn.com/2025/08/11/tech/aol-dial-up-internet-discontinue - PCWorld: Despite AOL's exit, dial-up internet isn't going away: https://www.pcworld.com/article/2874440/list-of-dialup-internet-providers-in-2025.html - Smithsonian Magazine: The Sounds of AOL Dial-Up Defined the Early Internet (2025): https://www.smithsonianmag.com/smart-news/the-sounds-of-aol-dial-up-defined-the-early-internet-now-the-service-is-shutting-down-for-good-180987177/ - B. Riley Financial Completes Acquisition of United Online (press release, July 2016): https://www.prnewswire.com/news-releases/b-riley-financial-completes-acquisition-of-united-online-300293239.html - CNNMoney: AOL hikes monthly fee, February 9, 1998: https://money.cnn.com/1998/02/09/technology/aol/ --- # How Many Sam Goody Stores Are Left? Exactly One URL: https://404memoryfound.com/posts/sam-goody-stores-left.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-13 Topics: Business Blunders, Music & Entertainment **Summary:** One Sam Goody store is left. The last location still trading in 2026 is inside the Rogue Valley Mall in Medford, Oregon, after the chain's other survivor in St. Clairsville, Ohio, closed following a December 2024 announcement. Sam Goody's parent company Musicland filed for Chapter 11 bankruptcy in January 2006 and closed 226 Sam Goody stores the next month, and Trans World Entertainment converted most of the rest to FYE between 2007 and 2008. **Key facts:** - Launched: 1951, New York City, by Sam "Goody" Gutowitz - Peak: More than 1,300 Musicland stores and over 300 million customer visits a year by December 2000 - Sold to Best Buy: January 2001, for $425.1 million in cash plus $271.2 million in assumed debt - What replaced it: FYE, which took over most Sam Goody stores between 2007 and 2008 - Stores left: One, at the Rogue Valley Mall in Medford, Oregon, still open in 2026 ## How a Manhattan record counter became a mall chain Sam "Goody" Gutowitz opened the first store carrying his name in New York City in 1951, when the long-playing record was still a new product and most Americans bought music at a department store counter. His method was volume and price. Goody sold records cheap, advertised hard, and treated music as something people bought on impulse rather than saved up for. In 1978 the American Can Company bought Sam Goody. American Can already owned Musicland, the Minneapolis chain that was Goody's biggest rival, so from that point the two competitors answered to the same parent company. Musicland is what pushed the name into every mall in America. It put Sam Goody stores beside food courts, ran the Suncoast Motion Picture Company for video, and built the Media Play superstores for everything at once. By the time Best Buy made its offer in December 2000, Musicland operated more than 1,300 stores that drew over 300 million customer visits a year. For a generation of American teenagers, a Sam Goody listening station was the closest thing to a streaming service. ## The 100,000 counterfeit tapes that put Sam Goody on trial Long before the mall era ended, the company nearly ended itself in a federal courtroom in Brooklyn. A sixteen-count indictment filed on February 28, 1980 charged Sam Goody, Inc., its president George Levy, and its vice president in charge of procurement, Samuel Stolon, with offenses arising from the way the company bought records and tapes. The government's case was that between June and October 1978 the company took delivery of more than 100,000 counterfeit eight-track and cassette tapes. The titles were the biggest sellers of the moment: thousands of copies of the soundtracks to "Saturday Night Fever" and "Grease", plus "The Stranger" by Billy Joel and "Slowhand" by Eric Clapton. The charges grew out of a three-year FBI investigation into counterfeit product moving through legitimate retailers, an operation the bureau called Mod Sound. A jury convicted the company and Stolon in 1981. The stores stayed open, but the largest record retailer in the country had spent two years explaining in court where its inventory came from. ## What Best Buy paid for Musicland, and what it lost Best Buy announced the Musicland purchase in December 2000 and completed it in January 2001, paying $425.1 million in cash and assuming $271.2 million in debt, a total near $700 million. The reasoning was distribution. Best Buy had large stores in the suburbs and nothing at all inside shopping malls, and Musicland came with 1,300 leases in the places where teenagers already spent their Saturdays. Richard Schulze, Best Buy's chairman and chief executive, said in the January 2001 announcement that the deal would let the company "deliver digital entertainment technologies to consumer segments not currently served by Best Buy's store format, particularly in rural areas, malls and the early technology adopters". The plan was to fill Sam Goody shelves with MP3 players, cell phones and digital cameras. What actually arrived was the collapse of the compact disc, accelerated by the file-sharing era that Napster (https://404memoryfound.com/posts/napster-destroyed-music.html) opened in 1999. Best Buy lost $85 million on Musicland in 2002 and started looking for an exit. By the time the group filed for bankruptcy, it belonged to the private equity firm Sun Capital Partners. ## How Chapter 11 took the name off the mall directory Musicland filed for Chapter 11 bankruptcy in January 2006. The following month it announced that it would close 226 Sam Goody stores and 115 Suncoast Motion Picture Company stores, and shut every Media Play location. Trans World Entertainment, which already ran the FYE chain, bought Sam Goody out of the bankruptcy in 2006. Between 2007 and 2008 it converted most of the surviving stores to FYE. That is why the chain seemed to vanish overnight without a farewell sale. In most malls it was not closed, it was renamed, and shoppers walked past the same shelves under a different sign. The broader business was leaving the mall regardless. Columbia House (https://404memoryfound.com/posts/is-columbia-house-still-around.html) had already lost its music club, physical album sales were falling every year, and the customer who used to browse for 40 minutes was now downloading in four. The FYE chain that absorbed Sam Goody was later sold by Trans World Entertainment to the Canadian retailer Sunrise Records, which runs it today. The Sam Goody name survived as a handful of stores that nobody bothered to repaint. ## Why the last store is in Medford, Oregon By 2024 exactly two Sam Goody stores were left: one at the Ohio Valley Mall in St. Clairsville, Ohio, and one at the Rogue Valley Mall in Medford, Oregon. In December 2024, national outlets reported that both were closing, and the Ohio store did shut down. Medford refused the script. Gavin Culver, the manager of the Oregon store, told KDRV NewsWatch 12 in December 2024 that the reports were wrong: "We're going to do the last Blockbuster kind of thing because we're cool like that out here." The comparison is a local one. The last Blockbuster (https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html) is also in Oregon, in Bend, and it has turned being the final survivor into its whole business model. It worked. Counts through November 2025 put the United States total at a single Sam Goody, and reporting in August 2026 found the Medford store still open and still selling records, movies and T-shirts. That leaves one store on the planet, in a mall in southern Oregon, a few doors down from the only Orange Julius left in the state. ## Where to find Sam Goody memorabilia today There is no Sam Goody product to collect, because the company sold other people's records. What circulates is the retail debris: yellow shopping bags, employee name badges and polo shirts, gift cards, promotional posters and standees, and the tall cardboard longbox packaging that CDs shipped in until 1993. Most of it reaches buyers through Etsy's vintage Sam Goody listings (https://www.etsy.com/search?q=sam+goody+vintage), generally from former staff or from people who cleared out a store when it closed. Prices track condition and sentiment rather than scarcity, since none of this was made to be rare. The other option costs a plane ticket. The Medford store still sells new stock over the counter, which makes a receipt from it the only Sam Goody artifact anyone can still buy new. ## Frequently Asked Questions ### Are there any Sam Goody stores left? Yes, one. A single Sam Goody store is still trading in 2026, at the Rogue Valley Mall in Medford, Oregon, and it is the last one anywhere. The chain's other survivor, at the Ohio Valley Mall in St. Clairsville, Ohio, closed after a December 2024 announcement. ### What happened to all the Sam Goody stores in malls? Musicland, the owner of Sam Goody, filed for Chapter 11 bankruptcy in January 2006 and announced the following month that it would close 226 Sam Goody stores and 115 Suncoast stores. Trans World Entertainment bought what was left in 2006 and converted most of the remaining Sam Goody locations to FYE between 2007 and 2008. ### How much did Best Buy pay for Sam Goody? Best Buy bought Musicland, the parent company of Sam Goody, in January 2001 for $425.1 million in cash plus $271.2 million in assumed debt. Best Buy lost $85 million on Musicland in 2002 and sold the business on, and the group filed for bankruptcy in January 2006 under its next owner, Sun Capital Partners. **Sources:** - Sam Goody, Wikipedia: https://en.wikipedia.org/wiki/Sam_Goody - Best Buy Co., Inc., Form 8-K exhibit: Musicland acquisition press release, January 2001 (SEC EDGAR): https://www.sec.gov/Archives/edgar/data/0000764478/000091205701003817/a2036797zex-99.htm - United States v. Sam Goody, Inc., 506 F. Supp. 380 (E.D.N.Y. 1981), Justia: https://law.justia.com/cases/federal/district-courts/FSupp/506/380/1654249/ - Could Medford be the home of the last Sam Goody?, KDRV NewsWatch 12, December 2024: https://www.kdrv.com/news/top-stories/could-medford-be-the-home-of-the-last-sam-goody/article_e6c2d60a-bf10-11ef-b544-0f3488cae422.html - Medford Mall Houses World's Last Sam Goody, Hoodline, August 2026: https://hoodline.com/2026/08/medford-mall-houses-world-s-last-sam-goody-next-to-oregon-s-only-orange-julius/ --- # When Did Netflix Stop Mailing DVDs? The Last Disc URL: https://404memoryfound.com/posts/netflix-dvd-mail-service-shutdown.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-13 Topics: Music & Entertainment, Internet Culture **Summary:** Netflix stopped mailing DVDs on September 29, 2023, closing DVD.com after 25 years and 5.2 billion discs shipped in the United States. The service is not coming back in 2026: subscriptions were cancelled automatically on that final shipping date, and Netflix has announced no plan to restart it. The last disc in the last red envelope was a Blu-ray of True Grit. **Key facts:** - Launched: April 14, 1998, as the first online DVD rental store - Price then: $4 per disc for seven days, plus $2 delivery on the first disc (1998) - Discs shipped: 5.2 billion in the United States over 25 years - Peak subscriptions: 13.93 million domestic DVD subscriptions, September 30, 2011 - Status today: Closed. Final discs mailed September 29, 2023, and never revived ## What did a Netflix rental cost in 1998? Netflix opened on April 14, 1998, describing itself as the first internet store for DVD rentals. The catalog held 925 titles, which at the time was close to every DVD released in the United States. The pricing looked nothing like the flat fee people remember. A seven day rental ran $4 for each of the first two discs, with delivery charged at $2 for the first disc and $1 for each additional disc in the same order. Late fees applied. The first disc the company ever mailed had gone out a month earlier, on March 10, 1998, and it was Beetlejuice. Netflix repeated that detail in 2023 when it published the history of the service. The flat monthly subscription with no due dates and no late fees arrived the following year, and that is the model that broke the video store. Blockbuster turned down the chance to buy the company in 2000 (https://404memoryfound.com/posts/blockbuster-netflix-acquisition.html), a decision that reads worse with every passing year. By the end, the red and white envelope was the part of Netflix people were sentimental about. Not the queue, not the star ratings, not the recommendation engine. ## How big did the red envelope business get? Netflix shipped 5.2 billion discs in the United States over the 25 year run of the service, a total the company published when it announced the closure in 2023. The subscriber peak sits in Netflix's own filings rather than in press coverage. The Form 10-Q for the quarter ended September 30, 2011 reported 13.93 million domestic DVD subscriptions. The mail business never got larger than that. At that size the operation was a logistics company wearing a movie company's name. Discs moved through regional centers timed around first class mail, which is how a disc dropped in a mailbox on Monday could be swapped for the next title in the queue within a couple of days. The most rented title across the whole 25 years was The Blind Side, the 2009 Sandra Bullock film, which Netflix confirmed when it released its all time rental list in 2023. Discs also paid for the pivot. While the streaming library was still thin and the licensing deals were still small, the envelopes were the part of the company that made money. ## Why Qwikster broke the DVD business in 2011 In July 2011 Netflix split the combined plan that cost $9.99 a month into two separate plans at $7.99 each, one for streaming and one for discs. Anyone who wanted both went from $9.99 to $15.98. The reaction was immediate and loud. On September 18, 2011, Reed Hastings, then chief executive, published an apology that opened with "I messed up. I owe everyone an explanation." In the same message he announced that the DVD half would be spun out under a new name, Qwikster, with its own website and its own queue. Subscribers would have kept two accounts, two ratings histories and two credit card charges. Netflix dropped the Qwikster plan within a month but kept the price split. Domestic DVD subscriptions never recovered. The 13.93 million on the books at the end of September 2011 was the high water mark, and the count fell in every year that followed. The episode is still taught as a case study in how a company can lose customers while solving none of the problems they actually raised. ## What did Netflix say when it pulled the plug? Netflix announced the end on April 18, 2023, alongside its first quarter results. Co-chief executive Ted Sarandos framed it as a service problem rather than a money problem: "Our goal has always been to provide the best service for our members but as the business continues to shrink that's going to become increasingly difficult." The money made the case regardless. The DVD business took in $145.7 million in 2022, down 20 percent from the year before, and that was roughly 0.5 percent of everything Netflix collected that year. The company called 2023 the final season of DVD.com and kept shipping for another five months so subscribers could work through their queues before the doors closed. Streaming accounts were untouched. Households that paid for both a streaming plan and a disc plan simply lost the disc half, and the streaming side carried on billing as normal. ## What was in the last red envelope? The final discs went into the mail on September 29, 2023. Netflix said the last one was a Blu-ray of True Grit, the 2010 Coen brothers western. The send off was unusually generous for a shutdown. Subscribers who opted into what Netflix called the finale surprise were mailed up to 10 extra discs pulled at random from their own queues, and nobody was billed for keeping them. Returns were accepted until October 27, 2023, but returning anything was optional. Millions of discs that had spent years cycling between strangers stayed wherever they last landed. DVD subscriptions were cancelled automatically on the final shipping date, so no subscriber had to remember to close an account or fight a renewal. It was a softer ending than physical rental usually gets. Redbox kiosks (https://404memoryfound.com/posts/is-redbox-still-around-kiosks.html) went dark in July 2024 in the middle of a Chapter 7 liquidation, with no farewell and no free discs. ## Is the Netflix DVD service coming back in 2026? No. DVD.com has been closed since September 29, 2023, Netflix has announced no plan to restart it, and nothing in the three years since suggests the company wants back into shipping and warehousing. What replaced it is partly streaming and partly nothing at all. Streaming rights expire and discs do not, so the disc library held thousands of older, foreign and out of print titles that are now on no service anywhere. Anyone who used DVD.com as a film school has noticed. Rental by mail did not die with Netflix, though it is a cottage industry now. Small operators still run the flat envelope model with no due dates, GameFly still mails games, and public libraries lend DVDs and Blu-rays free with a library card. Mail order media has a habit of hanging on: Columbia House (https://404memoryfound.com/posts/is-columbia-house-still-around.html) kept taking orders until 2026. Where to find one today: the discs themselves are worth almost nothing, so the collectible is the envelope. Original red mailers, disc sleeves and DVD.com promotional items turn up in ephemera shops, and Etsy sellers list them under Netflix red envelope (https://www.etsy.com/search?q=netflix+red+envelope). Check the seller's photos for the return address panel, which is what tells a real mailer from a reproduction. ## Frequently Asked Questions ### When did Netflix stop sending DVDs in the mail? Netflix mailed its last DVDs on September 29, 2023, ending a service that began on April 14, 1998. The company announced the closure on April 18, 2023, accepted disc returns until October 27, 2023, and cancelled every DVD subscription automatically on the final shipping date. ### What was the last DVD Netflix mailed? Netflix said the final disc it shipped on September 29, 2023 was a Blu-ray of True Grit, the 2010 Coen brothers western. The first disc Netflix ever mailed, on March 10, 1998, was Beetlejuice, and the most rented title across the 25 years of the service was The Blind Side. ### Can you still rent DVDs by mail in 2026? Yes, but not from Netflix, which closed DVD.com in September 2023 and has not revived it. Small rental by mail companies still ship discs in flat envelopes, GameFly still mails games, and public libraries in the United States lend DVDs and Blu-rays at no charge with a library card. **Sources:** - Netflix: First Online DVD Rental Store Opens (April 14, 1998): https://about.netflix.com/en/news/first-online-dvd-rental-store-opens - Netflix: DVD, The Final Season (2023): https://about.netflix.com/en/news/netflix-dvd-the-final-season - Variety: Netflix Is Shutting Down Its DVD Business (April 18, 2023): https://variety.com/2023/digital/news/netflix-dvd-business-shut-down-1235587325/ - Netflix Inc, Form 10-Q for the quarter ended September 30, 2011 (SEC): https://www.sec.gov/Archives/edgar/data/0001065280/000119312511284366/d222257d10q.htm - ABC News: Netflix CEO Apologizes for Price Increase, Announces Qwikster (September 19, 2011): https://abcnews.com/blogs/technology/2011/09/netflix-ceo-apologizes-for-price-increase-announces-qwikster-to-mail-dvds --- # Is Columbia House Still Around? The 2026 Shutdown URL: https://404memoryfound.com/posts/is-columbia-house-still-around.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-13 Topics: Money & Tech, Music & Entertainment **Summary:** Columbia House is shutting down: the mail-order club stops accepting new orders after September 15, 2026, 71 years after it launched as the Columbia Record Club in 1955. Its music club had already closed in 2010, leaving a small DVD and Blu-ray operation. At its 1996 peak, Columbia House had about 16 million members and roughly $1.4 billion in annual revenue. **Key facts:** - Launched: 1955, as the Columbia Record Club, by Columbia Records - Peak: About 16 million members and roughly $1.4 billion in revenue in 1996 - Final years: About $17 million in revenue in 2014, and 110,000 DVD club buyers - Owner today: John Lippman, who bought the brand out of bankruptcy in 2015 for about $1.5 million - Status today: Stops accepting new orders after September 15, 2026 ## What stops on September 15, 2026 Columbia House is closing after 71 years. A notice on its website told customers that "After 9/15/26, The Columbia House will no longer be accepting new orders," and a customer service representative confirmed the shutdown to CNN by phone in August 2026. There was no press release, no farewell campaign and no final catalog. For a company that mailed tens of millions of Americans a reply card every month, the ending arrived as one line of text on a web page most people assumed had stopped loading years ago. What closes is not the music club. Columbia House dropped CDs in 2010 and spent the next sixteen years selling DVDs and Blu-rays to a shrinking list of members. The business shutting down in September 2026 is a mail-order movie club that outlived the format war, the DVD player in most American living rooms and Blockbuster (https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html) itself. Existing memberships and outstanding purchases are being honored while the company winds down, so members with discs on order are not cut off on the fifteenth. After that date, the sign-up form is the part that goes away first. ## How a penny bought 12 CDs and still made money The offer was real. New members picked eight, ten or twelve titles for one cent, sent no money up front, and agreed to buy a set number of albums later at what the fine print called regular club prices, plus shipping and handling. The giveaway was the cheap part. Mental Floss reported that a free disc cost Columbia House roughly $1.50 to produce, and that the clubs paid music publishers 75 percent of the standard statutory royalty rather than the full rate, a discount written into the way record clubs licensed songs. The money came afterward. Every month the club shipped a Selection of the Month unless the member mailed back a card refusing it before a deadline, a practice regulators call negative option billing. Miss the deadline, and the album arrived anyway at full club price, with shipping charged per disc. On that arithmetic, Mental Floss estimated Columbia House made as much as $7.50 on each album it shipped. The penny was marketing. The postage, the deadlines and the albums nobody meant to order were the business, which is why a generation remembers the club as both a gift and a trap. ## Why 16 million members disappeared Columbia House began in 1955 as the Columbia Record Club, an experiment by Columbia Records in selling music by mail to Americans who lived nowhere near a record store. It took the Columbia House name in the early 1970s and turned into a fixture of late night television. Rolling Stone described those commercials in 2026 as coming on "like the cheesiest sitcom jingle ever," built on the claim that the club was "Big enough to entertain America ... one person at a time!" The CD boom made it enormous. By 1996 Columbia House had about 16 million members and roughly $1.4 billion in annual revenue, the high water mark every later figure gets measured against. Then the reasons to join evaporated one by one. Big box chains undercut club prices, Napster (https://404memoryfound.com/posts/napster-destroyed-music.html) made the catalog free in 1999, and paid downloads made a monthly reply card look absurd. The Sony Discman (https://404memoryfound.com/posts/what-happened-to-sony-discman-portable-cd-player.html) gave way to the iPod (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html), the shelf of club CDs stopped growing, and Columbia House stopped selling music altogether in 2010. ## Who owns Columbia House now? Filmed Entertainment Inc., the company behind the club, filed for Chapter 11 bankruptcy protection in Manhattan in August 2015. The filing put revenue at about $17 million in 2014, down from $1.4 billion in 1996, against $2 million in assets and $62 million in liabilities. The court papers also showed how small the club had become. In the year before the filing, 110,000 people bought something from the Columbia House DVD Club, which is roughly one customer for every 145 members it had at the 1996 peak. No strategic buyer turned up at the auction. John Lippman took the brand for about $1.5 million and announced a relaunch as a vinyl subscription service for 2016, betting that the record revival wanted a club of its own. "You can see a yearning and an interest to try a new format," he told The Wall Street Journal in 2015. The vinyl club never shipped a record. What Lippman kept running instead was the DVD business, quietly, for another decade. That is the Columbia House closing in September 2026, and it is still the same brand that sold 16 million Americans their first CDs. ## What the club left behind Columbia House pressed its own editions of other labels' albums for members, which is why so many American record collections hold copies that do not quite match the store version. Those pressings still circulate, along with the catalogs, the order forms and the sheets of stamps members peeled off to choose their twelve titles. Where to find one today: club-edition LPs and CDs, 1970s and 1980s catalogs and unopened mailers turn up in the vintage ephemera listings on Etsy (https://www.etsy.com/search?q=columbia+house+record+club). Club pressings were made in enormous quantities, so they are one of the cheaper ways to own a piece of the era, and condition matters far more than rarity. The club's longer legacy is the billing model. The monthly shipment you have to decline, the free trial that renews, the box that ships unless you remember to say no: that is negative option billing, and Columbia House ran it on American teenagers for four decades before subscription software made it ordinary. ## Frequently Asked Questions ### Is Columbia House still around in 2026? Only for a few more days. Columbia House stops accepting new orders after September 15, 2026, ending 71 years that began with the Columbia Record Club in 1955. Existing memberships and outstanding purchases are being honored while the company winds down, but no new sign-ups are taken after that date. ### When did Columbia House stop selling CDs? Columbia House closed its music club in 2010 and sold only DVDs and Blu-rays after that. The CD club had peaked in 1996 with about 16 million members and roughly $1.4 billion in annual revenue, then lost members through the download and streaming years until the music side was shut for good. ### Who owns Columbia House now? John Lippman has held the Columbia House brand since 2015, when he bought it out of the Chapter 11 bankruptcy of Filmed Entertainment Inc. for about $1.5 million. His announced plan to relaunch Columbia House as a vinyl subscription service in 2016 never shipped, and the DVD club he kept running is the business closing on September 15, 2026. **Sources:** - CNN Business: Columbia House, yes it still exists, is going out of business (2026): https://www.cnn.com/2026/08/19/media/columbia-house-shut-down - Rolling Stone: Farewell Columbia House, Music Club That Sold Pop Dreams for a Penny (2026): https://www.rollingstone.com/music/music-features/columbia-house-music-club-penny-1235610725/ - CNBC: Parent of Columbia House files for bankruptcy protection (2015): https://www.cnbc.com/2015/08/11/parent-of-columbia-house-files-for-bankruptcy-protection.html - CNN Money: Columbia House is back, and so is vinyl (2015): https://money.cnn.com/2015/12/24/news/companies/columbia-house-vinyl/ - Mental Floss: How Columbia House Made Money Giving Away Music: https://www.mentalfloss.com/article/28036/its-steal-how-columbia-house-made-money-giving-away-music --- # Scan Old Photos and Negatives: DIY vs Mail-In Cost URL: https://404memoryfound.com/posts/scan-old-photos-negatives-2026.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-13 Topics: Then vs Now, Hardware **Summary:** Scanning old photos in 2026 comes down to two routes. A flat-rate mail-in box from a bulk lab such as ScanMyPhotos digitizes an average of 1,800 4x6 prints for $145 with shipping included, while doing it at home means a film-capable flatbed scanner and roughly ten hours at the desk for the same pile. Scan prints at 400 ppi or better, the minimum US federal agencies must meet, and count negatives and slides separately, because labs bill film per frame. **Key facts:** - Cheapest bulk route: ScanMyPhotos flat-rate prepaid box, $145 for an average of 1,800 4x6 prints, shipping included (2026) - Home scanner: Epson Perfection V600 flatbed with two film holders, priced at $229.99 when DPReview tested it in 2019 - Free phone option: Google PhotoScan, which merges several frames to kill glare (method published April 2017) - Resolution to use: 400 ppi minimum for prints under 36 CFR Part 1236, Subpart E; 600 ppi at home - Status today: Costco closed every in-store Photo Center on February 14, 2021; mail-in labs and home scanners took the work ## What does a shoebox of prints cost to digitize? Two numbers decide this for most people. ScanMyPhotos sells a flat-rate prepaid box that holds an average of 1,800 4x6 prints and scans the lot for $145 with shipping included, which works out to about eight cents a picture in 2026. The home route has a different shape. The cost is not per photo, it is the scanner plus your evenings. DPReview priced the Epson Perfection V600 flatbed at $229.99 when it tested the scanner on film in 2019, and that machine takes one or two prints at a time. Run the arithmetic before buying anything. At a generous twenty seconds per print, counting the lid and the straightening, 1,800 pictures is ten hours at the desk. Under a few hundred prints, home scanning wins outright. You already own the evenings, and there is a decent chance the all-in-one printer in the corner already has a flatbed good enough for paper originals. Past a thousand prints the mail-in box costs less than the hardware and takes none of the evenings. That is the whole decision, and it turns on how big the pile is rather than on which service has the better website. Mixed closets change the answer again. If the same shelf holds home video, the calculation looks like the one in the guide to digitizing VHS tapes (https://404memoryfound.com/posts/how-to-digitize-vhs-tapes-2026.html), where finding working hardware is the hard part rather than the price. ## What resolution should you scan at? The federal government has already answered this for prints. Under 36 CFR Part 1236, Subpart E, agencies digitizing permanent photographic records must "produce image files at a minimum of 400 ppi sized to the source document." 400 ppi across a 4x6 print is a 1,600 by 2,400 pixel file. That reprints at the original size and survives a modest crop, which is all a family snapshot is ever asked to do. 600 ppi is the sensible setting at home, because the only thing the extra resolution costs is disk space and the scan is a one-time job. Going to 1,200 ppi on a drugstore print mostly resolves the texture of the paper. Save prints as JPEG at the highest quality the software offers. Reach for TIFF only for the handful of images that are genuinely irreplaceable, since those files are several times larger and a lot of photo apps handle them badly. Whatever the setting, choose it once and do not change it halfway through the box. A folder scanned at three different resolutions is a folder somebody scans again. ## Do negatives and slides need different gear? They do, and the reason is physical. A 35mm negative is roughly one inch by one and a half. A scanner has to magnify that original rather than copy it at size, so the resolution that works for a print is nowhere near enough for film. Federal guidance stops at the edge of the problem. Subpart E of 36 CFR Part 1236 covers paper and photographic prints and leaves negatives, transparencies and roll film out of scope, which is a fair signal that film is the harder job. What film needs is a scanner with a light source in the lid rather than only under the glass. The Epson Perfection V600 ships with two film holders for exactly that, and DPReview noted in 2019 that its highest resolution setting only appears once the software is switched into Pro mode. Dedicated 35mm scanners from Plustek and others do better on grain and on dust, at the cost of taking one strip or a few mounted slides at a time. They are slow in a way that matters when a carousel holds hundreds of slides. Mail-in labs price film per frame rather than per box. That is why a single slide carousel can cost more to digitize than an entire shoebox of prints, and why it is worth counting slides separately before ordering anything. ## Is the phone in your pocket good enough? For photographs that are not leaving somebody else's house, a phone is the only realistic answer. The problem with photographing a photograph is glare: gloss paper, album sleeves and glass frames all throw the ceiling light straight back into the lens. Google's free PhotoScan app gets around that by taking several frames instead of one. The user moves the phone through four corner positions and the app merges what it captured. Google researchers Ce Liu, Michael Rubinstein, Mike Krainin and Bill Freeman described the method on the Google Research blog in April 2017, as software that can "carefully align and combine several slightly different pictures of a print to separate the glare from the image underneath." What a phone cannot do is hold a measured resolution across the whole frame. Every capture sits at a slightly different distance, angle and white balance, so a phone-scanned album ends up looking like an album somebody photographed. Use it at a relative's kitchen table, where the alternative is no copy at all. Use a scanner or a lab for the box that is coming home in the car. ## Where can you still get photos scanned in 2026? Not at the warehouse club, and that still catches people out. Costco closed the photo counter in every one of its US warehouses on February 14, 2021, ending passport photos, photo restoration and the negative and slide digitizing that sat alongside them. The company was blunt about the reason. As PetaPixel reported in January 2021, Costco told members that "since the introduction of camera phones and social media, the need for printing photos has steeply declined, even though the number of pictures taken continues to grow." What is left in 2026 is three places. Mail-in labs work by the box and turn a closet into a download link. Independent camera shops work by the hour and charge like it, which is the right call for damaged or oversized originals. The third place is your own desk. Chain drugstores still print photographs in most US towns, but scanning a stack of old prints is not a counter service the way it was in the one-hour era. Where to find one today. A flat-rate prepaid box from ScanMyPhotos (https://www.scanmyphotos.com/photo-scanning/prepaid-photo-scanning-box.html) runs $145 in 2026 and holds an average of 1,800 4x6 prints with shipping included, which is the cheapest per-print route once the pile passes a thousand. Slides and negatives are billed per frame, so count those before ordering. ## How do you keep the files from vanishing? Scanning is the easy half of the job. One folder on one laptop is not an archive. It is a photograph of your photographs, waiting on a drive failure. Keep three copies: the working folder, an external drive, and a cloud account, with at least one of them outside the house. That covers the two failure modes that actually destroy family pictures, which are hardware death and fire. Name the files while the box is still open on the table. Something like 1987-summer-lake-01 costs ten minutes now and saves an hour of squinting later, and a date in the filename survives every software migration that follows. Then keep the paper. Prints have outlasted a long line of formats that were supposed to replace them, and the companies that bet against paper did not all come out well: Kodak (https://404memoryfound.com/posts/what-happened-to-kodak-digital-camera-bankruptcy.html) went through bankruptcy and Polaroid (https://404memoryfound.com/posts/what-happened-to-polaroid-camera.html) went through a long series of owners before the name came back. A scan is a copy, not a replacement. The negative in the sleeve still holds more detail than the file sitting next to it, and it will still be readable by anything with a light behind it in fifty years. ## Frequently Asked Questions ### How much does it cost to scan old photos in 2026? Scanning old photos costs about eight cents a print through a bulk mail-in box in 2026: ScanMyPhotos charges $145 for a flat-rate prepaid box that holds an average of 1,800 4x6 prints, with shipping included. Scanning at home costs nothing per photo but means buying a scanner first, and DPReview priced the film-capable Epson Perfection V600 flatbed at $229.99 back in 2019. ### What resolution should I use to scan old photos and negatives? Scan prints at 400 ppi or better. That is the minimum US federal agencies must meet under 36 CFR Part 1236, Subpart E, when they digitize permanent photographic records, and 600 ppi is a safe home setting. Negatives and slides need far more, because a 35mm frame is about one inch by one and a half and has to be magnified rather than copied at size. ### Can I scan old photos with my phone instead of a scanner? Yes, and for albums that are staying in somebody else's house it is the practical choice. Google's free PhotoScan app captures several frames of a print and merges them to remove the glare a single phone photo would leave, a method Google researchers described on the Google Research blog in April 2017. A phone still will not hold a measured 400 ppi across the frame the way a flatbed scanner does. **Sources:** - ScanMyPhotos: prepaid photo scanning box: https://www.scanmyphotos.com/photo-scanning/prepaid-photo-scanning-box.html - eCFR: 36 CFR Part 1236, Subpart E, Digitizing Permanent Federal Records: https://www.ecfr.gov/current/title-36/chapter-XII/subchapter-B/part-1236/subpart-E - DPReview: What it's like to scan film on the Epson Perfection V600: https://www.dpreview.com/reviews/1857649076/what-it-s-like-to-scan-film-on-the-epson-v600/ - Google Research Blog: PhotoScan, Taking Glare-Free Pictures of Pictures: https://research.google/blog/photoscan-taking-glare-free-pictures-of-pictures/ - PetaPixel: Costco is Closing All Photo Centers by February 14: https://petapixel.com/2021/01/13/costco-is-closing-all-photo-centers-by-february-14/ --- # Does Nokia Still Make Phones? Who Makes Them Now URL: https://404memoryfound.com/posts/does-nokia-still-make-phones.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-13 Topics: Hardware, Business Blunders **Summary:** Nokia does not make phones, and has not made them since it agreed to sell its handset business to Microsoft on September 3, 2013. Since 2016 the Nokia name on a phone has belonged to HMD, a separate Finnish company that discontinued Nokia-branded smartphones at the start of 2025 and stopped selling phones in the United States in July 2025. Nokia itself is a network equipment maker headquartered in Espoo, Finland. **Key facts:** - Sold its phone business: To Microsoft, announced September 3, 2013, for EUR 5.44 billion in cash - What Microsoft did with it: Wrote down $7.6 billion and announced 7,800 layoffs on July 8, 2015 - Brand licensee: HMD global, exclusive ten-year worldwide license signed May 18, 2016 - Status today: No new Nokia-branded smartphones since early 2025; no HMD or Nokia phones sold in the US since July 2025 - Owner today: Nokia Corporation, Espoo, Finland, a network equipment maker with EUR 19,889 million in 2025 net sales ## What did Nokia actually sell to Microsoft in 2013? On September 3, 2013, Nokia agreed to hand its entire phone-making operation to Microsoft. Microsoft would pay EUR 3.79 billion for substantially all of Nokia's Devices and Services business, plus EUR 1.65 billion to license Nokia's patents, for a total transaction price of EUR 5.44 billion in cash. The headcount says more than the price does. Roughly 32,000 people were expected to transfer to Microsoft at closing, including 4,700 in Finland and 18,300 working directly in the manufacturing, assembly and packaging of products around the world. What stayed in Espoo was the Nokia name, the patent portfolio, the mapping unit and the network equipment business. Nokia the corporation did not go anywhere. It stopped being a phone company and kept everything else. That distinction is what trips people up. The Nokia 3310 (https://404memoryfound.com/posts/what-happened-to-nokia-3310.html) and the Nokia N-Gage (https://404memoryfound.com/posts/what-happened-to-nokia-ngage-gaming-phone.html) came out of a division that has not belonged to Nokia since the sale went through. ## How badly did the Microsoft phone deal fail? Less than two years after that announcement, Microsoft quit. On July 8, 2015 it wrote down $7.6 billion against the Nokia acquisition, announced 7,800 layoffs and took a restructuring charge of roughly $800 million. The write-down was larger than the euro price Microsoft had paid for the hardware business. It had bought the factories, the supply chain and tens of thousands of workers, and the phones coming out of them were not selling. Microsoft said the job cuts were "primarily in the phone business", as TechCrunch reported on the day in 2015. The same report described a unit delivering lower revenue and lower unit volume than Microsoft had originally expected. So the sequence people usually get wrong runs like this. Nokia did not go bankrupt, and Microsoft did not shut Nokia down. Nokia sold one division, Microsoft ran that division into the ground in under two years, and the Nokia name was left sitting in Finland with nobody making a phone to put it on. ## Who put the Nokia name on phones after 2016? On May 18, 2016, Nokia signed the agreement that has confused shoppers ever since. Nokia Technologies granted HMD global Oy, a newly founded company based in Finland, an exclusive global license to create Nokia-branded mobile phones and tablets for the next ten years. Nokia did not build those phones, did not own HMD and did not staff it. It licensed a trademark and collected royalties. Ramzi Haidamus, then president of Nokia Technologies, said so in the announcement: "Instead of Nokia returning to manufacturing mobile phones itself, HMD plans to produce mobile phones and tablets that can leverage and grow the value of the Nokia brand." HMD spent the license years selling budget Android handsets and cheap 4G feature phones, with a run of retro reissues rebuilding the 3310 and the 3210 for buyers who remembered the originals. It was a real business, and it was nothing like the company that had once outsold everyone. Anyone shopping for a phone in a US carrier store in the years after 2016 and seeing the Nokia logo on the box was looking at an HMD product. The relationship was a naming deal, closer to how Motorola Razr V3 (https://404memoryfound.com/posts/motorola-razr-v3-coolest-phone-ever-made.html) nostalgia gets recycled than to a corporate revival. ## Can Americans buy a new Nokia phone in 2026? No. HMD discontinued the Nokia-branded smartphone range at the start of 2025 and moved the models into a legacy section of its site marked unavailable. Then it left the country outright. In July 2025 HMD confirmed it was pulling back from the United States and ending sales of both Nokia-branded and HMD-branded phones there, citing a "challenging geopolitical and economic environment", which in practice meant tariffs landing on hardware with very thin margins. HMD said it would keep honoring warranty coverage and supporting existing US customers through its global teams. The exclusive license that let HMD use the Nokia name on phones expires in 2026, and HMD has already replaced it with its own brand, shipping handsets badged HMD instead. Nokia-branded feature phones, the inexpensive 4G handsets sold in India, Africa and the Middle East, have continued past the smartphone cutoff under an extended arrangement, but they were never a US product and are not one now. So the answer to the status question is layered. Nokia has not made a phone since 2014, HMD stopped making Nokia smartphones in 2025, and no new Nokia handset is sold at retail in the United States. Where to find one today. The originals still turn up in quantity, because Nokia built them in enormous numbers and they refuse to break. Working 3310s, 8210s, 6310s and N-Gages circulate along with the parts that keep them alive: charger tips, back covers, swappable fascias and batteries. Most of it surfaces on Etsy's search page for vintage Nokia phones (https://www.etsy.com/search?q=vintage%20nokia%20phone). Price tracks condition and completeness far more than model, so a scuffed handset alone costs a fraction of the same phone boxed with its manual and charger, and any unit that has sat for 20 years needs a fresh battery before it holds a charge. ## What does Nokia the company do now? Nokia is headquartered in Espoo, Finland and sells network equipment: the radios, routers and optical gear that carriers use to run mobile and fixed broadband networks. Its customers are telecom operators and governments, not shoppers. For 2025 the company reported net sales of EUR 19,889 million, comparable operating profit of EUR 2.0 billion and free cash flow of EUR 1.5 billion, with roughly 78,000 employees on average across the year. Justin Hotard runs it as president and chief executive. From January 1, 2026 Nokia operates through two segments, Network Infrastructure and Mobile Infrastructure. Consumer handsets do not appear anywhere in that structure, and brand licensing is a rounding error against a business of that size. That is the whole answer in one line. Nokia is a live, profitable company that has not built a phone in over a decade, and every Nokia handset sold since the 2016 agreement came from a licensee. ## Frequently Asked Questions ### Are Nokia phones still being made in 2026? Nokia-branded smartphones are not. HMD, the Finnish company that has held the exclusive license to the Nokia name on phones since the agreement signed on May 18, 2016, discontinued the Nokia smartphone range at the start of 2025 and stopped selling phones in the United States in July 2025. Nokia-branded feature phones are still produced under an extension of that license for markets outside the US. ### Who owns the Nokia brand today? Nokia Corporation of Espoo, Finland owns the Nokia trademark and has never sold it. It licensed the name for phones and tablets to HMD global in a ten-year deal signed on May 18, 2016, and its own money comes from network equipment: EUR 19,889 million in net sales in 2025, with roughly 78,000 employees. ### Why did Nokia stop making phones? Nokia sold the business rather than closing it. On September 3, 2013 Nokia agreed to sell substantially all of its Devices and Services operation to Microsoft for EUR 3.79 billion, with a further EUR 1.65 billion for a patent license, and about 32,000 staff moved across. Microsoft wrote down $7.6 billion of that purchase on July 8, 2015 and announced 7,800 layoffs. **Sources:** - Microsoft to acquire Nokia's Devices & Services business, license Nokia's patents and mapping services (Microsoft, September 3, 2013): https://news.microsoft.com/source/2013/09/03/microsoft-to-acquire-nokias-devices-services-business-license-nokias-patents-and-mapping-services/ - Microsoft Writes Down $7.6B Of Its Nokia Acquisition, Announces 7,800 Layoffs (TechCrunch, July 8, 2015): https://techcrunch.com/2015/07/08/microsoft-writes-down-7-6b-of-its-nokia-acquisition-announces-7800-layoffs - Nokia signs strategic brand and intellectual property licensing agreement enabling HMD global to create a new generation of Nokia-branded mobile phones and tablets (Nokia, May 18, 2016): https://www.globenewswire.com/news-release/2016/05/18/841089/27866/en/Nokia-signs-strategic-brand-and-intellectual-property-licensing-agreement-enabling-HMD-global-to-create-new-generation-of-Nokia-branded-mobile-phones-and-tablets.html - HMD retreats from US market, ends Nokia phone sales amid economic pressures (TechSpot, July 2025): https://www.techspot.com/news/108640-hmd-global-retreats-us-market-ends-nokia-phone.html - Nokia Corporation Financial Report for Q4 2025 and full year 2025 (Nokia): https://www.nokia.com/newsroom/nokia-corporation-financial-report-for-q4-2025-and-full-year-2025/ --- # Who Owns Borders Now? The $13.9 Million Brand Sale URL: https://404memoryfound.com/posts/who-owns-borders-now-bookstore.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-13 Topics: Business Blunders, Then vs Now **Summary:** Borders is gone in the United States. The chain liquidated its last 399 stores in 2011, and Barnes & Noble has owned the Borders name ever since, bought at a bankruptcy auction for $13.9 million along with a database of 48 million customers. The only Borders-branded stores still trading are in the Gulf, run by Al Maya Group. **Key facts:** - Launched: 1971, Ann Arbor, Michigan, as a used bookstore run by Tom and Louis Borders - Peak size: 1,249 Borders and Waldenbooks stores in 2003 - Closed: September 18, 2011, after 399 stores were liquidated and 10,700 jobs went - Owner today: Barnes & Noble, which paid $13.9 million for the name, the domains and 48 million customer records in 2011 - Status today: No Borders stores in the US. Borders-branded shops still operate in the Gulf under Al Maya Group ## What Barnes & Noble bought out of the wreckage By the late summer of 2011 the most valuable thing Borders still owned was not a store. It was the name on the sign, the web address and a list of everyone who had ever signed up for a rewards card. Barnes & Noble (https://404memoryfound.com/posts/is-barnes-and-noble-still-open.html) bought all three at a bankruptcy auction on September 14, 2011, for $13.9 million. The package covered the Borders trademarks, the Waldenbooks and Brentano's names, the internet domains, and a customer database holding records on 48 million people. Al Maya International, the franchisee that had been running Borders stores in the Middle East, bought Borders trademarks for five countries at the same auction for $500,000. The customer list was the part that nearly broke the deal. Privacy advocates told the bankruptcy court that those 48 million people had given their names, email addresses and phone numbers to Borders, not to its largest competitor, and the judge required that each of them be offered a chance to opt out. So in October 2011 more than 40 million former Borders shoppers opened an email from Barnes & Noble chief executive William Lynch. "First of all let me say Barnes & Noble uniquely appreciates the importance bookstores play within local communities, and we're very sorry your Borders store closed," it began. Barnes & Noble has controlled borders.com ever since. The Borders online store stopped taking orders on September 27, 2011, and on October 14 the address began pointing at barnesandnoble.com. ## Are there any Borders stores left in 2026? Not in the United States. The last American Borders stores closed in September 2011, and nothing has carried the name in this country since. The brand kept trading abroad, because Borders Group had sold franchise rights overseas long before it failed. Al Maya Group, a distributor based in the United Arab Emirates, has run Borders-branded book, toy and stationery shops in the Gulf since 2006, and it bought the Middle East rights outright in 2015. Those Gulf stores are the clearest survivor of the brand. Malaysia had the other one. Berjaya Books ran Borders stores there under a licence bought from the liquidators, and they outlasted the American chain by 12 years before shutting in August 2023. In the United Kingdom the name went early. Borders UK was sold to a private investment group in 2007 and collapsed into administration in 2009, two years before its former American parent filed for Chapter 11. Where to find one today. Nothing new carries a Borders logo, so what is left is memorabilia: rewards cards, name badges, tote bags, bookmarks, staff shirts and store signage that walked out of the buildings during the 2011 going-out-of-business sales. Most of it surfaces on Etsy's search page for Borders bookstore items (https://www.etsy.com/search?q=borders%20bookstore). Price depends almost entirely on how much store branding is on the object, so a sign or a badge costs many times what a used gift card does. ## How Kmart stitched Borders and Waldenbooks together Borders began in 1971 as a used bookstore in Ann Arbor, Michigan, opened by the brothers Tom and Louis Borders. Its real advantage was software. The brothers built an inventory system that predicted which titles a particular town would buy, which let their stores stock tens of thousands of titles instead of the few thousand a mall bookshop carried. That caught the attention of Kmart (https://404memoryfound.com/posts/is-kmart-still-open-stores-left.html), which had bought the mall chain Waldenbooks in 1984 and wanted the superstore end of the market too. Kmart bought Borders in 1992 and ran both chains. The arrangement lasted three years. Kmart spun the business off in 1995 as Borders Group, listed in New York and headquartered in Ann Arbor, with Waldenbooks as a subsidiary. It worked for about a decade. Borders Group peaked in 2003 with 1,249 stores trading as Borders and Waldenbooks, a mix of superstores, mall shops, airport outlets and Borders Express units, with locations in the United Kingdom, Singapore and Australia as well. ## Why did Borders hand its website to Amazon? In 2001 Borders decided that running an online bookstore was someone else's job. It signed a deal that turned borders.com into a storefront operated by Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html), which handled the catalog, the checkout and the customer data. The arrangement ran until 2008. For seven years, a Borders customer who went looking for a book online was handed to the company that would eventually take the business. Barnes & Noble spent those same years building its own site, and in 2009 its own e-reader. Amazon had launched the Kindle in 2007. Borders did not open an e-bookstore of its own until July 2010, and that one ran on Kobo rather than anything Borders built. The rest of the floor plan was a second bet going the same way. Borders superstores devoted large areas to CDs and DVDs, which made sense in 1999 and was dead weight by 2008. ## How fast did the collapse actually happen? Borders lost money every year from 2006 on, and the cutting started at the small end. In January 2010 the company closed about 200 Waldenbooks stores, close to two-thirds of that chain. Borders Group filed for Chapter 11 bankruptcy protection on February 16, 2011, with 642 stores, and said it would close roughly 200 of them while it hunted for a buyer or a lender willing to fund a turnaround. Neither showed up. On July 18, 2011, the company asked the court for permission to liquidate the 399 stores still trading, a decision that ended 10,700 jobs. Going-out-of-business sales began on July 22. Borders Group president Mike Edwards said in the July 2011 announcement that the company had been fighting headwinds for some time, and that "the rapidly changing book industry, eReader revolution, and turbulent economy, have brought us to where we are now." The original Ann Arbor shop, known inside the company as Borders No. 1, closed that September. The last American Borders stores shut on September 18, 2011, five months short of the chain's 40th anniversary in its home town. ## Was Amazon really what killed Borders? Amazon takes the blame, and it was a real predator, but the order of events matters. Borders Group was still expanding in 2003, when Amazon was already eight years old. What broke the company was the shape of its own estate. Borders had signed long leases on very large buildings, sized for a shopper who would buy a book, a CD and a DVD in one trip. When music and video moved to downloads, a big share of that floor space stopped earning and the lease stayed. The digital side was a decision rather than an accident. Barnes & Noble kept its own website and its own customer list through the 2000s. Borders rented both from its biggest rival until 2008, then took them back after the online book market had already formed. Waldenbooks made the arithmetic worse. Mall traffic fell through the 2000s and Borders Group owned hundreds of small mall stores it could not fix, which is why 200 of them went in a single month in 2010. None of that was Amazon's doing. It only meant that when the book business moved, Borders had nothing left to move with. ## Frequently Asked Questions ### Is Borders still around in 2026? Borders is not around in the United States. The chain filed for Chapter 11 in February 2011, liquidated the last 399 stores that summer and closed its final American locations on September 18, 2011. Borders-branded shops still trade in the Gulf under Al Maya Group, which has held the Middle East rights to the name since 2015. ### Who owns the Borders name now? Barnes & Noble has owned the Borders name in the United States since 2011. It paid $13.9 million at a September 2011 bankruptcy auction for the Borders, Waldenbooks and Brentano's trademarks, the internet domains and a database of 48 million Borders customers. Al Maya International bought the Borders trademarks for five other countries at the same auction for $500,000. ### What happened to Waldenbooks? Waldenbooks was the mall bookstore chain Kmart bought in 1984 and brought under Borders Group in 1992. Borders closed about 200 Waldenbooks stores in January 2010, and the rest went in the liquidation that ended in September 2011. The Waldenbooks trademark has belonged to Barnes & Noble since that year. **Sources:** - CNNMoney: Borders to liquidate remaining stores (July 18, 2011): https://money.cnn.com/2011/07/18/news/companies/borders_liquidation/index.htm - AnnArbor.com: Borders' rise and fall, a timeline of the bookstore chain's 40-year history: https://www.annarbor.com/business-review/borders-rise-and-fall-a-timeline-of-the-bookstore-chains-40-year-history/ - Publishers Weekly: Barnes & Noble Big Winner in Borders IP Auction (2011): https://www.publishersweekly.com/pw/by-topic/industry-news/bookselling/article/48719-barnes-noble-big-winner-in-borders-ip-auction.html - AnnArbor.com: Barnes & Noble CEO: 'We're very sorry your Borders store closed' (2011): https://www.annarbor.com/business-review/barnes-noble-ceo-were-very-sorry-your-borders-store-closed/ - Wikipedia: Borders (retailer): https://en.wikipedia.org/wiki/Borders_(retailer) --- # Is Hollywood Video Still Around? Every Store Closed URL: https://404memoryfound.com/posts/is-hollywood-video-still-around.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-12 Topics: Business Blunders, Then vs Now **Summary:** Hollywood Video is not still around. Its parent company, Movie Gallery, told a bankruptcy court in May 2010 that it would close more than 1,900 remaining US stores and liquidate, and no Hollywood Video has opened since. What outlived the chain was its paperwork: about $244 million in disputed late fees owed by 3.3 million former customers. **Key facts:** - Launched: October 1988, Hollywood Entertainment Corporation - Peak size: 2,006 US stores and 715 Game Crazy departments at the end of 2004 - Revenue in 2004: $1.782 billion - Sold to: Movie Gallery, 2005, about $1 billion - Status today: Gone. The last US stores were liquidated in 2010 ## How an Oregon rental shop grew into Blockbuster's only real rival Hollywood Entertainment Corporation opened its first video store in October 1988. For most of the 1990s the company grew the obvious way, by putting a big, bright store across the road from whatever Blockbuster had already built. The strategy worked. As of December 31, 2004, Hollywood Entertainment operated 2,006 Hollywood Video stores in the United States, according to the annual report it filed with the Securities and Exchange Commission. Revenue that year was $1.782 billion. There was a second business running inside the same walls. The 2004 filing also counted 715 Game Crazy departments, the used-game counters where customers could trade in titles the way they did at GameStop (https://404memoryfound.com/posts/is-gamestop-still-in-business.html), without leaving the rental aisle. That combination made Hollywood Entertainment the number two video rental company in the country, and the only chain with enough locations to give Blockbuster a real fight. It was also, by 2004, a company whose entire model assumed people would keep driving somewhere to pick up a plastic disc and then drive back to return it. ## The $1 billion deal that took down both companies In 2005 the chain was bought by Movie Gallery, a rental company based in Dothan, Alabama, that had built its business on small towns and strip malls rather than dense suburbs. The price was about $1 billion. On paper the two halves fit. Movie Gallery had rural coverage, Hollywood Video had the metro stores, and together they were second only to Blockbuster (https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html) in the United States. The balance sheet is what broke. On October 16, 2007, Movie Gallery filed for Chapter 11 bankruptcy protection after months of struggling with the debt from the Hollywood Entertainment purchase. In its filing in the US Bankruptcy Court for the Eastern District of Virginia in Richmond, the company listed debts of between $500 million and $1 billion against assets of between $10 million and $50 million. Movie Gallery announced plans to shut roughly 520 unprofitable stores and asked the court for $150 million from Goldman Sachs to fund operations during the case. Its chief restructuring officer, Steve Moore, told the court the company faced "looming defaults" on its loan agreements. ## Why the second bankruptcy ended in a full liquidation Movie Gallery came out of that first bankruptcy in 2008 and went back in less than two years later. It filed for Chapter 11 again in February 2010, and this time the pressure was not coming from another chain of stores. It was coming from the mail, from vending machines and from broadband. Netflix was shipping discs to the door, Redbox kiosks (https://404memoryfound.com/posts/is-redbox-still-around-kiosks.html) were charging a dollar a night outside grocery stores, and streaming was starting to make the drive itself look silly. In May 2010 the company, by then based in Wilsonville, Oregon, filed a notice with the bankruptcy court saying it would close its more than 1,900 remaining stores and liquidate. The agreement told the court the shutdown was in the "best interests" of the company and its creditors. Game Crazy went with it. So did the rental accounts, the membership files and roughly two decades of late-fee records, which turned out to be the most durable asset the company had. ## What is left of the chain in 2026 Nothing operating. No Hollywood Video store has rented a movie in the United States since the 2010 liquidation, which makes 2026 the sixteenth year the brand has existed only as signage ghosts and search traffic. There is no successor chain either. Hollywood Entertainment was absorbed into Movie Gallery in 2005 and Movie Gallery was wound down in 2010, so the name is not attached to a working rental business anywhere. What survives is paper and plastic. Laminated membership cards, employee name badges, promotional standees and the yellow-and-black clamshell rental cases all left the stores during the going-out-of-business sales and ended up in collectors' hands. Where to find one today. Hollywood Video memorabilia turns up on Etsy's search page for Hollywood Video memorabilia (https://www.etsy.com/search?q=hollywood%20video%20memorabilia), mostly membership cards, badges and rental cases. Prices swing widely by condition and by how much store branding is on the item, so check recent sold listings before paying up for anything described as rare. ## The late fees that outlived the company When Hollywood Video and Movie Gallery were liquidated, the estate still held a mountain of unpaid rental charges. The approved plan created a liquidating trust to collect an estimated $244 million reportedly owed by 3.3 million customers. Those accounts were handed to debt collectors, and the complaints started almost immediately. Former customers said they were being billed for fees they had never been told about, and in many cases the first they heard of it was a rejected credit application. Mary Lobdell, an assistant attorney general in Washington state, said her office "received a flurry of complaints from Washington residents who told us they didn't owe the fees," in the 2011 announcement of the resulting settlement. That settlement, signed by the attorneys general of all 50 states and the District of Columbia, required the trust to rescind adverse credit reports, to stop collection whenever a consumer disputed a notice until the account was verified, and to collect only the lesser of a late fee or a product fee rather than both. ## Which video rental chain actually lasted the longest Not Blockbuster, and not Hollywood Video. The last national video rental chain in the United States was Family Video, owned by Highland Ventures, which kept renting movies for more than a decade after the 2010 liquidation. It stopped on January 6, 2021. Family Video announced it would end rentals and close its remaining stores, roughly 250 of them, after the pandemic emptied its aisles and the studios stopped releasing the new titles that drove rental traffic. Keith Hoogland, the chief executive of Highland Ventures, put it plainly in the announcement reported by Variety in January 2021: "The impact of COVID-19, not only in foot traffic but also in the lack of movie releases, pushed us to the end of an era." That is the honest end of the format. Hollywood Video lost to the mail and the kiosk in 2010, Blockbuster shrank to a single store, and the chain that outlasted both was finished off by a year with almost no new releases to rent. ## Frequently Asked Questions ### When did Hollywood Video close? Hollywood Video closed in 2010. Parent company Movie Gallery filed for Chapter 11 bankruptcy protection in February 2010 and then told the bankruptcy court in May 2010 that it would close its more than 1,900 remaining stores and liquidate. No Hollywood Video store has operated in the United States since that year. ### Who owned Hollywood Video? Hollywood Video was owned by Hollywood Entertainment Corporation, which opened the first store in October 1988 and ran 2,006 Hollywood Video stores in the United States as of December 31, 2004. Movie Gallery, based in Dothan, Alabama, bought Hollywood Entertainment for about $1 billion in 2005 and owned the chain until the 2010 liquidation. ### What was the last video rental chain in America? Family Video was the last national video rental chain in the United States. Owned by Highland Ventures, it stopped renting movies on January 6, 2021 and closed its remaining stores, roughly 250 of them, eleven years after Hollywood Video and Movie Gallery were liquidated in 2010. **Sources:** - Hollywood Entertainment Corporation, Form 10-K for 2004 (SEC EDGAR): https://www.sec.gov/Archives/edgar/data/0000905895/000090589505000003/r10k-04.txt - NBC News: Video giant Movie Gallery files for bankruptcy (2007): https://www.nbcnews.com/id/wbna21324720 - NBC News: No. 2 rental chain Hollywood Video to close (2010): https://www.nbcnews.com/id/wbna37068852 - Washington State Attorney General: Attorneys general rewind movie rental late-fees for customers nationwide (2011): https://www.atg.wa.gov/news/news-releases/attorneys-general-rewind-movie-rental-late-fees-customers-nationwide - Variety: Family Video, Last National Rental Chain, Is Shutting Down All Remaining Stores (2021): https://variety.com/2021/digital/news/family-video-shutting-down-all-stores-1234879655/ --- # Is Barnes and Noble Still Open? 720+ Stores in 2026 URL: https://404memoryfound.com/posts/is-barnes-and-noble-still-open.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-12 Topics: Then vs Now, Business Blunders **Summary:** Barnes & Noble is still open, and there is more of it than there was five years ago. The chain passed 720 US bookstores in 2026, up from fewer than 600 in 2023, and has a target of 60 new stores for the year. Elliott Investment Management has owned it since 2019, when it paid about $683 million. **Key facts:** - Owner today: Elliott Investment Management, since 2019 - Price paid: About $683 million, or $6.50 a share, in the deal announced June 7, 2019 - Stores today: More than 720 in the United States in 2026, up from fewer than 600 in 2023 - Opening in 2026: 60 new stores, with leases confirmed in nine states and Washington, DC - Status today: Open and expanding, with about 28 stores opened since January 2026 ## The chain that was supposed to go the way of Borders For most of the 2010s, the safe prediction about Barnes & Noble was that it would not make it through the decade. Borders (https://404memoryfound.com/posts/who-owns-borders-now-bookstore.html), the other big American bookstore chain, ran out of road first. It filed for bankruptcy protection in February 2011, failed to find a buyer, and on July 18, 2011 said it would liquidate. Borders still operated 399 stores and employed about 10,700 people on the day that decision was announced, and the last of those stores were gone by the end of September. Barnes & Noble looked like the next name on the list. Amazon had taken the price argument, then the selection argument, then the convenience argument, and the chain spent more than 15 years closing stores rather than opening them. That is the part most people remember, and it is why the question gets typed into a search box at all. The 2026 answer is not the one the 2011 headlines set up. Barnes & Noble is open, it runs more stores than it did five years ago, and it is the rare mall-era American retailer adding locations instead of shedding them. The last Blockbuster (https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html) is one store in Oregon. Barnes & Noble has hundreds. ## How many Barnes & Noble stores are open in 2026? More than 720, in every part of the country. The chain passed that mark in 2026 after falling below 600 stores as recently as 2023. The growth is not a rounding error either. Barnes & Noble has opened roughly 28 bookstores since the start of 2026, spread across 15 states and Washington, DC, and says dozens more are due before the end of the year. The target for 2026, announced in December 2025, is 60 new stores. Leases were confirmed in nine states plus the District of Columbia: California, Colorado, Florida, Georgia, Illinois, Ohio, Texas, Virginia and Washington state, along with Washington, DC. "Barnes & Noble is enjoying a period of tremendous growth as the strategy to hand control of each bookstore to its local booksellers has proven so successful," the company said in the statement to Fox Business announcing those openings. The same statement made the contrast explicit: sales are strong in the stores the chain already has, and the openings follow more than 15 years of declining store numbers. For a reader checking whether the branch they remember is still trading, the short version is that Barnes & Noble is opening bookstores rather than closing them for the first time in a generation. ## Who owns Barnes & Noble now? Elliott Investment Management, the activist investment firm, has owned Barnes & Noble since 2019. The deal was announced on June 7, 2019: $6.50 a share in cash, valuing the bookseller at about $683 million including debt. That price was a 43 percent premium to the volume weighted average closing price over the ten trading days to June 5, 2019, which says plenty about what the market thought the chain was worth by then. The purchase closed that August. Elliott already owned Waterstones, the largest bookstore chain in the United Kingdom, bought in 2018. Putting the two chains under one owner also put them under one bookseller: James Daunt, who ran Waterstones, took the chief executive job at Barnes & Noble as well and still holds both. Ownership matters more than usual in this story. An investment firm buying a fading retailer is normally the opening scene of a liquidation, the way it went at Kmart (https://404memoryfound.com/posts/is-kmart-still-open-stores-left.html) and Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html). Elliott kept the stores and changed how they were run instead. ## What James Daunt changed inside the stores Daunt's repair job was not an app or a loyalty scheme. It was taking buying decisions away from head office and handing them to the people standing in the shop. Under the old system, publishers paid for placement and a central team decided what every store in the country piled up by the front door. Under Daunt, individual stores pick much of what they stock and how it is displayed, which is why two Barnes & Noble branches in the same state no longer look identical. "We are trying to run really good bookstores, most of that effort is being done at the local level," Daunt told Fortune in 2026. His stated principle is shorter still: "bookselling is local." The change showed up in the accounts before it showed up in the store count. Daunt closed weak locations, cleared out the clutter that had crept into the floor plans, and led Barnes & Noble to its first year of profit since the 2008 recession, according to Fortune's August 2026 account of the turnaround. The leases came after that, not before. By 2026 the company is on course to operate more locations than it had in 2019, the year it was sold. ## Why the bookstore business itself came back Barnes & Noble is not growing against the market. It is growing with one. Independent bookshops are multiplying too. Some 605 new independent bookstores opened in the United States in 2025, up 87 percent on the year before, according to American Booksellers Association figures cited by Fortune in 2026. Daunt does not treat those shops as the enemy, which is a reversal for a chain that spent the 1990s cast as the villain putting neighborhood bookstores out of business. "The better our bookstores, the more books are sold, the more the market expands," he told Fortune in 2026, arguing that "it's not a zero-sum game." The rest of the answer is physical. A bookstore is one of the few shops left that people enter to spend an hour rather than to complete a transaction, and an hour is the thing a website cannot ship. Amazon still sells more books than anyone, and Amazon survived the dot-com crash (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html) partly by selling them cheaper than any store could. What it never built was somewhere to stand. ## Frequently Asked Questions ### Is Barnes & Noble going out of business? No. Barnes & Noble is not going out of business in 2026. The chain runs more than 720 bookstores in the United States, has opened about 28 of them since January 2026, and has a target of 60 new stores for the year, having fallen below 600 stores as recently as 2023. ### How many Barnes & Noble stores are left in 2026? More than 720 across the United States in 2026, up from fewer than 600 in 2023. Barnes & Noble confirmed 2026 leases in nine states and Washington, DC, among them California, Colorado, Florida, Georgia, Illinois, Ohio, Texas, Virginia and Washington state. ### Who owns Barnes & Noble now? Elliott Investment Management has owned Barnes & Noble since 2019, under a deal announced on June 7, 2019 at $6.50 a share and worth about $683 million including debt. Elliott also owns the British chain Waterstones, and James Daunt has run both booksellers since the takeover. **Sources:** - CNN Money: Borders to liquidate remaining stores (2011): https://money.cnn.com/2011/07/18/news/companies/borders_liquidation/index.htm - CNBC: Elliott Management to acquire Barnes & Noble for $683 million (2019): https://www.cnbc.com/2019/06/07/elliott-management-to-acquire-barnes-noble-for-683-million.html - Fox Business: Barnes & Noble to open 60 new stores in 2026: https://www.foxbusiness.com/retail/barnes-noble-open-60-new-stores-2026 - Fortune: Barnes & Noble's surprising comeback hinges on a radical idea (2026): https://fortune.com/2026/08/04/barnes-nobles-comeback-radical-idea-james-daunt/ - Yahoo Finance: Barnes & Noble tops 720 stores in major revival (2026): https://ca.finance.yahoo.com/news/barnes-noble-tops-720-stores-233615268.html --- # Is an Anbernic Handheld Worth It for Game Boy? URL: https://404memoryfound.com/posts/anbernic-handheld-worth-it-game-boy.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-12 Topics: Gaming, Hardware **Summary:** An Anbernic handheld is worth it for Game Boy, Game Boy Color and Game Boy Advance games, which are the easiest systems these cheap Linux machines emulate, and the newest clamshell model launched under $60 in August 2026. The catch is not performance but supply and law: Anbernic suspended all orders shipping from China to the United States in April 2025 over tariffs, and the microSD cards these devices often arrive with are full of games nobody licensed. Buy the hardware bare, supply your own files, and check which warehouse ships the order before you pay. **Key facts:** - What it is: A line of budget Linux and Android emulation handhelds from Anbernic, a Chinese manufacturer - Price today: Under $60 for Anbernic's newest Game Boy Advance SP-style clamshell, August 2026 - Status today: Still selling in 2026, but direct China-to-US shipping was suspended on April 21, 2025 and US orders depend on warehouse stock - What you do not get: Games. Anbernic ships no licensed titles; preloaded microSD cards come from resellers, not from the maker - Cheapest licensed alternative: Nintendo Switch Online at $19.99 a year for Game Boy, $49.99 a year with the Expansion Pack for Game Boy Advance ## What is an Anbernic, and what does one cost in 2026? Anbernic is a Chinese manufacturer that builds small emulation handhelds and sells them direct. The devices look like Nintendo hardware from the 1990s and early 2000s, run Linux or Android, and arrive with no licensed games on them. The pitch is price. In August 2026 Anbernic released a clamshell model styled after the Game Boy Advance SP and priced it under $60, which is less than most people pay for a single working vintage handheld in good condition. That money buys a color screen, a rechargeable battery, a microSD slot and a menu that sorts emulators by system. It does not buy a warranty you can use easily from the United States, and it does not buy games. Anbernic also ships new models constantly, several a year, with overlapping names built from the same three or four letters. The practical effect for a buyer is that the device in a review from last season may already have been replaced by one with a different chip and almost the same name. ## How well does an Anbernic handle Game Boy and Game Boy Advance games? Very well, and that is the honest core of the answer. Emulating the Game Boy, Game Boy Color and Game Boy Advance is the easiest job any of these machines gets asked to do. Those consoles are decades old and thoroughly documented, so even the cheapest current Anbernic chips run them at full speed with the sound intact. Screens matter more than processors here. Game Boy Advance output is a low resolution by modern standards, and it scales cleanly onto the small 4:3 and square panels Anbernic favors, so the picture looks sharp rather than smeared. The trouble starts above that line. Nintendo 64, Dreamcast (https://404memoryfound.com/posts/why-sega-stopped-making-consoles-dreamcast.html), PlayStation Portable and Nintendo DS emulation on the cheap models runs anywhere from fine to a slideshow depending on the game, and the product listings rarely say which. Reviewers test the same handful of titles, which hides the problem. If your library is Game Boy, Game Boy Color, Game Boy Advance, NES, SNES and Genesis, a cheap Anbernic does the job with room to spare. The same holds for Sega Genesis games (https://404memoryfound.com/posts/sega-genesis-games-how-to-play-2026.html) and for the Sega Game Gear (https://404memoryfound.com/posts/what-happened-to-sega-game-gear-handheld.html) library. If you bought one expecting smooth Dreamcast, you bought the wrong tier of device. ## Can you still buy an Anbernic in the United States? This is the part of the answer that changed, and it changed on April 21, 2025, when Anbernic told customers it was cutting off direct shipping from China to the United States. "Due to changes in U.S. tariff policies, we will be suspending all orders shipping from China to the United States starting today," the company wrote in a notice reported by Engadget in 2025. Two things forced it. Tariffs on Chinese goods had reached 145%, and the de minimis exemption that let packages worth under $800 enter the United States duty free was being withdrawn. A handheld that leaves Shenzhen priced under $60 does not survive that arithmetic as a direct-mail purchase. Anbernic told buyers to prioritize stock already sitting in its United States warehouse, which was not subject to the new import duties. That is still the shape of the problem in 2026. The product exists, the price on the page looks normal, and whether the order can actually be filled depends on which warehouse the listing ships from. So check the shipping origin on the product page before paying, and treat any United States warehouse listing as limited stock rather than a standing supply. ## Is it legal to put Game Boy ROMs on one? The hardware is legal and emulators are legal. The game files are where it goes wrong, and the trade around these handhelds has leaned on that ambiguity for years. Plenty of resellers ship the devices with a microSD card already loaded with thousands of games. Nobody licensed those, and the card is usually assembled by the store rather than by the manufacturer. The risk is not theoretical. On April 15, 2025 Italy's Guardia di Finanza raided the home and office of the retro gaming reviewer known as Once Were Nerd and seized more than 30 emulation handhelds, Anbernic models among them, over reviews of devices that shipped carrying copyrighted games. The Italian copyright provision cited allows penalties of up to 15,000 euros and three years in prison, and no formal charges had been disclosed when Android Authority reported the case in 2025. That case is Italian and United States law is different, but the buying advice holds on both sides of the Atlantic. Order the handheld without the loaded card, and put on it only what you can account for: homebrew, games you bought as digital downloads, or cartridges you own and dumped yourself. ## Is an Analogue Pocket or Switch Online a better answer? Those are the two real alternatives, and they solve different problems. The Analogue Pocket is not an emulator. It uses programmable logic to behave like the original hardware, and it takes real cartridges, so the games already in your drawer work in it. It launched at $199.99 in December 2021 and rose to $219.99, and in March 2026 Analogue posted that "Analogue Pocket is now $239.99 due to recent tariff announcements," as Time Extension reported that month. That is four times the price of a cheap Anbernic, for one family of systems rather than a dozen. Nintendo Switch Online is the licensed route. A standard membership costs $19.99 a year and carries a Game Boy library. The Expansion Pack costs $49.99 a year and adds Game Boy Advance. You do not pick the catalog, and the library disappears when you stop paying. The three routes line up cleanly. Switch Online is cheapest and fully licensed but narrow. The Analogue Pocket is the most faithful and the most expensive. An Anbernic sits in the flexible middle, and that flexibility is exactly what makes filling it legally your job. A real Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html) is a fourth option, and the one that usually needs a screen replacement before anyone enjoys it indoors. ## Where to find one today Anbernic sells direct from its own store, where the range in 2026 starts under $60 for the newest Game Boy Advance SP-style clamshell released in August 2026 and climbs through larger Android models that cost a good deal more. Before ordering from the United States, open the product page and confirm which warehouse fills it, because the China route has been closed to direct American orders since April 2025 and warehouse stock comes and goes. The current lineup is listed at Anbernic's handheld game console page (https://anbernic.com/collections/handheld-game-console). Buy it bare and bring your own files. ## Frequently Asked Questions ### Is an Anbernic handheld worth it in 2026? For Game Boy, Game Boy Color and Game Boy Advance games, an Anbernic handheld is worth it: those systems run at full speed on even the cheap models, and the newest Game Boy Advance SP-style clamshell launched under $60 in August 2026. It is worth much less if you expect a budget model to handle Dreamcast or PlayStation Portable smoothly, or if you are not prepared to source your own game files, because Anbernic sells no licensed games with the hardware. ### Do Anbernic handhelds come with games already on them? Anbernic sells its handhelds without licensed games, but many resellers include a microSD card preloaded with thousands of copyrighted titles that nobody paid for. That card is the legally risky part of the purchase: on April 15, 2025 Italian financial police seized more than 30 emulation handhelds from the reviewer known as Once Were Nerd over reviews of preloaded devices, under a statute allowing up to 15,000 euros in fines and three years in prison. ### Does Anbernic ship to the United States? Anbernic suspended all orders shipping from China to the United States on April 21, 2025, citing changes in US tariff policy, and directed American buyers to stock held in its United States warehouse instead. In 2026 the shipping origin of each listing is what decides whether a US order can be filled, so check that on the product page before paying rather than assuming the site ships domestically. **Sources:** - Retro handheld maker Anbernic stops US shipments due to tariffs, Engadget, 2025: https://www.engadget.com/gaming/retro-handheld-maker-anbernic-stops-us-shipments-due-to-tariffs-220217833.html - Anbernic's latest GBA SP-inspired handheld is here, and costs less than $60, Time Extension, 2026: https://www.timeextension.com/news/2026/08/relive-the-pure-joy-of-the-2d-golden-era-anbernics-latest-gba-sp-inspired-handheld-is-here-and-costs-less-than-usd60 - Analogue Pocket gets hit with a price increase 'due to recent tariff announcements', Time Extension, 2026: https://www.timeextension.com/news/2026/03/analogue-pocket-gets-hit-with-a-price-increase-due-to-recent-tariff-announcements - Retro gaming YouTuber faces possible jail time for reviewing gaming handhelds, Android Authority, 2025: https://www.androidauthority.com/once-were-nerd-youtuber-copyright-lawsuit-3577995/ - Compare Nintendo Switch Online membership plans, Nintendo official site: https://www.nintendo.com/us/online/compare-memberships/ --- # Who Owns Claire's Now? The $140 Million Buyout URL: https://404memoryfound.com/posts/who-owns-claires-now.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-12 Topics: Business Blunders, Then vs Now **Summary:** Claire's is owned by Ames Watson, a private investment firm that bought the North American business out of Chapter 11 bankruptcy in September 2025 for about $140 million. The chain still runs roughly 900 stores across the United States and Canada and still pierces ears, but every standalone Claire's in the United Kingdom and Ireland closed in April 2026. **Key facts:** - Launched: 1961 as Fashion Tress Industries, a wig company, renamed after buying Claire's in 1973 - Peak size: About 3,000 stores as of March 3, 2007 - Owner today: Ames Watson, since September 2025 - Stores left: About 900 in the US and Canada; all 154 UK and Ireland stores closed in April 2026 - Status today: Open and still piercing ears, with more than 131 million piercings claimed since 1978 ## How a wig company ended up piercing America's ears Claire's did not begin as a jewelry store, and there was never a founder named Claire. In 1961 Rowland Schaefer started Fashion Tress Industries, a company built to sell wigs at a moment when wigs were a mass market product. The wig business faded. In 1973 Fashion Tress bought Claire's, a Chicago jewelry chain with 25 stores, and took the smaller company's name for the whole operation. The decision that actually mattered came in 1978, when Claire's started piercing ears inside its stores. That one service turned a rack of inexpensive earrings into a rite of passage, and it gave the chain something no catalog and later no website could copy. A ten year old came in for a pair of studs and left with a permanent reason to remember the brand. Claire's now claims more than 131 million piercings performed since 1978, a number the company repeated in its own August 2026 announcement. For a very large share of American women, the first jewelry they ever owned came from a Claire's counter in a shopping mall. ## What the $3.1 billion Apollo buyout did to the balance sheet By the middle of the 2000s the chain was close to saturation. A company filing shows Claire's operating approximately 3,000 stores as of March 3, 2007, across the United States, Canada, Puerto Rico, the Virgin Islands and a long list of European countries. That scale attracted private equity. Apollo Global Management took Claire's private in a merger valued at about $3.1 billion, and the transaction closed on May 29, 2007. The timing was poor. The buyout loaded debt onto a mall retailer months before the recession, and mall foot traffic in the United States began a decline that never reversed. Claire's spent the following decade servicing that debt rather than rebuilding stores or catching up online. Its first Chapter 11 filing arrived in 2018. The company cut debt, changed hands and kept trading, but the structural problem stayed put. The same arithmetic emptied Toys R Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html) and shrank Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html) to a rounding error. ## Why Claire's filed for Chapter 11 again in August 2025 On August 6, 2025, Claire's Holdings filed for Chapter 11 protection in the United States Bankruptcy Court for the District of Delaware. It was the second bankruptcy since 2018, and this time liquidation was openly on the table. Chief executive Chris Cramer did not dress it up. "This decision is difficult, but a necessary one," he said in a statement reported by CBS News in 2025. In the same statement Cramer named the causes: increased competition, consumer spending trends, the ongoing shift away from brick and mortar retail, the company's debt obligations, and macroeconomic factors. That last phrase covered a lot of ground in 2025, including new United States tariffs landing on a business that imports most of what it sells. Competition had also changed shape. Cheap accessories shipped straight from overseas sellers undercut the mall price and reached the same shoppers through a phone. Specialty mall retail had already proved how fast that can go wrong, as anyone following GameStop (https://404memoryfound.com/posts/is-gamestop-still-in-business.html) and its shrinking store count can attest. ## Who Ames Watson is and what the firm actually paid Claire's found a buyer inside bankruptcy rather than on the courthouse steps. Ames Watson, a private investment firm, agreed to take over the North American business for about $140 million and closed the purchase in September 2025. The deal covered a store fleet the buyer said it would hold at a minimum of roughly 800 locations, with room to keep as many as 950. Stores outside that count went through closing sales. Ames Watson framed the purchase as a brand rescue. "Claire's is one of those rare brands that defines a stage of life," partner and co-founder Lawrence Berger said in a 2025 statement reported by Commercial Observer. His partner made the staffing case. "Every turnaround we've done begins with people," co-founder Tom Ripley said in the same 2025 announcement, pointing to store employees with twenty years or more behind the piercing counter. That distinction matters for anyone asking whether the brand survived. A liquidator sells the fixtures and licenses the trademark. Ames Watson bought the operating business, the stores and the staff, which is why Claire's still opens its doors every morning. ## How many Claire's stores are left in 2026? About 900 in the United States and Canada. Claire's used that figure itself in August 2026, when it announced a chainwide overhaul of the piercing service covering all 900 locations, with the rollout due to finish in the fall of 2026. The British and Irish side did not survive. In April 2026 administrators closed every standalone Claire's in the United Kingdom and Ireland, 154 shops in total, cutting around 1,300 jobs. The brand's concession counters inside other retailers were not part of that closure. North America went the other way. On May 18, 2026, Centric Brands and Claire's announced a licensing partnership intended to put Claire's product into more than 7,000 additional retail locations across the United States and Canada, including Walmart, Kohl's and CVS. Read together, those two moves describe the strategy. The mall store count stops shrinking at roughly 900, the piercing chair stays as the thing competitors cannot replicate, and the earrings and hair clips go wherever people already shop. Claire's is smaller than it has been since the 1990s, and it is no longer betting its future on the mall. ## Frequently Asked Questions ### Is Claire's closing all of its stores? No. Claire's filed for Chapter 11 bankruptcy in August 2025 and closed hundreds of locations, but the North American business was sold to Ames Watson in September 2025 and about 900 Claire's stores were still trading in the United States and Canada in 2026. The exception is the United Kingdom and Ireland, where all 154 standalone Claire's shops closed in April 2026. ### Who owns Claire's now? Claire's North American business has been owned by the private investment firm Ames Watson since September 2025, when it completed a roughly $140 million purchase out of Chapter 11 bankruptcy. Before that, Claire's had been controlled by lenders following its 2018 bankruptcy, and by Apollo Global Management after a $3.1 billion buyout that closed in May 2007. ### Does Claire's still pierce ears in 2026? Yes. Ear piercing remains the core of the business, and in August 2026 Claire's announced a rebuilt piercing service rolling out across all 900 of its United States and Canadian stores, with new training, updated clinical standards and a new earring range. Claire's says it has performed more than 131 million piercings since it began offering the service in 1978. **Sources:** - SEC EDGAR: Claire's Stores, Inc. Form 8-K exhibit on the Apollo merger (2007): https://www.sec.gov/Archives/edgar/data/0000034115/000095012307004241/y32335aexv99w2.htm - FundingUniverse: History of Claire's Stores, Inc.: https://www.fundinguniverse.com/company-histories/claire-s-stores-inc-history/ - CBS News: Claire's, a jewelry retailer for teens, files for Chapter 11 bankruptcy (2025): https://www.cbsnews.com/news/claires-jewelry-retailer-bankruptcy-chapter-11/ - Commercial Observer: Ames Watson Acquires Claire's for $140M Following Bankruptcy (2025): https://commercialobserver.com/2025/09/ames-watson-acquires-claires-bankruptcy/ - RTE: Claire's shuts all UK and Ireland stores (2026): https://www.rte.ie/news/2026/0427/1570474-claires-uk-ireland/ - Business Wire: Centric Brands and Claire's Announce Strategic Licensing Partnership (2026): https://www.businesswire.com/news/home/20260518119685/en/Centric-Brands-and-Claires-Announce-Strategic-Licensing-Partnership - PR Newswire: Claire's Reimagines Piercing Experience for a New Generation (2026): https://www.prnewswire.com/news-releases/claires-reimagines-piercing-experience-for-a-new-generation-302844349.html --- # How Many Boston Market Locations Are Left in 2026? URL: https://404memoryfound.com/posts/boston-market-locations-left.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-12 Topics: Business Blunders, Then vs Now **Summary:** Boston Market listed six restaurants in 2026, and Restaurant Business reached only three of them by phone. The chain peaked at 1,143 locations in 1998, and almost everything left has closed since 2022 under owner Jay Pandya, mostly over unpaid rent, taxes and wages. The name survives on frozen dinners in supermarkets. **Key facts:** - Launched: 1985 as Boston Chicken in Newton, Massachusetts - Peak size: 1,143 US restaurants in 1998 - Owner today: Engage Brands, controlled by Jay Pandya since April 2020 - Stores left: Six listed in 2026; Restaurant Business confirmed three open - What replaced it: Supermarket rotisserie chicken and licensed Boston Market frozen dinners ## How a Newton chicken shop turned into a Wall Street story Boston Market began in 1985 as Boston Chicken, a single storefront in Newton, Massachusetts, opened by Steven Kolow and Arthur Cores, two Northeastern University students selling rotisserie chicken at fast-food speed. The pitch was a home-cooked dinner nobody had to cook. It worked in one neighborhood, and then it attracted investors who knew how to scale retail fast. Those investors came out of video rental. Scott Beck and Saad Nadhir had run Blockbuster (https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html) franchises before they took control of the chicken chain, and they moved its headquarters out of Massachusetts and started signing area developers across the country. The growth was close to vertical. The Washington Post counted it in 1998: 34 stores in the Northeast in May 1992, 1,143 restaurants nationwide six years later. The name became Boston Market in 1995, once turkey, meatloaf and ham joined the chicken. Wall Street paid up for it. The stock was priced at $20 a share in November 1993 and shot to nearly $50, and it split two-for-one in 1994. For about four years, a family dinner chain was one of the hottest equities in the country. ## Why did Boston Chicken file for bankruptcy in 1998? Because the stores were financed with debt that the parent company had issued to its own franchise developers, and the sales could not cover it. Boston Chicken lent money to the area developers who opened its restaurants, then booked the interest on those loans. That made the parent look profitable while the restaurants underneath it did not earn enough to repay anything. Sales flattened in 1997 and 1998 as supermarkets moved into the exact business Boston Market had invented: a hot roasted chicken and two sides, bought on the way home. The Washington Post attributed the weak results to competition from grocery chains selling ready-to-eat meals. On October 5, 1998, the company filed for Chapter 11 and closed 178 restaurants the same day. It could not finish its restructuring before October 17, when about $283 million in debt came due. Roughly a third of the chain was gone within days of the filing, and the rest was sold as a brand rather than rebuilt as a business. ## Who owns Boston Market now? Engage Brands, a company controlled by the real estate investor and restaurant franchisee Jay Pandya, has owned Boston Market since April 2020. McDonald's bought the brand out of its troubles in 2000 and never made it grow, then sold it to the private equity firm Sun Capital Partners in 2007. Sun Capital held it for 13 years and sold it to Pandya's Engage Brands, when the chain still had more than 300 US restaurants. The last two decades read like Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html): each owner smaller and further from the original operator, each one treating a familiar name as an asset rather than a kitchen. Pandya also bought Corner Bakery, and he has filed personal bankruptcy twice. Both filings were thrown out, one because he did not produce insurance documents for properties he owned, the second because he missed court deadlines, and a judge barred him from filing again for at least six months. Nobody at Engage Brands has announced a shutdown. The restaurants simply stopped, one address at a time, without press releases. ## Rent, wages and stop-work orders: what actually closed the stores Not consumer taste. Unpaid bills. Landlords evicted Boston Market over rent, suppliers sued over invoices, and state agencies shut restaurants over taxes and payroll. The clearest record is in New Jersey. In August 2023 the state Department of Labor and Workforce Development issued stop-work orders at 27 Boston Market restaurants after complaints from workers, and the investigation found wages that had gone unpaid or been paid late. The order was lifted in September 2023, after the company paid more than $630,000 in back wages to 314 workers. New Jersey Labor Commissioner Robert Asaro-Angelo put the message plainly in the department's announcement that month: "if you don't pay your workers, we will shut you down." Colorado took a similar route. Revenue officials seized the company's headquarters in Golden and its three remaining restaurants in the state in May 2023 over unpaid sales and payroll taxes, then released them once the bill was settled. Closures that arrive by sheriff's notice do not get a farewell sign, which is why so many Boston Market regulars found a locked door and an unchanged menu board. ## Six addresses and three answered phones In 2026 the count is small enough to verify by hand, and Restaurant Business did exactly that. The company's own website listed six restaurants: two in New Jersey and one each in Texas, Pennsylvania and New York. The trade publication called all six. Three were open and operating, two calls went unanswered, and the sixth returned a busy signal. That is the end of a very steep line. Boston Market went into the pandemic with just under 400 restaurants and finished 2025 with 15, according to Technomic. Massachusetts, where the first store opened, has had no Boston Market since 2025. The chain that started in Newton has nothing left in its home state. For comparison, Kmart (https://404memoryfound.com/posts/is-kmart-still-open-stores-left.html) is down to a handful of US stores and still maintains an accurate store locator. Boston Market's locator kept pointing customers at restaurants that had already closed. The name itself is not dead. Boston Market frozen dinners are still stocked in American supermarkets, made under license by Bellisio Foods and run as a separate business from the restaurants. A franchised store also opened in Delhi in December 2024, and it closed after its landlord sued over unpaid rent. ## The $38 million judgment nobody turned up to fight The most detailed account of how the money moved comes from a lawsuit over the other chain Jay Pandya bought. SSCP Restaurant Investors, which now owns Corner Bakery, took Pandya, his son Ronak, Boston Market and Pandya Restaurant Growth Brands to court in 2024. It accused them of transferring $34 million out of Corner Bakery "without justification." The filing described "shocking, systemic breaches of fiduciary duties" and "a near complete dereliction of the finance, accounting and record-keeping functions" that a company is obliged to perform. Boston Market and Pandya Restaurant Growth Brands never answered the complaint and were declared in default, and a federal judge awarded SSCP $38 million. Process servers could not reach Jay Pandya himself, so he was dismissed as a defendant; Ronak, who was given the title of president and chief executive of Corner Bakery in 2020, stayed in the case. A $38 million judgment against a company with six listed restaurants is not a penalty anyone expects to collect. It is a record of what was left. ## Frequently Asked Questions ### Is Boston Market still open in 2026? Barely. In 2026 Boston Market listed six restaurants on its own website, two in New Jersey and one each in Texas, Pennsylvania and New York, and Restaurant Business confirmed only three of them were open when it called. There has been no Boston Market location in Massachusetts, where the chain started in 1985, since 2025. ### Who owns Boston Market now? Boston Market has been owned since April 2020 by Engage Brands, a company controlled by Jay Pandya. McDonald's owned the chain from 2000 until 2007, when it sold to the private equity firm Sun Capital Partners, and Sun Capital sold to Engage Brands while the chain still had more than 300 US restaurants. ### Why did Boston Market close so many locations? Unpaid bills rather than empty dining rooms. Landlords evicted Boston Market restaurants over rent and states acted over taxes and payroll, including stop-work orders at 27 New Jersey locations in August 2023 that were lifted only after the company paid more than $630,000 in back wages to 314 workers. Boston Market had 1,143 restaurants in 1998 and about six listed in 2026. **Sources:** - The Washington Post: Boston Chicken Seeks Bankruptcy Protection (1998): https://www.washingtonpost.com/archive/business/1998/10/06/boston-chicken-seeks-bankruptcy-protection/8ab3c194-ff83-41a7-b4b1-2db9d03d11f5/ - New Jersey Department of Labor: NJDOL recoups more than $630K in unpaid wages for 314 Boston Market workers (2023): https://nj.gov/labor/lwdhome/press/2023/20230914_wages.shtml - Restaurant Business: How many Boston Market locations are left? That's a good question: https://www.restaurantbusinessonline.com/financing/how-many-boston-market-locations-are-left-thats-good-question - Restaurant Business: Corner Bakery's owner wins $38M judgment against Boston Market: https://www.restaurantbusinessonline.com/financing/corner-bakerys-owner-wins-38m-judgment-against-boston-market - The Boston Globe: The last Boston Market in Mass. has closed (2026): https://www.bostonglobe.com/2026/05/04/lifestyle/boston-market-chicken-downfall/ --- # Does Chuck E. Cheese Still Have Animatronics? URL: https://404memoryfound.com/posts/chuck-e-cheese-animatronics-still-there.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-12 Topics: Then vs Now, Hardware **Summary:** Chuck E. Cheese still has animatronics in 2026, but only at five locations in the United States: Northridge, California; Springfield, Illinois; Hicksville, New York; Charlotte, North Carolina; and Nanuet, New York. CEC Entertainment announced the retirement of Munch's Make Believe Band in November 2023 and finished pulling the stages out of every other store by the end of 2024. The robots that came out went to collectors or to a parts warehouse in Topeka, Kansas. **Key facts:** - Launched: 1977, founded by Atari co-founder Nolan Bushnell - Status today: Animatronics in 5 US locations; removed chain-wide by the end of 2024 - Owner today: CEC Entertainment, about 90 percent held by six private equity groups - Chain size: Nearly 700 Chuck E. Cheese and Peter Piper Pizza venues, 47 states, 17 countries (2026) - What replaced the band: Video walls, interactive dance floors and trampoline zones ## Which stores kept a working stage? Five locations in the United States still had animatronics running when the removals finished at the end of 2024. IEEE Spectrum named them: Northridge, California; Springfield, Illinois; Hicksville, New York; Charlotte, North Carolina; and Nanuet, New York. Four of the five run Munch's Make Believe Band, the group of Chuck E. Cheese, Mr. Munch, Helen Henny, Jasper T. Jowls and Pasqually. Northridge, in the San Fernando Valley, is the one fans drive across a state to see, because for several months it was the only store the company planned to spare. Nanuet is the odd one out. It keeps a solo Chuck E. set rather than the full band, so a family arriving there expecting five singing robots finds one. Everywhere else the stage is now a wall. CEC Entertainment operated 559 Chuck E. Cheese locations when it left bankruptcy in December 2020, so five surviving stages works out to fewer than one store in a hundred. The rest of the chain got the same package: screens, a dance floor and a trampoline zone. ## Why did CEC Entertainment take the robots out? The company announced the retirement of Munch's Make Believe Band in November 2023 and finished the removals by the end of 2024, as part of a rebrand that also changed the menu, the lighting and the logo. David McKillips, then the chief executive of CEC Entertainment, put the decision down to how children spend their attention now. "Kids, really of all ages, are consuming their entertainment on a screen," he told TODAY in 2025, describing the overhaul as the company's "most aggressive transformation." The replacements are physical rather than theatrical. IEEE Spectrum described the swap plainly: the moving figures came out and video screens, dance floors and trampolines went in. A stage takes floor space that a trampoline zone can monetize by the hour, and a video wall can be updated from a laptop. There is also a supply problem behind the decision. Nobody has manufactured these robots for decades, and the five stages still standing are kept alive with parts pulled from the ones that were taken down. ## Where did the retired animatronics end up? A large number of them went to Topeka, Kansas. Fortune reported in November 2023 that dozens of Chuck E. Cheese robots, in various states of disrepair, sit in a warehouse there and serve as the parts bin for the bands still performing. Others went to private collectors. Complete stages and single characters move through hobbyist groups and classified listings rather than any retail channel, which is why there is no published price for one. The company held some back for itself. A Chuck E. Cheese spokesperson told Fortune in 2023, "We have preserved some for our 50th anniversary," which lands in 2027, the fiftieth year since the first store opened. A few survivors are back in public view. Chuck's Arcade, the spin-off aimed at adults rather than at eight-year-olds, began opening in shopping malls in 2025 with a retired character standing in each one as a display piece instead of a performer. That is the same nostalgia trade running through the return of the coin-op arcade (https://404memoryfound.com/posts/what-happened-to-arcades-golden-age-gaming.html), aimed at the people who were children the first time around. ### Where to find one today Complete stages almost never reach a storefront, and the ones that do change hands privately between collectors. The affordable end of this hobby is the small stuff: arcade tokens stamped with the year, prize-counter toys, birthday badges, cups and show tapes, which turn up in quantity on Etsy's vintage Chuck E. Cheese listings (https://www.etsy.com/search?q=chuck+e+cheese+vintage). Tokens are the usual starting point, and the common years are still easy to find. ## How one surviving store turned into five The original plan was a single exception. Fortune reported in November 2023 that the band would play on in Northridge, California, and nowhere else in the chain. That went badly. Adults who had spent birthdays in front of the stage complained loudly enough that the company reopened the question in 2024, and by the time the removals were complete at the end of that year, IEEE Spectrum counted five locations with animatronics rather than one. The four additions were not a restoration. Nothing was rebuilt or brought back from the warehouse; the company simply stopped removing four stages that were already there, and designated those stores as the places where the show continues. Nothing protects them beyond company policy. There is no trust, no landmark status and no contract, which is why fans track the five stores the way collectors track a discontinued format. ## Who owns Chuck E. Cheese in 2026? CEC Entertainment owns the chain, and CEC Entertainment is owned by the investors who held its debt. The company filed for Chapter 11 protection in June 2020 after the pandemic closed its dining rooms, and emerged on December 30, 2020, having eliminated 705 million dollars of debt. It counted 559 Chuck E. Cheese and 122 Peter Piper Pizza locations on the day it came out. By 2026 that had hardened into private equity ownership. D Magazine reported in August 2026 that six private equity groups together hold about 90 percent of CEC Entertainment, and that the company is not publicly traded. The leadership turned over as well. Scott Drake, who joined as chief financial officer in 2024, became president and chief executive in February 2026. Under him the company counts nearly 700 venues across 47 states and 17 countries, is opening its first store in the United Kingdom, and expects five or six more Adventure World locations by January 2027. The company that started this was a side project of the arcade business. Chuck E. Cheese was founded in 1977 by Nolan Bushnell, the co-founder of Atari (https://404memoryfound.com/posts/who-owns-atari-now.html), as a room where a family could eat pizza while the machines paid the rent. The chain outlived the 1983 video game crash (https://404memoryfound.com/posts/how-nintendo-nes-saved-video-games-1983-crash.html) that took most of that industry down with it, and the robots outlived the arcade cabinets by forty years. ## Frequently Asked Questions ### Does Chuck E. Cheese still have animatronics in 2026? Yes, but only at five Chuck E. Cheese locations in the United States: Northridge, California; Springfield, Illinois; Hicksville, New York; Charlotte, North Carolina; and Nanuet, New York. Every other store lost its stage by the end of 2024, when CEC Entertainment finished the removal it announced in November 2023. ### Where did the Chuck E. Cheese animatronics go? Dozens of retired Chuck E. Cheese robots went to a warehouse in Topeka, Kansas, where Fortune reported in November 2023 that they supply spare parts for the stages still running. Others went to private collectors, and a set was preserved by the company for the chain's 50th anniversary in 2027. ### Why did Chuck E. Cheese get rid of Munch's Make Believe Band? CEC Entertainment retired Munch's Make Believe Band as part of a rebrand completed at the end of 2024, replacing the stages with video walls, dance floors and trampoline zones. Chief executive David McKillips told TODAY in 2025 that children of all ages now take their entertainment from a screen. **Sources:** - Fortune: Chuck E. Cheese animatronic band removal (2023): https://fortune.com/2023/11/17/chuck-e-cheese-animatronic-band-removal - IEEE Spectrum: Chuck E. Cheese's Animatronics Band Bows Out: https://spectrum.ieee.org/chuck-e-cheese-animatronics - TODAY: Chuck E. Cheese gets a rebrand, which means no more animatronic band: https://www.today.com/food/restaurants/chuck-e-cheese-animatronic-band-rcna195724 - Nation's Restaurant News: CEC Entertainment emerges from bankruptcy: https://www.nrn.com/family-dining/chuck-e-cheese-parent-cec-entertainment-emerges-from-bankruptcy - D Magazine: Chuck E. Cheese CEO Scott Drake reveals what's next for the brand: https://www.dmagazine.com/business-economy/2026/08/chuck-e-cheese-scott-drake/ --- # Is GameStop Still in Business? 1,598 Stores Left URL: https://404memoryfound.com/posts/is-gamestop-still-in-business.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-11 Topics: Gaming, Business Blunders **Summary:** GameStop is still in business in 2026. The chain closed 727 stores in the United States during the fiscal year that ended January 31, 2026, including more than 470 in January alone, and finished with 1,598 US stores. What is left sells more collectibles than video games. **Key facts:** - Founded: 1984 in Dallas, Texas, as Babbage's - Renamed GameStop: 1999 - Stores left: 1,598 in the United States as of January 31, 2026 - Closed in fiscal 2025: 727 US stores, including more than 470 in January 2026 - Status today: Still in business; video game software is under 12 percent of sales ## What was GameStop called before 1999? The chain that became GameStop opened in Dallas in 1984 under a different sign. Two Harvard Business School classmates, Gary Kusin and James McCurry, started a software store called Babbage's in the NorthPark Center mall and named it after Charles Babbage, the mathematician who designed the first mechanical computer. Ross Perot, the Texas businessman and later presidential candidate, was an early investor. Babbage's sold boxed software for home computers years before console games paid the rent. It merged with Software Etc. in the mid 1990s, passed through the collapse of the NeoStar Retail Group, and ended up owned by Barnes & Noble. The GameStop name arrived in 1999, fifteen years after the first store opened. By then the business had found the model it would ride for two decades: buy used games cheap, sell them again at a markup, and keep a wall of new releases by the door. ## How fast has the store count fallen since 2024? In February 2024, GameStop ran 2,915 stores in the United States. Twelve months later it ran 2,325, part of a worldwide total of 3,203 that also covered Canada, Australia and Europe. Then came the year that redrew the map. GameStop closed 727 stores in the United States during fiscal 2025, the year that ended on January 31, 2026, and finished with 1,598 US stores standing. That is a cut of 1,317 stores, or 45 percent of the US footprint, in 24 months. Nearly half the counters where a generation traded in a stack of discs for store credit are gone, and most closed quietly rather than with a liquidation sale. The shape of the decline is familiar to anyone who watched Blockbuster (https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html) fall from a store in every strip mall to a single location in Oregon. The difference is the speed. GameStop shed more than a thousand American stores in two years while still opening its doors every morning. ## Which GameStop stores closed in January 2026? More than 470 locations across 43 states went dark in January 2026 alone, one of the largest single waves in the chain's history. GameStop did not publish a list. Reporters assembled one by watching the company's own store locator flip hundreds of addresses to closed, which is how shoppers in Ohio, Florida and Virginia found out their store was on it. Investors had been warned in general terms. In a filing with the Securities and Exchange Commission, GameStop said it would close "a significant number of additional stores in fiscal 2025", the year that ran to January 31, 2026. The damage was spread thin rather than concentrated. Forty-three states lost at least one location in that January, which is why the closure list read like a national retreat instead of an exit from one weak region. A mall that kept its GameStop through 2025 had no particular reason to expect it would keep it through 2026. ## Does GameStop still sell many games? Not many. In the first quarter of fiscal 2026, collectibles, a category that now leans on graded trading cards and Pokemon singles, brought in 41.8 percent of GameStop's revenue. Hardware and accessories accounted for close to 40 percent. Video game software, the product the store is named for, is under 12 percent of the business, chief executive Ryan Cohen said in 2026. Cohen went further than the numbers. After Sony said it would stop producing discs for games released after January 2028, he called physical software "totally irrelevant" to GameStop's results, which is a strange sentence from the man running the largest game store chain in the country. That leaves the modern GameStop as a trading floor for cards, figures, consoles and controllers with a games shelf attached. Anyone who wants to play Sega Genesis games (https://404memoryfound.com/posts/sega-genesis-games-how-to-play-2026.html) in 2026 is better served by a specialist than by a mall counter. ## Why did GameStop shut down Game Informer? On August 2, 2024, GameStop closed Game Informer (https://404memoryfound.com/posts/who-owns-game-informer-now.html) and laid off the entire staff the same day. The magazine had run for 33 years and was the last large print games magazine in the United States. The farewell posted in its place said "it is with a heavy heart that we announce the closure of Game Informer". GameStop gave no public reason. The timing lined up with the cost cutting that was emptying hundreds of stores at the same time, and the magazine was an expense that did not sell a controller. The bigger loss was the archive. The Game Informer website was replaced by the farewell note, and decades of reviews and features went offline with it. The title has since come back under different ownership and is no longer part of GameStop. Readers who followed the slow hollowing of RadioShack (https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html) will recognize the sequence: cut the parts that made the store worth a visit, then measure the visits. ## Is it worth walking into a GameStop in 2026? GameStop is still in business. It operated 1,598 stores in the United States as of January 31, 2026, plus locations in Canada, Australia and Europe, and it has told investors it does not expect to close a significant number of stores in fiscal 2026. The company now describes the remaining footprint as part of its logistics network rather than as a chain of game shops. So the answer people search for is yes, with a condition attached. Your nearest GameStop may be gone even though the company is not, and checking the store locator before driving over is a reasonable habit in 2026. Inside, the trade counter is still the center of the store, but what crosses it has changed. Cards get graded, consoles get valued, and the wall of new releases keeps shrinking because fewer of them ship on a disc at all. ### Where to buy retro games now Used shelves thinned out along with the store count, so the dependable route for older cartridges and discs is a specialist retro dealer rather than a mall chain. Stone Age Gamer (https://stoneagegamer.com/) stocks tested cartridges, repair parts and flash carts for the consoles GameStop stopped carrying years ago. ## Frequently Asked Questions ### Is GameStop going out of business? No. GameStop is still operating in 2026, although it closed 727 stores in the United States during the fiscal year that ended January 31, 2026, including more than 470 in January 2026. The company has told investors it does not expect to close a significant number of additional stores in fiscal 2026. ### How many GameStop stores are left? GameStop had 1,598 stores in the United States as of January 31, 2026, down from 2,915 in February 2024. It also runs stores in Canada, Australia and Europe, which brought the worldwide total to 3,203 as of February 2025. ### Who owns GameStop now? No parent company owns GameStop. It is publicly traded on the New York Stock Exchange under the ticker GME and is run by chief executive Ryan Cohen. The Babbage's and Barnes & Noble chapters of its ownership ended decades before the 2026 store closures. **Sources:** - Retail Dive: GameStop closed over 700 stores last year: https://www.retaildive.com/news/gamestop-closed-seven-hundred-stores-2025/815827/ - Fast Company: GameStop store closures 2026, the full list of over 470 locations across 43 states: https://www.fastcompany.com/91479231/gamestop-stores-closing-in-2026-full-list-of-doomed-locations-states - GameSpot: Physical Games Are 'Irrelevant' To GameStop's Business, Says CEO: https://www.gamespot.com/articles/physical-games-are-irrelevant-to-gamestops-business-says-ceo/ - Variety: Game Informer Shuts Down, GameStop Magazine Ends After 33 Years: https://variety.com/2024/digital/news/game-informer-shuts-down-gamestop-magazine-1236094109/ - Wikipedia: GameStop: https://en.wikipedia.org/wiki/GameStop --- # How to Run 90s PC Games on Windows 11 in 2026 URL: https://404memoryfound.com/posts/run-90s-pc-games-windows-11.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-11 Topics: Software & Apps, Gaming **Summary:** Most 90s PC games will not start on Windows 11 because Windows 11 ships only in a 64-bit edition, and 64-bit Windows runs no 16-bit code. The fix in 2026 is emulation or a re-release: DOSBox Staging for DOS games, ScummVM for adventure games, and a pre-configured store version such as GOG's for everything else. All three routes work, and two of them are free. **Key facts:** - Why it breaks: Windows 11 is 64-bit only, and 64-bit Windows runs no 16-bit code - Free route: DOSBox Staging 0.83.0 (August 28, 2026) for DOS games, ScummVM 2.9.0 for adventures - Paid route: Pre-configured re-releases, such as Doom + Doom II at $9.99 - Games preserved: GOG Preservation Program: 100 at launch in November 2024, past 300 in 2026 - Status today: All three routes work on Windows 11 in 2026, and two of them cost nothing ## Why does a 1996 game refuse to open on Windows 11? Windows 11 is the first version of Windows that Microsoft never shipped in a 32-bit edition. It installs on 64-bit processors only, and that single decision is what breaks the 1990s. A Microsoft support article states the limit in its own title: "64-bit versions of Windows don't support 16-bit components, 16-bit processes, or 16-bit applications." NTVDM, the DOS compatibility layer Microsoft built into Windows NT, was never ported to 64-bit Windows, because a 64-bit processor running in long mode cannot drop into the virtual-8086 mode that DOS software expects. So a game written for MS-DOS has nowhere to execute. Double-click it on Windows 11 and the refusal arrives before a single frame is drawn. The trap catches games that are not DOS games at all. Plenty of titles from the late 1990s are 32-bit programs sitting behind a 16-bit setup stub, a small launcher whose only job was to inspect the machine and unpack the real installer. The game underneath would run fine on a modern PC. The installer never gets far enough for anyone to find out. ## Does compatibility mode fix anything, or is it folklore? Right-click the executable, open Properties, choose the Compatibility tab, pick Windows 95 (https://404memoryfound.com/posts/the-night-windows-95-launched.html) from the list. It is the first thing everyone tries, and for one class of game it genuinely works. The setting applies a shim, which is a set of small lies Windows tells the program: a different version number, an older privilege model, a fixed color depth, display scaling switched off. A 32-bit Windows game from 1999 that quits because it does not recognize the operating system will often start after nothing more than that. What a shim cannot do is build a DOS machine. It does not restore the processor mode that DOS software needs, it does not bring back the copy protection drivers Microsoft removed, and it does not hand a game the direct hardware access a 1994 sound routine assumes it has. It also cannot resurrect Glide, the graphics API that 3dfx Voodoo cards (https://404memoryfound.com/posts/what-happened-to-3dfx-voodoo-graphics-card.html) used and that dozens of late-90s games were built against. The practical order is simple. Try the Compatibility tab first, because it costs about thirty seconds. When the game is DOS-era, skip it and go straight to emulation. ## Which free tool matches which kind of game? Two free projects cover most of the decade between them, and they split the work along a clean line. DOSBox emulates the whole machine: an x86 processor, a Sound Blaster (https://404memoryfound.com/posts/what-happened-to-creative-labs-sound-blaster.html), a VGA card, a CD-ROM drive and a working DOS prompt. It is open source and costs nothing. The version to install in 2026 is DOSBox Staging, the actively maintained fork, whose 0.83.0 release landed on August 28, 2026 with an offline manual and named compatibility fixes for Tomb Raider, King's Quest IV, Phantasmagoria and the 1997 Grand Theft Auto. ScummVM works the other way around. It emulates nothing, because it reimplements the game engines themselves. The project describes the trick plainly: "ScummVM just replaces the executables shipped with the games, allowing you to play them on systems for which they were never designed." Point-and-click adventures from LucasArts, Sierra On-Line and Revolution Software then run as ordinary Windows 11 programs, with modern resolutions and save states attached. Release 2.9.0 shipped on December 22, 2024, and more than 1,600 games are detected and supported. The dividing question is what the game was written for, not what year it came out. DOS shooters, strategy and simulation go to DOSBox Staging. The LucasArts and Sierra adventure catalog goes to ScummVM. Neither program includes any game data, which is the next problem. ## What can you do with the original discs in the closet? Owning the disc helps less than it should, and three separate things go wrong. The first is the drive. Laptops stopped including optical drives years ago, so the disc needs an external USB reader before anything else can happen. The second is copy protection. Discs pressed with SafeDisc or SecuROM depend on a Windows driver called secdrv.sys, which Microsoft stopped shipping with Windows 10 and later switched off on older versions of Windows through a security update. A game you paid for can refuse to run on a machine more than capable of running it. The third is the setup stub described above, which fails on Windows 11 whatever the box promises. For DOS-era discs none of this matters much. Copy the game folder off the disc to the hard drive, point DOSBox Staging at it, and the disc has done its job. For Windows games of the SafeDisc years it matters a great deal, and it is the honest reason so many people who already own a game end up buying it a second time. ## When is paying ten dollars the smarter move? Emulation is free, but somebody still has to do the configuring. A re-release is how you pay for that work to have been done already. GOG built a business on exactly this. In November 2024 the store launched the GOG Preservation Program, re-releasing 100 classics under a "Preserved by GOG" stamp with a written commitment attached: "If a game is part of the Preservation Program, it means that we commit our own resources to maintaining its compatibility with modern and future systems." The list passed 300 titles in 2026. Publishers run their own version of the same idea, and Doom (https://404memoryfound.com/posts/how-doom-changed-pc-gaming-internet.html) is the clearest example. Doom + Doom II arrived in August 2024 at $9.99 on Steam, GOG, the Epic Games Store and the Microsoft Store, and anyone who already owned either game got the new version as a free upgrade. That is worth checking before paying twice, because the modern port is sometimes free to existing owners. ### Where to find one today Start free. DOSBox Staging and ScummVM cost nothing, install on Windows 11 without special permissions, and for DOS games they are the better experience anyway. Pay when you want the configuration handled for you: Doom + Doom II is $9.99, and prices across the classics shelf sit in that neighborhood rather than at new-release money. GOG sells its classics as DRM-free offline installers, the format that survives a storefront closing: GOG's catalog of classic PC games (https://www.gog.com/en/games). Humble Bundle discounts the same era in its store, usually as Steam keys rather than standalone installers. ## Will these routes still work in 2030? The free side looks stable. DOSBox Staging and ScummVM are open source, so the code outlives any single maintainer and anyone can build it again. Both are still shipping: ScummVM 2.9.0 in December 2024, DOSBox Staging 0.83.0 in August 2026. Neither depends on a company staying in business. The paid side carries a storefront's usual risk, which is exactly why the file format matters more than the shop. A GOG purchase downloads as an offline installer with no launcher and no account check, so a copy saved to an external drive today still installs years from now, whatever happens to the store. Steam and the Microsoft Store do not work that way. The preservation work is ongoing rather than finished. GOG's program started at 100 games in November 2024 and passed 300 in 2026, and every addition is a set of fixes applied to a game that would otherwise drift out of compatibility as Windows changes underneath it. The answer for 2026, then, and probably for some years after: all three routes work, two of them are free, and the safest move a reader can make is to keep personal copies of the installers on a drive of their own. ## Frequently Asked Questions ### Can you play MS-DOS games on Windows 11? Yes, through emulation rather than directly. Windows 11 ships only in a 64-bit edition, and 64-bit Windows supports no 16-bit components, processes or applications, so an MS-DOS game from the 1990s will not launch on its own. The standard answer in 2026 is DOSBox Staging, a free emulator whose 0.83.0 release arrived on August 28, 2026, or a pre-configured re-release bought from a store such as GOG. ### Is DOSBox free to use? Yes. DOSBox and its maintained fork DOSBox Staging are free and open source, and DOSBox Staging 0.83.0, released on August 28, 2026, is the build most Windows 11 users should install. Neither program includes any game data, so the game files still have to come from your own discs or from a store purchase. ### Do old PC game CDs still work on Windows 11? Often they do not. Discs pressed with SafeDisc or SecuROM copy protection rely on a Windows driver called secdrv.sys that Microsoft stopped shipping with Windows 10 and later disabled on older Windows versions through a security update, and many late-1990s discs use a 16-bit setup stub that 64-bit Windows 11 cannot run at all. DOS-era discs are the exception, because their files can be copied to the hard drive and run in DOSBox Staging. **Sources:** - 64-bit versions of Windows don't support 16-bit components, 16-bit processes, or 16-bit applications (Microsoft Learn): https://learn.microsoft.com/en-us/troubleshoot/windows-client/application-management/x64-windows-not-support-16-bit-programs - DOSBox Staging 0.83.0 release notes (August 2026): https://www.dosbox-staging.org/releases/release-notes/0.83.0/ - ScummVM: project site, the 2.9.0 release and the compatibility list: https://www.scummvm.org/ - GOG Preservation Program (GOG.com): https://www.gog.com/en/gog-preservation-program - DOOM + DOOM II announced for PS5, Xbox Series, PS4, Xbox One, Switch, and PC (Gematsu, August 2024): https://www.gematsu.com/2024/08/doom-doom-ii-announced-for-ps5-xbox-series-ps4-xbox-one-switch-and-pc --- # Why Did Skype Shut Down? What Replaced It in 2025 URL: https://404memoryfound.com/posts/skype-shut-down-what-replaced-it.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-11 Topics: Software & Apps, Then vs Now **Summary:** Skype shut down on May 5, 2025, fourteen years after Microsoft bought it for $8.5 billion in cash. Microsoft replaced it with Microsoft Teams Free, which took over Skype logins, contacts and chat history. Leftover Skype Credit still works inside Teams, but Skype itself is gone and the old message archives were deleted in 2026. **Key facts:** - Launched: 2003, by Niklas Zennström and Janus Friis - Owner today: Microsoft, which paid $8.5 billion in cash in May 2011 - Users: More than 300 million monthly in 2016, about 36 million by 2023 - Status today: Retired on May 5, 2025; the apps and skype.com sign-in now point to Microsoft Teams Free - What replaced it: Microsoft Teams Free, with the Skype Dial Pad kept alive for leftover Skype Credit ## The free call that broke the long distance business Skype arrived in 2003 as a peer to peer experiment. Niklas Zennström and Janus Friis, the pair behind the file sharing program Kazaa (https://404memoryfound.com/posts/what-happened-to-kazaa-p2p-wars.html), pointed the same distributed network idea at voice calls and then gave the software away. The pitch needed no explanation. International calls were billed by the minute in 2003, and Skype made them free between computers. Students called home from dorm rooms, grandparents met grandchildren over webcams, and freelancers ran whole businesses through a laptop microphone. By the time Microsoft came calling, Skype was not a novelty. Microsoft's own announcement said the service had 170 million connected users who logged more than 207 billion minutes of voice and video conversations in 2010. No telephone company had built anything close to that in seven years. The name became a verb, which is the rarest thing a piece of software can do. People did not say they would place a video call. They said they would Skype you, and they said it on a desktop crowded with chat programs that have since died too, next to MSN Messenger (https://404memoryfound.com/posts/what-happened-to-msn-messenger.html) and AOL Instant Messenger (https://404memoryfound.com/posts/what-happened-to-aim-aol-instant-messenger.html). ## What Microsoft got for $8.5 billion in 2011 Microsoft announced the purchase on May 10, 2011. The price was $8.5 billion in cash, paid to an investor group led by Silver Lake, and it was the largest acquisition Microsoft had ever made at that point. "Skype is a phenomenal service that is loved by millions of people around the world," Steve Ballmer said in the press release Microsoft filed with the Securities and Exchange Commission that day. Ballmer was Microsoft's chief executive at the time, and Skype chief executive Tony Bates was named president of a new Microsoft Skype Division. The plan was distribution. Skype went into Windows, into Xbox and into Outlook.com, and for the next decade it was how Microsoft handled consumer calling. What Microsoft did not buy was a business that paid for itself at that price. Skype earned its money from the minority of users who bought credit to dial real phone numbers, while the free app to app calling was the part almost everyone actually used. Fourteen years later Microsoft retired the product, and it has never claimed the $8.5 billion came back. ## How 300 million users turned into 36 million Skype peaked above 300 million monthly users in 2016, according to the figures CNBC published on the day the shutdown was announced in 2025. By 2023 that count was down to roughly 36 million. The decline had three causes, and no single bad decision among them. Phones came first. WhatsApp and FaceTime put free calling inside apps people already carried, with no separate account to remember and no desktop client to install. Then Microsoft split its own attention. Teams shipped as the company's serious communication product and took the engineering budget, the enterprise contracts and the marketing, while Skype was redesigned, criticized, and redesigned back again. Then came 2020. The year the world moved onto video calls should have belonged to the company that had been selling video calls since 2003. Instead Zoom became the shorthand for a work meeting, and Microsoft answered it with Teams rather than with Skype. By the time offices reopened, Skype was software people still had installed rather than software they opened. ## Can you still sign in to Skype in 2026? No. Skype stopped working on May 5, 2025, and nothing done in 2026 will bring the app back. Microsoft announced the retirement on February 28, 2025. Jeff Teper, the Microsoft president responsible for Microsoft 365 collaborative apps and platforms, wrote on the Microsoft 365 blog that the company would "be retiring Skype in May 2025 to focus on Microsoft Teams (free)." Microsoft Teams Free is what replaced it, and the handover was meant to be mechanical. Microsoft told users to sign in to Teams with the same Skype username and password, and said contacts and chat history would come across with them. The Skype apps and the sign-in page on skype.com now point at Teams instead of a login screen. Teper also wrote the eulogy in that same post: "Skype has been an integral part of shaping modern communications and supporting countless meaningful moments, and we are honored to have been part of the journey." One clarification, because it confuses people: Skype for Business was a separate corporate product on its own timetable. It is not the thing that shut down in May 2025. ## What happened to Skype Credit and Skype Numbers This is the part that costs real money, so it is worth being exact. Skype Credit did not vanish. A balance that was live on shutdown day moved across to Microsoft Teams Free, where it still pays for calls to landlines and mobile numbers through the Skype Dial Pad. Microsoft kept that dial pad running for exactly this reason. What stopped was buying more. Microsoft closed new Skype Credit purchases and new subscriptions, so a surviving balance can only shrink. There is no top up and no way to reopen the tap. Skype Numbers, the rented phone lines that let people reach you from an ordinary telephone, were the harder loss. Microsoft said subscriptions would keep renewing until April 3, 2025 and then retire, with no purchase, renewal or reactivation after that date. Anyone who wanted to keep a number had to port it to another carrier before the subscription lapsed. Numbers nobody ported went back into the pool, and in 2026 there is no process for claiming one back. ## Your old chats are gone, and here is the deadline that took them Microsoft's first promise was that Skype data would stay available until January 2026. That date moved, which is why plenty of people who checked once and relaxed lost everything anyway. The final schedule published on Microsoft's export page was stricter than it sounded. Deletion of Skype chat and call history began on April 1, 2026. The export portal stayed open until June 15, 2026, with a warning that anything downloaded after April 1 might already be incomplete. Both dates are now in the past. One group kept its history. People who were active on Skype between December 2024 and February 2025, and who then signed in to Microsoft Teams Free before December 1, 2025, had their Skype chats and call logs carried into Teams, where they remain. Everyone else is out of luck. If you never moved to Teams Free and never ran an export, those messages are not sitting in an archive waiting for a support ticket. Microsoft deleted them. ## Frequently Asked Questions ### Is Skype shut down for good? Yes. Microsoft retired Skype on May 5, 2025 and has not reversed the decision since. The Skype apps and the skype.com sign-in page send users to Microsoft Teams Free instead, and Skype has taken no new sign-ups and sold no new Skype Credit since the shutdown. ### What replaced Skype? Microsoft Teams Free replaced Skype when Skype was retired on May 5, 2025. Skype users were told to sign in to Teams with the same credentials to keep their contacts and chat history, and the Skype Dial Pad survives inside Teams Free so leftover Skype Credit can still call landlines and mobile numbers. ### Can I still get my old Skype messages back? Almost certainly not. Microsoft began deleting Skype chat and call history on April 1, 2026 and closed the Skype export portal on June 15, 2026. The only surviving copies sit inside Microsoft Teams Free, for people who were active on Skype between December 2024 and February 2025 and signed in to Teams Free before December 1, 2025. **Sources:** - The next chapter: Moving from Skype to Microsoft Teams (Microsoft 365 Blog, February 28, 2025): https://www.microsoft.com/en-us/microsoft-365/blog/2025/02/28/the-next-chapter-moving-from-skype-to-microsoft-teams/ - Skype is retiring in May 2025: what you need to know (Microsoft Support): https://support.microsoft.com/en-us/skype/skype-is-retiring-in-may-2025-what-you-need-to-know - How do I export or delete my Skype data? (Microsoft Support): https://support.microsoft.com/en-us/skype/how-do-i-export-or-delete-my-skype-data-84546e00-2fef-4c45-8ef6-3a27f83242cc - Microsoft to Acquire Skype: press release filed as Exhibit 99.1 to Microsoft's Form 8-K (SEC, May 2011): https://www.sec.gov/Archives/edgar/data/789019/000119312511134416/dex991.htm - Microsoft is retiring Skype in May, encouraging people to adopt Teams (CNBC, February 28, 2025): https://www.cnbc.com/2025/02/28/microsoft-is-retiring-skype-in-may-encouraging-people-to-adopt-teams.html --- # Is Redbox Still Around? Why Kiosks Are Still There URL: https://404memoryfound.com/posts/is-redbox-still-around-kiosks.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-11 Topics: Business Blunders, Then vs Now **Summary:** Redbox is not still around. Its parent company, Chicken Soup for the Soul Entertainment, was liquidated in a Chapter 7 bankruptcy on July 10, 2024, and all of Redbox's roughly 24,000 kiosks stopped working that month. Thousands of the red machines are still bolted outside American stores because removing one costs a retailer around $500, and nobody in the bankruptcy was left to pay for it. **Key facts:** - Launched: 2002 inside McDonald's Ventures; DVD kiosks from 2004 - Price then: $1 per night for a DVD - Kiosks at shutdown: About 24,000 across the United States - Status today: Shut down. Parent liquidated in Chapter 7 on July 10, 2024 - Owner today: No operator. Trademarks auctioned by the bankruptcy trustee on April 23, 2025 ## How a McDonald's side project became 24,000 red boxes Redbox did not begin as a movie company. It began in 2002 inside McDonald's Ventures, the unit the burger chain used to test new reasons for people to pull into the parking lot. The first machines sold convenience-store goods. By 2004 the idea had narrowed to DVDs, and the whole pitch fit into one number: a dollar a night. That number did the work. A new release at a Blockbuster store (https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html) cost several dollars for several days, and a late fee waited at the end of it. Redbox charged a dollar, took the disc back at any kiosk in the network, and never mailed anyone a bill for forgetting. Coinstar, the company behind the supermarket coin-counting machines, bought its way to full ownership of Redbox in 2009 and pushed the kiosks into every grocery store, pharmacy and gas station that would take one. Renting a movie stopped being a trip and became something you did while waiting for a prescription. When the company finally shut down, about 24,000 kiosks were still standing outside American stores. ## What actually killed Redbox Streaming did the slow part, and it had help. The DVR (https://404memoryfound.com/posts/what-happened-to-tivo-dvr-pioneer.html) had already moved the video store into the living room, and the HD DVD and Blu-ray format war (https://404memoryfound.com/posts/what-happened-to-hd-dvd-format-war.html) ended just as the disc itself started shrinking as a business. Redbox answered with a streaming app, free ad-supported channels and a digital storefront nobody was waiting for. The fast part was the balance sheet. Apollo Global Management bought Redbox's parent company, Outerwall, in 2016 and took it private. In October 2021 Redbox went public again through a merger with the blank-check company Seaport Global Acquisition Corp., landing on Nasdaq under the ticker RDBX at an enterprise value of about $693 million. It traded as a meme stock for a few weeks and then came apart. In August 2022 Chicken Soup for the Soul Entertainment bought Redbox in a deal worth roughly $375 million: about $50 million in stock plus $325 million of assumed Redbox debt. That is the sentence that mattered. Chicken Soup for the Soul Entertainment had not bought a rental chain so much as a pile of borrowed money with kiosks attached, and it never found a way to pay it down. ## The Delaware hearing that switched off every kiosk Chicken Soup for the Soul Entertainment filed for Chapter 11 protection in the U.S. Bankruptcy Court for the District of Delaware in late June 2024, carrying close to $1 billion in debt. Chapter 11 is supposed to be a reorganization. This one lasted under two weeks. HPS Investment Partners, the lender behind the Redbox purchase, declined to put in more money, and on July 10, 2024 the court converted the case into a Chapter 7 liquidation. There was no soft landing built into it. About 1,000 employees lost their jobs, many of them already owed wages. "There is no means to continue to pay employees, to pay any bills," Judge Thomas Horan said from the bench, in a hearing reported by Deadline in 2024. The kiosks went with the company. A Redbox machine is a terminal: it checks the card, the account and the title against servers that had just stopped existing. All 24,000 of them quit taking rentals at once, and the app, the website and the free streaming channels went dark on the same schedule. ## Why the machines are still bolted to the pavement Here is the part nobody planned for. In a Chapter 7 liquidation a trustee sells what has value and walks away from what does not. Twenty-four thousand dead vending machines have negative value, so nobody came to collect them. That left the bill with the stores. A Redbox kiosk can weigh as much as 900 pounds, is usually bolted into concrete and is often hardwired into the building's power, which makes removal a contractor job. Retailers have put the cost at roughly $500 per unit once labor, mechanical work and disposal are counted. Walgreens took the problem to the court in August 2024. The chain said it was still powering 5,400 Redbox kiosks across about 3,800 locations at a cost of $184,000 a month in electricity, for machines that no longer sold anything. "Walgreens should not be required to continue to 'store' and power Redbox kiosks across the country without any form of payment," the filing read. CVS asked the same court for permission to scrap the kiosks sitting at its stores. Two years on, plenty of them are still in place, unplugged or quietly drawing power, some of them still loaded with the discs that were inside on the day the servers died. ## Who owns the Redbox name in 2026? No one is operating it. The Chapter 7 trustee put the estate's intellectual property up for sale, and the Redbox and Crackle trademarks went to auction on April 23, 2025, alongside film rights held by the parent company's other units. Buying a trademark is not the same as buying a business. Whoever holds the Redbox name holds a logo and a shade of red, not a kiosk network, not the servers, not the disc supply deals with the studios and not the retail contracts with Walgreens, CVS and the grocery chains. Every one of those was unwound in the bankruptcy. So the answer to the question people type while standing in front of one of these machines is short. Redbox is not still around. The company was liquidated in 2024, the kiosks are inert, and as of September 2026 no buyer has switched the network back on. What replaced it is the thing that beat it. Renting a movie now means a digital storefront, and physical rental in the United States has narrowed to libraries, a few independents and the last Blockbuster store in Bend, Oregon. ## Frequently Asked Questions ### Is Redbox still in business in 2026? No. Redbox stopped operating in July 2024, when parent company Chicken Soup for the Soul Entertainment converted its Delaware bankruptcy into a Chapter 7 liquidation and all of its roughly 24,000 DVD kiosks stopped taking rentals. The Redbox website, app and streaming channels shut down at the same time, and no company has restarted the service since. ### Why are Redbox kiosks still outside stores? Because nobody was left to pay for hauling them away. When Redbox's parent company was liquidated in July 2024, the Chapter 7 trustee did not remove the kiosks, which left them with the retailers that had hosted them. Each machine can weigh up to 900 pounds, is often bolted into concrete, and costs a retailer around $500 to take out, so thousands have simply been left where they stood. ### Who owns Redbox now? No operating company owns Redbox. The Redbox trademark was put up for sale by the Chapter 7 trustee at an auction held on April 23, 2025, together with the Crackle brand and film rights from the same parent company, and buying the name bought nothing else: the kiosks, the servers and the retail contracts were all dismantled in the 2024 liquidation. **Sources:** - Variety: RIP Redbox: DVD Kiosk-Rental Business Is Shutting Down With Parent's Bankruptcy Liquidation (2024): https://variety.com/2024/digital/news/redbox-shutting-down-bankruptcy-liquidation-chicken-soup-for-the-soul-1236067161/ - Deadline: Redbox Owner Chicken Soup For The Soul To Liquidate In Chapter 7 Bankruptcy Shift (2024): https://deadline.com/2024/07/redbox-liquidation-bankruptcy-chicken-soup-for-the-soul-chapter-7-1236006525/ - Media Play News: Big Box Retailers Grapple With Costs Associated With Abandoned Redbox Kiosks (2024): https://www.mediaplaynews.com/big-box-retailers-grapple-with-abandoned-redbox-kiosks-dvd-movies/ - Variety: Redbox Is Set to Go Public After Officially Closing SPAC Merger (2021): https://variety.com/2021/digital/news/redbox-public-offering-closes-spac-deal-1235095923/ - Deadline: Redbox, Crackle And Other Chicken Soup For The Soul Entertainment Assets Set For Post-Bankruptcy Auction (2025): https://deadline.com/2025/04/redbox-crackle-chicken-soup-for-the-soul-entertainment-bankruptcy-auction-1236363107/ --- # Dial-Up Sound Explained: What Each Screech Meant URL: https://404memoryfound.com/posts/dial-up-modem-sound-explained.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-11 Topics: Then vs Now, Internet Culture **Summary:** The dial-up sound was not random static. It was a scripted handshake in which two modems disabled the network's echo cancellers, traded capability menus under ITU-T Recommendation V.8, and probed the phone line with 21 tones before agreeing on a speed. You can still hear it in 2026 on the dial-up services NetZero and Juno sell, but not on AOL, which shut its dial-up down on September 30, 2025. **Key facts:** - What it was: A modem handshake: dial tone, DTMF digits, answer tone, capability menus, line probe, training - The first tone: 2100 Hz, phase-reversed every 450 ms to disable network echo cancellers - The line probe: 21 tones spaced 150 Hz apart, from 150 Hz to 3750 Hz (ITU-T V.34) - Why it stopped: The Hayes default M1 mutes the speaker once a carrier is detected - Status today: AOL ended dial-up September 30, 2025; about 100,277 US households still use it ## What are you hearing in the first five seconds? The dial-up sound was never one noise. It was a fixed sequence of steps, each one doing a specific job, and the order barely changed from call to call. It opened with the dial tone, then the touch tones of the phone number itself, punched out as DTMF digits. People who connected every day could recognize their own provider's access number by ear, the way you recognize a ringtone. Then the far end spoke for the first time: a steady tone at 2100 Hz. That tone was not a greeting. It was a demand. The phone network put echo suppressors and echo cancellers on long trunks because they help human conversation, and they would have wrecked a data call. ITU-T Recommendation V.8 specifies that the answering modem flip the phase of that 2100 Hz tone by 180 degrees every 450 ms, which is the exact signal network echo cancellers are built to stand down for. The same tone carried a second message. Amplitude-modulated by a 15 Hz sine wave, it became ANSam, and the wobble was a flag that said this end speaks V.8 and is ready to negotiate. A flat 2100 Hz tone with no wobble meant an older machine at the other end and a slower call ahead. ## Why two modems had to argue before they could talk Nothing about a residential phone line was guaranteed. The copper between a house and the local exchange had its own length, its own noise and its own quirks, and it could change between one call and the next. Two modems could not assume anything about each other or about the wire in between, so they had to find out in public, out loud. That is the negotiation V.8 defines. After the answer tone, the calling modem sends CM, the call menu, listing every modulation, data rate and error-correction scheme it can handle. The answering modem replies with JM, the joint menu, naming the best combination the two of them share. The caller sends CJ to acknowledge, and the argument is settled. Oona Räisänen, the Finnish engineer whose 2012 spectrogram of a handshake is still the clearest picture anyone has made of it, put the whole exchange in one line on her blog absorptions: "The modems are trying to find a common language and determine the weaknesses of the telephone channel originally meant for human speech." That is why the sequence sounded like an argument. It was one. ## What the harsh middle stretch was actually measuring The ugliest part of the sound, the burst that came after the negotiation and before the hiss, was a measuring instrument. ITU-T Recommendation V.34, the standard behind the 28,800 bit/s modems of the mid-1990s, defines a line probing signal built from 21 tones spaced 150 Hz apart, running from 150 Hz all the way up to 3750 Hz, with the tones at 900, 1200, 1800 and 2400 Hz deliberately left out. The gaps matter as much as the tones, because they let the receiving end hear what the line itself adds. The probe goes out twice. L1 is sent 6 dB above the normal power level for 160 ms. L2 repeats the same tones at normal power for no longer than 550 ms plus the round trip delay. The listening modem runs a spectrum analysis on both and reads the damage: which frequencies the line swallows, how much it delays them, how nonlinear it is. From that it picks a symbol rate and a carrier frequency, then both ends send scrambled data at each other to train their equalizers and echo cancellers. That training is the long featureless hiss at the end. By then the modems were not negotiating anymore. They were rehearsing. ## Why the screech stopped the second you were online Here is the part almost nobody gets right: the modem did not go quiet because the noise ended. The noise never ended. The speaker was switched off. Consumer modems inherited the Hayes command set, and its default speaker setting, M1, means the speaker stays on until a carrier is detected and then cuts out. Once the two modems agreed on a scheme and locked onto each other, the connection was made, the speaker went dead, and everything that followed happened silently on the same wire. The default was a deliberate design choice, not an accident. Leaving the speaker on through the handshake turned the modem into a diagnostic tool that anyone could use without reading a manual. A busy signal, a recorded message, a wrong number, a person picking up an extension upstairs: all of it came through the speaker, and all of it told you the call had failed before the software did. The moment the call worked, the information stopped being useful, so the speaker stopped. That also explains why the sound never got faster as modems got faster. The handshake was not the data. It was the setup, and the setup was bound by physics and by protocol timers, not by the dial-up speed (https://404memoryfound.com/posts/dialup-speed-then-vs-now.html) you were negotiating toward. ## Can you still hear a modem connect in 2026? Barely, and almost never by accident. AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html) shut down the service that produced the sound for more Americans than any other. In a notice posted in August 2025 the company said it "routinely evaluates its products and services and has decided to discontinue Dial-up Internet," and the service, along with the AOL Dialer software and the AOL Shield browser, went dark on September 30, 2025. Dial-up itself is not extinct. NetZero (https://404memoryfound.com/posts/dial-up-internet-still-available-2026.html) and Juno still sell it, and the Census Bureau's American Community Survey counted about 100,277 US households with dial-up as their only home internet subscription in the 2024 data released in September 2025. Those households hear the handshake every time they connect. Everyone else has to go looking. Fax machines, alarm panels and some payment terminals still run the same negotiation over the same copper. Räisänen's 2012 recording and spectrogram are still online. So are plenty of the websites those modems used to load (https://404memoryfound.com/posts/90s-websites-still-online.html), which is a stranger kind of survival than the sound itself. ## Frequently Asked Questions ### What was the dial-up sound actually doing? The dial-up sound was two modems setting up a data call over a phone line built for speech. It ran in order: dial tone, the DTMF touch tones of the number, a 2100 Hz answer tone whose 450 ms phase reversals switched off the network's echo cancellers, a V.8 exchange of capability menus, a 21-tone line probe spanning 150 Hz to 3750 Hz under ITU-T Recommendation V.34, and a final training hiss. ### Why did the dial-up sound stop after you connected? The dial-up sound stopped because the modem muted its own speaker, not because the signaling ended. The Hayes command set that every consumer modem of the 1990s inherited defaults to M1, which keeps the speaker on only until a carrier is detected. The point of hearing the handshake was to catch a busy signal, a wrong number or a human voice before the software reported a failure, and once the call connected there was nothing left to diagnose. ### Can you still hear a dial-up modem connect in 2026? Yes, but you have to seek it out. AOL ended its dial-up service on September 30, 2025, while NetZero and Juno still sell dial-up access, and the Census Bureau's American Community Survey counted roughly 100,277 US households relying on dial-up alone in its 2024 data. Fax machines and alarm panels still perform the same handshake, and recordings such as Oona Räisänen's 2012 spectrogram remain online. **Sources:** - ITU-T Recommendation V.8: Procedures for starting sessions of data transmission over the PSTN: https://www.itu.int/rec/T-REC-V.8/en - ITU-T Recommendation V.34: Modem operating at data signalling rates of up to 33 600 bit/s: https://www.itu.int/rec/T-REC-V.34/en - Oona Räisänen, 'The sound of the dialup, pictured', absorptions, 2012: https://www.windytan.com/2012/11/the-sound-of-dialup-pictured.html - CNN Business: AOL will stop offering dial-up internet service after more than 30 years in business (2025): https://www.cnn.com/2025/08/11/tech/aol-dial-up-internet-discontinue - Reviews.org: The U.S. Digital Divide Is Shrinking, But Nearly 8 Million Households Remain Offline: https://www.reviews.org/internet-service/how-many-us-households-are-without-internet-connection/ --- # Who Owns Yahoo Now? The $5 Billion Apollo Takeover URL: https://404memoryfound.com/posts/who-owns-yahoo-now.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-11 Topics: Business Blunders, Money & Tech **Summary:** Yahoo is owned by funds managed by affiliates of Apollo Global Management, which announced a roughly $5 billion purchase of Verizon Media on May 3, 2021 and completed it later that year. Verizon kept a 10 percent stake, and Yahoo has had no publicly traded shares since. The company is still running in 2026 under chief executive Jim Lanzone, with Yahoo Finance, Yahoo Mail, Yahoo Sports and Yahoo Search still drawing billions of visits a month. **Key facts:** - Owner today: Funds managed by affiliates of Apollo Global Management; Verizon kept a 10 percent stake - Price paid: About $5 billion, announced May 3, 2021: $4.25 billion in cash plus $750 million in preferred interests - Previous owner: Verizon, which closed its $4.48 billion purchase of Yahoo's operating business on June 13, 2017 - Offer refused: $44.6 billion from Microsoft, rejected by Yahoo's board on February 11, 2008 - Status today: Private and still operating in 2026: Yahoo Finance, Yahoo Mail, Yahoo Sports, Yahoo News and Yahoo Search ## Who are the Apollo funds that bought Yahoo? Yahoo is owned by funds managed by affiliates of Apollo Global Management, a New York investment firm that runs private equity and credit money for pension plans, insurers and endowments. Apollo announced the purchase on May 3, 2021 and valued it at about $5 billion. The asset on the table was Verizon Media, the division that held Yahoo and AOL. Verizon took $4.25 billion in cash and $750 million in preferred interests, and held on to a 10 percent stake in the business. The deal closed later that year. Apollo renamed the whole company Yahoo at closing, which is why the AOL brand stopped being the parent name and went back to being one property among several. The firm was explicit about the thesis. "We are big believers in the growth prospects of Yahoo and the macro tailwinds driving growth in digital media, advertising technology and consumer internet platforms," David Sambur, co-head of private equity at Apollo, said in the May 2021 announcement. That ownership split has not moved since. Apollo controls the large majority through its funds, Verizon keeps its minority slice, and there are no public shareholders, no ticker and no quarterly earnings calls. ## Why Verizon gave up on the portal it had just bought Verizon owned Yahoo for four years and sold it for roughly what it paid, which tells you how the experiment went. The purchase closed on June 13, 2017 at $4.48 billion, a price Verizon had renegotiated downward after Yahoo disclosed two enormous data breaches during the deal process. Chief executive Marissa Mayer resigned the day it closed, leaving with a package CNBC reported in 2017 at about $23 million. Verizon then poured the Yahoo sites into a new subsidiary called Oath, alongside AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html), and put former AOL chief Tim Armstrong in charge. The idea was to build a third advertising pillar next to Google and Facebook out of two brands that had already peaked. It did not work. Oath was later renamed Verizon Media, the advertising share never arrived, and by 2021 Verizon was a phone company that wanted to be a phone company. The corporate shell left over from the 2017 sale kept the Alibaba and Yahoo Japan stakes, renamed itself Altaba, sold down those holdings and wound itself up. The operating business and the name went one way, the investments went another. ## How a $44.6 billion offer turned into a $4.48 billion sale The number that haunts every Yahoo ownership question is the one the company said no to. On February 1, 2008 Microsoft made an unsolicited offer to buy Yahoo for $44.6 billion in cash and stock. Ten days later the board turned it down, and it said so in a press release filed with the Securities and Exchange Commission. "After careful evaluation, the Board believes that Microsoft's proposal substantially undervalues Yahoo!" the statement read on February 11, 2008, listing the global brand, the worldwide audience, recent investments in advertising platforms and the stakes in other companies as the reasons the price was too low. Microsoft eventually walked away. Yahoo stayed independent, cycled through chief executives, and nine years later sold its entire operating business to Verizon for $4.48 billion, roughly a tenth of the rejected offer. Four years after that it changed hands again for about $5 billion. The Microsoft bid was not the only chance Yahoo let go. The company also passed on the search engine that became Google (https://404memoryfound.com/posts/yahoo-rejected-google.html), a decision that shaped everything that followed. ## What has Apollo sold off since 2021? Private equity owners tend to sort a portfolio into the parts that earn and the parts that distract, and Apollo has done exactly that with Yahoo. The biggest piece to go was AOL. In October 2025 Apollo agreed to sell it to Bending Spoons, an Italian software group that had already bought Vimeo and Evernote, for roughly $1.5 billion. That is a striking price for a brand that had been the headline asset in a much larger deal a generation earlier. Other properties that arrived with the 2021 purchase have been sold to media investors as well, leaving Apollo with the Yahoo-branded consumer sites and the advertising technology underneath them. Some of what the old portal carried is simply gone rather than sold. Yahoo Answers (https://404memoryfound.com/posts/does-yahoo-answers-still-exist.html) is closed, and the directory, the chat rooms and the personal home pages went years before it. What remains is narrower and, by every outside account, a great deal healthier than the sprawling collection Verizon tried to run. ## Is Yahoo still a real business in 2026? Yes, and the traffic is the argument. Semafor reported in May 2025, citing Similarweb, that Yahoo was among the internet's five busiest destinations, with about 3 billion visits in the previous month, more than Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html). That traffic does not come from people browsing a 1990s directory. It comes from Yahoo Finance, Yahoo Mail, Yahoo Sports, Yahoo News and Yahoo Search, products that millions of Americans open by habit every morning without thinking of them as retro. Jim Lanzone, Yahoo's chief executive, has spent his tenure pushing the company into artificial intelligence rather than defending the portal, and he has been open about the stakes. "If we don't innovate and get out ahead of it, we die," he told Semafor in 2025, comparing executives who ignore the shift to the characters in Don't Look Up who tune out a warning. So the answer to the ownership question has two halves. Apollo Global Management owns Yahoo, and Yahoo is not a museum piece. It is a private, profitable-enough media and advertising business that happens to carry one of the oldest brand names on the web. ## Frequently Asked Questions ### Who owns Yahoo now? Yahoo is owned by funds managed by affiliates of Apollo Global Management, which announced the roughly $5 billion purchase of Verizon Media on May 3, 2021 and completed it later that year. Verizon retained a 10 percent stake. Yahoo has had no publicly traded shares since the deal closed. ### Is Yahoo still a company in 2026? Yes. Yahoo is a privately held company in 2026, still operating Yahoo Finance, Yahoo Mail, Yahoo Sports, Yahoo News and Yahoo Search under chief executive Jim Lanzone. Semafor reported in May 2025 that Yahoo was one of the internet's five busiest destinations, with about 3 billion visits in a month. ### Did Microsoft try to buy Yahoo? Microsoft made an unsolicited offer of $44.6 billion for Yahoo on February 1, 2008. Yahoo's board rejected it on February 11, 2008, saying the proposal substantially undervalued the company, and Microsoft later withdrew. Yahoo's operating business was sold to Verizon for $4.48 billion in 2017 and to Apollo Global Management for about $5 billion in 2021. **Sources:** - Verizon Media to be Acquired by Apollo Funds, Apollo Global Management, 2021: https://www.apollo.com/insights-news/pressreleases/2021/05/verizon-media-to-be-acquired-by-apollo-funds-130247212 - Verizon completes its $4.48 billion acquisition of Yahoo; Marissa Mayer leaves with $23 million, CNBC, 2017: https://www.cnbc.com/2017/06/13/verizon-completes-yahoo-acquisition-marissa-mayer-resigns.html - Yahoo! Inc. Form 8-K exhibit: board statement on Microsoft's proposal, SEC, February 11, 2008: https://www.sec.gov/Archives/edgar/data/0001011006/000095013408002131/f37925exv99w1.htm - If we don't innovate, we die: Yahoo CEO Jim Lanzone on reviving a 30-year-old dot-com star, Semafor, 2025: https://www.semafor.com/article/05/29/2025/if-we-dont-innovate-we-die-yahoo-ceo-jim-lanzone-on-reviving-a-30-year-old-dot-com-star - AOL to be sold to Bending Spoons for roughly $1.5B, Axios, 2025: https://www.axios.com/2025/10/29/aol-bending-spoons-deal --- # Sega Genesis Games in 2026: The Best Way to Play URL: https://404memoryfound.com/posts/sega-genesis-games-how-to-play-2026.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-10 Topics: Gaming, Hardware **Summary:** The best way to play Sega Genesis games in 2026 depends on how much hardware you want in the room. A used Sega Genesis Model 2 cost about $65 loose in September 2026 and still gives the most faithful result, especially with a flash cart and a decent video upscaler. If you want no old hardware at all, Nintendo Switch Online plus the Expansion Pack costs $49.99 a year and includes a Sega Genesis library, while Sega's own Genesis Mini, launched September 19, 2019 at $80, is a used purchase now. **Key facts:** - Launched: August 14, 1989 in North America, by Sega, with Altered Beast packed in - Price then: $189.99 at the 1989 North American launch - Used price today: About $65 loose and $120 complete for a Model 2, September 2026 (PriceCharting) - Cheapest legal route: Nintendo Switch Online plus Expansion Pack, $49.99 a year, includes a Sega Genesis library - Status today: No new Sega Genesis hardware from Sega since the Genesis Mini 2 in 2022; everything official is a used purchase ## Which route to pick, and what each one costs There are five honest ways to play Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html) Genesis games in 2026, and they split by budget and patience rather than by taste. A used console and a shelf of cartridges is the most faithful. PriceCharting listed a loose Sega Genesis Model 2 at about $65 in September 2026, with a complete boxed one near $120. A subscription is the cheapest way in. Nintendo Switch Online plus the Expansion Pack costs $49.99 a year for an individual membership and carries a Sega Genesis library alongside the Nintendo 64 one. The Sega Genesis Mini, Sega's own plug-and-play box, sits in the middle: no cartridges to hunt, HDMI in the box, and emulation Sega supervised itself. The last two routes are for people who want more than a starter set. A flash cart puts a whole backed-up library on one SD card inside a real console. A Linux handheld from a maker such as Anbernic (https://404memoryfound.com/posts/anbernic-handheld-worth-it-game-boy.html) runs the same games on a screen the size of a Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html), at the cost of doing the file work yourself. ## Is a used Sega Genesis still worth buying? Yes, if you want the real machine and you know what to check. The console launched in North America on August 14, 1989 at $189.99 with Altered Beast packed in, and it was built solidly enough that a large share of them still power on 37 years later. Prices have stayed within reach. PriceCharting had the Model 2 at roughly $65 loose and about $120 complete in September 2026, well under half of the 1989 sticker price before any inflation math. Three things separate a good unit from a bad one. Look at the cartridge slot for bent or corroded pins, confirm the seller has the correct Sega power supply for that model, because the Model 1 and Model 2 use different adapters, and check which AV connector the console carries, since the two models do not share a cable. Cartridges are the other half of the cost. Genesis carts were sold in hard plastic clamshell cases, so complete copies survive in real numbers, and the common sports and platform titles are still some of the cheapest 16-bit software anywhere. ## How do you get a 1989 console onto a 2026 television? This is the part that surprises people who dig the box out of a closet. The Genesis outputs analog video only, and most televisions sold in the last decade have dropped composite inputs entirely. There are three fixes, in rising order of cost. A generic composite to HDMI converter box works, but it adds input lag and cannot invent detail that was never in the signal. A dedicated retro upscaler is the enthusiast answer and produces a sharp, low lag picture from the console's RGB or S-video output. A clone console with a real cartridge slot and an HDMI port avoids the question altogether, and several models are still manufactured new. If you keep the original hardware, the single best upgrade is the cable. RGB SCART or component out of a modified console looks dramatically cleaner than the composite lead that came in the box in 1989, and it costs less than the console did. Sound is worth a thought too. The Genesis ran a Yamaha FM synthesis chip, and cheap converter boxes can flatten it into mush. ## Is the Sega Genesis Mini still around, and should you buy one? The Sega Genesis Mini arrived on September 19, 2019 at $80, and it was the first Sega branded plug-and-play box that reviewers took seriously. Kotaku wrote in 2019 that the earlier licensed clones were "full of emulation problems and mediocre games," while the Mini's software came from M2, the Japanese studio known for accurate retro ports. A follow-up, the Genesis Mini 2, came in 2022. Sega has not released another Genesis box since, and the company's classic hardware producer explained why when he was asked about a Saturn or Dreamcast version. "The development of new boards has been stagnant due to the Coronavirus and, of course, it would be a fairly expensive product in terms of cost," Yosuke Okunari told Famitsu in 2022. The tooling costs he described have not gone away, which is the same commercial logic that ended Sega's own console business after Sega stopped making consoles (https://404memoryfound.com/posts/why-sega-stopped-making-consoles-dreamcast.html) in 2001. So the Mini is a used purchase now, like the 1989 hardware it copies. It is the right pick for someone who wants a set of well chosen games on a modern television tonight, and the wrong pick for someone who already owns cartridges. ## What does a flash cart add that a mini console cannot? A flash cart is a cartridge with a microSD slot in the end of it. You load game files onto the card, put the cart in a real Genesis, and the console runs them off original hardware with original controllers and original video output. The better models do more than that. Depending on the cart, they add region switching for Japanese and European releases, save state support, in game cheats, and compatibility with the add-on libraries, so a single cartridge can also cover Master System, Sega CD (https://404memoryfound.com/posts/what-happened-to-sega-cd-add-on.html) and 32X (https://404memoryfound.com/posts/what-happened-to-sega-32x-add-on.html) software. The rule the specialist shops repeat is worth repeating here: use backups of cartridges you actually own. Downloading game files you never bought is not something a flash cart makes legal, whatever a forum tells you. ### Where to find one today A loose Genesis Model 2 ran about $65 in September 2026 and a complete boxed one about $120, according to PriceCharting. Specialist retro dealers charge more than a garage sale but sell tested consoles, cleaned cartridges and flash carts with support behind them, which is the difference between playing tonight and troubleshooting a dead capacitor: Stone Age Gamer's Genesis flash carts (https://stoneagegamer.com/flash/genesis/). ## Can you play Genesis games without owning any Sega hardware? Yes, and it is the cheapest legal start. Nintendo Switch Online plus the Expansion Pack costs $49.99 a year for an individual plan and includes a Sega Genesis library that any subscriber can play on a Switch, with rewind and save states built in. Sega also sells a large official Genesis collection on PC, which bundles dozens of the same games with save states, rewind and online play, and it goes on sale often enough that patience pays. Handheld emulation boxes from Anbernic and its competitors are the fourth option. A 1989 console is trivial work for a modern mobile chip, so Genesis games run at full speed on hardware that costs less than a new game, but the devices ship empty and you supply the files, with the same ownership rule as a flash cart. None of these three routes gives you the cartridge slot, the click of the power switch, or the resale value of the real console. All three get you playing in an afternoon. ## Frequently Asked Questions ### What is the best way to play Sega Genesis games in 2026? For most people in 2026, the best way to play Sega Genesis games is a used Sega Genesis Model 2, which PriceCharting listed at about $65 loose in September 2026, connected through a retro upscaler or an HDMI capable clone console. If you do not want old hardware, Nintendo Switch Online plus the Expansion Pack costs $49.99 a year and includes a Sega Genesis library. ### How much is a Sega Genesis worth today? A Sega Genesis Model 2 console was worth roughly $65 loose and about $120 complete in its box in September 2026, according to PriceCharting. That is well below the $189.99 the console cost at its North American launch on August 14, 1989, and common Genesis cartridges remain among the cheapest 16-bit games on the market. ### Can you still buy a new Sega Genesis Mini? Sega no longer manufactures the Sega Genesis Mini, which launched on September 19, 2019 at $80, or its 2022 follow-up the Genesis Mini 2, so both are used purchases in 2026. Sega's classic hardware producer Yosuke Okunari told Famitsu in 2022 that new board development had stalled and that mini consoles had become expensive to build. **Sources:** - Sega Genesis: release date, models and history, Video Game Console Library: https://www.videogameconsolelibrary.com/console/sega-genesis/ - The Genesis Mini Makes Up For Years Of Crappy Sega Clones, Kotaku, 2019: https://kotaku.com/the-genesis-mini-makes-up-for-years-of-crappy-sega-clon-1837110882 - Sega producer says high costs are preventing a Dreamcast or Saturn mini console, Video Games Chronicle, 2022: https://www.videogameschronicle.com/news/sega-says-high-costs-are-preventing-a-dreamcast-or-saturn-mini-console/ - Nintendo Switch Online + Expansion Pack, Nintendo official site: https://www.nintendo.com/us/online/nintendo-switch-online/expansion-pack/ - Sega Genesis Model 2 Console prices, PriceCharting: https://www.pricecharting.com/game/sega-genesis/sega-genesis-model-2-console --- # Is Habbo Hotel Still Around? The Pixel Hotel in 2026 URL: https://404memoryfound.com/posts/is-habbo-hotel-still-around.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-10 Topics: Internet Culture, Gaming **Summary:** Habbo Hotel is still around in 2026. The virtual hotel launched by Finnish studio Sulake in 2000 is owned today by the Dutch group Azerion, which bought Sulake outright in January 2021, and Sulake said in August 2025 that the site had passed 300 million registered accounts with hundreds of thousands of monthly active users. A separate retro version, Habbo Hotel: Origins, has recreated the 2005 client since June 2024. **Key facts:** - Launched: August 2000 in Finland as Hotelli Kultakala; UK hotel January 2001, US hotel September 2004 - Company: Sulake, founded in Helsinki in 2000 by Sampo Karjalainen and Aapo Kyrola - Peak: 200 million registered accounts by January 2011, more than 13 million monthly unique visitors - Owner today: Azerion, which took 51 percent in 2018 and the remaining 49 percent in January 2021 - Status today: Active in 2026: 300 million registered accounts, hundreds of thousands of monthly active users, plus Habbo Hotel: Origins ## How a Finnish goldfish tank turned into a virtual hotel Habbo Hotel started as a side project in Helsinki. Sampo Karjalainen and Aapo Kyrola founded Sulake in 2000, and their first version of the idea was called Hotelli Kultakala, Finnish for Hotel Goldfish, which went live on a Finnish internet provider's portal in August 2000. The concept was simple enough to explain in a sentence. You picked a pixel character, walked around isometric rooms drawn at an angle, typed into a speech bubble, and bought furniture for a room of your own. The first hotel outside Finland opened in the United Kingdom in January 2001, built with the British entrepreneur Dee Edwards. The American hotel, at habbo.com, opened in September 2004, and it became one of the busiest rooms in the chain almost immediately. Entry was always free. The money came from virtual furniture, called furni by everyone who played, and from a paid membership called Habbo Club. Sulake dropped the word Hotel from the branding in 2006, though almost nobody who grew up on it ever stopped saying Habbo Hotel. By the end of the decade there were separate hotels for different countries and languages, each with its own staff, its own economy, and its own cliques. It was one of the largest virtual worlds on the web before Second Life (https://404memoryfound.com/posts/what-happened-to-second-life-virtual-world.html) convinced the press that virtual worlds were a story worth covering. ## Why 4chan closed the pool in July 2006 On July 12, 2006, users from the 4chan message board flooded Habbo Hotel with hundreds of identical avatars: dark suits, gray skin, and large afros. They arranged themselves into blockades around the swimming pools in the busiest rooms and announced that the pool was closed. The tactic worked because of a design detail. Avatars in Habbo Hotel could not walk through each other, so a line of characters standing shoulder to shoulder was a physical wall that moderators had to remove by hand. The raids were repeated for years afterward, and the phrase outlived the site that produced it. Pool's Closed became one of the first internet memes to escape its origin completely, showing up on forums and in image macros long after the people sharing it had any idea what Habbo Hotel was. For Sulake the raids were an early warning. A world with millions of teenagers in it, run by a company of a few hundred people, was going to be very hard to police. That problem came back six years later in a far more serious form. ## What the 2012 Channel 4 News report cost Sulake On June 12, 2012, Channel 4 News in the United Kingdom aired the results of a two-month investigation into Habbo Hotel. Reporters found sexually explicit chat on a site used by children, and concluded that moderation was far weaker than Sulake had claimed publicly. Sulake's response the next day was drastic. On June 13, 2012, the company muted every user in every hotel worldwide at once, an event players still call the Great Mute. Chat, private messages and the console all went silent. Chief executive Paul LaFontaine said the company had "taken the decision to mute all conversations across the site," in a statement reported by ITV News in 2012. The financial damage arrived within hours. Balderton Capital gave up its 13 percent stake in Sulake and resigned its board seat, as TechCrunch reported on June 12, 2012, and the investment firm 3i pulled its 16 percent stake the following day. British retailers pulled Habbo gift cards from shelves. Chat came back on June 19, 2012 with tighter filters and a chat system that limited what younger accounts could type. The site survived, but the growth curve never came back. ## How many people still log in to Habbo Hotel Habbo Hotel is still running in 2026, and the honest answer about its size is that it is a fraction of what it was, but a fraction of a very large number. The peak is well documented. Sulake announced that Habbo Hotel had passed 200 million registered accounts in January 2011, and VentureBeat reported in February 2011 that the site was drawing more than 13 million unique visitors a month. The current figures come from Sulake itself. In August 2025, around the site's 25th anniversary, Sulake product director Mika Timonen told Forbes that Habbo Hotel had passed 300 million total registered users and had hundreds of thousands of monthly active users across nine language communities. That puts it in unusual company. It is smaller than it was, but it is still open, which is more than can be said for Club Penguin (https://404memoryfound.com/posts/what-happened-to-club-penguin.html), and it never needed a rescue campaign from its own players to stay online. Timonen was blunt about the modern competition. "When we see platforms like Roblox, Fortnite, and others rising, we don't feel threatened, we feel proud," he told Forbes in 2025. ## Who owns Habbo Hotel now, and who runs it Sulake still develops and operates Habbo Hotel from Helsinki, but Sulake itself is no longer independent. The Dutch company Azerion bought a 51 percent controlling stake in 2018, under the name Orange Games, and acquired the remaining 49 percent in January 2021. Azerion is a digital entertainment and advertising group listed in Amsterdam, and Habbo Hotel is one of the older assets in its portfolio. That ownership change coincided with the biggest technical rebuild in the site's history. Habbo Hotel ran on Flash for two decades, and Flash was being switched off. Sulake opened a Unity client in beta on December 23, 2020 and retired the Flash version on January 1, 2021, the same deadline that killed thousands of browser games built on Macromedia Flash (https://404memoryfound.com/posts/definitive-history-macromedia-flash.html). The reaction was ugly. Long-time players complained that the Unity client had lost features they used daily, the hashtag SaveHabbo spread across social media, and in February 2021 Sulake added an Adobe AIR client alongside Unity to bring some of them back. ## What Habbo Hotel: Origins brought back The most interesting thing Sulake has done in years is admit that a lot of people preferred the old version. In June 2024 it launched Habbo Hotel: Origins, a recreation of the 2005 client, with the old navigation, the old furniture catalog, and the visual style people actually remember. Origins is rated for adults rather than children, which is the point. The audience that filled the hotels in 2005 is now in its thirties and forties, and Sulake built a version for them instead of trying to win back a teenage market that has gone to Roblox. It runs as a separate service with its own hotels, alongside the main Unity version of Habbo Hotel, and it later added a release on Steam. Both versions are live in 2026. That is the short answer to what happened to Habbo Hotel. It did not shut down, it did not get sold for parts, and it did not quietly become a shell. It got smaller, changed owners, rebuilt itself twice, and then reopened a replica of its own past next door. ## Frequently Asked Questions ### Is Habbo Hotel still online in 2026? Yes. Habbo Hotel is still online in 2026, operated by Sulake in Helsinki and owned by the Dutch group Azerion, which took full control of Sulake in January 2021. Sulake said in August 2025 that the site had passed 300 million registered accounts and had hundreds of thousands of monthly active users. ### Why did Habbo Hotel mute every user in 2012? Habbo Hotel muted every user worldwide on June 13, 2012, the day after a Channel 4 News investigation in the United Kingdom found explicit chat and weak moderation on a site used by children. Chat returned on June 19, 2012 with stricter filters, but investors Balderton Capital and 3i had already given up their stakes in Sulake. ### Can you still play the 2005 version of Habbo Hotel? Yes. Sulake launched Habbo Hotel: Origins in June 2024, a recreation of the 2005 Habbo Hotel client with the original look and catalog, rated for adult players. It runs as a separate service from the main Habbo Hotel and is still live in 2026. **Sources:** - Habbo (history, hotels, Flash to Unity, Origins): https://en.wikipedia.org/wiki/Habbo - Virtual world Habbo hits more than 200M registered users, VentureBeat, 2011: https://venturebeat.com/2011/02/03/virtual-world-habbo-hits-more-than-200m-registered-users - Trouble At Habbo Hotel: Balderton Dumps Its 13% Stake In Owner Sulake, TechCrunch, 2012: https://techcrunch.com/2012/06/12/trouble-at-habbo-hotel-balderton-dumps-13-stake-in-owner-sulake-over-dodgy-child-content/ - 'Greatly saddened' Habbo Hotel CEO suspends chats following accusations, ITV News, 2012: https://www.itv.com/news/update/2012-06-13/greatly-saddened-habbo-hotel-ceo-suspends-chats-following-accusations/ - How Social MMO Habbo Has Thrived and Survived for Over 25 Years, Forbes, 2025: https://www.forbes.com/sites/davidjagneaux/2025/08/15/how-social-mmo-habbo-has-thrived-and-survived-for-over-25-years/ --- # Does Yahoo Answers Still Exist? Why It Shut Down URL: https://404memoryfound.com/posts/does-yahoo-answers-still-exist.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-10 Topics: Internet Culture, Business Blunders **Summary:** Yahoo Answers no longer exists. Yahoo took the site read-only on April 20, 2021 and shut it down on May 4, 2021, more than 15 years after the December 2005 launch, and answers.yahoo.com now sends visitors to Yahoo's home page. The old questions survive only in web archives, and the Yahoo brand itself has belonged to Apollo Global Management since September 2021. **Key facts:** - Launched: December 8, 2005, as a public beta by Yahoo - Shut down: May 4, 2021, read-only from April 20, 2021 - Peak scale: 300 million questions asked by March 2012 - Status today: Closed. answers.yahoo.com redirects to Yahoo's home page - Owner today: Apollo Global Management, which closed its $5 billion purchase of Yahoo on September 1, 2021 ## What Yahoo switched on in December 2005 Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) Answers opened to the public as a beta on December 8, 2005 at answers.yahoo.com. Yahoo's own announcement that day framed the service as a way for people to "tap into the collective intelligence of Web users" for the everyday questions a search box handled badly. The design took one sentence to explain. You asked a question in plain English, the site filed it in a category, strangers answered, and after a few days you picked one reply as the Best Answer. Points ran everything. Asking spent them, answering earned them, and a Best Answer earned more. Nobody could cash them in for anything, which is part of why people chased them so hard. The timing was right. Search engines in 2005 were built for keywords rather than sentences, and Ask Jeeves (https://404memoryfound.com/posts/what-happened-to-ask-jeeves-search-engine.html) had already shown that Americans wanted to type questions as questions. Yahoo Answers gave those questions somewhere to land where a person replied. ## How big did the Q and A site get at its peak? Large, and faster than almost anyone predicted. By March 2012 the site had collected 300 million questions since launch, Danny Sullivan reported in Search Engine Land. The rate figures in that report describe what the site felt like from the inside. Roughly 2 questions were asked and 6 answered every second, which works out to about 7,000 questions and 21,000 answers an hour. Even that was the sound of a site past its best. Two years earlier Yahoo had counted more than 34,000 questions and answers an hour, so the 2012 rate was down 17 to 18 percent. Nothing else in English ran a general question board at that volume. Reddit had not yet become the default place to ask a stranger anything, and Stack Overflow only wanted programming questions. Yahoo Answers took all of it: homework, breakups, car noises, symptoms, religion, tax panic, and an enormous amount of nonsense. ## When Stephen Hawking asked the internet a question In July 2006 Yahoo ran a promotion called Ask The Planet, in which ten well-known people posted questions for the site to answer. One of them was Stephen Hawking. His question read: "In a world that is in chaos politically, socially and environmentally, how can the human race sustain another 100 years?" NBC News reported that nearly 17,000 people had answered within days. The replies were what a general audience produces when handed a cosmologist's question. NBC News quoted one that ran "get rid of nuclear weapons" and another that read "I don't think it is possible unless we expand into space." The University of Cambridge, where Hawking taught, confirmed to NBC News in 2006 that he had written the message and said he would not comment further. That week is the high point of the whole idea. A famous physicist put a serious question to the general public in public, and the general public answered, sincerely and clumsily and 17,000 times over. ## Why the answers got worse than the questions Yahoo Answers had no gate. Anyone with a Yahoo account could post, questions reached category front pages without review, and the reward for a careful answer and a joke answer was identical. Trolls found that irresistible and spammers found those front pages valuable. Moderation ran on user reports routed through a corporate review process that could not keep pace with thousands of new posts an hour. Screenshots of the worst spelling and the strangest questions turned into internet-wide jokes. That brought more people to the site as an audience rather than as askers, and a comedy site has no need of correct answers, so fewer people bothered to write them. Google changed the other half of the equation. As search results got better at answering plain-language questions on the results page itself, the reason to open a question thread at all got thinner every year. Yahoo had bigger problems by then. The company that let Flickr (https://404memoryfound.com/posts/what-happened-to-flickr-yahoo-photo-sharing.html) drift and later sold Tumblr (https://404memoryfound.com/posts/what-happened-to-tumblr-yahoo-billion-dollar-mistake.html) for a fraction of what it paid was never going to spend its remaining engineering budget on the manners of a message board. ## The 29 days from the notice to the shutdown On April 5, 2021 a short notice went up on the Yahoo Answers home page. "While Yahoo Answers was once a key part of Yahoo's products and services, it has become less popular over the years as the needs of our members have changed," it said. The schedule was brisk. Posting stopped on April 20, 2021, when the site went read-only. The site closed completely on May 4, 2021. Users had until June 30, 2021 to download an archive of what they had posted. Engadget reported the announcement the day it appeared and laid out those dates. Twenty-nine days separated the notice from the shutdown, which is not much runway for 15 years of accumulated content. There was no sale, no spin-off, and no handoff of the archive to a library or a university. Yahoo simply took the content offline instead of parking it somewhere read-only. Volunteer archivists reacted the way they usually do. Through April 2021 they raced the clock, pulling question pages into public web archives before the deadline, which is the only reason anything is readable now. ## Can you still read the old questions in 2026? Not on Yahoo. The address answers.yahoo.com no longer serves questions and sends visitors to Yahoo's home page instead. There is no replacement product and no announced plan for one. What survives sits in the Internet Archive. Pages captured before May 4, 2021 can still be read through the Wayback Machine one URL at a time, and the coverage is uneven: heavily linked questions survived, obscure ones often did not. The Yahoo brand changed hands in the same year the site died. Apollo Global Management completed its acquisition of what was then Verizon Media on September 1, 2021 in a deal worth $5 billion, renamed the business Yahoo, and left Verizon holding about 10 percent. Apollo has owned it ever since. The questions themselves went somewhere else rather than stopping. Reddit absorbed most of the ask-a-stranger traffic, Quora took the ones people wanted an expert to answer, and Stack Exchange kept the technical ones. Search engines and chatbots now handle the rest before another human ever sees them. ## Frequently Asked Questions ### Does Yahoo Answers still exist in 2026? No. Yahoo Answers stopped accepting posts on April 20, 2021 and shut down completely on May 4, 2021, and in 2026 the address answers.yahoo.com sends visitors to Yahoo's home page. Yahoo has not launched a replacement question and answer service. ### Why did Yahoo Answers shut down? Yahoo closed Yahoo Answers on May 4, 2021 because use of the site had been falling for years. The notice Yahoo posted on April 5, 2021 said the service "has become less popular over the years as the needs of our members have changed," and Search Engine Land had already reported in 2012 that question and answer volume was down 17 to 18 percent from two years earlier. ### Where can you read old Yahoo Answers questions? Old Yahoo Answers pages survive mainly in the Internet Archive's Wayback Machine, where volunteers and crawlers saved question threads in April 2021 before the May 4, 2021 shutdown. Yahoo removed the content from its own servers and gave account holders only until June 30, 2021 to download an archive of their own questions and answers. **Sources:** - Yahoo press release: Yahoo! Introduces Yahoo! Answers Beta (December 8, 2005): https://altaba.gcs-web.com/news-releases/news-release-details/yahoo-introduces-yahoo-answers-beta-service-enables-people-tap - Search Engine Land: Yahoo Answers Hits 300 Million Questions, But Q&A Activity Is Declining (2012): https://searchengineland.com/yahoo-answers-hits-300-million-questions-but-qa-activity-is-declining-127314 - NBC News: Hawking searches for answers... on Yahoo (2006): https://www.nbcnews.com/id/wbna13781538 - Engadget: Yahoo Answers is shutting down on May 4th (2021): https://www.engadget.com/yahoo-answers-shutdown-may-4th-210240460.html - Apollo Global Management: Apollo Funds Complete Acquisition of Yahoo (September 1, 2021): https://www.apollo.com/insights-news/pressreleases/2021/09/apollo-funds-complete-acquisition-of-yahoo-161530593 --- # Betamax vs VHS: Why Sony Lost the First Format War URL: https://404memoryfound.com/posts/betamax-vs-vhs-format-war.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-10 Topics: Hardware, Then vs Now **Summary:** Betamax lost to VHS because a VHS cassette recorded two hours to Betamax's one, the first VHS deck sold in America undercut Sony by $295, and JVC licensed its format to anyone who would build it. Sony's picture-quality edge was 250 lines against 240, and it traded most of that away to add recording time. Sony sold about 18 million Betamax recorders, stopped building them in 2002, and shipped the last blank tapes in March 2016. **Key facts:** - Launched: May 10, 1975 in Japan, February 1976 in the United States - Company: Sony - Launch price: $1,295 for the SL-7200 in 1976, about $7,600 in 2026 dollars - Units sold: About 18 million Betamax recorders worldwide - Status today: Beaten by VHS, recorders ended in 2002, blank cassettes in March 2016 ## Why did Betamax lose to VHS? Betamax lost to VHS on three things: tape length, price, and the number of factories building machines. Sony launched Betamax in Japan on May 10, 1975 with a recorder that held one hour of video. JVC answered on October 31, 1976 with the HR-3300, the first VHS deck, and its cassette held two. One hour was enough for a sitcom. It was not enough for a film, a baseball game, or an evening out of the house. Sony had designed the format around a cassette small enough to slip into a coat pocket, and that choice locked in the running time before anyone knew what buyers would actually do with the machine. Sony spent the rest of the war catching up. The SL-8200 arrived in 1977 with a half-speed mode that fit two hours onto an L-500 cassette. VHS decks were already recording four hours by then, and later six and eight. Every time Sony matched the number, the number had moved. ## What a Betamax cost in 1976, and what VHS cost in 1977 The first stand-alone Betamax deck sold in the United States, the Sony SL-7200, reached stores in February 1976 at $1,295. Adjusted with the consumer price index, that is roughly $7,600 in 2026 money for a machine that recorded one hour at a time. RCA announced its answer on August 23, 1977 and put it on sale that October. The SelectaVision VBT200 was built by Matsushita, ran VHS tapes, and recorded four hours. It sold for $1,000, and RCA advertised it with a line that made the entire argument in five words: four hours, $1,000, SelectaVision. A shopper standing in a store in late 1977 was therefore looking at a $295 price gap and a four-to-two gap in recording time, both running the same direction. Sony's answer was that Beta looked slightly better. That is a hard sell against a cheaper box that can record a whole football game unattended. ## How JVC's open licence beat Sony's closed one Sony kept Betamax as a Sony product and licensed it narrowly. JVC treated VHS as a standard and licensed it to anyone willing to build it, including its parent company Matsushita, plus RCA, Sharp, Zenith, and Magnavox. The two formats were not really competing products. They were competing supply chains. The gap shows up in the tape aisle. By 1984 roughly 40 companies were manufacturing VHS cassettes against 12 making Betamax cassettes. More manufacturers meant lower prices, more shelf space, and more rental stock, because video shops carried whatever the most customers could play at home. Market share followed. Beta slipped to around 25 percent of VCR sales by 1981 and 7.5 percent by 1986. In 1988 Sony began selling VHS recorders under its own name, which settled the question without anyone using the word surrender. Sony would run the closed-format play again with MiniDisc (https://404memoryfound.com/posts/what-happened-to-minidisc-sony.html), and would finally win one two decades later in the HD DVD fight (https://404memoryfound.com/posts/what-happened-to-hd-dvd-format-war.html), by signing up studios and hardware partners before the shooting started. ## Was Betamax actually the better format? Slightly, at first, and by less than the legend claims. At its original speed Betamax resolved 250 horizontal lines against 240 for VHS. Ten lines, on a 1977 television, across a living room. Sony then traded that edge away on its own. The two-hour Beta II mode that answered VHS ran the tape at half speed and gave up picture quality to do it, so the setting nearly everyone actually used was no longer the sharper one. Sony did eventually build the machine the legend describes. ED Beta, launched in 1988, recorded about 500 lines of luminance resolution and outclassed everything else on the consumer market. It shipped the same year Sony started making VHS decks, to a public that had stopped listening a decade earlier. The one place the engineering paid off was professional: Betacam, a broadcast derivative using a similar cassette shell, became standard equipment in television news long after consumer Beta was finished. ## Did adult films really decide the format war? This is the version everyone repeats, and it does not survive contact with the timeline. Sony did have rules about what it would attach its own name to, but that governed what Sony itself produced and distributed. It never stopped anyone from recording whatever they liked onto a blank Beta cassette, and adult titles were duplicated onto both formats, because distributors wanted revenue from both sets of owners. The bigger problem with the story is that it is not needed. Beta was already losing on the two numbers buyers compared in the shop, and losing badly on the number of companies making tapes and decks. Two hours beat one, $1,000 beat $1,295, and 40 factories beat 12. Nothing about the format war requires a more colourful explanation than that. ## The 1984 Supreme Court case Sony won Universal City Studios and Walt Disney Productions sued Sony in Los Angeles federal court in 1976. Their argument was that a machine built to record television is a machine built to infringe copyright, and that Sony should be liable for what its customers did with it. They asked the court to stop the sale of the Betamax. The case ran for eight years. On January 17, 1984 the Supreme Court ruled 5 to 4 for Sony in Sony Corp. of America v. Universal City Studios, 464 U.S. 417. Recording a broadcast at home to watch it later, which the court called time-shifting, was fair use, and a device capable of substantial non-infringing uses does not make its manufacturer a contributory infringer. That holding outlasted the format by decades. It is the legal ground under the cassette deck, the CD burner, the DVR (https://404memoryfound.com/posts/what-happened-to-tivo-dvr-pioneer.html), and most of the recording hardware sold since. Sony lost the format and won the case that kept the entire home video business legal, including the VHS trade that had beaten it. ## Where to find a Betamax today Working Sony Beta decks still turn up on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html) most weeks. Refurbished players commonly sell for about $50 to $200, while clean high-end Super Beta Hi-Fi machines are listed in the high hundreds and occasionally above $1,000. Tapes are cheap, generally a few dollars to around $20, with rare horror and cult titles reaching $50 and up. Check the belts and the loading tray before paying: forty-year-old rubber is the usual point of failure, and spare parts are scarce. ## Frequently Asked Questions ### Why did Betamax fail if the picture was better? The picture advantage was 250 lines against 240 at Betamax's original speed, and Sony gave most of it back when it introduced the slower two-hour Beta II mode in 1977. Buyers were comparing a $1,295 one-hour Sony against a $1,000 four-hour RCA, and the tape length and price decided it. ### When did Betamax stop being made? Sony stopped manufacturing Betamax recorders in 2002, then kept producing blank Betamax cassettes for another fourteen years. The last tapes shipped in March 2016, roughly 41 years after the format launched in Japan. ### Are Betamax tapes worth anything? Most are not. Common blanks and mainstream titles sell for a few dollars to about $20, while scarce horror, sci-fi, and cult releases can reach $50 or more in good condition. The machines are usually worth more than the tapes, with refurbished decks typically listed between $50 and $200. **Sources:** - Betamax: https://en.wikipedia.org/wiki/Betamax - Sony's First Betamax VCR, the SL-7200 from 1976: https://www.cedmagic.com/history/betamax-sl-7200-1976.html - VBT200: The First RCA SelectaVision VHS Video Cassette Recorder: https://www.cedmagic.com/history/vbt200.html - Sony Corp. of America v. Universal City Studios, Inc., 464 U.S. 417 (1984): https://supreme.justia.com/cases/federal/us/464/417/ - Value of 1976 US dollars today (CPI inflation calculator): https://www.in2013dollars.com/us/inflation/1976 --- # Who Owns Napster in 2026? Every Owner Since 2001 URL: https://404memoryfound.com/posts/who-owns-napster-now.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-10 Topics: Business Blunders, Music & Entertainment, Internet Culture **Summary:** Napster is owned by Napster Corporation, the company called Infinite Reality until May 2025, which paid $207 million for the brand that March. It is the sixth owner of the name since the original file-sharing service closed in 2001, and on January 1, 2026 it shut Napster's music streaming down for good. The name now sells artificial-intelligence software. **Key facts:** - Brand launched: 1999, as a file-sharing app by Shawn Fanning and Sean Parker - Owners since 2001: Six: Roxio, Best Buy, Rhapsody International, MelodyVR, Hivemind and Algorand, Infinite Reality - Owner today: Napster Corporation, formerly Infinite Reality, since March 2025 - Last sale price: $207 million in March 2025, against about $5.3 million in 2002 - Status today: Music streaming shut down January 1, 2026; the brand now sells AI software ## How a bankrupt file-sharing site turned into a brand for sale The Napster that shut down in 2001 and the Napster that exists in 2026 share a name and almost nothing else. The original service, built by Shawn Fanning and Sean Parker in 1999, was closed by court order, and the company behind it went into bankruptcy in 2002. What went up for sale was not the network. It was the word, the cat-with-headphones logo and a stack of patents. Roxio, a California software company known for CD-burning tools, bought those assets out of the bankruptcy in 2002 for about $5.3 million. Roxio left the lawsuits behind. The copyright liabilities stayed with the dead company while the trademark walked out clean, which is the whole reason the brand still exists. It was cheap, famous and legally detached from everything it had done. Roxio relaunched Napster in 2003 as a paid download and subscription store, then renamed the entire company Napster, Inc. The free-for-all it had started moved on without it, to Kazaa and the second wave of file sharing (https://404memoryfound.com/posts/what-happened-to-kazaa-p2p-wars.html), while the original Napster story (https://404memoryfound.com/posts/napster-destroyed-music.html) became something a marketing department owned. ## Why Best Buy paid $121 million for a music brand in 2008 In September 2008 Best Buy agreed to buy Napster, Inc. for $121 million, or $2.65 a share, and closed the deal that October. The purchase came with roughly 700,000 paying subscribers and a store running a distant second to Apple's iTunes. The retail logic was to sell music subscriptions next to the hardware that played them. It did not work. In October 2011, three years after buying it, Best Buy handed Napster to Rhapsody, the subscription service majority owned by RealNetworks. Terms were not disclosed, and Best Buy took a minority stake in Rhapsody rather than cash. Rhapsody absorbed the subscribers and retired the Napster name in the United States, keeping it alive in Europe where it tested better than Rhapsody did. In 2016 the company reversed itself and put the Napster name back on the American service, then on everything. A brand that had been buried was dug up because it was more recognizable than the company that owned it. The same trick explains who owns the Atari name today (https://404memoryfound.com/posts/who-owns-atari-now.html). ## The $70 million VR deal and the crypto owners who came next In August 2020 MelodyVR, a British virtual-reality concert company, agreed to buy Rhapsody International, by then trading as Napster, from RealNetworks for about $70 million. The deal closed at the end of that year. The price was less impressive than the headline. Only part of it was cash. A large share was MelodyVR taking on money Napster already owed to record labels and other partners, so the buyer inherited a subscriber base of around 3 million and a bill. The pitch was to bolt virtual-reality concerts onto a streaming catalog, and that combination never turned into a business. In 2022 Napster changed hands again, to the crypto investment firm Hivemind Capital Partners and the blockchain company Algorand, who announced a web3 relaunch with artist tokens. The price was not disclosed. Very little of it shipped. By 2025 Napster was what it had been for two decades: a mid-sized music subscription app carrying a name far more famous than its market share. ## What Infinite Reality bought for $207 million in 2025 In March 2025 Infinite Reality, a media and technology company, bought Napster for $207 million. That is roughly forty times what Roxio paid for the same trademark in 2002, and it was paid for a service that no longer led anything. Two months later the buyer took the name for itself. Infinite Reality announced in May 2025 that it would rebrand as Napster Corporation and gather its artificial-intelligence products under a division called Napster AI. The music service was not the reason for the purchase. The word was. The record labels noticed when the payments stopped. In August 2025 Sony Music sued Napster, its parent Rhapsody International and Infinite Reality, claiming $9.2 million in unpaid royalties and seeking up to $37.5 million more in copyright damages. Sony had terminated its license in June 2025 and said the catalog stayed up anyway. By the end of 2025 Napster was a streaming service missing a major label, owned by a company that had decided its future was software agents. ## What happened to Napster on January 1, 2026 Subscribers learned about it the way this story deserves. On January 1, 2026, playback in the Napster app stopped mid-listen and a notice took over the screen. "Napster is no longer a music streaming service. We've become an AI platform for creating and experiencing music in new ways," the message read, as reported by Digital Music News in January 2026. There was no wind-down and no migration inside the app. The notice pointed people to TuneMyMusic, a third-party playlist transfer tool, which in practice meant exporting your library to a competitor. Twenty-five years after Napster handed the record industry's catalog to anyone with a modem, the brand left licensed music altogether. What remains is Napster Corporation, which sells AI companions and agent software for businesses. The service people actually paid for is gone, and nothing has been announced about bringing it back. ## The $3 billion investor who never existed The oddest part of Napster's current ownership has nothing to do with music. In 2025 the company then called Infinite Reality told reporters it had raised more than $3 billion from an investor it would not name. In June 2026 federal authorities said the money was never there. The Department of Justice charged Charles J. Cole, a North Carolina man, with fraud and conspiracy, and the Securities and Exchange Commission filed parallel civil claims that also named his attorney, Torben M. Welch. Cole had obtained 239 million shares of Infinite Reality, roughly a quarter of the company, by claiming access to $55 billion in cash. Regulators say the proof was manufactured, including forged bank statements and a fake website built to mirror a real Malaysian bank. "Cole, with Welch's assistance, then used the fraudulently obtained shares in Infinite Reality to secure a $1 million loan from a private third-party lender that he never repaid," the SEC said in its 2026 litigation release. So the ownership answer carries a footnote. A quarter of the parent company's stock went to someone the government says had nothing to pay with. ## Frequently Asked Questions ### Is Napster still around in 2026? The Napster name is still in use, but the music service is not. Napster shut down its licensed streaming service on January 1, 2026, and the brand now belongs to Napster Corporation, which sells AI companion and agent software instead of music subscriptions. ### Who owns Napster now? Napster is owned by Napster Corporation, the company called Infinite Reality until May 2025, which bought the brand for $207 million in March 2025. Before that, the Napster name passed through Roxio in 2002, Best Buy in 2008, Rhapsody International in 2011, MelodyVR in 2020 and a crypto group led by Hivemind Capital Partners and Algorand in 2022. ### Can you still stream music on Napster? No. Napster stopped streaming licensed music on January 1, 2026, and directed subscribers to the third-party tool TuneMyMusic to move their playlists elsewhere. Sony Music had already terminated its license in June 2025 and sued Napster that August over $9.2 million in unpaid royalties. **Sources:** - Napster (streaming service), ownership history: https://en.wikipedia.org/wiki/Napster_(streaming_service) - Music Ally: Best Buy acquires Napster for $121 million (2008): https://musically.com/2008/09/16/best-buy-acquires-napster-for-121-million/ - Music Business Worldwide: Sony Music sues Napster over $9.2m in unpaid royalties (2025): https://www.musicbusinessworldwide.com/sony-music-sues-napster-over-9-2m-in-unpaid-royalties-up-to-37-5m-in-copyright-claims/ - Digital Music News: Napster music streaming service abruptly shuts down, pivots to AI (2026): https://www.digitalmusicnews.com/2026/01/02/napster-music-streaming-shut-down-ai-pivot/ - SEC litigation release: SEC v. Charles J. Cole, Torben M. Welch, et al. (2026): https://www.sec.gov/enforcement-litigation/litigation-releases/lr-26563 --- # How to Digitize VHS Tapes in 2026: DIY vs Legacybox URL: https://404memoryfound.com/posts/how-to-digitize-vhs-tapes-2026.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-10 Topics: Then vs Now, Hardware **Summary:** There are two honest ways to digitize VHS tapes in 2026. At home, a secondhand VCR and a USB capture device cost about $30 to $80 for the capture hardware and nothing per tape after that; by mail, services such as Legacybox run roughly $15 to $30 a tape. Nobody has built a new VCR since Funai stopped in July 2016, and the standard reference on magnetic tape puts its useful life near 30 years, so the tapes in the closet are already overdue. **Key facts:** - Last VCR made: End of July 2016, by Funai Electric, after 750,000 units in its final year - DIY cost in 2026: About $30 to $80 for a USB capture device, plus $30 to $300 for a used VCR - Mail-in cost in 2026: Roughly $15 to $30 a tape, sold as fixed-size kits - Tape life expectancy: About 30 years at 68F and 50 percent humidity (CLIR, 1995) - Status today: Still doable at home; Legacybox, Southtree and Kodak Digitizing are one Chattanooga company ## Are your tapes already too old to save? Magnetic tape has a shelf life, and home video sat on the cheapest magnetic tape ever mass produced. Even a cassette filled on New Year's Eve 1999 is 26 years old in 2026. The reference number comes from a 1995 document. In "Magnetic Tape Storage and Handling: A Guide for Libraries and Archives", written by John Van Bogart for the Commission on Preservation and Access, the verdict is flat: "thirty years appears to be the upper limit for magnetic tape products". That ceiling assumes good storage. Van Bogart's estimates give videotape roughly 30 years at 68 degrees Fahrenheit and 50 percent relative humidity, about 10 years if the room runs at 76 degrees, and about 5 years at 80 percent humidity. A box in a Phoenix garage and a box in a basement closet are not the same box. The damage is chemical as well as magnetic. The binder that glues magnetic particles to the polyester backing absorbs moisture and breaks down, which is what makes a neglected cassette squeal, shed brown dust and stick to the heads rather than simply look washed out. None of that means a 1992 birthday tape is lost. It means the copy made in 2026 will be better than the copy made in 2031, and that the order of operations is capture first, tidy up later. ## The cheap route: a used VCR and a USB capture stick The do-it-yourself chain has four parts: a working deck, a capture device, a computer and free software. Nobody has built a new VCR since Funai Electric shut its line at the end of July 2016. Funai once sold about 15 million VCRs a year worldwide and was down to 750,000 in its final year, and the end came from the supply chain rather than the customers. A Funai spokesman told AFP in 2016 that a supplier had quit, and that "we can't make them without that part". So the deck is secondhand. Capture's 2026 buying guide puts working players at roughly $30 to $300, the low end being thrift shops and estate sales, the high end being sellers who service the transport and belts before shipping. The capture device is the cheap part. The same guide puts useful USB capture hardware at about $30 to $80 and warns that the sub-$20 sticks introduce artifacts, audio drift and dropped frames that no amount of editing fixes afterward. The software can be free: OBS Studio records whatever the capture device sends it straight to an MP4 file. Budget the time, not just the money. Every capture runs in real time, so a two-hour tape takes two hours, and 40 tapes is a month of evenings. Under roughly ten tapes, the math favors paying somebody. Over 30, the gear pays for itself on the first box. ## What a digitized tape actually looks like on a 4K TV VHS is an analog standard-definition format, and capturing it does not add detail that was never recorded. A clean transfer looks exactly like the tape looked in 1997 on a 27-inch tube, stretched across a screen four times the size. That is the honest baseline, and it is worth setting before spending money. A $200 capture card does not produce a sharper image than a $50 one; it produces a more faithful one, with fewer dropped frames and audio that stays in sync across a three-hour recording. The bigger quality lever is the deck. A player with clean heads and working tracking control recovers detail that a rattling thrift-store unit turns into noise, which is why professional shops keep rooms full of serviced machines. Cleanup, if it happens at all, happens after capture. Save the raw file first, then experiment with deinterlacing or noise reduction on a copy. Home video was never broadcast quality, and it was never meant to be watched on the hardware that replaced the DVR generation (https://404memoryfound.com/posts/what-happened-to-tivo-dvr-pioneer.html) that followed the VCR. ## Hi8, VHS-C and MiniDV are three different jobs The word "tapes" hides at least four formats in the average shoebox, and they do not all take the same route. VHS and S-VHS play in a standard deck. VHS-C, the compact cassette used by camcorders, does not: it needs a motorized full-size adapter, or the original camcorder, and the adapters fail more often than the tapes do. Video8 and Hi8 are analog, like VHS, but they only play in an 8mm camcorder, so the camcorder becomes the deck and its analog output feeds the same capture chain. MiniDV is the exception, because MiniDV was already digital. A MiniDV camcorder connected over FireWire or i.Link hands the computer the recorded data rather than a re-recorded picture, which is the one home format that transfers without a generation of loss. The catch in 2026 is the connector: FireWire ports vanished from laptops years ago, and adapters are a hunt of their own. Store-bought movie tapes are a separate question and rarely worth digitizing. Nearly everything Hollywood released on VHS is on disc or streaming, and the last Blockbuster (https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html) still keeps a wall of them for anyone who misses the format. ## What mail-in services charge, and who does the work Capture's 2026 guide puts professional digitizing at $15 to $30 a tape, which is where the DIY math stops making sense for small collections. Mail-in kits are sold by item count rather than per tape, so the quoted price covers a box of a fixed size whether it arrives full or half empty. Legacybox is the name most Americans have seen, usually in a holiday ad. It runs out of Chattanooga, Tennessee, where founders Nick Macco and Adam Boeselager started the business in 2009 after transferring tapes to DVD as college students. Macco told the Chattanooga Times Free Press in 2026 that the plant holds "the largest working collection of VCRs in the world", which is less of a boast than a job requirement. Comparison shopping is narrower than it looks. Legacybox, Southtree and Kodak Digitizing are brands of the same Chattanooga company, AMB Media, and the Kodak name is used under a trademark license. Eastman Kodak (https://404memoryfound.com/posts/what-happened-to-kodak-digital-camera-bankruptcy.html) is not the company handling the tapes. Set expectations on speed. Reviewers in 2026 report waits of roughly three months for standard processing, longer around the holidays, and that clock starts when the box arrives at the factory rather than when it is ordered. Anyone with a deadline, a memorial or an anniversary should send the box early or drive to a local shop instead. Where to find one today: mail-in digitizing kits are priced by the number of items, and the per-item price drops as the kit gets bigger, so it pays to gather everything in the house before ordering. Prices and kit sizes are listed at Legacybox (https://legacybox.com/pages/pricing). Whichever route wins, label the box, count the tapes and photograph the labels before anything ships, because a handwritten "Christmas 94" is the only index that exists. ## Frequently Asked Questions ### How much does it cost to digitize VHS tapes in 2026? Doing it at home in 2026 costs about $30 to $80 for a USB capture device plus roughly $30 to $300 for a secondhand VCR, after which every tape is free apart from the time. Mail-in services such as Legacybox charge in the range of $15 to $30 a tape, sold as fixed-size kits rather than per cassette. ### Can you still buy a VCR in 2026? Not a new one. Funai Electric, the last manufacturer, ended VCR production at the end of July 2016 after selling 750,000 units in its final year, down from about 15 million a year at the peak. Every VCR available in 2026 is used or refurbished, and working decks run roughly $30 to $300. ### Do VHS tapes go bad if you never play them? Yes. VHS tapes degrade in storage because the binder holding the magnetic particles absorbs moisture and breaks down. The 1995 CLIR guide by John Van Bogart estimates about 30 years of life at 68 degrees Fahrenheit and 50 percent humidity, dropping to about 10 years at 76 degrees and about 5 years at 80 percent humidity. **Sources:** - Magnetic Tape Storage and Handling: Life Expectancy (John Van Bogart, CLIR, 1995): https://www.clir.org/pubs/reports/pub54/4life_expectancy/ - End of an era: VCRs headed for outdated tech heaven (AFP, 2016): https://phys.org/news/2016-07-era-vcr-factory-line.html - Chattanooga-based company hires Vimeo cofounder as it prepares for growth (Chattanooga Times Free Press, 2026): https://www.timesfreepress.com/news/2026/mar/27/chattanooga-based-company-hires-vimeo-cofounder/ - Best VHS to Digital Converters (Capture, 2026): https://www.capture.com/blog/best-vhs-to-digital-converters - Kodak Digitizing vs Legacybox: Same Company, Licensed Name (Legacy Digital, 2026): https://legacydigital.net/versus/kodak-digitizing-vs-legacybox/ --- # Who Was the AOL 'You've Got Mail' Voice? Elwood Edwards URL: https://404memoryfound.com/posts/aol-youve-got-mail-voice-elwood-edwards.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-10 Topics: Internet Culture **Summary:** The AOL "You've Got Mail" voice was Elwood Edwards, a television announcer who recorded four phrases on a cassette deck in his own living room in 1989 and was paid $200, once, with no residuals. He died on November 5, 2024, at 74. AOL never commissioned a replacement, so the alert playing in 2026 is still his. **Key facts:** - Recorded: 1989, on a home cassette deck - Fee: $200, paid once, no residuals - Lines recorded: Welcome / You've got mail / File's done / Goodbye - Died: November 5, 2024, New Bern, North Carolina, age 74 - Status today: Still AOL Mail's alert; no replacement voice announced ## The $200 cassette recorded in a living room in 1989 Elwood Edwards was a working broadcaster, not a celebrity. He started in radio in 1964, while he was still in high school, and he spent the rest of his career inside announce booths and control rooms rather than in front of an audience. In 1989 his wife, Karen, was a customer service representative at Quantum Computer Services, the company that became America Online that same year. She overheard AOL's Steve Case talking with programmers about adding a voice to the software, and she put her husband's name forward without asking him first. There was no studio session and no audition. "Karen volunteered me, and on a cassette deck in my living room, I recorded, Welcome! You've got mail. File's done, goodbye," Edwards told the podcast Twenty Thousand Hertz in 2019. The fee was $200, paid once. There were no residuals, no per-play rate and no clause covering what would happen if the four phrases outlasted the company that ordered them. The couple had met in a chat room on Q-Link, AOL's predecessor service, and married in December 1988, so the whole thing started as a favor between spouses. ## Does AOL still play that recording in 2026? Yes. AOL never announced a replacement voice, and the greeting has now outlived the dial-up business that made it famous. AOL shut off dial-up internet (https://404memoryfound.com/posts/dialup-speed-then-vs-now.html) on September 30, 2025, and told customers the change "will not affect any other benefits in your AOL plan." Mail stayed, and so did the alert. AOL did sell subscribers on alternatives for years, offering celebrity reads in place of the original. Most people declined. "Fewer than 20% of the AOL subscribers, throughout the years, had elected to change from my voice," Edwards said in 2019. That number is the part people miss. The recording was never protected by exclusivity or scarcity. It survived because subscribers actively chose to keep it when the software offered them something newer, year after year, which is a rarer kind of durability than a contract would have bought. So the search-box question has an unusually clean answer. AOL still exists in 2026 (https://404memoryfound.com/posts/does-aol-still-exist-today.html), the free mail service still runs, and the sound it makes is a cassette recorded in a North Carolina broadcaster's living room 37 years ago. ## How four phrases became a Warner Bros. movie title By the late 1990s the line had escaped the software entirely. Warner Bros. released You've Got Mail on December 18, 1998, directed by Nora Ephron and starring Tom Hanks and Meg Ryan, and built the whole romance around two people meeting through AOL screen names. The film cost $65 million and took $115.8 million in the United States and Canada, plus roughly $135 million elsewhere, for $250.8 million worldwide. It became Ephron's highest-grossing picture, and it turned a notification sound into shorthand for the entire online era. Edwards appeared in the film in a cameo. He did not get a screen credit as the voice of the title, and the $200 he had been paid nine years earlier remained the total. The same era produced the other AOL product everyone remembers, AOL Instant Messenger (https://404memoryfound.com/posts/what-happened-to-aim-aol-instant-messenger.html), and the same corporate confidence produced the merger with Time Warner (https://404memoryfound.com/posts/aol-time-warner-merger-worst-deal-in-history.html). Of the three, only the four-word greeting is still running. ## The WKYC years and the Uber rides in Ohio Edwards spent his working life in Cleveland television at WKYC, where he was a booth announcer and supervised news graphics. He retired from television in 2016, after more than five decades in broadcasting that had started with a high school radio shift in 1964. He then drove for Uber in Ohio. Passengers who recognized the voice in the front seat would ask him to say the line, and the clips they filmed traveled further than anything he had recorded on purpose. His recurring complaint was not the driving. It was the assumption that the recording had made him rich. He kept getting asked to perform it anyway, and he kept saying yes. He read ordinary sentences in the AOL voice on The Tonight Show Starring Jimmy Fallon on March 4, 2015. In October 2022 he turned up in a Shopify commercial announcing "You've got sales." He treated the whole thing as luck rather than achievement, and said so plainly: "Our world is full of people who were in the right place at the right time, and I'm glad to be one of those." ## What his death in November 2024 left behind Edwards died on November 5, 2024, at his home in New Bern, North Carolina, from complications of a stroke. He was 74, and he died one day before his 75th birthday. He had been born in Glen Burnie, Maryland, on November 6, 1949. AOL marked it with a statement that read less like a press release than an epitaph for the whole dial-up decade. "Edwards's voice made AOL feel a little friendlier, a little more welcoming at a time when the internet was a big, new world for most people," the company said in 2024. The tributes came from an odd mix of places: the Cleveland station where he had actually worked, national obituary desks, and thousands of people who had never known his name and could still do the impression. The obituaries all led with the same detail, which is the reason the story travels: one of the most reproduced sounds in American consumer software cost $200 and was captured on consumer tape hardware in a house. Nobody involved in 1989 thought it was worth a contract. Where to find one today: the cassette itself is not for sale, but the disks AOL mailed out by the millions are easy to collect. Sealed and opened America Online install disks, floppies and later CD-ROMs, show up regularly in the vintage computing listings on Etsy (https://www.etsy.com/search?q=vintage+aol+disk), usually alongside the cardboard sleeves they arrived in. ## Frequently Asked Questions ### How much was Elwood Edwards paid for the AOL "You've Got Mail" recording? Elwood Edwards was paid $200 in 1989, one time, for recording all four AOL phrases: "Welcome," "You've got mail," "File's done" and "Goodbye." The deal included no residuals, so the fee never grew no matter how many times the sound played over the following three decades. ### Was Elwood Edwards in the 1998 movie You've Got Mail? Elwood Edwards had a cameo in the Warner Bros. film You've Got Mail, released on December 18, 1998 and directed by Nora Ephron. The movie took its title from the AOL alert he had recorded in 1989, and it grossed $250.8 million worldwide. ### What did Elwood Edwards do for a living besides the AOL greeting? Elwood Edwards worked in broadcasting from 1964 onward, spending years at WKYC in Cleveland as a booth announcer and news graphics supervisor before retiring from television in 2016. He drove for Uber in Ohio afterward, and died on November 5, 2024, at 74. **Sources:** - Twenty Thousand Hertz: You've Got Mail, the voice behind AOL (2019): https://www.20k.org/episodes/youvegotmail - NPR: Elwood Edwards, the voice of AOL's 'You've got mail' alert, dies at 74 (2024): https://www.lpm.org/news/2024-11-09/elwood-edwards-the-voice-of-aols-youve-got-mail-alert-dies-at-74 - Wikipedia: Elwood Edwards: https://en.wikipedia.org/wiki/Elwood_Edwards - Wikipedia: You've Got Mail (1998 film): https://en.wikipedia.org/wiki/You%27ve_Got_Mail - ABC News: AOL to end its dial-up internet service (2025): https://abcnews.com/GMA/Living/aol-set-pull-plug-iconic-dial-internet-service/story?id=124539332 --- # Is Sega Still Making Games? Life After the Dreamcast URL: https://404memoryfound.com/posts/is-sega-still-making-games.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-09 Topics: Gaming, Business Blunders **Summary:** Sega is still making games. It sits inside Sega Sammy Holdings, formed in October 2004, whose Entertainment Contents division sold 326.6 billion yen in the fiscal year ended March 31, 2026. What Sega no longer makes is consoles, and nothing in its filings suggests that is changing. **Key facts:** - Status today: Still publishing games, no home console since the Dreamcast - Owner today: Sega Sammy Holdings, formed October 2004 - Games and content sales: 326.6 billion yen, fiscal year ended March 2026 - Full games sold: 22.95 million copies, fiscal year ended March 2026 - Biggest recent hit: Sonic the Hedgehog 3, 492.2 million dollars worldwide ## Is Sega still making games in 2026? Yes. Sega is owned by Sega Sammy Holdings, and the division that holds the game business, Entertainment Contents, reported sales of 326.6 billion yen for the fiscal year ended March 31, 2026, up from 321.5 billion yen the year before. The consumer games area inside it accounted for 219.9 billion yen. Sega Sammy reports in yen, so most figures here stay in yen. What matters for the question is the direction. Sega shipped 22.95 million full game copies in that fiscal year, down from 31.45 million a year earlier, a fall of 27 percent. Segment operating income dropped to 32.4 billion yen from 40.8 billion yen. So the answer is yes, with a wobble. Sega walked away from the console business after the Dreamcast (https://404memoryfound.com/posts/why-sega-stopped-making-consoles-dreamcast.html) and turned itself into a publisher, and the publisher is still shipping. Sega Sammy forecasts 357.0 billion yen in Entertainment Contents sales for the year ending March 2027, built on four new full game releases from its main franchises. ## Who owns Sega now, and why does a pachinko company matter? Sega Sammy Holdings was established in October 2004 through the management integration of Sega and Sammy Corporation. Sammy built pachislot and pachinko machines, the Japanese gambling cabinets, and that side of the house has been the group's profit engine for most of the years since. The split in the fiscal year ended March 2026 is worth sitting with. The Pachislot and Pachinko Machines segment sold 132.0 billion yen and delivered 32.1 billion yen of operating income. Entertainment Contents sold 326.6 billion yen and delivered 32.4 billion yen. Games bring in roughly two and a half times the revenue for about the same profit. Group wide, Sega Sammy booked 487.5 billion yen in sales and 47.1 billion yen in operating income, then closed the year with a net loss of 5.7 billion yen after 58.8 billion yen of extraordinary losses. The company that lost the Saturn generation (https://404memoryfound.com/posts/what-happened-to-sega-saturn-console.html) and then watched the PlayStation 2 (https://404memoryfound.com/posts/playstation-2-best-selling-console-ever.html) bury its last console now runs a games publisher inside a gambling machine business. ## What Sega bought after it stopped building hardware Two acquisitions define modern Sega. In 2014 Sega Sammy took over the game division of the collapsed Index Corporation and rebuilt it as Atlus, which brought the Persona and Shin Megami Tensei series in house. That purchase aged well. Atlus role playing games are now among the most reliable earners Sega has. The second one has not aged well at all. Sega acquired Rovio Entertainment, the Finnish studio behind Angry Birds, as a subsidiary in 2023. Rovio's sales were 309 million euros in the fiscal year ended March 2023, before the deal closed. By the fiscal year ended March 2026 they were 181 million euros, and Sega Sammy forecasts 158 million euros for the year ending March 2027. That decline is a large part of why the group posted 58.8 billion yen in extraordinary losses in the fiscal year ended March 2026, driven by impairment charges on Rovio and on Stakelogic, an online casino supplier Sega Sammy bought the same year. The mobile bet has cost real money. ## Sonic earns more at the box office than most Sega games do The clearest sign that Sega changed shape is where Sonic makes his money. Sonic the Hedgehog 3 opened in the United States on December 20, 2024 and finished its run on March 27, 2025 with 236.1 million dollars domestic and 492.2 million dollars worldwide. No Sega game release in that window came close to that kind of gross. The films feed a licensing business that has grown every year. Sega Sammy's licensing revenue reached 20.8 billion yen in the fiscal year ended March 2026, up from 17.5 billion yen, spread across 748 licensees. On Sega Sammy's own chart that line rises without a down year, from 5.7 billion yen at the left edge to 20.8 billion yen at the right. The pipeline is already on the calendar. Sega Sammy's own results materials list The Angry Birds Movie 3 for December 23, 2026 and Sonic the Hedgehog 4 for March 19, 2027, with a Golden Axe project listed as licensing out. Sega is now a character company that also publishes games. ## The 2024 layoffs and the studios Sega let go The restructuring has been real. On March 28, 2024 Sega announced roughly 240 job cuts across Sega Europe, Total War maker Creative Assembly and mobile studio Hardlight, and confirmed that Relic Entertainment, the Company of Heroes developer, would leave the group as an independent studio backed by an outside investor. The cuts did not stop there. In May 2026 Sega Sammy confirmed it had canceled Super Game, the multi year attempt at a big online service title, and lowered the priority of free to play development after new titles underperformed. More than 100 developers were moved off free to play work and onto full game teams built around the main franchises. Read together, the two moves say the same thing. Sega tried to be a live service company, it did not work, and it has retreated to what it is good at, which is making finished single purchase games with characters people already know. ## Will Sega make a new console? Nothing in the company's own filings points that way. Sega Sammy reports three segments, Entertainment Contents, Pachislot and Pachinko Machines, and a Gaming segment covering overseas online casino operations that sold 25.3 billion yen and lost 7.2 billion yen in the fiscal year ended March 2026. There is no consumer hardware line anywhere in the structure. The closest Sega has come since 2001 is plug and play nostalgia. The Sega Genesis Mini arrived in the United States on September 19, 2019 at 79.99 dollars with 42 built in games, and had sold more than 300,000 units by the end of March 2020. A Genesis Mini 2 followed worldwide on October 27, 2022 with 60 Genesis and Sega CD (https://404memoryfound.com/posts/what-happened-to-sega-cd-add-on.html) games. Before that Sega had licensed the name to AtGames, whose 2017 Sega Genesis Flashback was panned, and dropped that partnership to build the Mini itself. Handheld rumors surface every year or two, but Sega has announced nothing. If you want one of the Minis today, Sega no longer sells either model, so the second-hand market is the route; specialist shops such as Stone Age Gamer (https://stoneagegamer.com/) also carry flash carts and accessories that make original hardware easier to live with. The original listed at 79.99 dollars in 2019, and the Mini 2 was made in smaller numbers, so sealed second models are the harder find. Check that a listing includes the power supply and a controller. ## Frequently Asked Questions ### Is Sega still around in 2026? Yes. Sega operates as part of Sega Sammy Holdings, which reported 487.5 billion yen in group sales for the fiscal year ended March 31, 2026. Its Entertainment Contents division, which contains the game business, accounted for 326.6 billion yen of that. ### Does Sega still own Sonic? Yes. Sega owns Sonic outright and licenses him out for films and merchandise. Sonic the Hedgehog 3 grossed 492.2 million dollars worldwide after its December 20, 2024 release, and Sega Sammy's materials list Sonic the Hedgehog 4 for March 19, 2027. ### What was the last Sega console? The Dreamcast was the last home console Sega manufactured, and the company has not shipped a successor. The nearest thing since is the Sega Genesis Mini line, a plug and play reissue that launched in the United States in September 2019 at 79.99 dollars. **Sources:** - SEGA SAMMY HOLDINGS, Full-year results presentation for the fiscal year ended March 2026: https://www.segasammy.co.jp/cms/wp-content/uploads/pdf/en/ir/20260512_q4_presentation_E.pdf - SEGA SAMMY HOLDINGS, Corporate history: https://www.segasammy.co.jp/en/corp/history/ - Box Office Mojo, Sonic the Hedgehog 3: https://www.boxofficemojo.com/release/rl886211329/ - Game Developer, Sega sells off Relic Entertainment, will axe 240 jobs: https://www.gamedeveloper.com/business/sega-sells-off-relic-entertainment-will-axe-240-jobs - Wikipedia, Sega Genesis Mini: https://en.wikipedia.org/wiki/Sega_Genesis_Mini --- # Is RuneScape Still Popular in 2026? OSRS vs RS3 URL: https://404memoryfound.com/posts/is-runescape-still-popular-2026.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-09 Topics: Gaming, Internet Culture **Summary:** RuneScape is still popular in 2026, and the popular half is the 2007 version. Old School RuneScape peaked above 240,000 concurrent players in August 2025, against roughly 30,000 for RuneScape 3. Jagex, the Cambridge studio behind both games, has been owned by CVC Capital Partners and Haveli Investments since February 2024. **Key facts:** - Owner today: CVC Capital Partners and Haveli Investments, since February 2024 - Peak players: 240,851 at once in Old School RuneScape, August 2025 - Subscribers: 2.4 million paid plus 1 million free-to-play (2024) - Cost today: $14.99 a month or $131.88 a year in the US (2026) - Status: Both games live, with Old School roughly eight times larger than RuneScape 3 ## Is RuneScape still popular in 2026? Yes, and the healthy half of the franchise is the version Jagex restored from a 2007 server backup. On a Sunday in early August 2025, Old School RuneScape passed 240,000 people logged in at the same moment, with player screenshots putting the peak at 240,851. The game had crossed 200,000 concurrent players only weeks earlier. Mainline RuneScape, the version players call RuneScape 3, runs at roughly 30,000 concurrent players by the same reporting. That is a gap of about eight to one between a 2007 snapshot and the modern game it was copied from. The commercial side agrees. When Carlyle announced on February 9, 2024 that it had agreed to sell Jagex, it described a community of 2.4 million active subscribers and one million free-to-play users, sitting on more than 300 million lifetime RuneScape accounts and $1.5 billion in lifetime franchise revenue. A browser game from the era of Neopets and other click-and-wait virtual worlds (https://404memoryfound.com/posts/what-happened-to-neopets-virtual-world.html) is not supposed to be setting attendance records two decades later. ## Who owns Jagex now? Jagex is owned by CVC Capital Partners Fund VIII and Haveli Investments. Carlyle announced the agreed sale on February 9, 2024, and the two buyers have run the studio since. CVC is a European private equity firm with approximately 188 billion euros of assets under management, better known outside gaming for its stakes in soccer and rugby competitions. Haveli is an Austin firm whose chief investment officer, Brian Sheth, was quoted on the deal. Neither side disclosed a price, though PocketGamer.biz reported that Reuters valued the sale at as much as 1 billion pounds, about $1.25 billion. The seller was Carlyle, which had held Jagex since 2021 through Carlyle Partners VII and Carlyle Europe Technology Partners IV, a 1.35 billion euro fund. At the time of the sale Carlyle described Jagex as a Cambridge company employing more than 700 people worldwide, with Phil Mansell as chief executive. So the plain answer to who owns RuneScape: two private equity firms, one European and one Texan, neither of which was anywhere near the game when it launched. ## How Jagex kept getting sold The current owners are the sixth set in roughly a decade, a churn PocketGamer.biz laid out when the 2024 deal landed. Insight Venture Partners, a New York firm, took a majority share in 2012. In 2016 the studio passed to Hongtou, a Chinese investment vehicle later absorbed by Zhongji Holdings, which renamed itself Fukong Interactive Entertainment. In 2020 it moved again, to Macarthur Fortune Holdings, a US asset manager. Carlyle bought it in 2021, a deal PocketGamer.biz put at $530 million. Then CVC and Haveli in 2024. Reported prices along that chain are inconsistent, partly because several of the transactions were never officially disclosed and partly because the figures were reported by different outlets working from different filings. What matters for players is the pattern rather than any single number. Every owner arrives with a return to earn and an exit to plan, and on a subscription game the lever closest to hand is the subscription. ## Why Jagex dug a 2007 backup out of storage In early 2013, Jagex went digging through old server backups because forum threads kept circling back to the 2006 and 2007 builds of the game. On February 15, 2013 the studio posted that it had recovered a complete, verified snapshot of RuneScape from August 2007. It also admitted that it had changed its backup systems that October, so nothing after August 2007 survived in a usable state. Rather than simply switch the old servers on, Jagex put it to a vote, in the same spirit as an earlier poll on restoring the Wilderness and free trade that had drawn 1.4 million votes. The thresholds were blunt. At 50,000 votes the old servers would return with no content updates and an extra fee of around $15. At 250,000 they would get a small development team and roughly $5 extra. At 500,000 they would get a dedicated team and no extra fee at all. The old servers came back as Old School RuneScape, with no surcharge and a dedicated team. The rule that players vote on new content dates from that same promise. ## How Old School RuneScape outgrew RuneScape 3 The 2007 build was meant to be a nostalgia annex. It became the main game instead. Part of that is combat. The modern version had moved to an ability-driven system, the Evolution of Combat, and a large share of the audience never accepted it. Part of it is governance, because content in Old School RuneScape has to clear a player poll, so the game changes slowly and in public. Part of it is simply that the older client is lighter, easier to explain to a newcomer, and runs on a phone. The result is the trick Counter-Strike pulled on Half-Life (https://404memoryfound.com/posts/what-happened-to-counter-strike-half-life-mod.html), where the offshoot ends up bigger than the thing it came from. By August 2025 Old School RuneScape was running above 240,000 concurrent players against roughly 30,000 for RuneScape 3. Jagex maintains both, plus a co-op survival game set in the same world called RuneScape: Dragonwilds. An MMO that outlasted Second Life and the first metaverse rush (https://404memoryfound.com/posts/what-happened-to-second-life-virtual-world.html) managed it by shipping a copy of its own past. ## What RuneScape membership costs in 2026 Membership is $14.99 a month or $131.88 a year in the United States, following an increase Jagex announced on March 10, 2026. In the UK it is 10.99 pounds a month, up by a pound. That is the third increase in four years. In May 2022 membership was $12.49 a month or $79.99 a year. In September 2024 it moved to $13.99 and $99.48. The 2026 change took the monthly price to $14.99, level with World of Warcraft in the US, and pushed the annual plan up by more than $32 in a single step. The six-month plan was dropped, which Jagex presented as simplifying the options. Totaled up, the annual price has risen more than 60 percent since May 2022 while the monthly price has risen about 20 percent. The free tier still exists, and Carlyle counted a million people on it in 2024, so trying RuneScape costs nothing. Players have noticed the direction of travel. PCGamesN framed its report on the 2026 increase around the question of when the hikes stop. ## Frequently Asked Questions ### Is Old School RuneScape bigger than RuneScape 3? Yes, by a wide margin. In August 2025 Old School RuneScape peaked above 240,000 concurrent players while mainline RuneScape sat near 30,000, a gap of roughly eight to one. Both are made by Jagex and both are covered by the same membership. ### Can you still play RuneScape for free? Yes. Both games keep a free tier with a reduced set of skills, quests and areas. Carlyle counted one million free-to-play users alongside 2.4 million paying subscribers when it agreed to sell Jagex in February 2024. ### Is Old School RuneScape the same game as the 2007 original? It started as one. Jagex recovered a complete backup of RuneScape as it stood in August 2007 and relaunched it as a separate game. Years of new content have gone in since, but each addition has to pass a player poll first, which is why it still plays like the older game. **Sources:** - Carlyle Agrees to Sell Jagex to CVC Capital Partners and Haveli Investments (February 9, 2024): https://www.carlyle.com/media-room/news-release-archive/carlyle-agrees-to-sell-jagex-to-cvc-capital-partners-and-haveli - PocketGamer.biz: Jagex changes hands for the sixth time in ten years: https://www.pocketgamer.biz/jagex-sold-again/ - Jagex: 2007 - Old School RuneScape... You Vote! (February 15, 2013): https://secure.runescape.com/m=news/a=427/2007---old-school-runescape-you-vote - GamesRadar+: With a new record 240,000 players online, Old School RuneScape is roughly 8x bigger than its sister MMO (August 4, 2025): https://www.gamesradar.com/games/mmo/with-a-new-record-240-000-players-online-old-school-runescape-is-roughly-8x-bigger-than-its-sister-mmo-and-would-be-the-fourth-biggest-game-on-all-of-steam-if-not-for-a-technicality/ - PCGamesN: Old School RuneScape membership price increase, March 2026: https://www.pcgamesn.com/old-school-runescape/membership-price-increase-march-2026 --- # Is MySpace Still Around? What Is Left of the Site URL: https://404memoryfound.com/posts/is-myspace-still-around-today.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-09 Topics: Internet Culture **Summary:** MySpace is still around in 2026, but not as a social network. myspace.com is online as a music news site owned by Viant Technology Inc., the Nasdaq-listed ad tech company run by the Vanderhook brothers, and reporting since 2024 describes it as read-only. About 50 million songs uploaded before 2016 were destroyed in a 2019 server migration and never came back. **Key facts:** - Status today: Online at myspace.com, reported read-only since 2024 - Owner today: Viant Technology Inc. (Nasdaq: DSP), Irvine, California - Bought for: $580 million by News Corporation in 2005 - Sold for: $35 million to Specific Media in June 2011 - Music lost: About 50 million songs in the 2019 server migration ## Is MySpace still around in 2026? Yes. myspace.com loads, and on September 8, 2026 the front page was still publishing music news, including a piece on Stevie Wonder touring an album for its fiftieth anniversary. Those articles carry a SPIN byline rather than anything written by a MySpace newsroom. What is gone is the social network. Reporting through August 2026 describes the site as read-only, which means old pages render but nothing new gets posted to them. Tom Anderson's profile, the automatic first friend for a whole generation of American teenagers, still sits at myspace.com/tom with its counters for photos, videos, mixes and connections intact. The footer on the homepage still reads "© 2014 Myspace LLC." Nothing sums up the level of attention the front end has had since then more efficiently than that. ## Who owns MySpace now? Viant Technology Inc., an advertising technology company listed on Nasdaq under the ticker DSP, with offices on Michelson Drive in Irvine, California. Viant's annual report for fiscal 2025, filed with the SEC on March 11, 2026, names Myspace.com as one of the sites it owns outright through its subsidiary Myspace LLC. Viant was founded in 1999 by Tim, Chris and Russ Vanderhook, and the same brothers still run it. The Vanderhook parties hold a majority of the voting power in the company, which makes Viant a "controlled company" under Nasdaq's listing standards and lets it skip several governance requirements that apply to most public companies. MySpace is a rounding error inside that business. Viant sells a demand side platform for programmatic advertising, and its 2025 filing splits its issued patents into 35 covering the platform and its household identification technology, 15 covering Myspace.com, and 15 covering IRIS.TV. The site's main function on the balance sheet is as a legal risk, listed under a heading about liabilities arising from owning and operating it. ## How MySpace went from $580 million to $35 million Rupert Murdoch's News Corporation bought MySpace in 2005 for $580 million. On June 29, 2011 it sold the site to the ad targeting firm Specific Media for $35 million in stock and cash, below the $100 million News Corp had been seeking, and held on to a stake of less than 5 percent. Specific Media's chief executive was Tim Vanderhook. Justin Timberlake joined as a co-owner and public face of the deal. The chain gets stranger after that. In 2016 Time Inc. acquired a 60 percent interest in Viant, which carried MySpace along with it. Meredith Corporation inherited that interest when it bought Time Inc. in 2018. In 2019 Viant bought the 60 percent stake back and became independent again, then completed its initial public offering on February 12, 2021. So a social network passed from a newspaper empire to an ad tech startup to a magazine publisher to another magazine publisher and back to the ad tech startup, all within fourteen years. The story of how it lost its audience before any of that happened is in the rise and fall of MySpace (https://404memoryfound.com/posts/rise-and-fall-of-myspace.html). This is what came after. ## What the 2019 server migration destroyed In March 2019 users noticed the music player had stopped working. MySpace then confirmed that roughly 50 million songs uploaded between 2003 and 2015 were gone, along with photos and videos, affecting around 14 million artists. The company's statement ran to one sentence: "As a result of a server migration project, any photos, videos, and audio files you uploaded more than three years ago may no longer be available on or from Myspace." No recovery plan came with it. Viant's own language is drier. The fiscal 2025 annual report tells investors that "as a result of a server migration project in 2019, older photo, video or audio files of some users were lost," filed under operational and performance issues. Twelve years of uploads went with it, and most of it existed nowhere else. Garage bands, bedroom demos, first mixes by producers who later got famous, all posted by people who treated a free website as permanent storage. Anyone who put a childhood into a MySpace profile or a GeoCities (https://404memoryfound.com/posts/history-of-geocities-websites.html) page learned the same lesson twice: a free host owes you nothing. ## Can you get an old MySpace profile back? Partly. The sign in and password reset pages still respond in September 2026, and old profile URLs still resolve, so an account tied to an email address you still control is worth an attempt. The media is a different matter. Anything uploaded before 2016 sat in the batch lost during the migration, and MySpace told users at the time there was no way to recover it. One partial rescue exists. On April 1, 2019 the Internet Archive published a collection called The Myspace Dragon Hoard (2008-2010): 490,000 MP3 files pulled off MySpace's content delivery network by an anonymous academic study between 2008 and 2010. It runs to more than 1.3 terabytes packed into 144 ZIP files, with a search and playback tool layered on top. Set against 50 million lost songs, that is roughly 1 percent. There is a security footnote too. Viant has disclosed that in 2016 it found a third-party attack in which about 360 million MySpace account email addresses, usernames and hashed passwords were taken from the old platform, all of it stolen before June 11, 2013. If that password is still in use anywhere else, change it before you go looking for your profile. ## Will MySpace relaunch? The owners say yes and give no date. In a 2026 documentary about the site, Tim Vanderhook said, "We still own Myspace. We are stewards of the Myspace brand at this point, and we are going to relaunch Myspace," adding that they were waiting for the right moment to do it. They have already tried once. The brothers rebuilt the site from scratch during their ownership and have said the attempt lost a little over $150 million, because advertisers kept their budgets with Facebook and YouTube instead. The obstacles have grown since. Music licensing has to be renegotiated from zero. Myspace LLC is bound by a 2012 Federal Trade Commission consent order that requires independent privacy assessments every two years and does not expire until August 2032. And TikTok and Instagram now hold the attention that MySpace would need to win back. Nostalgia alone has not revived a dead social network yet, as Friendster (https://404memoryfound.com/posts/what-happened-to-friendster-social-network.html) demonstrated. ## Frequently Asked Questions ### Can you still log in to MySpace? The sign in and password reset pages still load as of September 2026, so an account attached to an email address you still have may open. Expect an empty shell, because reporting since 2024 describes the site as read-only and every photo, video and song uploaded before 2016 was destroyed in the 2019 migration. ### How much did News Corp lose on MySpace? News Corporation paid $580 million for MySpace in 2005 and sold it for $35 million in stock and cash in June 2011, a gap of $545 million before six years of operating costs are counted. It had been asking $100 million and kept a stake of under 5 percent in the buyer. ### Where can I find MySpace songs that were deleted? The Internet Archive's Myspace Dragon Hoard holds 490,000 MP3 files collected from the site between 2008 and 2010, with a search tool built over the top. The files are named using MySpace's content delivery network keys, so tracking down one specific band means working through the collection's metadata file rather than searching by name. **Sources:** - Viant Technology Inc., Annual Report on Form 10-K for fiscal year 2025 (SEC, filed March 11, 2026): https://www.sec.gov/Archives/edgar/data/1828791/000182879126000019/dsp-20251231.htm - NPR: News Corp. Takes Huge Loss, Selling Myspace For $35 Million (June 29, 2011): https://www.npr.org/sections/thetwo-way/2011/06/29/137509647/news-corp-takes-huge-loss-selling-myspace-for-35-million - The FADER: MySpace confirms 12 years of music lost in server migration (March 18, 2019): https://www.thefader.com/2019/03/18/myspace-server-migration-loss-music - Internet Archive: The Myspace Dragon Hoard (2008-2010): https://archive.org/details/myspace_dragon_hoard_2010 - Tubefilter: MySpace's owners want to bring it back. But does it belong in 2026? (August 10, 2026): https://www.tubefilter.com/2026/08/10/myspace-2026-relaunch-tim-chris-vanderhook/ --- # Is Kmart Still Open? The 3 Stores Left in 2026 URL: https://404memoryfound.com/posts/is-kmart-still-open-stores-left.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-09 Topics: Business Blunders **Summary:** Kmart is still open in 2026, but only three stores remain: a small one in Kendale Lakes, Florida, a big-box store in Tamuning, Guam, and one in Charlotte Amalie in the U.S. Virgin Islands. The chain is owned by Transform SR Brands LLC, which trades as Transformco, and the last full-size mainland store closed in Bridgehampton, New York, on October 20, 2024. **Key facts:** - Peak size: 2,486 stores worldwide in 1994 - First bankruptcy: Chapter 11 on January 22, 2002, with 2,114 stores open - Owner today: Transform SR Brands LLC (Transformco), tied to Edward Lampert - Stores left: 3, in Kendale Lakes, Florida; Tamuning, Guam; and Charlotte Amalie, U.S. Virgin Islands - Status: Open, plus Kmart.com shipping to the 50 states and APO/FPO addresses ## Is Kmart still open in 2026? Yes, barely. As of 2026 there are three Kmart stores left, and only one of them is in the 50 states. The mainland survivor is in Kendale Lakes, an area of southwest Miami. It runs out of the building's former garden center and carries a limited selection of clothing, appliances and household goods. It is a fragment of a store rather than a discount department store. The second is a big-box store in Tamuning, Guam, which opened in May 1995 and covers about 170,000 square feet. That makes the largest Kmart on earth a store most Americans will never see. The third is a big-box store in Charlotte Amalie, in the U.S. Virgin Islands, at Tutu Park Mall on St. Thomas. Kmart.com is still trading and still ships to the 50 states plus APO and FPO addresses, so for almost every American the answer to "is Kmart still open" is now a website rather than a parking lot. ## Who owns Kmart now? Transform SR Brands LLC, which does business as Transformco, owns Kmart. It is privately held, so it publishes no store counts and no sales figures. Kmart got there through Edward Lampert. His hedge fund, ESL Investments, bought up a large share of Kmart's debt while the retailer was in Chapter 11, and that debt converted into control when the company came out. In February 2019 a U.S. bankruptcy judge approved the sale of the most valuable pieces of Sears Holdings to Lampert, and those assets, including 202 remaining Kmart stores, went into Transform SR Brands. Transformco has spent the years since subtracting rather than adding. Closures cut the chain to under a dozen stores by early 2022. In July 2026 its logistics arm filed a Pennsylvania layoff notice to shut a 2.2 million square foot warehouse at 1 Kresge Road in Falls Township, cutting 147 jobs, with a final day of September 16, 2026. The address is a fossil of the founding company, S. S. Kresge, incorporated in 1899. ## How Kmart went from 2,486 stores to three Kmart hit its peak in 1994 with 2,486 stores worldwide. Then Walmart beat it on price, Target beat it on presentation, and the stores themselves got old. On January 22, 2002, Kmart Corporation and 37 of its U.S. subsidiaries filed voluntary Chapter 11 petitions in the U.S. Bankruptcy Court for the Northern District of Illinois, case number 02-B02474. All 2,114 Kmart stores stayed open that day, held up by a $2 billion debtor in possession financing package. The company blamed a collapse in fourth quarter sales, the evaporation of the surety bond market and a loss of supplier confidence. Chief executive Charles C. Conaway said Kmart would use the filing to "make a fresh start and reposition Kmart for the future." In 2005 Kmart and Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html) merged into Sears Holdings Corporation, which filed for bankruptcy of its own in 2018. Two chains that had each been the largest retailer in America ended up losing together. ## The last full-size Kmart closed on October 20, 2024 The final full-size Kmart in the mainland United States stood in Bridgehampton, New York, on the east end of Long Island. It closed on October 20, 2024. The setting was odd enough to make national news. A discount chain built for the Midwest ended its mainland run in the Hamptons, in a store that had outlived more than 2,000 American Kmarts. Shoppers cleared the shelves of toys and shoes in the final weeks, and the red sign came down. What was left after that was the small Miami store, plus the stores in Guam and the U.S. Virgin Islands. That pattern is familiar from the rest of the era's big-box collapse. RadioShack (https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html) and Circuit City (https://404memoryfound.com/posts/what-happened-to-circuit-city-electronics.html) both ended as brand names attached to a website and a handful of licensed outposts, and Kmart followed the same track: the logo survives the leases. ## The U.S. Virgin Islands lost two Kmarts in one summer In 2024 the U.S. Virgin Islands still had three Kmart stores, more than the entire mainland. That did not last. Transform KM LLC issued a notice dated May 9, 2025, saying it would permanently cease operations at the Sunny Isle store on St. Croix, affecting about 96 employees, with severance set at one week per full year of service. The store closed in early July 2025, roughly a month ahead of what most of the staff expected. It was already the second St. Croix closure, after the Sunshine Mall store in Frederiksted shut in 2023 following about 30 years on the island. The Lockhart Gardens store on St. Thomas went next, closing on August 16, 2025 after a liquidation sale that included the fixtures. That left Tutu Park Mall as the only Kmart in the territory, and one of only three anywhere. ## What happened to the Blue Light Special The Blue Light Special ran from 1965 to 1991. A rotating blue light was rolled out to a rack or a bin, a loudspeaker said "Attention, Kmart shoppers," and whatever sat under the light was cheap for a few minutes. It was a flash sale decades before anyone used that phrase. Kmart retired it in 1991 and then kept exhuming it, with revivals running as late as 2015. None of them worked, because the thing that made the original land was scarcity of information. In 1978 a shopper had no way of knowing whether $6.88 for a set of towels was a real deal. By 2015 the same shopper could check three prices on a phone while standing under the light. That, more than any single bad quarter, is what replaced Kmart. Walmart took the price position, Target took the taste position, and Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html) took the browsing. The blue light was left announcing discounts to people who could already see every price in the country. ## Frequently Asked Questions ### How many Kmart stores are left? Three, as of 2026. One small store in Kendale Lakes, Florida, one big-box store in Tamuning, Guam, and one big-box store in Charlotte Amalie in the U.S. Virgin Islands. ### Where is the last Kmart in the United States? The last Kmart in the 50 states is in Kendale Lakes, Florida, in the Miami area. It operates out of the former garden center of a much larger Kmart and sells a narrow range of clothing, appliances and household goods. The Guam and U.S. Virgin Islands stores are in U.S. territories, not states. ### Can you still shop at Kmart online? Yes. Kmart.com is still operating and ships to all 50 states plus APO and FPO addresses. For most of the country the website is the only Kmart there is. **Sources:** - Kmart Corporation, Form 8-K and Chapter 11 press release, January 22, 2002 (SEC EDGAR): https://www.sec.gov/Archives/edgar/data/56824/000095017202000116/s588140.txt - Fortune: Kmart, which once had over 2,000 locations, just closed its last major U.S. store (September 24, 2024): https://fortune.com/2024/09/24/kmart-which-once-had-over-2000-locations-just-closed-its-last-major-u-s-store/ - VI Consortium: With Sunny Isle Closed and Lockhart Gardens Set to Follow, Tutu Park Stands as Kmart's Final Holdout in USVI (July 2025): https://viconsortium.com/vi-economy/virgin-islands-with-sunny-isle-closed-and-lockhart-gardens-set-to-follow--tutu-park-stands-as-kmart-s-final-holdout-in-usvi - LevittownNow: Owner Of Kmart, Sears To Close Falls Twp. Warehouse, Lay Off 147 (July 22, 2026): https://levittownnow.com/2026/07/22/owner-of-kmart-sears-to-close-falls-twp-warehouse-lay-off-147/ - Wikipedia: Kmart (store list and corporate history): https://en.wikipedia.org/wiki/Kmart --- # Is CompUSA Still Around? What Happened to the Chain URL: https://404memoryfound.com/posts/is-compusa-still-around.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-09 Topics: Hardware, Business Blunders **Summary:** CompUSA is not still around, as a store or as a website. The Dallas chain sold its last 103 stores to liquidator Gordon Brothers in December 2007, and Systemax folded the brand into TigerDirect in the fourth quarter of 2012. In 2026 the CompUSA trademarks belong to Source Brand Group, LLC and compusa.com loads an error page. **Key facts:** - Launched: 1984 as Soft Warehouse, Inc. in Dallas; renamed CompUSA in March 1991 - Peak scale: 211 superstores in 42 states, $6.32 billion in net sales, fiscal 1999 - Stores left: None. The last 103 stores were sold to a liquidator in December 2007 - Owner today: Source Brand Group, LLC holds the live COMPUSA.COM trademark - What replaced it: TigerDirect, which absorbed the brand in the fourth quarter of 2012 ## Is CompUSA still around in 2026? No. There is no CompUSA store anywhere in the United States, and there has not been one since 2008. The chain's last 103 locations were sold in December 2007 to Specialty Equity LLC, an affiliate of the Boston restructuring firm Gordon Brothers Group, which ran store-closing sales through the holidays and emptied the buildings after. The name outlived the stores by five years. Systemax, the parent company of TigerDirect, bought the CompUSA brand and website in January 2008 and put the logo back on retail stores and an online shop. That ended in the fourth quarter of 2012, when Systemax folded its United States consumer business into the TigerDirect name and wrote the CompUSA brand off its books. Type compusa.com into a browser in 2026 and you do not reach a store. The domain resolves to a hosting error page. The CompUSA trademarks, on the other hand, are still live at the United States Patent and Trademark Office, registered to a company that does not sell computers. ## How big did CompUSA actually get? Bigger than most people remember. The business started in 1984 as Soft Warehouse, Inc., was incorporated in Delaware in 1988, and operated under the Soft Warehouse name until March 1991. After that it was CompUSA, run out of offices on North Dallas Parkway in Dallas, with common stock on the New York Stock Exchange under the ticker CPU. By the close of fiscal 1999 on June 26 of that year, CompUSA ran 211 computer superstores across 42 states. Net sales for the year were $6.32 billion, up 19.6 percent from $5.29 billion the year before. That is the high-water mark in the company's own SEC filings, and it is why the chain shows up in so many first-PC stories: for a stretch in the late 1990s there was one within driving distance of most American metro areas. Some of that reach was bought rather than built. On August 31, 1998, CompUSA acquired Computer City from Tandy (https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html) Corporation for roughly $175 million, taking on 96 stores in the United States. It closed 55 of them, reopened 37 as CompUSA superstores by fiscal year end, and sold seven Canadian locations to Future Shop for about $7 million. This was the same period in which Gateway (https://404memoryfound.com/posts/what-happened-to-gateway-2000-cow-box-computer.html) was opening its own storefronts and Best Buy was scaling nationally. Selling personal computers in a warehouse still looked like a growth industry. ## What did Carlos Slim do with CompUSA? He bought it, then spent years paying for it. The Mexican financier Carlos Slim Helu took his first stake in CompUSA in 1999 and took the whole company private the next year in an $800 million buyout, which is why the detailed public filings stop after fiscal 1999. Ownership was never the problem. Margin was. PC prices fell year after year, and the profit in consumer electronics moved toward flat-panel televisions, where Circuit City (https://404memoryfound.com/posts/what-happened-to-circuit-city-electronics.html) and Best Buy were already cutting each other's prices. CompUSA cycled through several chief executives and several turnaround plans, including a late attempt to aim the stores at gadget buyers and small-business customers rather than at everyone. In the spring of 2007 the company closed more than half its stores and took a $440 million cash infusion to restructure around the ones that were left. That bought roughly nine months. By December the owner had stopped funding it. ## Why did CompUSA shut down in December 2007? Because the smaller CompUSA still lost money. With 103 stores left after the spring cuts, the company was sold in December 2007 to Specialty Equity LLC, the Gordon Brothers affiliate, and those stores went straight into holiday liquidation sales rather than a Christmas selling season. Gordon Brothers is a restructuring and liquidation firm, not a retailer, and it treated CompUSA as a parts bin. It shopped individual stores in strong retail markets, the CompUSA TechPro technical-services business, and the CompUSA.com online operation as separate lots, and left it to whoever bought each piece to decide whether the CompUSA name came with it. That is what makes the ending different from the chains it competed against. There was no drawn-out Chapter 11, no reorganization plan, no fight over the whole company. A liquidator bought it in one December transaction, and the warehouse-sized stores, the same format Fry's Electronics (https://404memoryfound.com/posts/what-happened-to-frys-electronics.html) built its business on, went dark over the following weeks. ## Who owns the CompUSA name in 2026? Systemax moved first. It bought CompUSA's e-commerce business for $18.9 million in cash on January 10, 2008, then took sixteen retail store leases and their fixtures for about $11.7 million during February and March. Most of what it paid for was the name itself: $17.0 million of the purchase was booked as trademarks and trade names and $8.0 million as domain names. The revival was real for a while. The deal gave Systemax 29 retail storefronts across North America and Puerto Rico, run next to its TigerDirect business, and CompUSA-branded operations produced $63.0 million of sales in the third quarter of 2008 and $139.4 million over the first nine months of that year. Then Systemax decided one consumer brand was enough. In the fourth quarter of 2012 it concluded that its North American consumer business would work better consolidated under TigerDirect, and wrote off the intangible assets and goodwill of CompUSA and Circuit City together, about $35.3 million. Systemax finished 2012 with 41 retail stores, 17 of them in Florida and 6 in Texas. The paperwork survived all of it. The COMPUSA.COM trademark, registration number 2582709, filed February 12, 1999 and registered June 18, 2002, is listed at the USPTO as live and was renewed in January 2023. The owner of record is Source Brand Group, LLC of New York. Somebody is still paying to keep the name, and nobody is selling computers with it. If you want something with the logo on it, the only supply is retail debris. Employee polo shirts with the embroidered CompUSA logo, 1990s store charge cards, plastic shelf talkers and the occasional piece of store signage turn up on Etsy (https://www.etsy.com/search?q=compusa+vintage) and at estate sales, usually from former staff or from people who cleared out a closing location. Prices move with condition and sentiment, and none of it is scarce enough to treat as an investment. ## Frequently Asked Questions ### When did CompUSA close? CompUSA sold its remaining 103 stores in December 2007 to Specialty Equity LLC, an affiliate of Gordon Brothers Group, and those stores ran closing sales through the holidays before shutting in 2008. More than half the chain had already closed in the spring of 2007, alongside a $440 million cash infusion that failed to turn the business around. ### Did TigerDirect buy CompUSA? Systemax, which owned TigerDirect, bought CompUSA's e-commerce business for $18.9 million in January 2008 and sixteen store leases for about $11.7 million over the next two months. It ran CompUSA as a separate brand until the fourth quarter of 2012, when it consolidated its United States consumer operations under TigerDirect and wrote off roughly $35.3 million of CompUSA and Circuit City intangibles. ### Does compusa.com still work? Not as a store. The address loads a hosting error page rather than a shop, and no CompUSA retail site has operated there for years. The COMPUSA.COM trademark itself is still registered and active, held by Source Brand Group, LLC and renewed in January 2023, so the name is owned rather than abandoned. **Sources:** - CompUSA Inc., Form 10-K for fiscal 1999, U.S. Securities and Exchange Commission: https://www.sec.gov/Archives/edgar/data/0000880323/000104746999035798/0001047469-99-035798.txt - Associated Press via NBC News, "CompUSA is sold; will close all stores," December 2007: https://www.nbcnews.com/id/wbna22153236 - Systemax Inc., Form 10-Q for the quarter ended September 30, 2008, U.S. Securities and Exchange Commission: https://www.sec.gov/Archives/edgar/data/0000945114/000110465908068665/a08-25647_110q.htm - Systemax Inc., Form 10-K for fiscal 2012, U.S. Securities and Exchange Commission: https://www.sec.gov/Archives/edgar/data/945114/000114036113012535/form10k.htm - USPTO Trademark Status and Document Retrieval, COMPUSA.COM, Reg. No. 2582709: https://tsdr.uspto.gov/statusview/sn75639885 --- # Windows 95 Startup Sound: Brian Eno's 3.25 Seconds URL: https://404memoryfound.com/posts/windows-95-startup-sound-brian-eno.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-09 Topics: Software & Apps, Music & Entertainment **Summary:** The Windows 95 startup sound was composed by Brian Eno in 1994 and 1995, working to an agency brief that asked for something optimistic and futuristic and exactly 3.25 seconds long. He made 84 versions, on a Macintosh, for a fee of $35,000. You can still download it free from the Internet Archive, and in April 2025 the Library of Congress added it to the National Recording Registry. **Key facts:** - Composer: Brian Eno, commissioned by Microsoft in 1994 - The brief: 3.25 seconds, “optimistic, futuristic, sentimental, emotional” - What shipped: The Microsoft Sound.wav, about 6 seconds - Fee: $35,000, confirmed by Eno in 2025 - Status today: Free CC0 download at the Internet Archive; National Recording Registry, 2025 ## Can you still hear the Windows 95 startup sound in 2026? Yes, and it is free. The Internet Archive hosts the original Windows 95 system sounds as downloadable files, released under a CC0 public domain dedication, with the startup chime available as WAV, FLAC or MP3. A second Archive item offers a cleaned-up stereo transfer of the same recording. What you cannot do is get it from Microsoft. The company replaced Brian Eno's piece with a new startup sound in Windows 98 and has never shipped it again in a consumer release of Windows. The file to look for is named The Microsoft Sound.wav. On a 1995 machine it lived alongside Chimes, Chord, Ding and Tada, and it only played at all if the PC had a sound card installed (https://404memoryfound.com/posts/what-happened-to-creative-labs-sound-blaster.html). Plenty of office PCs in 1995 did not, which is why some people who used Windows 95 every day have no memory of the sound whatsoever. ## Who wrote the Windows 95 startup sound? Brian Eno, the English producer and ambient composer behind "Ambient 1: Music for Airports" and records by U2 and David Bowie. Microsoft designers approached him in 1994, and the piece first appeared in a Windows build in May 1995, three months before Windows 95 went on sale at midnight (https://404memoryfound.com/posts/the-night-windows-95-launched.html) on August 24. Eno described the commission to San Francisco Chronicle critic Joel Selvin in a June 1996 interview. The brief, he said, came from an agency and asked for "a piece of music that is inspiring, universal, blah-blah, da-da-da, optimistic, futuristic, sentimental, emotional," a whole list of adjectives, "and then at the bottom it said, and it must be three and a quarter seconds long." He took it because he was stuck. "The idea came up at the time when I was completely bereft of ideas," he told Selvin. "I really appreciated someone coming along and saying, Here's a specific problem, solve it." Microsoft's own proposal, quoted in the Library of Congress essay on the recording, asked for something that would "capture the spirit of innovation, feel contemporary, and convey a more approachable, human side of the brand." ## Eighty-four versions, written on a Macintosh Eighty-four is the number Eno gave Selvin in 1996. The Library of Congress essay by Eno biographer David Sheppard puts it at more than 80. The scale of the job is the part he kept talking about afterward. "It's like making a tiny little jewel," he said. "I got completely into this world of tiny, tiny little pieces of music. I was so sensitive to microseconds at the end of this that it really broke a logjam in my own work." When he went back to normal work, "pieces that were like three minutes long, it seemed like oceans of time." His own 1985 album "Thursday Afternoon" is a single unbroken piece lasting an hour. He also did not own a PC. Asked by BBC Radio 4 in 2009 whether he had composed the Windows music on a Windows machine, Eno said he wrote it on a Mac, adding, "I've never used a PC in my life; I don't like them." ## Why the brief said 3.25 seconds and the file runs six The version Microsoft shipped is roughly twice the length it asked for. Download The Microsoft Sound.wav from the Internet Archive and it clocks in at about 6.1 seconds. Wikipedia describes it as a six-second start-up sound, and the German archive winhistory.de lists the Windows 95 chime at about 6 seconds against 7 for Windows 98 and 4 for Windows XP. The Library of Congress addressed the gap directly when it took the recording in. The finished clip was twice as long as Microsoft's engineers had asked for, and the company used it anyway because it conveyed the sense of welcome and progress they wanted for the launch. So 3.25 seconds is the brief, not the artifact. The piece Sheppard describes as the chosen one, "all 3.25 seconds of it," is a bell-like attack followed by a decaying electric-piano figure over a synth pad. What plays when the machine boots is that idea with room to breathe. One footnote on the famous number. Microsoft's own developer blog, most reprints and the Library of Congress essay all quote the brief as three and a quarter seconds. An archived transcript of the 1996 Chronicle interview reads "three-and-a-half seconds." The 3.25-second figure is the one Microsoft and the Library use. ## Why later versions of Windows dropped the startup sound Windows 98 replaced Eno's piece with a seven-second crescendo by Microsoft sound engineer Ken Kato. Windows XP got a shorter one. For Windows Vista, Microsoft flew in King Crimson guitarist Robert Fripp, and its Channel 9 developer channel published behind-the-scenes footage of the session in January 2006. That Vista sound carried into Windows 7. The same era gave Windows the Bliss wallpaper (https://404memoryfound.com/posts/windows-xp-wallpaper-bliss-story.html) and a house style that treated boot-up as an event. Then it stopped. Jensen Harris, the lead user-interface designer on Windows 8, killed the startup sound partway through development. He told the podcast Twenty Thousand Hertz that opening a laptop with a sleeping baby on his arm and having the machine announce itself convinced him that a device used in bedrooms and libraries should not make noise by default. Dropping it also saved boot time and memory, which mattered as Microsoft pushed Windows onto smaller devices. The sound file stayed in Windows 8, buried and renamed Windows Logon, and stayed off by default through Windows 8.1 and Windows 10. Windows 11 brought it back in 2021, on by default, playing when the PC reaches the lock screen. Accessibility drove the reversal, because blind and low-vision users needed an audible signal that the machine had finished booting. The new chime is called Petals, designed by Phi Bui, and Microsoft has described it as a small idea blooming into something greater. ## What happened to the Microsoft Sound in 2025 On April 9, 2025, the Library of Congress named the Windows start-up chime to the National Recording Registry, in a class of 25 recordings that also included the original Broadway cast album of "Hamilton," Amy Winehouse's "Back to Black" and the Minecraft soundtrack. The entry is filed as "Windows Start-Up Chime, Brian Eno (1995)," with an essay by David Sheppard, who published the Eno biography "On Some Faraway Beach" in 2008. The Library's stated reason had less to do with the music than with the machine. Windows 95 moved more of the computer's operation under a graphical interface and made a home PC usable by people who were not specialists, and the chime is the sound that announced it. Six weeks later the fee became public. In May 2025 Eno published an open letter criticizing Microsoft's supply of cloud and AI services to Israel's Ministry of Defense and said he would give his original $35,000 composition fee to Palestinian aid. His work for the company in the early 1990s, he wrote, "represented a gateway to a promising technological future." Sheppard's essay ends on a flat note for anyone hoping for a sequel: those waiting for a new Brian Eno computer startup sound are probably in for an infinite wait. ## Frequently Asked Questions ### Was the Windows 95 startup sound really 3.25 seconds long? The brief was. The shipped file was not. Microsoft's agency asked for three and a quarter seconds, and Eno delivered pieces to that spec, but the version that ended up on the Windows 95 disc runs about six seconds. The Library of Congress noted that the final clip was twice as long as the engineers requested. ### How much was Brian Eno paid for the Windows 95 sound? $35,000. Eno confirmed the figure himself in May 2025, when he announced he was donating that fee to Palestinian aid as part of an open letter criticizing Microsoft's contracts with Israel's Ministry of Defense. ### Does Windows 11 still have a startup sound? Yes. Windows 11 restored the startup sound in 2021 and turns it on by default. It is a new composition called Petals by Phi Bui, not Eno's, and you can switch it off in the Sounds control panel by clearing the "Play Windows Startup sound" checkbox. **Sources:** - Library of Congress, “Windows Start-Up Chime, Brian Eno (1995),” National Recording Registry essay by David Sheppard: https://www.loc.gov/static/programs/national-recording-preservation-board/documents/Windows-Start-Up-Chime_Sheppard.pdf - San Francisco Chronicle, “Q&A With Brian Eno” by Joel Selvin, June 2, 1996 (archived): https://www.moredarkthanshark.org/eno_int_sfc-jun96.html - Twenty Thousand Hertz, “Why Windows Got Rid of Startup Sounds”: https://www.20k.org/episodes/windowslogonwav - Internet Archive, “The Original Windows 95 sounds in WAV format”: https://archive.org/details/win95sounds - Engadget, “The Windows 95 start-up chime has been added to the Library of Congress”: https://www.engadget.com/entertainment/music/the-windows-95-start-up-chime-has-been-added-to-the-library-of-congress-203909673.html --- # Who Owns Sears Now, and How Many Stores Are Left? URL: https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-09 Topics: Business Blunders, Money & Tech **Summary:** Sears still exists in 2026. The brand belongs to Transformco, the private company Eddie Lampert's hedge fund ESL Investments created to buy Sears out of Chapter 11 in February 2019, and the last published count put the chain at five full-line department stores. That is down from roughly 3,500 stores at the time of the 2005 Kmart merger. **Key facts:** - Owner today: Transform Holdco LLC, trading as Transformco, controlled by Eddie Lampert's ESL Investments - Stores left: 5 full-line Sears stores, per USA Today's December 2025 count - Status: Open. Not in bankruptcy since the February 2019 sale - Bankruptcy: Chapter 11 filed October 2018; $5.2 billion sale approved February 7, 2019 - Catalog run: 1888 to January 25, 1993, 105 years ## Who owns Sears now? Sears is owned by Transform Holdco LLC, which does business as Transformco. It is a private company set up in 2019 by ESL Investments, the hedge fund run by Edward Lampert, who was chairman of Sears Holdings and, in its final years, its chief executive. The handover happened in a courtroom. On February 7, 2019, Judge Robert Drain of the U.S. Bankruptcy Court in White Plains, New York, approved ESL's $5.2 billion bid for the salvageable parts of Sears Holdings. The sale closed on February 11, 2019. Transformco took 223 Sears stores, 202 Kmart (https://404memoryfound.com/posts/is-kmart-still-open-stores-left.html) stores and roughly 45,000 jobs, plus the names that were worth more than the buildings: Kenmore, DieHard, Craftsman, Sears Home Services and Sears Auto Centers. Because Transformco is private, it files no annual report and publishes no store count. Every number for how many Sears are left comes from reporters counting doors rather than from the company itself, which is why the figure below carries a date attached to it. ## How many Sears stores are left in 2026? Five, and that is a ceiling rather than a promise. USA Today counted the survivors in December 2025 and found full-line Sears department stores still trading in Braintree, Massachusetts; Concord, California; El Paso, Texas; Orlando, Florida; and Coral Gables, Florida. Most of them sit inside malls owned by Simon Property Group, the largest mall operator in the country. The collapse reads cleanly as a list. Sears Holdings ran 1,002 stores in the United States at February 3, 2018, made up of 547 Sears full-line stores, 23 Sears specialty stores and 432 Kmarts, according to its last annual report on Form 10-K. The 2019 bankruptcy sale carried 425 of those forward. By 2022 the Sears half was down to 22 locations. By December 2025 it was five. Set that against roughly 3,500 stores at the time of the 2005 Kmart merger and the shape of the thing is obvious. The same stretch of years emptied out RadioShack (https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html) and Circuit City (https://404memoryfound.com/posts/what-happened-to-circuit-city-electronics.html), but neither of those had spent most of a century as the default American store. Transformco does not announce closures in advance and does not confirm a count, so the honest answer for 2026 is five or fewer. ## What happened in the 2018 Sears bankruptcy Sears Holdings filed for Chapter 11 protection in October 2018. The scale of the retreat was already sitting in the filings: revenue of $16.7 billion for fiscal 2017 and a net loss of $383 million attributable to shareholders, on a store base that had been cut every year for a decade. The auction that followed produced exactly one bid that would keep the chain running instead of liquidating it, and it came from the sitting chairman. Judge Drain approved that bid, having pressed Lampert in open court on whether his legacy would end up looking like Jay Gould's or Barney Fife's. Worth saying plainly, because the search results rarely do: Sears did not go out of business in 2018. Chapter 11 is a reorganization, not a funeral. The company that walked out of it in February 2019 was smaller by design, at 425 stores against the 1,002 of a year earlier. Everything since has been a slower version of the same process, run privately, with no filings and no warning to the towns losing an anchor tenant. ## The catalog that was Amazon before Amazon Richard Sears mailed his first catalog in 1888. It sold watches and jewelry and promised "THE LOWEST PRICES ON EARTH." The book kept growing until it ran well over 1,000 pages and carried more than 100,000 items: tools, hardware, apparel, appliances, furniture, sporting goods. Between 1908 and 1940 it sold houses. More than 100,000 mail-order homes went out as numbered kits of lumber, priced from $107 at the bottom of the range to $3,506 at the top, which Smithsonian magazine puts at roughly $3,500 and $113,000 in current money. People ordered a house out of a book and it arrived on a railcar. Sears opened its first physical store in 1925, and over the following decades the stores quietly became the whole company. The Big Book was retired on January 25, 1993, after a 105-year run, right as the web was arriving. So the line about Sears being Amazon before Amazon is not a blogger's flourish. Sears had the catalog, the distribution, the credit operation and the national reach, and it traded them for anchor space in malls. The company that later rebuilt that model online only had to survive the dot-com crash (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html) to inherit it. ## The $175 million settlement that closed the case The Chapter 11 case outlived the 2019 sale by three years. Unsecured creditors, mostly former suppliers left holding unpaid invoices, sued Lampert along with other former executives and investors, alleging they had stripped Sears Holdings of billions of dollars in value to enrich themselves in the years before the filing. It settled for $175 million, announced on September 1, 2022 and approved by a federal bankruptcy judge in New York. Insurers paid the bulk of it. Lampert and the other defendants admitted no liability, and the plaintiffs acknowledged that the defendants had acted in good faith. Transform Holdco, the company that owns Sears today, was one of the named defendants. What survives now is mostly a brand and a service business. Sears.com still takes orders, Sears Home Services still books appliance repairs, and Kenmore, DieHard and Craftsman are still sold, though not always through Sears. On the stores, Dan Hamilton Rice, a marketing professor at Louisiana State University, told USA Today what most of the retail industry assumes: "With so few stores, there's serious doubt from industry experts and former insiders that they could be profitable." ## Frequently Asked Questions ### Is Sears still in business in 2026? Yes. Sears still operates as a retail brand under Transformco, with a small number of department stores, an online store and a home services arm. It is not currently in bankruptcy: the October 2018 Chapter 11 case was a reorganization, and the surviving business was sold in February 2019. ### Where are the last Sears stores? The last published count, from USA Today in December 2025, listed five full-line stores: Braintree, Massachusetts; Concord, California; El Paso, Texas; Orlando, Florida; and Coral Gables, Florida. Transformco publishes no store directory and gives no notice before a closing, so call ahead before driving to one. ### Is Kmart owned by the same company as Sears? Yes. Kmart bought Sears in 2005 and the two have shared an owner ever since, first as Sears Holdings and now as Transformco. Kmart shrank on the same curve, going from 432 stores in early 2018 to the 202 that were carried into the 2019 bankruptcy sale. **Sources:** - Sears Holdings Corporation, Form 10-K for fiscal 2017 (U.S. Securities and Exchange Commission): https://www.sec.gov/Archives/edgar/data/0001310067/000131006718000006/shld201710k.htm - Retail Dive: Sears back in Eddie Lampert's hands: https://www.retaildive.com/news/sears-back-in-eddie-lamperts-hands/547983/ - Retail Dive: Sears Holdings' $175M settlement with Lampert has court approval: https://www.retaildive.com/news/sears-holdings-175m-settlement-with-lampert-has-court-approval/631045 - USA Today: Sears once dominated retail. Now only 5 stores remain.: https://www.yahoo.com/news/articles/sears-once-dominated-retail-now-195606317.html - Smithsonian Magazine: Before Folding 30 Years Ago, the Sears Catalog Sold Some Surprising Products: https://www.smithsonianmag.com/innovation/before-folding-30-years-ago-the-sears-catalog-sold-some-surprising-products-180981504/ --- # Who Owns Atari Now? The Brand That Refuses to Die URL: https://404memoryfound.com/posts/who-owns-atari-now.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-09 Topics: Gaming, Business Blunders **Summary:** Atari is owned by Atari SA, a French company listed in Paris and led by Wade Rosen, whose holding company IRATA LLC held 39.96% of the shares in August 2026. Atari is still in business and reported about $65 million in revenue for the year to March 2026, its best in more than a decade. The name itself has been bought and sold five times since 1984, and once went for $5 million. **Key facts:** - Owner today: Atari SA, listed on Euronext Growth Paris as ALATA - Controlling shareholder: Wade Rosen's IRATA LLC, 39.96% of capital (August 2026) - Status: Active and operationally profitable; about $65 million revenue for the year to March 2026 - Cheapest the brand ever sold for: $5 million, to Hasbro Interactive, 1998 - Current hardware: Atari 2600+ (2023) and Atari 7800+ (2024, $129.99) ## Who owns Atari in 2026? Atari is owned by Atari SA, a French holding company that spent most of its life under a different name: Infogrames Entertainment. Its shares trade on Euronext Growth in Paris under the ticker ALATA and on the OTC Pink market in the United States as PONGF. The controlling shareholder is Wade Rosen, an American investor who became chairman of the board in April 2020 and chief executive in April 2021. His holding company, IRATA LLC, held 39.96% of the capital and 40.2% of the voting rights as of August 17, 2026. The remaining 60% or so is public float. So nobody owns Atari outright. Rosen controls it the way a large minority holder controls a small listed company, and the Atari name is a corporate asset that has now changed hands five times. ## How the Atari name changed hands five times Nolan Bushnell and Ted Dabney founded Atari in Sunnyvale, California on June 27, 1972. Warner Communications bought the company in 1976 for $28 million. After the video game crash of 1983, Warner broke Atari apart. In 1984 it handed the consumer division to Jack Tramiel, who had just walked out of Commodore (https://404memoryfound.com/posts/what-happened-to-commodore-computers.html), for $240 million in promissory notes and stock. Almost no cash changed hands. Tramiel renamed his half Atari Corporation and spent the next decade shipping computers, the Lynx handheld (https://404memoryfound.com/posts/what-happened-to-atari-lynx-handheld.html) and finally the Jaguar (https://404memoryfound.com/posts/what-happened-to-atari-jaguar-64-bit-console.html). In 1996 Atari Corporation merged with JTS Inc., a hard drive manufacturer, which left one of the most famous names in gaming parked inside a disk drive company. JTS sold that name and its assets to Hasbro Interactive in 1998 for $5 million. Infogrames acquired Hasbro Interactive in 2001, took the brand with it, and in 2009 renamed the entire French parent company Atari SA. That is the company answering to the name today. ## Is Atari still in business, and is it profitable? Yes on both counts, and it took a bankruptcy to get there. Atari's North American subsidiaries filed for Chapter 11 protection in 2013 and emerged about a year later. For the fiscal year that ended March 31, 2026, Atari reported revenue of roughly 56 million euros, about $65 million, its highest in more than a decade. Current operating income was 0.9 million euros, against a 0.8 million euro loss the year before. Operations threw off 11.0 million euros in cash. Those are small numbers next to any major publisher. They are also the first genuinely healthy ones Atari has posted since the Infogrames years, and they come from a company that spent most of the 2010s losing money on licensing deals and a crowdfunded console nobody bought. ## What Atari actually sells today Two things, mostly: an old catalog and modern-retro hardware. Games revenue rose 67.7% to 46.1 million euros in the year to March 2026, driven by new digital and physical releases, platform partnerships and steady management of a library that reaches back past the crash of 1983 (https://404memoryfound.com/posts/how-nintendo-nes-saved-video-games-1983-crash.html). Hardware revenue grew 83% to 7.3 million euros, and that line rests on the Plus consoles. The Atari 2600+ arrived in 2023, a shrunken replica of the original machine with an HDMI output and a working cartridge slot. The Atari 7800+ followed in 2024 at $129.99, made with the German publisher Plaion, playing both 2600 and 7800 cartridges and bundled with a new game called Bentley Bear's Crystal Quest. Where to find one today: both Plus consoles are current products rather than collectibles, so the cheapest route is a new unit at or below the $129.99 the 7800+ launched at, with used ones showing up on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html).com for less. Original consoles from the late 1970s and early 1980s are everywhere on eBay too, where condition and a working switch matter far more than which revision you get. ## The buying spree that turned Atari into a retro company Rosen's Atari has spent the 2020s buying the people who preserve old games rather than the people who make new ones. Nightdive Studios, which rebuilds classics like System Shock and Quake (https://404memoryfound.com/posts/what-happened-to-quake-online-fps.html) for modern hardware, and Digital Eclipse, the studio behind Atari's anniversary compilations, are both Atari subsidiaries now. On May 23, 2024 Atari bought the Intellivision brand and the rights to more than 200 games from Intellivision Entertainment, which kept its Amico console business and licensed the games back. Rosen called it "a very rare opportunity to unite former competitors." A console rivalry that started in 1979 ended with one company owning both logos. The shopping did not stop there. Atari lifted its stake in the Swedish publisher Thunderful to 97%, and after the fiscal year closed it paid an initial $29.3 million for Hipster Whale, the Australian studio behind Crossy Road. That one is a working mobile developer, not a museum piece, which suggests where Rosen wants to take the money the catalog is finally producing. ## What happened to the Atari hotels? On January 27, 2020 Atari announced a licensing deal with GSD Group to build Atari Hotels, with the first set to break ground in Phoenix in the middle of that year. Atari was not paying to build them, it was renting out the name to a developer who would. Nothing was built. The groundbreaking slipped, the pandemic arrived, and more than six years later there is still no Atari hotel open anywhere in the United States. A Phoenix project has since been revived on paper with a different development team and an opening date pushed years past the one originally promised. The hotels are worth remembering because they show what the brand had been reduced to: a logo rented to whoever wanted instant 1980s shorthand. Under Rosen the company has pulled back toward products it controls end to end. That is why the 2600+ exists, why Atari owns the studios that restore its own games, and why the hotel still does not exist. ## Frequently Asked Questions ### Is Atari still making consoles in 2026? Yes, in a narrow sense. The Atari 2600+ and 7800+ are cartridge-compatible reissues of 1970s and 1980s hardware, not new platforms, and they helped hardware revenue reach 7.3 million euros in the year to March 2026, up 83%. Atari is not building anything meant to compete with PlayStation or Xbox. ### Who is the CEO of Atari? Wade Rosen. He joined the board as chairman in April 2020 and took over as chief executive in April 2021. His holding company IRATA LLC held 39.96% of Atari SA's capital as of August 17, 2026, which makes him both the boss and the largest shareholder. ### Is the Atari 2600+ worth buying? It depends on whether you own cartridges. The 2600+ and the 7800+ read original 2600 and 7800 games through a real slot, and that slot is the entire reason to buy one; the 7800+ launched at $129.99. If you have no cartridges sitting in a closet, a digital compilation gives you far more games for less money. **Sources:** - Atari SA investor relations: governance and shareholding: https://atari-investisseurs.fr/en/about-governance/ - Reuters: Once-bankrupt Atari returns to profit with decade-high revenue (August 3, 2026): https://kfgo.com/2026/08/03/once-bankrupt-atari-returns-to-profit-with-decade-high-revenue/ - Atari press release: Atari Acquires Intellivision Brand (May 23, 2024): https://www.globenewswire.com/news-release/2024/05/23/2887219/0/en/Atari-Acquires-Intellivision-Brand.html - Atari and Plaion announce the Atari 7800+ game console (August 20, 2024): https://seekingalpha.com/pr/19823637-get-your-pixel-fix-like-it-s-1986-atari-and-plaion-announce-the-atari-7800-plus-game-console - Wikipedia: Atari (ownership history and corporate timeline): https://en.wikipedia.org/wiki/Atari --- # Is Toys R Us Still in Business? What Survived 2018 URL: https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-09 Topics: Business Blunders **Summary:** Toys R Us is still in business as a brand, but the retailer that ran the superstores is not. All 735 US stores closed on June 29, 2018 after a Chapter 11 filing in 2017. WHP Global bought control of the name in March 2021 and now runs it through Macy's shop-in-shops, a few flagships and 1,664 stores across 35 countries. **Key facts:** - Status today: Brand active; the original US retailer is gone - Owner today: WHP Global, controlling stake since March 15, 2021 - US stores closed: 735, all shut by June 29, 2018 - Stores worldwide: 1,664 in 35 countries as of 2026 - Where to find one: Macy's shop-in-shops, plus flagships like American Dream ## Is Toys R Us still in business in 2026? Yes, but not in the form most Americans remember. The company that ran the big blue box stores was wound down in 2018 and no longer exists. The Toys R Us name, the Babies R Us name and Geoffrey the Giraffe were pulled out of that wreckage and are now brands owned by a New York firm called WHP Global. The distinction matters. When people ask whether Toys R Us is still open, they usually mean the 40,000 square foot store off the highway with the bikes stacked by the door. Those are gone in the United States. Most chains that lost their stores that way, Circuit City (https://404memoryfound.com/posts/what-happened-to-circuit-city-electronics.html) included, never got them back. What replaced them is a mix of departments inside Macy's, a small number of mall flagships, seasonal pop-up shops and stores on Navy bases. Outside the US the brand never stopped trading at all. ## Who owns Toys R Us now? WHP Global, a brand management company, announced on March 15, 2021 that it had acquired a controlling interest in Tru Kids Inc., the entity holding Toys R Us, Babies R Us, Geoffrey the Giraffe and more than 20 related toy and baby brands. The deal came with a $350 million equity commitment from Oaktree Capital Management and a leverage facility from BlackRock. Other shareholders included funds managed by Solus Alternative Asset Management and the private equity group of Ares Management. At the time, Toys R Us and Babies R Us together generated more than $2 billion in global retail sales a year across nearly 900 branded stores and websites in more than 25 countries. WHP took over management of that global business and its expansion. So the honest answer to who owns Toys R Us is a licensing company, not a retailer. WHP holds the trademarks. The stores are mostly run by partners: Macy's, Go! Retail Group, the Navy Exchange Service Command and overseas licensees. ## How a $6.6 billion buyout set up the 2017 bankruptcy On March 17, 2005, a consortium of Bain Capital Partners, Kohlberg Kravis Roberts and Vornado Realty Trust announced a $6.6 billion leveraged buyout of Toys R Us. The debt used to buy the company landed on the company's own balance sheet. By September 2017 Toys R Us was paying roughly $400 million a year just to service that debt. That was money not going into stores, wages or a working website, in the exact decade when Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html) and Walmart were fighting over the same toy shoppers on price. Toys R Us filed for Chapter 11 protection on September 18, 2017 with a reported workforce of about 64,000 people. The stated plan was to restructure, keep the stores open and come out the other side. That plan lasted about six months. ## What happened in 2018, and the comeback that failed The holiday season of 2017 went badly enough that a reorganization stopped looking credible to lenders. In the spring of 2018 the company moved to wind down its US business instead, and on June 29, 2018 the last American Toys R Us stores closed after 70 years of operations. The scale was severe: 735 US stores shut and more than 30,000 people lost their jobs, in most cases with no severance at all. After months of public pressure from worker groups and members of Congress, Bain Capital and KKR each contributed $10 million to a $20 million fund for former employees in November 2018. Individual payouts ran from $200 to $12,800. Vornado, the third buyout partner, did not contribute. Anyone who followed the end of RadioShack (https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html) will recognize the shape of it: a chain taken apart from the balance sheet outward while the staff found out last. What was left of the estate emerged on January 20, 2019 as Tru Kids Inc., which held the trademarks and set out to license them back into stores. Tru Kids opened two small US stores that same year, one at Garden State Plaza in Paramus, New Jersey on November 27, 2019 and one at The Galleria in Houston on December 7, 2019. Both were built around demonstrations and play space rather than aisles of boxed stock. Both closed in January 2021 after the pandemic emptied out American malls. For a stretch of that winter the number of Toys R Us stores in the United States was zero. The Macy's partnership had been announced only months before, and it is what put the name back on the map. ## How many Toys R Us stores are left in the US? It depends on what you count, which is why published numbers disagree. Fox Business reported on September 1, 2026 that the retailer's own website listed more than 30 standalone and flagship locations nationwide. A stricter tally splits the formats apart. Wikipedia's current count puts it at 8 permanent standalone US stores, alongside 451 branded concessions inside Macy's, 13 seasonal holiday shops and 7 stores on military bases. The Macy's arrangement is the backbone of the US business. Toys R Us sections opened in every Macy's location in the country by October 15, 2022, more than 400 of them, which put the brand back in front of American shoppers without signing a single big-box lease. The showpiece is at American Dream in East Rutherford, New Jersey, where a two-story, 20,000 square foot flagship opened on December 16, 2021. That is a fraction of the floor space of a 1990s superstore, and it is the closest thing left to the original. ## What the brand looks like worldwide Globally, Toys R Us is larger now than it was on the day the US stores closed. In 2026 the brand counted 1,664 stores across 35 countries, according to Jamie Uitdenhowen, executive vice president of Toys R Us and Babies R Us at WHP Global. 2025 was the biggest international expansion year in the brand's history, with openings in six new countries and more than 100 new locations worldwide. Panama and Ecuador are the next targets in South America. In the US the growth comes from partners rather than company-owned real estate. Go! Retail Group operates the flagships and the holiday shops, and the Navy Exchange Service Command deal puts stores on bases. There is also a Toys R Us Studios arm developing content, including a movie project. It is an unusual second life, closer to what happened at Polaroid (https://404memoryfound.com/posts/what-happened-to-polaroid-camera.html) than to a normal retail turnaround. The name outlived the retailer that made it famous, the stores got smaller, and the leases and the risk now belong to somebody else. Whether that counts as Toys R Us being back depends on what you wanted back. ## Frequently Asked Questions ### Did Toys R Us go bankrupt or go out of business? Both, in stages. Toys R Us filed for Chapter 11 bankruptcy protection on September 18, 2017 intending to reorganize, then wound down its US operations and closed all 735 remaining stores on June 29, 2018. The brands were later sold, and WHP Global has run them since March 2021. ### Where can you find a Toys R Us store today? In the United States the most reliable option is the Toys R Us section inside a Macy's, since those opened in every US Macy's store by October 2022. There are also a handful of standalone flagships, including American Dream in New Jersey, plus seasonal holiday shops and stores on Navy bases. ### Is Babies R Us still around? Yes. Babies R Us was part of the same brand portfolio WHP Global took control of on March 15, 2021, and the two names are marketed together. It exists as shops and licensed retail rather than as the standalone US superstore chain it was before 2018. **Sources:** - WHP Global Acquires Controlling Stake in Toys“R”Us (PR Newswire, March 15, 2021): https://www.prnewswire.com/news-releases/whp-global-acquires-controlling-stake-in-toysrus-301247254.html - $20 million severance fund started for Toys R Us workers (CBS News, November 2018): https://www.cbsnews.com/news/20-million-severance-fund-started-for-tens-of-thousands-of-toys-r-us-workers/ - State of the Industry Q&A 2026: Jamie Uitdenhowen, Toys “R” Us, WHP Global (The Toy Book): https://toybook.com/state-of-the-industry-qa-2026-jamie-uitdenhowen-toys-r-us-whp-global/ - Toys R Us expands brick-and-mortar comeback to over 30 US locations (Fox Business, September 2026): https://www.foxbusiness.com/retail/beloved-toy-retailer-continues-comeback-new-california-store-nostalgic-shoppers-rejoice - Toys “R” Us (Wikipedia): https://en.wikipedia.org/wiki/Toys_%22R%22_Us --- # Is Palm Still a Company? From Pilot to a $349 Phone URL: https://404memoryfound.com/posts/is-palm-still-a-company.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-09 Topics: Hardware, Business Blunders **Summary:** Palm is not a company anymore. HP completed its roughly $1.2 billion purchase of Palm on July 1, 2010, shut the hardware down in August 2011, sold webOS to LG in February 2013 and sold the Palm trademark in October 2014 to a holding company tied to TCL. TCL still owns the name, and the last Palm phone was the 3.3-inch, $349 Android device sold from November 2018. **Key facts:** - Founded: 1992, by Jeff Hawkins with Donna Dubinsky and Ed Colligan - Bought by HP: July 1, 2010, for about $1.2 billion ($5.70 a share) - Owner today: TCL, which bought the Palm trademark from HP in October 2014 - webOS today: LG Electronics, which bought the software in February 2013 and runs it on TVs - Status: Dormant brand, no new Palm phone since the 3.3-inch model of November 2018 ## Is Palm still a company in 2026? No. Palm stopped being an independent business on July 1, 2010, the day Hewlett-Packard completed its purchase of the company. Everything that carried the name after that was owned by somebody else. Three pieces of Palm went three different ways. The hardware and phone business was shut down by HP in August 2011. The webOS software was sold to LG Electronics in February 2013. The Palm trademark itself was sold in October 2014 to a holding company tied to the Chinese electronics maker TCL, which still owns it. So the status today is split. Palm the company is gone, Palm the name is dormant under a Chinese owner, and Palm's software is alive under a third company, running television sets. ## Who owns Palm now? TCL. In October 2014, HP sold the Palm trademark to Wide Progress Global Limited, a shelf company controlled by Nicolas Zibell, at the time a regional president at TCL. TCL confirmed the purchase publicly on January 6, 2015 and talked about building a team in Silicon Valley to work out what to do with the name. What it did with the name took another three and a half years, and then it stopped. A single Palm-branded Android phone was unveiled on October 15, 2018, built by TCL, with Golden State Warriors guard Stephen Curry attached as an investor and the face of the launch. Since then, nothing. Palm's own store at palm.com is still online in 2026, still describing that 3.3-inch phone in the present tense, with no product available to buy. The trademark is maintained. The company behind it is a portfolio line item. ## What did HP get for $1.2 billion? HP announced the deal on April 28, 2010: $5.70 a share in cash, an enterprise value of roughly $1.2 billion. Todd Bradley, who ran HP's Personal Systems Group, said Palm's operating system was "an ideal platform to expand HP's mobility strategy." The deal closed on July 1, 2010. What HP bought was a company that had already been through four owners and two identities. Palm was founded in 1992 by Jeff Hawkins, with Donna Dubinsky and Ed Colligan, sold to U.S. Robotics in 1995, absorbed into 3Com in 1997, spun back out as a public company on March 1, 2000, split into a software arm and a hardware arm in 2002 and 2003, and reassembled in 2005 when the hardware side paid $30 million for the other half of its own trademark. That history is the subject of our piece on the rise and fall of the Palm Pilot (https://404memoryfound.com/posts/what-happened-to-palm-pilot-pda.html). By 2010 the good part was the software. webOS launched with the Palm Pre on June 6, 2009 on Sprint, and reviewers liked its card-based multitasking well enough that HP bought the company mostly to get it. ## Why HP killed webOS in 13 months On August 18, 2011, HP announced it would end production and support of Palm and webOS devices. That was less than 14 months after the acquisition closed and roughly 16 months after HP told investors it was buying its way into the smartphone business. The accounting caught up in November. In its fiscal 2011 results, released November 21, 2011, HP excluded $3.3 billion in after-tax costs, or $1.56 per diluted share, tied to winding down the webOS device business plus the related goodwill impairment, amortization, restructuring and acquisition charges. The fourth quarter alone carried $2.1 billion of that. For a company that had spent the previous decade digesting its purchase of Compaq (https://404memoryfound.com/posts/compaq-lost-pc-business-hp-merger.html), it was a familiar shape of problem: buy a strong brand, inherit a losing platform war, discover the war is already over. Apple and Android had taken the market by 2011, and the same squeeze was already flattening BlackBerry (https://404memoryfound.com/posts/what-happened-to-blackberry-smartphone.html). ## Where webOS ended up: LG television sets On February 25, 2013, LG Electronics announced it had acquired webOS from HP. The deal covered the source code, the documentation, the engineering team and the related websites, plus licenses to HP patents originally filed by Palm covering operating system and user interface work. LG also took over stewardship of the Open webOS and Enyo open source projects, and picked up HP's Sunnyvale and San Francisco research sites. HP kept the Palm cloud assets: the source code, the staff, the infrastructure and the contracts behind Palm's online services. Neither company expected the deal to move its numbers, and neither disclosed a price. LG's plan was smart TVs, and that is where webOS still lives. LG has since turned it into a licensing business, selling the platform to other television manufacturers under the webOS Hub name. Palm's software now ships on sets made by companies that never sold a handheld. ## The 2018 Palm phone was a $349 accessory The one product that used the Palm name in the smartphone era was strange by design. Announced October 15, 2018 and sold from November through Verizon, the Palm phone had a 3.3-inch screen, ran Android, and could not be bought unlocked at first. It cost $349 up front, or about $15 a month financed, and it needed a Verizon Connected Device plan at $10 a month plus the carrier's NumberShare feature so it could borrow your real phone number. In other words, it was a $349 second phone for the kind of person who already owned a first phone. Reviewers struggled to explain who that was, the idea did not spread, and no Palm phone has followed it in the seven years since. Where to find one today: nothing Palm-branded is sold new, so the secondhand market on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html).com is the only route, where the small model turns up unlocked, boxed, or as a carrier-locked Verizon unit. Use the launch price as the anchor, $349 in November 2018, and check the battery, because a cell that small, sitting in a drawer since 2019, is the part that fails first. ## Frequently Asked Questions ### Does Palm still make phones? No. The last Palm-branded phone was the 3.3-inch Android device sold from November 2018, and no successor has been announced. TCL owns the trademark and has kept it dormant. ### What happened to webOS after HP gave up? LG Electronics bought it on February 25, 2013, including the source code, the engineers and the patent licenses, and moved it into smart TVs. LG now sells it to other television makers under the webOS Hub name, which is why webOS is still shipping in 2026 while the Palm brand sits idle. ### How much did HP lose on Palm? HP paid about $1.2 billion in 2010 and shut the hardware business on August 18, 2011. In its fiscal 2011 results HP set aside $3.3 billion in after-tax costs for the wind-down and the related impairment, amortization and restructuring charges. **Sources:** - HP: HP to Acquire Palm for $1.2 Billion (April 28, 2010): https://investor.hp.com/news-events/news/news-details/2010/HP-to-Acquire-Palm-for-12-Billion/default.aspx - HP: HP Reports Fourth Quarter and Full Year 2011 Results (November 21, 2011): https://investor.hp.com/news-events/news/news-details/2011/HP-Reports-Fourth-Quarter-and-Full-Year-2011-Results/default.aspx - LG Electronics Acquires webOS from HP to Enhance Smart TV (February 25, 2013): https://www.prnewswire.com/news-releases/lg-electronics-acquires-webos-from-hp-to-enhance-smart-tv-193071841.html - 9to5Google: Palm Phone goes official as a companion to your standard phone, $349 Verizon exclusive (October 15, 2018): https://9to5google.com/2018/10/15/palm-phone-launch-date-verizon/ - Palm, Inc. (company history, trademark sale to TCL): https://en.wikipedia.org/wiki/Palm,_Inc. --- # Is Netscape Still Around? What Happened After AOL URL: https://404memoryfound.com/posts/is-netscape-still-around.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-09 Topics: Software & Apps, Internet Culture **Summary:** Netscape is not still around as a company or a working browser. AOL bought it in a stock deal announced at $4.2 billion in November 1998 and worth about $10 billion when it closed on March 17, 1999, then ended browser support on March 1, 2008. The name survives only as a live US trademark held by AOL Membership Services LLC, and netscape.com now redirects to aol.com. **Key facts:** - Bought by AOL: Announced November 24, 1998 at $4.2 billion, closed March 17, 1999 at about $10 billion - Browser status: Discontinued. Support ended March 1, 2008 - Netscape ISP: Launched January 8, 2004 at $9.95 a month, shut down November 30, 2025 - Owner today: AOL, acquired by Bending Spoons on January 2, 2026 - Trademark: US Reg. No. 2027552, live, AOL Membership Services LLC ## Is Netscape still around in 2026? Not as a company, and not as a browser anyone should be running. Netscape Communications stopped existing as an independent business on March 17, 1999, the day AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html) closed its acquisition. AOL then ended support for every Netscape browser and client product on March 1, 2008. What survived was the name. AOL kept the trademark and attached it to a discount dial-up service, a news portal and, more recently, a rebadged Chromium browser. The last of those, Netscape Internet Service, was discontinued on November 30, 2025. So the honest answer to "does Netscape still exist" is that the brand is legally alive and commercially retired. Type netscape.com into any browser in 2026 and the server answers with a permanent redirect to aol.com. There is no Netscape branding waiting on the other side, no archive page, no farewell note. ## What AOL actually bought in 1999 AOL announced the deal on November 24, 1998 as a tax-free stock swap valued at $4.2 billion. AOL shares kept climbing through the regulatory review, so by the time the transaction closed on March 17, 1999 it was worth roughly $10 billion. Netscape shareholders were paid in AOL paper, not cash, which is why the price moved so much between the handshake and the closing. The headline number hid what AOL was really shopping for. Netscape's browser share was already collapsing under Internet Explorer, the fight covered in how Netscape lost the browser war (https://404memoryfound.com/posts/how-netscape-lost-the-browser-war.html). The assets with real revenue attached were Netcenter, then one of the busiest portals on the web, and Netscape's enterprise server software. AOL split them up. The server business went into an alliance with Sun Microsystems. The portal was fed into AOL's own traffic machine. The browser, the one thing every user associated with the name, got the smallest budget and the least attention. By July 2003 the Netscape division was disbanded, engineers were laid off and the logo came off the Mountain View building. Four years after the biggest software acquisition of the dot-com boom, the product had no team. ## Who owns the Netscape name now? AOL does, and AOL itself changed hands again in 2026. Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html), controlled by Apollo Global Management, sold the AOL business to the Italian software group Bending Spoons, which reported in a US securities filing that it acquired AOL Holdco I LLC on January 2, 2026. The trademark file is the clearest evidence that somebody still pays the maintenance bills. US Registration No. 2027552 for the word mark NETSCAPE, registered December 31, 1996, is listed by the USPTO as live, issued and active, with the status "the registration has been renewed." The owner of record is AOL Membership Services LLC of 770 Broadway, New York. That is the same corporate lineage that shut down AOL Instant Messenger (https://404memoryfound.com/posts/what-happened-to-aim-aol-instant-messenger.html) and quietly wound down one legacy product after another. Bending Spoons said AOL arrives with about 8 million daily and 30 million monthly active users, concentrated in email and news. Netscape is a line item inside that portfolio, not a product with a roadmap. ## What happened to the Netscape browser after AOL The post-1999 release history is short and unhappy. Netscape 6 arrived on November 14, 2000 and shipped with enough bugs to burn whatever goodwill was left. Netscape 7.0 followed in August 2002, and version 7.2, released August 18, 2004, was the last built on the old Mozilla suite. Then AOL outsourced. Netscape Browser 8.0 through 8.1.3, released between 2005 and 2007, was Windows only and built on Firefox. Netscape Navigator 9, the final version, launched October 15, 2007 on top of Firefox 2 and was the first release produced in house since 2004. The last update, 9.0.0.6, shipped February 20, 2008. The numbers explain the decision. In November 2007, Net Applications measured Internet Explorer at 77.4% of the browser market, Firefox at 16.0% and Netscape at 0.6%. On December 28, 2007 AOL announced it would stop developing and supporting the browser, setting the cutoff at February 1, 2008 and then extending it to March 1, 2008 to allow one last security update. AOL told the remaining users to install Flock or Firefox instead. ## The Mozilla spin-off that outlived the company The most valuable thing Netscape did in its final independent year was give the code away. In February 1998, about a year before the AOL deal closed, Netscape released the source of its browser and created the Mozilla Organization to coordinate development. That project rewrote the browser around the Gecko rendering engine. When AOL scaled back its involvement in the early 2000s, the Mozilla Organization set up the Mozilla Foundation in July 2003 as an independent nonprofit, with financial and other assistance from AOL on the way out the door. Gecko still powers Firefox. So the practical answer to what replaced Netscape is not a competitor that beat it, but a nonprofit that inherited its code and its engineers. AOL paid roughly $10 billion for a browser and ended up funding the exit of the only part of it that lasted. It was not the last time AOL turned a famous name into a write-down. The AOL and Time Warner merger (https://404memoryfound.com/posts/aol-time-warner-merger-worst-deal-in-history.html) announced in January 2000 followed the same pattern on a much larger scale. ## What netscape.com is today The main domain is a redirect and nothing more. The one address still serving a page is isp.netscape.com, which loads a generic news aggregator wearing the Netscape ISP logo, with headlines pulled from AOL's feeds and a "Copyright 2026 AOL Media LLC" line at the bottom. Buried in that footer is a "Download Browser" link. It installs a Chromium fork carrying the Netscape name, developed for AOL by the UK security firm SentryBay, with AOL Search wired in as the default. It shares nothing with the browser that shipped in 1994 except six letters. The ISP that page was built to sell is gone. Netscape Internet Service launched on January 8, 2004 at $9.95 a month for unlimited dial-up, ran for more than two decades, and was discontinued on November 30, 2025. Subscribers kept their netscape.net and netscape.com email addresses, which now run through AOL Mail at no charge. If you want a physical piece of it, shrink-wrapped Navigator and Communicator boxes and the retail CD-ROMs bundled with 1990s modems still turn up regularly on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html).com. Condition of the box, not the disc, is what collectors argue about, since the software itself has been a free download from archive sites for years. ## Frequently Asked Questions ### Does Netscape still exist as a company? No. Netscape Communications ceased to be an independent company when AOL completed the acquisition on March 17, 1999, and AOL disbanded the Netscape division in July 2003. Only the trademark survives, held by AOL Membership Services LLC. ### Can you still download the Netscape browser? You can download a Chromium-based browser that carries the Netscape name from the old Netscape ISP site, but it is not the original software. The real Netscape browser has been unsupported since March 1, 2008, and its last build, version 9.0.0.6, dates from February 20, 2008. ### Is Netscape the same thing as Firefox? They are not the same product, but they share ancestry. Netscape released its browser source code in February 1998 and started the Mozilla Organization, which became the Mozilla Foundation in July 2003 and ships Firefox on the Gecko engine that grew out of that code. **Sources:** - Netscape (Wikipedia): https://en.wikipedia.org/wiki/Netscape - TechCrunch: A Sad Milestone, AOL To Discontinue Netscape Browser Development (December 28, 2007): https://techcrunch.com/2007/12/28/a-sad-milestone-aol-to-discontinue-netscape-browser-development/ - USPTO Trademark Status and Document Retrieval: NETSCAPE, US Reg. No. 2027552: https://tsdr.uspto.gov/statusview/sn74574057 - Bending Spoons S.p.A., SEC Form DRS/A (2026): https://www.sec.gov/Archives/edgar/data/0002004711/000110465926054644/filename1.htm - Bending Spoons to acquire AOL following $2.8B debt financing: https://finance.yahoo.com/news/bending-spoons-buy-aol-yahoo-080640555.html --- # Is Blockbuster Still Open? The Last Store in 2026 URL: https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-09 Topics: Business Blunders, Music & Entertainment **Summary:** Blockbuster is still open, but only as a single franchise store in Bend, Oregon, which has been the last one on earth since March 2019. The chain once had 9,000 stores worldwide before its 2010 bankruptcy. The brand itself is owned by Dish Network, which Dish bought at auction in 2011 and which has been a subsidiary of EchoStar since the end of 2023. **Key facts:** - Status: Open, one store, in Bend, Oregon - Stores left: 1, down from 9,000 worldwide at its height - Owner today: Dish Network, a wholly owned subsidiary of EchoStar - Brand sold for: About $320 million at a bankruptcy auction, April 6, 2011 - Last in the world since: March 2019, when the Perth, Australia store closed ## Is Blockbuster still open in 2026? Yes, in the narrowest possible sense. Exactly one Blockbuster store is still trading, in Bend, Oregon, and it has been the only one left on earth since March 2019. Everything corporate is long gone. Blockbuster filed for bankruptcy in 2010, and the last company-owned stores were shut by 2014. What survived was a single franchise that kept paying to use the name over the door. The Bend shop is not a display case. It still rents discs, still sells merchandise, and still signs up new members. In October 2025 it threw a street party for the 40th anniversary of the chain's founding, held on October 18 outside the store, with an 80s costume contest, karaoke with a live band, and a share of the day's rental and merchandise takings going to the Humane Society of Central Oregon, The Bulletin reported. So the accurate answer to the question people type into Google is this: Blockbuster is open, if you are willing to drive to central Oregon. ## Who owns Blockbuster now? The brand belongs to Dish Network, and Dish Network belongs to EchoStar. Dish won Blockbuster's assets at a bankruptcy auction on April 6, 2011. The winning bid was valued at about $320 million, and after adjustments for available cash and inventory Dish expected to hand over roughly $228 million in cash at closing, according to the announcement it filed with the SEC. The package included more than 1,700 store locations, the brand, and the distribution channels attached to it. Dish never rebuilt a retail chain out of that. Retail Dive reported that Dish said in November 2021 it licenses the Blockbuster brand to apparel and games manufacturers and to the store in Bend, and that blockbuster.com works mainly as a funnel pointing visitors toward Dish's own programming packages. The same reporting noted that Blockbuster gets no mention in Dish's 10-K filings, and that Dish counsel sent a cease and desist in November 2020 to Free Blockbuster, the volunteer project that stocks donated movies in repainted newspaper boxes. The parent company changed again at the end of 2023. EchoStar completed its merger with Dish Network on December 31, 2023, with Dish surviving as a wholly owned subsidiary. Each share of Dish Class A stock converted into 0.350877 EchoStar shares. The combined company trades on Nasdaq as SATS and is run out of Englewood, Colorado. ## How 9,000 stores became one Blockbuster once had 9,000 stores worldwide, in the blue and yellow livery that sat in more or less every American strip mall. The collapse from that number to one took roughly a decade, and it happened in stages rather than in a single crash. The 2010 bankruptcy took out the balance sheet. The Dish purchase in 2011 kept about 1,700 doors open on paper, but Dish closed and sold down the estate rather than running it, and the last company-owned locations were gone by 2014. After that, every remaining Blockbuster was a franchise, an independent operator writing a check for the right to use the sign. Those went one at a time, usually over rent. Blockbuster stores in Anchorage and Fairbanks, Alaska, closed in July 2018, which left Bend as the last Blockbuster in the United States. A store in Perth, Australia, closed in March 2019, and Bend became the last one anywhere. The chain that passed on buying Netflix (https://404memoryfound.com/posts/blockbuster-netflix-acquisition.html) ended as a single storefront run by people who know their regulars by name. That arc is not unique. RadioShack (https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html) and Circuit City (https://404memoryfound.com/posts/what-happened-to-circuit-city-electronics.html) both went through the same sequence of bankruptcy, asset sale and brand-in-a-drawer afterlife, but neither left a working store behind as a landmark. ## What the last Blockbuster actually sells today Rentals still happen. The store carries about 26,000 movies, and staff sign up new membership cards daily for locals and for visitors who flew in from other countries, according to The Bulletin's reporting in October 2025. The most-rented title is still "The Goonies," helped along by the fact that it was shot in Astoria, up on the Oregon coast. The rest of the business is merchandise. The store sells branded shirts, hoodies and even branded beer, and that side of the counter is what keeps the lights on rather than the disc rentals. The customer base is a nostalgia audience, and the store knows it. Visit Bend's own surveys show visitors to the store skew older and remember life before streaming, which is a polite way of saying the shop sells a memory as much as a movie. General manager Sandi Harding has run the place since 2004 and has become the public face of the brand. Her line to visitors is straightforward. Renting a disc is the point, and buying a T-shirt on the way out is what makes renting the disc possible. ## Why this store survived when the rest did not Three things kept Bend open. It is a franchise, so it never depended on Dish's corporate strategy. It sits in a tourist town, so the foot traffic is not only local. And it leaned into being the last one instead of pretending the chain still existed. The tourist angle turned out to be the business model. Back in March 2019, when the Perth closure made Bend the last store in the world, Harding told reporters the store had become a tourist attraction and was not closing anytime soon. That was seven years ago and it still holds. Rumors that the store is shutting circulate on social media every year or two, and every time the store and local outlets knock them down. The one real risk has always been the same one that killed the Alaska stores, which is the lease rather than the rentals. Dish, for its part, has no reason to pull the license. The brand costs it nothing to sit on, the store generates free publicity for a name Dish paid $320 million for, and the alternative is owning a trademark with no visible use at all. ## Where to find Blockbuster stuff today The store runs its own online shop, so you can buy Blockbuster-branded merchandise without going to Oregon. That is the only retail channel officially licensed to use the name. Beyond that, the secondary market carries the physical leftovers. Clamshell rental cases, laminated membership cards, drop-box signage and store display standees all turn up regularly on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html).com. Prices swing hard by title and condition, so check completed listings rather than asking prices before you bid, and be aware that reproduction membership cards are common. ## Frequently Asked Questions ### Can you still rent movies from Blockbuster? Yes, at the Bend, Oregon store, which carries about 26,000 movies and issues new membership cards to walk-in customers. It is the only place in the world where a Blockbuster rental is still possible, and it has held that status since March 2019. ### How many Blockbuster stores are left? One. Blockbuster once had 9,000 stores worldwide. The Alaska stores in Anchorage and Fairbanks closed in July 2018, and the Perth, Australia store closed in March 2019, leaving Bend alone. ### Does Dish Network still own Blockbuster? Yes. Dish bought Blockbuster's assets at auction on April 6, 2011 with a bid valued at about $320 million, and it still licenses the brand to the Bend store and to merchandise makers. Dish Network has been a wholly owned subsidiary of EchoStar since December 31, 2023. **Sources:** - DISH Network press release on the Blockbuster auction (SEC Form 8-K exhibit, April 6, 2011): https://www.sec.gov/Archives/edgar/data/0001001082/000100108211000002/exhibit99_1.htm - NBC News: Oregon city will have the last Blockbuster store in the world (March 2019): https://www.nbcnews.com/news/us-news/oregon-city-will-have-last-blockbuster-store-world-n980401 - Retail Dive: Blockbuster is trapped in brand limbo. Will it ever get out?: https://www.retaildive.com/news/blockbuster-is-trapped-in-brand-limbo-will-it-ever-get-out/609054/ - EchoStar Corporation Completes Merger with DISH Network Corporation: https://www.prnewswire.com/news-releases/echostar-corporation-completes-merger-with-dish-network-corporation-302024076.html - The Bulletin: Rewind to 1985, Bend's Blockbuster celebrates 40 years (October 9, 2025): https://bendbulletin.com/2025/10/09/rewind-to-1985-bends-blockbuster-celebrates-40-years-of-movie-magic/ --- # Does AOL Still Exist? What You Get for $10 a Month URL: https://404memoryfound.com/posts/does-aol-still-exist-today.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-09 Topics: Internet Culture, Business Blunders **Summary:** AOL still exists in 2026. It runs free AOL Mail, a news portal and paid plans starting at $9.99 a month, and since January 2, 2026 it has been owned by the Italian software company Bending Spoons, which bought it from Apollo-backed Yahoo for roughly $1.5 billion. The one thing that is gone is dial-up, which AOL switched off on September 30, 2025. **Key facts:** - Status today: Active: AOL Mail, news portal and paid plans - Owner today: Bending Spoons (Milan), since January 2, 2026 - Last sale price: About $1.5 billion, down from $4.4 billion in 2015 - What it costs: AOL Desktop Gold $9.99 a month, 30-day trial - Dial-up: Discontinued September 30, 2025 ## Does AOL still exist in 2026? Yes. AOL still exists, and it is not a shell or a parked domain. It runs a free webmail service, a news and lifestyle portal, and a set of paid subscription plans that real people renew every month. The company that once mailed America hundreds of millions of free trial discs is now a mid-size internet business with a straightforward job: keep tens of millions of email accounts working. When Bending Spoons agreed to buy AOL, chief executive Luca Ferrari put the audience at roughly 8 million daily active users and 30 million monthly active users, and described AOL as one of the ten most-used email providers in the world. What no longer exists is the part everyone pictures. The modem handshake is gone. AOL shut off dial-up internet access on September 30, 2025. Email, the portal and the paid plans all survived that cut. ## Who owns AOL now? AOL is owned by Bending Spoons, a Milan-based software company that buys aging internet products and rebuilds them. The same firm owns Evernote, Vimeo, Meetup and WeTransfer. The deal was announced on October 29, 2025 and reported at roughly $1.5 billion by the Wall Street Journal and Axios, with Reuters putting it closer to $1.4 billion. It closed on January 2, 2026, when Bending Spoons acquired AOL Holdco I LLC, a fact the company later confirmed in its Form F-1 registration statement filed with the U.S. Securities and Exchange Commission. Ferrari said "Bending Spoons has never sold an acquired business" and framed the purchase as a long-term hold rather than a flip. ## How AOL got passed down four times The ownership chain is short and expensive. Verizon bought AOL in 2015 for $4.4 billion, then bought Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) in 2017 for $4.8 billion, then wrote down about half the combined value of both brands in 2018. In 2021 Verizon sold the whole media group to private equity firm Apollo Global Management for $5 billion. Apollo renamed it Yahoo and kept AOL inside it. Verizon held on to a 10 percent stake. Five years later Apollo sold AOL alone for about $1.5 billion. That is a third of what Verizon paid for AOL by itself in 2015, and a rounding error against the AOL Time Warner merger (https://404memoryfound.com/posts/aol-time-warner-merger-worst-deal-in-history.html) that valued the company in the hundreds of billions in 2001. Each owner bought AOL for a different reason. Verizon wanted an ad business. Apollo wanted cash flow. Bending Spoons wants a subscription base it can price and improve. ## What does an AOL subscription cost in 2026? AOL Desktop Gold, the paid browser and mail client that most subscribers actually use, is listed at $9.99 a month on AOL's own product page, with a 30-day free trial. That is the $10 a month figure people ask about. For that price the plan includes anti-keylogging and screen-grab protection, anti-phishing tools, mail filtering that collects every attachment or image from a sender in one place, two-step verification, encryption of information stored in AOL Desktop, and live phone or chat support advertised as available 24 hours a day, seven days a week. The wider AOL Advantage plans are harder to price from the outside. The comparison site Allconnect lists AOL Advantage at $9.99 a month, while AOL's own pages have pointed to bundles running from $11.99 to $16.99 a month, mostly built around technical support and identity theft protection from partners such as LifeLock and McAfee. None of that includes internet access. It is a software and support subscription that happens to carry the AOL name. A subscriber in 2026 is buying antivirus-style peace of mind and a phone number to call, wrapped in an interface they have used since the Clinton administration. That is also the reason the business is worth buying. The plans renew quietly, the customers rarely leave, and the cost of keeping the lights on is low compared with what people pay. ## Why did AOL kill dial-up on September 30, 2025? AOL announced in August 2025 that dial-up would end on September 30 of that year, along with the AOL Dialer and AOL Shield software that supported it. The company said only that it "routinely evaluates its products and services" and had decided to discontinue dial-up internet. The demand had drained away years earlier. Verizon inherited about 2.2 million dial-up subscribers when it bought AOL in 2015. By 2021 roughly 1.5 million people still paid AOL something every month, but almost none of them were dialing in. Reporting on the shutdown cited 2023 Census Bureau data showing about 163,400 Americans relied on dial-up as their only internet connection, across every provider, not just AOL. The product itself had stopped being competitive long before that. Reporting at the time of the shutdown pegged AOL dial-up at roughly 0.056 megabits per second, which is not enough to load a modern web page, let alone stream anything. AOL charged $25.90 a month for dial-up in 2006, according to a CNET report from the time, which was the same price as its DSL service. If you want a sense of what those subscribers were actually getting, the math on how slow dial-up really was (https://404memoryfound.com/posts/dialup-speed-then-vs-now.html) has not aged kindly. ## Can you still get dial-up internet anywhere? Yes, just not from AOL. NetZero (https://404memoryfound.com/posts/dial-up-internet-still-available-2026.html) still advertises dial-up, including a free tier capped at 10 hours a month and a paid accelerated plan reported at $11.95 a month. EarthLink has offered dial-up at $9.95 a month for an introductory period and $24.95 a month afterward. So dial-up is not extinct in 2026. It is a niche product for people outside broadband coverage, sold by companies that never made it the center of their identity the way AOL did. AOL's other 1990s assets did not all make it this far. AIM was shut down in 2017 (https://404memoryfound.com/posts/what-happened-to-aim-aol-instant-messenger.html), and the AOL Time Warner content empire was broken up and sold in pieces. Email is the one thing that kept compounding, because changing an email address is genuinely annoying and most people never bother. ## Frequently Asked Questions ### Is AOL Mail still free? Yes. AOL Mail remains free to use with ads in the inbox, and it does not require any paid AOL plan. The paid plans add software such as AOL Desktop Gold, security tools and live support, not access to the mailbox itself. ### Can I still get AOL dial-up internet in 2026? No. AOL ended dial-up on September 30, 2025 and retired the AOL Dialer and AOL Shield software with it. Other providers including NetZero and EarthLink still sell dial-up, so the technology exists, just not under the AOL brand. ### Why do people still pay AOL every month? Mostly for support and security rather than internet access. Roughly 1.5 million people were still paying AOL in 2021, and the plans they bought were built around technical help, identity theft protection and the AOL Desktop Gold software, which AOL lists at $9.99 a month. **Sources:** - Bending Spoons S.p.A., Form F-1 registration statement (U.S. SEC): https://www.sec.gov/Archives/edgar/data/0002004711/000110465926071170/tm2613674-7_f1.htm - AOL: You've got a new owner: AOL is sold in reputed $1.5B deal: https://www.aol.com/articles/ve-got-owner-aol-sold-201343341.html - AOL Desktop Gold product page (pricing and features): https://www.aol.com/products/browsers/desktop-gold - Currently: How much AOL dial-up service costs in 2025: https://currently.att.yahoo.com/att/much-aol-dial-costs-2025-171205880.html - Yahoo News: AOL to shut down dial-up Internet in September: https://www.yahoo.com/news/articles/aol-shut-down-dial-internet-125631260.html --- # How Compaq Lost the PC Business It Helped Invent URL: https://404memoryfound.com/posts/compaq-lost-pc-business-hp-merger.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-09 Topics: Hardware, Business Blunders **Summary:** Compaq lost the PC business it helped invent by defending a dealer network that Dell's direct model had already made obsolete. It was the world's largest PC maker by 1994, fell behind Dell in 2001, and disappeared into Hewlett-Packard in a $25 billion all-stock deal that closed on May 3, 2002. HP retired the Compaq brand in 2013 and now licenses it to other manufacturers. **Key facts:** - Founded: February 1982, Houston, by three ex-Texas Instruments managers - First product: Compaq Portable, March 1983, $2,995 - Peak: World's largest PC maker from 1994 - Bought by: Hewlett-Packard, $25 billion all-stock deal closed May 3, 2002 - Status today: Brand retired by HP in 2013, now licensed to other makers ## What was Compaq, and why did it matter? Compaq Computer Corporation was founded in February 1982 by Rod Canion, Jim Harris and Bill Murto, three managers who had just walked out of Texas Instruments. The plan they sketched over pie in Houston was simple and expensive: build a portable machine that ran every piece of software written for the IBM PC, without copying a line of IBM's code. The obstacle was the BIOS, the small program that tells an IBM PC how to talk to its own hardware. Compaq solved it with a clean-room project. One team studied IBM's BIOS and wrote a specification of what it did. A second team, which had never seen IBM's code, wrote fresh software to match that specification. The Compaq Portable was announced in November 1982 and shipped in March 1983 at $2,995 with a single 360 KB floppy drive, or $3,590 with two. It ran Lotus 1-2-3 (https://404memoryfound.com/posts/what-happened-to-lotus-1-2-3-spreadsheet.html), which was the only compatibility test most business buyers cared about. The entire PC clone industry starts at that machine. ## How did Compaq get so big so fast? In 1983, its first full year of production, Compaq shipped 53,000 portables and booked more than $111 million in revenue, the largest first year any American company had recorded to that point. In 1986 it joined the Fortune 500 in its fourth year of existence, faster than any company before it. That September it shipped the Deskpro 386, the first personal computer built around Intel's 32-bit 80386 processor. IBM, which was supposed to set the pace for the standard it had created, did not ship a 386 machine of its own until the PS/2 Model 80 in August 1987. That was the moment the PC standard stopped belonging to IBM. Compaq decided what a PC was, and the rest of the industry followed. Buyers noticed. In the first half of 1994 Compaq held 10.4 percent of the world PC market against IBM's 8.5 percent, which made a twelve-year-old company the largest PC maker on earth. It got there with premium engineering sold through a dealer network that earned a healthy margin on every box. Both halves of that sentence became the problem. ## What went wrong in 1991? Cheap clones got good. Dell, Gateway (https://404memoryfound.com/posts/what-happened-to-gateway-2000-cow-box-computer.html) and AST were shipping machines that ran the same software for hundreds of dollars less, and buyers stopped paying extra for the Compaq badge. After sales of $3.6 billion and profits of $455 million in 1990, Compaq posted a $70 million loss in the third quarter of 1991, the first quarterly loss in its history. Canion cut about 1,400 jobs and asked the board for eighteen months to build a low-cost line. He did not get them. Chairman Ben Rosen had quietly commissioned his own team to find out how cheaply a Compaq could be built, and on October 25, 1991 the board replaced Canion with chief operating officer Eckhard Pfeiffer. Co-founder Jim Harris resigned two weeks later. Pfeiffer cut prices hard and went after ordinary shoppers. On August 27, 1993 Compaq launched the Presario at $1,399, a complete machine with a monitor and a modem in the box, stacked in electronics stores rather than sold by a reseller. More than 100,000 went out in sixty days, and the line brought in $500 million before the year ended. ## Why did buying Digital Equipment backfire? Pfeiffer wanted Compaq to be a full-service computer company rather than a box shop, so on January 26, 1998 he agreed to buy Digital Equipment Corporation for $9.6 billion, the largest acquisition the computer industry had seen. Digital shareholders received $30 in cash and about 0.945 Compaq shares for each share they held. The deal closed in June 1998 at an aggregate purchase price of $9.1 billion, and the accounting was brutal. Compaq wrote off roughly $3.2 billion of purchased in-process technology, booked $393 million in restructuring charges, and accrued $1.1 billion for separation benefits covering about 19,700 employees worldwide. It also bought a culture that did not fit. Digital sold minicomputers and services to engineers through its own sales force. Compaq sold consumer PCs through retailers. Revenue climbed from roughly $25 billion in 1997 to $31 billion in 1998, and the profits went missing on the way. Early in 1999 Compaq warned that first-quarter earnings would miss expectations. On April 19, 1999 the board pushed out Pfeiffer along with chief financial officer Earl Mason. Rosen ran the company himself until Michael Capellas took the job. ## How did Dell beat Compaq at its own game? Dell built each machine after the customer ordered it and sold it direct. No dealer markup, no warehouse full of last quarter's configurations, no third party's margin to protect. Compaq could not copy that without wrecking the thousands of resellers who carried its products, so for years it half-copied it and the price gap stayed open. Selling beige boxes was a fine business until beige boxes became interchangeable, a point Apple made loudly with the iMac G3 (https://404memoryfound.com/posts/how-imac-g3-saved-apple-killed-beige-box.html) in 1998. Dell passed Compaq as the largest PC seller in the United States in the third quarter of 1999. Worldwide it took the lead in the first quarter of 2001, holding 13 percent of the market to Compaq's 12 percent. Dell's shipments that quarter rose 30 percent to 4.1 million units. Compaq's fell 4.7 percent. Compaq was not a small company in 2001. It was a slow one, in a market that had stopped paying for anything except cost. ## What happened when HP bought Compaq? On September 3, 2001 Hewlett-Packard announced it would buy Compaq in an all-stock deal worth about $25 billion. Walter Hewlett, an HP director and the son of one of its founders, opposed it and ran a proxy campaign against his own board. Shareholders approved the merger by a narrow margin at the vote on March 19, 2002. Hewlett contested the result in Delaware court, lost, and the deal closed on May 3, 2002. On June 4, 2002 HP said it would cut 15,000 jobs from a combined workforce of roughly 150,000, with 10,000 going by that November and 5,000 more in 2003. The Compaq name lived on for another decade on budget desktops and Presario laptops, until HP retired the brand in 2013. It survives as a licence. HP owns the trademark and rents it out, which is why Compaq-branded televisions sell in India and Compaq-branded laptops sell in Brazil, made by companies with no connection to Houston. Where to find one today. Original 1983 Compaq Portables show up on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html) most weeks. Untested or parts-only units are the cheapest way in, working examples that still boot from a floppy typically run a few hundred dollars, and complete systems with the original carry case and manuals get listed in four figures. Presario desktops from the 1990s are far more common and often sell for little more than the cost of shipping them. ## Frequently Asked Questions ### Is Compaq still around? Not as a company. Compaq was absorbed into Hewlett-Packard on May 3, 2002, and HP stopped putting the name on its own products in 2013. HP still owns the trademark and licenses it to other manufacturers, so new Compaq-branded televisions and laptops are sold in markets including India and Brazil. ### How much did the first Compaq computer cost? The Compaq Portable shipped in March 1983 at $2,995 with one 360 KB floppy drive, or $3,590 with two. Compaq sold 53,000 of them that year for more than $111 million in revenue, the biggest first year of any American company at the time. ### Why did HP buy Compaq? HP wanted scale in PCs and servers to compete with Dell and IBM, and Compaq had become cheap enough to buy after years of falling margins. The all-stock deal was worth about $25 billion when it was announced in September 2001, and HP cut 15,000 jobs after it closed in May 2002. **Sources:** - Compaq Computer Corporation, Handbook of Texas, Texas State Historical Association: https://www.tshaonline.org/handbook/entries/compaq-computer-corporation - Compaq: Gone But Not Forgotten, TechSpot: https://www.techspot.com/article/2112-compaq/ - Compaq Computer Corp, Form 10-Q, 1998, SEC EDGAR: https://www.sec.gov/Archives/edgar/data/0000714154/000071415498000030/0000714154-98-000030.txt - Dell steals Compaq's crown, The Register, April 2001: https://www.theregister.com/2001/04/20/dell_steals_compaqs_crown/ - HP to cut 15,000 jobs by the end of 2003, CNN Money, June 2002: https://money.cnn.com/2002/06/04/technology/hp/index.htm --- # Why Iridium Failed: The $5 Billion Satellite Phone URL: https://404memoryfound.com/posts/iridium-satellite-phone-failure.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-09-09 Topics: Business Blunders, Hardware, Money & Tech **Summary:** Iridium failed because it spent about $5 billion and 11 years building a satellite phone network aimed at a market that ordinary cellular coverage reached first. Service opened on 1 November 1998 with $3,000 handsets and airtime at $3 to $8 a minute, and by August 1999 the company had roughly 20,000 subscribers against loan covenants that required 52,000. It filed for Chapter 11 on 13 August 1999, and the satellites were later sold out of bankruptcy for $25 million. **Key facts:** - Launched: Commercial service 1 November 1998, Iridium LLC, backed by Motorola - Price then: $3,000 handset, $3 to $8 per minute of airtime - What it cost to build: About $5 billion and 11 years for 66 active satellites - The collapse: Chapter 11 on 13 August 1999; satellites sold for $25 million in November 2000 - Status today: Iridium Communications (Nasdaq: IRDM), 2,537,000 billable subscribers at the end of 2025 ## Why did Iridium fail? Iridium failed because it sold a 1990 idea to a 1999 customer. The handset specification, the pricing and the subscriber forecasts all came from market work done by Motorola engineers at the start of the decade, when a business traveller landing in another country often had no way to use their phone at all. By the time the network switched on in November 1998, ordinary cellular coverage had spread across most of the places those travellers actually went, and roaming agreements had filled in much of the rest. The product made that gap obvious. An Iridium handset cost $3,000 and calls ran $3 to $8 a minute. The phone needed a clear view of the sky, which meant stepping outside a building and away from tall structures to place a call that a $50 handset could make from a hotel lobby for a fraction of the price. The subscriber numbers followed. Iridium reported 10,294 subscribers for the first quarter of 1999 against an internal target of 27,000. By August the base had reached roughly 20,000, well under the 52,000 its loan covenants required. Nine months separated the first commercial call from the bankruptcy filing. ## How a complaint from the Bahamas turned into 66 satellites The idea started in 1985, when Motorola engineer Bary Bertiger was on holiday in the Bahamas and his wife could not reach her clients on her cellular phone. Bertiger, Raymond J. Leopold and Ken Peterson worked the problem into a design in late 1987, and Motorola patented it in 1988. Their answer was to skip the towers entirely. Instead of a network of ground stations that stopped at the edge of the coverage map, a ring of low-orbit satellites would pass overhead everywhere on Earth, poles and oceans included, and hand calls between each other in space. The first design called for 77 satellites, which is where the name came from: iridium is element 77 on the periodic table, and the constellation orbiting the planet looked like electrons around a nucleus. Engineers later worked out that 66 satellites would cover the globe. Element 66 is dysprosium, which did not have the same ring to it, so the name stayed. Building it took about 11 years and roughly $5 billion. That is the number worth holding on to, because everything that happened after 1999 is a story about what that $5 billion turned out to be worth. ## What did an Iridium phone cost in 1998? Commercial service opened on 1 November 1998. The first call went from US Vice President Al Gore to Gilbert Grosvenor, chairman of the National Geographic Society and a descendant of Alexander Graham Bell, a piece of staging that told you exactly how the company saw itself. A handset cost $3,000. Airtime ran from $3 to $8 a minute depending on where the call went. The phone itself was a brick with a thick swivelling antenna, closer in size to a 1980s car phone than to the pocketable handsets that were selling by then. Motorola would spend the next few years proving it knew how to make small phones instead, a lesson that showed up in the Motorola RAZR V3 (https://404memoryfound.com/posts/motorola-razr-v3-coolest-phone-ever-made.html). The economics only worked for customers with no alternative: ships, oil rigs, remote mining sites, aid agencies, film crews and armies. Those customers existed, and Iridium still serves them. There were just never enough of them at those prices to service the debt on a $5 billion constellation. The sales operation also had to explain, country by country, that a phone bought in one place could be used everywhere, which ran into local licensing rules and distribution partners who had little reason to hurry. ## How fast did the collapse happen? Fast enough that the satellites were still being commissioned while the company was running out of money. Chief executive Edward Staiano, who had come from Motorola, announced he was leaving in April 1999 after a disagreement with the board over strategy. The first quarter subscriber number, 10,294, was published the following month. By the summer, Iridium was in default. Two days after missing payments on $1.5 billion of loans, on 13 August 1999, Iridium LLC filed for Chapter 11 protection in New York. Contemporary reports placed it among the 20 largest bankruptcies in United States history at that point. The filing relieved the company of interest payments on about $5 billion of debt, which cut its costs sharply, but it could not manufacture subscribers. Attempts to restructure through 1999 and into 2000 failed to produce a buyer at anything close to the money spent. What made this failure different from the dot-com collapses that followed is that Iridium had shipped. The technology worked. Calls connected from the middle of the Pacific. The thing that did not work was the arithmetic, in the same way it did not for Excite@Home's broadband bet (https://404memoryfound.com/posts/what-happened-to-excite-at-home-broadband.html). ## How did $5 billion of satellites sell for $25 million? Because the alternative was burning them. Motorola, which operated the constellation, had prepared to bring the satellites down through the atmosphere rather than keep paying to fly them for a company that no longer existed. A group led by Dan Colussy, a former chief executive of UNC Inc, spent 2000 assembling a bid. In November 2000, US Bankruptcy Judge Cornelius Blackshear approved the sale of the 72 orbiting satellites to the Colussy group for $25 million. Three rival bids were rejected, including one for $100 million, because the bidders had not posted the required $1.5 million deposit. The piece that made the numbers work was a customer. In December 2000 the Defense Information Systems Agency awarded a two-year, $72 million contract, worth $36 million a year, giving 20,000 government users unlimited airtime on the network. Commercial service relaunched on 30 March 2001 under the new owner. So the constellation that cost about $5 billion to build changed hands for roughly half a cent on the dollar, kept flying, and has been in continuous service ever since. ## Is Iridium still around in 2026? Yes, and it is public. A special purpose acquisition company, GHL Acquisition Corp, bought Iridium Holdings on 29 September 2009 and renamed itself Iridium Communications, trading on Nasdaq as IRDM. The original satellites are gone. Between January 2017 and 11 January 2019, SpaceX launched all 75 Iridium NEXT replacements across eight Falcon 9 missions, giving the company 66 active satellites plus nine in-orbit spares. The last of the original satellites was deorbited on 5 December 2019, which also ended the Iridium flares, the brief mirror-bright flashes the old antenna panels threw across the night sky at up to magnitude -8. The business is unrecognisable from 1999. Iridium ended 2025 with 2,537,000 total billable subscribers and reported service revenue of $634.0 million for the year, most of it from tracking devices, ships, aircraft and machine-to-machine hardware rather than voice calls. Putting satellite calling directly into ordinary phones has been harder. Qualcomm announced Snapdragon Satellite with Iridium at CES in January 2023, then terminated the agreement in November 2023 after handset makers declined to build the feature in. ### Where to find one today The original Motorola-built handsets turn up regularly on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html). Recent listings ran from about $70 for parts-only units up to roughly $425 for a working 9505 with battery and antenna, which is a strange fate for a $3,000 phone that once needed a $5 billion network behind it. Dedicated hardware being swallowed by the phone in your pocket is the same arc that finished off standalone GPS units (https://404memoryfound.com/posts/what-happened-to-standalone-gps-tomtom-garmin-magellan.html). ## Frequently Asked Questions ### Why did Iridium go bankrupt? Iridium filed for Chapter 11 on 13 August 1999, two days after defaulting on $1.5 billion of loans, because it had signed up roughly 20,000 subscribers against loan covenants requiring 52,000. Ordinary cellular networks had expanded faster than the 1990 forecasts the business plan was built on, and $3,000 handsets with $3 to $8 per minute airtime could not compete with them. ### How much did an Iridium satellite phone cost in 1998? The handset cost $3,000 when commercial service began on 1 November 1998, and calls were billed at $3 to $8 a minute depending on the route. Used Motorola 9505 handsets now sell on eBay for roughly $70 to $425 depending on condition. ### Is Iridium still in business today? Yes. Iridium Communications trades on Nasdaq as IRDM, flies 66 active Iridium NEXT satellites launched by SpaceX between 2017 and 2019, and ended 2025 with 2,537,000 billable subscribers and $634.0 million in annual service revenue. **Sources:** - Iridium files for Chapter 11, CNN Money, 13 August 1999: https://money.cnn.com/1999/08/13/companies/iridium/ - Iridium Files for Chapter 11, The Washington Post, 14 August 1999: https://www.washingtonpost.com/archive/business/1999/08/14/iridium-files-for-chapter-11/29828512-3dce-4a0a-8e8e-b5725c72cad5/ - Judge to approve sale of Iridium satellites, The Baltimore Sun, 16 November 2000: https://www.baltimoresun.com/news/bs-xpm-2000-11-16-0011160450-story.html - Defense Department Exercises First Renewal Option of Contract With Iridium Satellite LLC, SpaceNews: https://spacenews.com/defense-department-exercises-first-renewal-option-of-contract-with-iridium-satellite-llc/ - Iridium Announces 2025 Results; Issues 2026 Outlook, 12 February 2026: https://investor.iridium.com/2026-02-12-Iridium-Announces-2025-Results-Issues-2026-Outlook --- # The Beanie Babies Bubble: How a $5 Toy Went Bust URL: https://404memoryfound.com/posts/beanie-babies-bubble-what-happened.html Author: Dana Reyes (404 Memory Found) Published: 2026-09-09 Topics: Money & Tech, Business Blunders **Summary:** Beanie Babies were $5 plush toys from Ty Inc. that became a speculative bubble between 1996 and 1999, when scarcity, retired designs and eBay pushed some bears to hundreds of dollars. Ty's sales reached an estimated $1.4 billion in 1998. The market collapsed in 1999 and 2000, and today most Beanie Babies are worth a few dollars. **Key facts:** - Launched: 1993, Ty Inc. (founded by H. Ty Warner in 1986) - Price then / now: About $5 in stores, roughly $10 in 2026 dollars; hundreds on the secondary market at the peak - Peak: 1998, with Ty sales estimated at $1.4 billion; McDonald's sold 100 million Teenie Beanies in 1997 - Bust: Ty announced a retirement of the whole line for 31 December 1999; prices collapsed through 2000 - Status today: Ty still sells Beanie Babies; most 90s bears trade for $1 to $5 ## How did a $5 plush toy become an investment? Ty Warner started Ty Inc. in 1986 selling stuffed cats, and in 1993 he introduced nine small, under-stuffed animals filled with plastic pellets so they could sit and slump like real pets. Each had a name, a birthday and a poem on a heart-shaped tag, and each cost about $5, which is roughly $10 in 2026 money. Three decisions turned a toy into an asset. Warner refused to sell to big chains such as Toys "R" Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html) and Walmart, so Beanie Babies only appeared in small gift shops and Hallmark stores, in limited quantities per order. He "retired" designs without warning, which meant any bear could become unavailable overnight. And he kept changing tags and details between production runs, which gave collectors variants to hunt. By 1996 the pattern was set: a retirement announcement, a sold-out shelf, and a price on the secondary market that had nothing to do with the cost of fabric and pellets. ## What made 1997 and 1998 the peak? Two things poured fuel on the fire. In April 1997 McDonald's put miniature Teenie Beanie Babies in Happy Meals and sold about 100 million of them in a matter of weeks, with restaurants running out and adults buying meals just for the toy. Later that year Ty released Princess, a purple bear honouring Diana, Princess of Wales, with proceeds going to her memorial fund. Princess was so scarce at first that it became the emblem of the whole craze. The other accelerant was eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html), which had launched in 1995 and gave every collector in America a live price for every bear. Suddenly a retired Beanie Baby had a quote, like a stock, and magazines and price guides published "portfolios" of which animals to hold. Ty's revenue in 1998 was estimated at $1.4 billion, and Warner, who owned the company outright, became a billionaire on a product that retailed for the price of a sandwich. The numbers behind the mania were never about the toy. A bear cost Ty a dollar or two to make and sold to shops for around $2.50, so the $5 shelf price was already a healthy margin. Everything above that, the $50 for a retired Garcia bear or the $300 for an early Peanut the elephant in royal blue, was money changing hands between collectors, and Ty saw none of it. What Ty did get was free demand: every resale price printed in a newspaper sent more people into gift shops asking for the next release. ## Why did the Beanie Babies bubble burst? Bubbles end when the supply of new buyers runs out, and by 1999 almost everyone who wanted Beanie Babies already had bins of them. Ty had also kept producing. Retiring a design creates scarcity only if the factory stops, and the factory did not stop; it moved on to new animals in growing quantities, so the total number of bears in circulation kept climbing while the number of people paying $50 for one shrank. On 31 August 1999 Ty announced that every Beanie Baby would be retired on 31 December 1999. Collectors read it as a last chance and prices briefly spiked, then Ty ran an online vote on whether the line should continue, the public voted yes, and the company kept making them. The retirement that was supposed to lock in value turned out to be a marketing event, and the market lost faith. Through 2000 secondary prices fell to a fraction of their peak, and the bears that had been "worth" hundreds went back to being worth about what they cost new. It was the same lesson the dot-com companies (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html) were about to learn a few months later: a price that depends on the next buyer paying more is not a value, it is a queue. ## What happened to Ty Warner? Ty Inc. survived the crash and still sells Beanie Babies, Beanie Boos and other plush lines today; Warner remained the sole owner and one of the richest people in the toy industry. His reputation took a different kind of hit in 2013, when he pleaded guilty to tax evasion for hiding income in a Swiss bank account. In January 2014 a federal judge in Chicago sentenced him to two years of probation and community service, and he paid a civil penalty of more than $53 million on top of back taxes. The craze itself became a subject in its own right. Zac Bissonnette's 2015 book The Great Beanie Baby Bubble told the business story, the 2021 documentary Beanie Mania interviewed the collectors, and the 2023 film The Beanie Bubble dramatised Ty's rise with Zach Galifianakis as Warner. ## What are Beanie Babies worth today? For almost every bear from the 90s, the honest answer is $1 to $5, and often less if the tag is bent or missing. The listings that go viral, asking $10,000 for a Princess bear or a Valentino with a "rare" typo, are asking prices, not sales. Every Princess bear has the tag errors people describe as rare, because Ty made millions of them across several factories. Completed eBay sales for Princess typically land between $10 and $30. A small number of genuine rarities do sell for real money: early bears with first-generation tags in mint condition, prototypes, and a few limited editions made for employees or events. Those are the exceptions that keep the myth alive. If a childhood collection is sitting in a closet, the most valuable thing about it is probably the memory. ## Frequently Asked Questions ### Are any Beanie Babies actually valuable? A few are: first-generation bears with early tags in mint condition, prototypes, and limited editions made for employees or events can sell for hundreds or occasionally thousands. The Princess bear, despite what listings claim, usually sells for $10 to $30 because Ty made millions of them. ### When did the Beanie Babies craze end? The peak was 1998. Ty's announcement on 31 August 1999 that the whole line would retire, followed by a public vote to keep it going, broke collectors' confidence, and secondary prices collapsed through 2000. ### Does Ty still make Beanie Babies? Yes. Ty Inc. never stopped, and it still sells Beanie Babies alongside newer lines such as Beanie Boos. The company remains privately owned by Ty Warner. **Sources:** - Beanie Babies creator Ty Warner sentenced to two years' probation for tax evasion (U.S. Department of Justice, 2014): https://www.justice.gov/usao-ndil/pr/beanie-babies-creator-ty-warner-sentenced-two-years-probation-tax-evasion - Ty Warner profile (Forbes): https://www.forbes.com/profile/ty-warner/ - Ty Inc. official site: https://www.ty.com/ - Beanie Babies (Wikipedia): https://en.wikipedia.org/wiki/Beanie_Babies --- # Why the NSA Banned Furby: The 1998 Toy Craze Explained URL: https://404memoryfound.com/posts/furby-history-nsa-ban.html Author: Marcus Vale (404 Memory Found) Published: 2026-09-09 Topics: Hardware, Internet Culture **Summary:** Furby was a $35 animatronic toy from Tiger Electronics that launched in October 1998, sold about 1.8 million units before Christmas, and reached 40 million sales within three years. In January 1999 the National Security Agency banned it from its Maryland headquarters over fears that its chip could record conversations. It could not, but the ban made Furby a legend. **Key facts:** - Launched: October 1998, Tiger Electronics (owned by Hasbro since early 1998) - Inventors: Dave Hampton and Caleb Chung - Price then / now: $35 in 1998, roughly $68 in 2026 dollars; resold for $100 and more that Christmas - Sales: 1.8 million in 1998, 14 million in 1999, more than 40 million in three years - Status today: Relaunched by Hasbro in 2012, 2016 and 2023; 1998 originals trade on eBay ## What was Furby, exactly? Furby was a hamster-sized robot covered in fur that blinked, wiggled its ears, spoke a made-up language called Furbish, and appeared to learn English the longer you played with it. Tiger Electronics showed it at the American International Toy Fair in February 1998 and put it in stores that October at a suggested price of $35. The idea came from Dave Hampton, an electronics engineer, and Caleb Chung, a toy designer and former street performer. Hampton had watched the Tamagotchi (https://404memoryfound.com/posts/tamagotchi-digital-pet-that-made-us-feel.html) craze of 1997 and wanted a pet you could actually touch. The pair built the prototype in about nine months, and Hasbro, which had bought Tiger Electronics earlier in 1998, put its distribution muscle behind it. Inside the fur was a single motor driving cams for the eyes, ears, mouth and body, a small microcontroller with about 80 KB of program memory, a light sensor, a tilt sensor, touch switches on the back and belly, a microphone that detected loud sounds, and an infrared port so two Furbys could "talk" to each other. ## Why did every store sell out in 1998? Tiger built the launch around scarcity, on purpose or not. Furby arrived in stores in October 1998 with limited stock, television news ran segments on parents queuing before dawn, and by December the toy had become the year's hard-to-find gift, the way Cabbage Patch Kids had been in 1983 and Tickle Me Elmo in 1996. About 1.8 million Furbys sold in the last three months of 1998. On the secondary market the $35 toy routinely changed hands for $100, and newspaper reports from that December cited asking prices of $300 for rare colours. The following year Tiger shipped 14 million, and by the end of 2000 total sales had passed 40 million units. For a toy whose only real trick was talking, that put it among the best-selling electronic toys of the decade. Adjusted for inflation, the $35 sticker is about $68 today. That makes Furby cheaper in real terms than most of the 90s gadgets people remember as expensive (https://404memoryfound.com/posts/90s-tech-inflation-calculator.html), which is part of why parents bought three or four of them. ## Did Furby actually learn English? No. Every Furby left the factory with the same vocabulary: roughly 200 words, about half of them Furbish ("kah" for me, "may-may" for love, "u-nye" for you) and half English. The English words were unlocked gradually by an internal counter that advanced with play time, so a Furby that had been switched on for a few weeks used more English than one fresh from the box. Owners read this as learning, and Tiger's marketing let them. The infrared port added to the illusion. Two Furbys placed face to face exchanged short signals and appeared to hold a conversation, and a "sick" Furby could catch a sneeze from another. None of it involved recording or processing speech. The microphone only measured volume, so the toy could react to a clap or a shout. ## Why did the NSA ban Furby? In January 1999 an internal memo circulated at the National Security Agency's headquarters at Fort Meade, Maryland, warning staff not to bring Furbys into the building. The memo, reported by the BBC and American newspapers on 13 January 1999, described the toy as having an embedded chip that could record and repeat what it heard, and asked anyone who found one on the premises to contact the security office. The fear was that a Furby carried home from a classified area might replay fragments of what it had overheard. Tiger Electronics responded within days. Roger Shiffman, the company's president, said Furby had no recording capability at all: it could not store audio, and everything it said was pre-programmed. The NSA never withdrew the memo, and the story ran for weeks, which was the best advertising the toy ever got. Whether the memo reflected a real assessment or an over-cautious security officer was never confirmed. What is certain is that a toy with less processing power than a graphing calculator became, for a month, the most famous piece of "surveillance equipment" in America. ## What happened to Furby after the craze? Demand fell as fast as it rose. By 2000 stores were discounting the original, Tiger released Furby Babies and licensed variants, and by 2001 the line was quietly wound down. Hasbro brought it back in 2005 as the larger Emoto-Tronic Furby with better voice recognition, but sales were modest. The 2012 relaunch was the one that stuck. Hasbro replaced the plastic eyes with small LCD screens, added a companion app for phones and tablets, and sold it at $54 during another strong Christmas. Furby Boom followed in 2013 and Furby Connect, with Bluetooth, in 2016. In 2023, for the 25th anniversary, Hasbro released a new Furby at $69.99 with five voice-activated modes and, this time, no app. It is still on sale. ## Where to find an original Furby today The 1998 and 1999 Furbys are cheap and common. Working examples typically list on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html) for $20 to $80 depending on colour and condition, with boxed or unusual editions asking more. Buyers should ask whether the motor still runs through a full cycle: the cam mechanism is the first thing to fail, and a Furby that only blinks is a Furby that needs surgery. The batteries are four AA cells in the base, so nothing exotic is required to wake one up. ## Frequently Asked Questions ### Could Furby record conversations? No. The original Furby had a microphone that only measured loudness, so it could react to noise, but it had no memory for audio and no way to play anything back. Everything it said was stored in its read-only program memory at the factory. ### How much was a Furby in 1998? The suggested retail price was $35, roughly $68 in 2026 money. During the shortage in December 1998, resellers regularly asked $100, and some rare colours were advertised at $300. ### Is Furby still being made? Yes. Hasbro relaunched Furby in 2023 for its 25th anniversary at $69.99, following earlier revivals in 2005, 2012, 2013 and 2016. Original 1998 units are no longer produced but are easy to find second-hand. **Sources:** - NSA: a Furby-free zone (BBC News, 13 January 1999): http://news.bbc.co.uk/2/hi/americas/254094.stm - Furby brand page (Hasbro): https://www.hasbro.com/en-us/brands/furby - Furby (Wikipedia): https://en.wikipedia.org/wiki/Furby --- # Sega Channel Explained, the 1994 Cable Game Service URL: https://404memoryfound.com/posts/what-happened-to-sega-channel.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-18 Updated: 2026-09-18 Topics: Gaming, Hardware, Then vs Now **Summary:** Sega Channel was a cable subscription service that broadcast Sega Genesis games into a special adaptor, launching on December 12, 1994 with TCI and Time Warner and shutting down on July 31, 1998 after peaking at just over 250,000 subscribers. The service is gone and an original adaptor receives nothing today, but the library survived, because the Video Game History Foundation recovered 144 Sega Channel ROMs and Gaming Alexandria published them for free in December 2025. **Key facts:** - Launched: December 12, 1994, by Sega of America with TCI and Time Warner - Price: About $15 a month plus a one-time $25 activation fee that included the adaptor - Peak subscribers: Just over 250,000, against Sega's target of one million in year one - Shut down: July 31, 1998, after the partners announced the withdrawal in late 1997 - Status today: Dead since 1998, but 144 recovered ROMs were published free by Gaming Alexandria on December 15, 2025 ## How did games travel down a coaxial cable in 1994? Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html) Channel used the cable line that was already in the wall. A technician split the feed, ran one branch to the television and the other into a tall black adaptor that sat in the Sega Genesis cartridge slot, and the console booted into an on-screen menu instead of a game. Behind that menu was a broadcast loop, not a download server. Sega sent game data to cable headends through Hughes Communications' Galaxy 7 satellite, uploading at 1.435 GHz and downloading at 1.1 GHz, and each headend pushed the loop out to subscribers around the clock. Nothing was on demand. Once a player picked a title, the console "would wait until the selected game appeared in the loop and then capture its data," as Wikipedia's technical summary of the service puts it in 2026. The adaptor held 4 MB of RAM. The captured game ran out of that RAM exactly as a cartridge would, and it vanished the moment the console was switched off, so playing the same title the next day meant catching it on the loop again. Scientific Atlanta built the first adaptor and General Instrument built a later version. In Canada and South America the satellite stage was skipped entirely and data went straight into the cable headends. ## What did a Sega Channel subscription actually cost? About $15 a month in the United States, plus a one-time $25 activation fee that covered the adaptor and the installation. The figure moved a little by operator, and Canadian subscribers paid roughly Can$19 a month when the service crossed the border in late 1995. Set against 1995 retail, that was cheap. A new Genesis cartridge ran close to $50, and a weekend rental cost a few dollars for two days, so fifteen dollars bought a rotating shelf of about fifty games. Sega made the pitch in its own words. A full-page advertisement in Electronic Gaming Monthly in August 1995 sold the service as "the all day, all night video game channel" with "up to 50 games a month" for one monthly fee. The same advertisement pushed the parts a rental store could not match, including secret codes, insider tips, and a Test Drives section where subscribers played slices of games that were not in stores yet. By 1997 the library had grown to roughly 70 titles, refreshed every two weeks instead of monthly. ## Which games could you play nowhere else? A handful of titles were genuinely locked to the service in North America. Alien Soldier, Pulseman and Mega Man: The Wily Wars all shipped as cartridges in Japan or Europe but reached American players only through Sega Channel. For years the service was the only legal route to those three games in the United States, which is a large part of why their import cartridges still carry the prices they do among collectors in 2026. The monthly reset also did something retail could not. A kid who would never have spent $50 on an unknown platformer would try it on a Tuesday afternoon because it was already sitting on the menu, and several cult favourites in the Genesis catalogue built their audience exactly that way. Sega also broadcast versions of games trimmed to fit the filesize limits of the loop, so some Sega Channel builds are not byte-identical to the cartridge release. Those variants became one of the main targets for preservationists three decades later. Two further exclusives, Garfield: Caught in the Act - The Lost Levels and a Sega Channel build of The Flintstones, were considered lost outright until 2025. ## Why did Sega, TCI and Time Warner pull the plug? The partners announced the end in late 1997, and the reason they gave was hardware rather than demand. Telecompaper reported in December 1997 that Sega of America, Tele-Communications Inc. and Time Warner Entertainment would withdraw the service by June 1998 because it ran on a 16-bit platform at a point when 32-bit and 64-bit consoles had become the standard. The subscriber numbers backed that up. Sega had aimed for a million subscribers inside the first year. The service peaked at just over 250,000, even though participating operators could reach more than 20 million American households. Timing did the rest. Sega Channel launched on December 12, 1994, five years into the Genesis lifecycle and after the company had already spent its attention on add-ons like the Sega 32X (https://404memoryfound.com/posts/what-happened-to-sega-32x-add-on.html). The Sega Saturn (https://404memoryfound.com/posts/what-happened-to-sega-saturn-console.html) reached North America in 1995 and the Sony PlayStation followed, and parents who had just bought a 32-bit console were not signing up for a monthly bill to keep playing 16-bit games. The shutdown slipped by a month. The announced date was June 30, 1998, and the service was finally discontinued on July 31, 1998. Sega itself left the hardware business a few years later, a retreat traced in detail in the story of why Sega stopped making consoles after the Dreamcast (https://404memoryfound.com/posts/why-sega-stopped-making-consoles-dreamcast.html). ## Who rescued the library, and can you play it in 2026? You can play it, though not the way subscribers did. The Video Game History Foundation spent two years on the recovery, starting in 2024 after meeting Michael Shorrock, the former Sega Channel vice president of programming, and working from tape backups tracked down by a collector who had contacted former staff. In December 2025 the foundation published the result, which was 144 recovered ROMs including close to a hundred unique system builds covering most versions of the service shipped between 1994 and mid-1997. The haul included the two games thought lost and a prototype Genesis web browser Sega had been building to push compressed web pages down the same cable pipe. The files did not stay in a vault. "We've donated the data from the 144 new ROMs we recovered to the team at Gaming Alexandria," the foundation wrote in its December 2025 write-up, and Gaming Alexandria posted the collection on December 15, 2025 as free direct downloads and a torrent, playable in any Genesis emulator or on real hardware with a flash cartridge. If you want the standard routes to the rest of the catalogue, the options for playing Sega Genesis games in 2026 (https://404memoryfound.com/posts/sega-genesis-games-how-to-play-2026.html) are separate from this archive. A fan effort called Sega Channel Revival goes further and rebuilds the menu itself month by month for MiSTer FPGA hardware, Raspberry Pi and Analogue Pocket, hosted on the Internet Archive. An original adaptor, though, is dead weight. The broadcast it listens for stopped in the summer of 1998 and no cable system has carried it since. ## Frequently Asked Questions ### When did Sega Channel launch and when did it shut down? Sega Channel launched on December 12, 1994 as a joint venture between Sega of America, Tele-Communications Inc. and Time Warner. The partners announced its withdrawal in late 1997 for June 30, 1998, and the service was finally discontinued on July 31, 1998, a run of roughly three and a half years. ### How much did Sega Channel cost per month? Sega Channel cost about $15 a month in the United States in the mid 1990s, plus a one-time $25 activation fee that included the cartridge adaptor and the installation visit. Canadian subscribers paid around Can$19 a month after the service launched there in late 1995, and the subscription covered roughly fifty rotating Sega Genesis games. ### Can you play Sega Channel games in 2026? Yes. The original Sega Channel broadcast ended in 1998 and an adaptor plugged into a Genesis today receives nothing, but the Video Game History Foundation recovered 144 Sega Channel ROMs and Gaming Alexandria published them as free downloads on December 15, 2025. The separate Sega Channel Revival project rebuilds the on-screen menu for MiSTer FPGA, Raspberry Pi and Analogue Pocket through the Internet Archive. **Sources:** - Video Game History Foundation, The Secrets of Sega Channel (2025): https://gamehistory.org/segachannel/ - Gaming Alexandria, Sega Channel and Prototype Sega Genesis ROMs (December 2025): https://www.gamingalexandria.com/wp/2025/12/sega-channel-prototype-sega-genesis-roms/ - Telecompaper, Sega, TCI, TWE withdraw Sega Channel service (December 1997): https://www.telecompaper.com/news/sega-tci-twe-withdraw-sega-channel-service--124330 - Electronic Gaming Monthly issue 73, August 1995 (Internet Archive): https://archive.org/details/Electronic_Gaming_Monthly_073_August_1995_U - Wikipedia, Sega Channel: https://en.wikipedia.org/wiki/Sega_Channel --- # Is HyperCard Still Around? Apple Killed It in 2004 URL: https://404memoryfound.com/posts/what-happened-to-hypercard-apple.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-05-18 Updated: 2026-09-18 Topics: Software & Apps, Internet Culture, Then vs Now **Summary:** HyperCard is discontinued. Apple shipped it free with every Macintosh from August 1987, moved it to Claris, stopped updating it in 1998 and withdrew it from sale in March 2004 without ever porting it to Mac OS X. In 2026 the program itself is dead, but more than 3,500 HyperCard stacks run inside a web browser through the Internet Archive's Macintosh emulation. **Key facts:** - Launched: August 11, 1987, Apple Computer, written by Bill Atkinson - Price: $49.95 list, bundled free with every new Macintosh - Last update: 1998, never ported to Mac OS X - Status today: Discontinued March 2004, stacks emulated free at the Internet Archive - What replaced it: The web browser, plus LiveCode and SuperCard as direct descendants ## Why did Apple give away a $49.95 program with every Mac? Bill Atkinson, the Apple programmer behind QuickDraw and MacPaint, showed HyperCard at the MacWorld Expo in Boston on August 11, 1987. It listed at $49.95, and Apple bundled it free on every new Macintosh anyway, which was the condition Atkinson attached to handing the program over. The pitch was that a teacher, a doctor or a 12 year old could assemble working software out of cards, buttons and fields without ever seeing a compiler. Atkinson described the goal in a 2016 interview quoted by Vice's Motherboard as a way to "give programming abilities to people with a passion, rather than trying to give programmers a passion." The scripting language underneath was HyperTalk, which read close to English. A button that jumped to another card needed one line. That single design choice turned millions of Mac owners into part time programmers inside a year. ## What did people actually build out of stacks? A stack was a file full of cards, and cards linked to other cards. The Internet Archive, which now hosts the largest public pile of them, called the format "the Web-Before-The-Web version of hyperlinks and document reading" in a 2020 post about its emulation work. The range was enormous. The Electronic Whole Earth Catalog, published on CD-ROM in 1989, packed more than 9,000 individual pages and 430 megabytes into HyperCard. Schools shipped chemistry tutors on floppies. Museums ran kiosks on it. Mac user groups traded shareware stacks by mail, and the Berkeley Macintosh Users Group collection alone filled hundreds of megabytes. Most of it was amateur work, and that was the point. The Internet Archive's HyperCard Stacks collection holds more than 3,500 examples, from sound samplers and animations to choose-your-own-adventure stories and reference guides built by people who were never paid to write software. ## Was Myst really built on it? Yes. Rand and Robyn Miller built the original Myst at Cyan on a modified copy of HyperCard, described by UploadVR in a 2016 interview with Rand Miller as "a card-based slideshow/development program built into early Macs." Each Age of the island was a stack, and the navigation was HyperTalk scripts behind buttons. HyperCard was black and white, so the brothers had to add color support themselves. Myst reached the Macintosh in 1993 and became the best selling PC game of the decade, which means the top selling title of the 1990s was authored in a tool Apple gave away for free. Cyan's earlier children's titles, including The Manhole, came out of the same workflow. So did prototypes of business software, hospital record systems and airline kiosks that never carried a HyperCard credit on the box. ## Why did Apple let it die? The damage started with ownership. Apple moved HyperCard to its Claris software subsidiary, which tried to sell it as a product instead of bundling it, and Engadget's 2007 retrospective marks that handoff as the moment the momentum broke. Apple took the program back in 1993, and by then the free copy on every Mac was gone. A version that could have saved it existed. HyperCard 3.0 was demonstrated at Apple's developer conference in 1996 with the ability to show stacks inside a web browser. It never shipped. The last real update landed in 1998, and the program was folded in under the QuickTime group. Then came the company's own survival fight. Apple was cutting everything that did not sell hardware in the years it came within weeks of running out of money (https://404memoryfound.com/posts/when-apple-almost-went-bankrupt-1997.html), and the rebuilt company that followed the iMac G3 turnaround (https://404memoryfound.com/posts/how-imac-g3-saved-apple-killed-beige-box.html) never ported HyperCard to Mac OS X. Apple pulled it from sale in March 2004 without a press release. ## How much of HyperCard is still running inside the modern web? More than Apple ever admitted. The original 1989 proposal for the World Wide Web cites HyperCard by name as prior art, and Robert Cailliau, who co-wrote the proposal with Tim Berners-Lee, worked from it directly. What HyperCard lacked was a network. Its links pointed at cards on one hard drive, so the first real hypertext network arrived instead through the Mosaic browser in 1993 (https://404memoryfound.com/posts/what-happened-to-ncsa-mosaic-browser.html). The scripting side spread further. Brendan Eich has credited HyperTalk among the influences on JavaScript, and Ward Cunningham traced the wiki back to a HyperCard stack he built in the late 1980s to link ideas, people and projects. Every page you edit on Wikipedia descends from that stack. Direct descendants are still commercial products. SuperCard and LiveCode both grew out of the HyperCard idea and both still sell, three decades after Apple stopped caring. ## How do you open a HyperCard stack in 2026? The easiest route costs nothing and takes about a minute. The Internet Archive runs emulated Macintosh systems inside a web browser, so the 3,500 stacks in its HyperCard collection boot in a tab with no install, no Apple hardware and no license. The Electronic Whole Earth Catalog runs the same way, though the browser downloads all 430 megabytes of the disc first. For anything not already uploaded there, the standard method is a vintage Macintosh emulator running System 7 with a copy of HyperCard, which is how preservation projects handle stacks that need a full operating system around them. Mac owners with a pile of old floppies usually go this way. The third option is a rebuild. ViperCard, an open source recreation of the 1987 program by Ben Fisher, runs at vipercard.net and lets anyone make new stacks in HyperTalk in a browser. It is not the same software, but it answers the same question HyperCard answered in 1987, which is what a person with an idea and no programming training is supposed to do next. ## Frequently Asked Questions ### Is HyperCard still available from Apple? No. Apple withdrew HyperCard from sale in March 2004 and never ported it to Mac OS X, so it will not run natively on any Mac sold since. Its final significant update was released in 1998. ### What games were made with HyperCard? The original Myst, released for the Macintosh in 1993, was built on a modified version of HyperCard by Rand and Robyn Miller at Cyan, along with their earlier title The Manhole. Thousands of smaller adventure and puzzle stacks were traded through Mac user groups in the same era. ### Can you still run old HyperCard stacks today? Yes. The Internet Archive emulates classic Macintosh systems in a web browser and hosts more than 3,500 HyperCard stacks that run without any download or Apple hardware. Stacks that are not in that collection can be opened in a vintage Mac emulator running System 7. **Sources:** - Internet Archive Blogs, The Whole Earth on CD-ROM, in HyperCard, in your browser (2020): https://blog.archive.org/2020/07/08/the-whole-earth-on-cd-rom-in-hypercard-in-your-browser/ - Engadget, HyperCard's history (2007): https://www.engadget.com/2007-07-24-hypercards-history.html - Vice Motherboard, Someone Recreated HyperCard, Apple's 80s Programming Tool: https://vice.com/en/article/59jg73/vipercard-recreated-hypercard-apple-80s-acid - UploadVR, Cyan's Rand Miller on Myst and Obduction (2016): https://www.uploadvr.com/cyan-worlds-rand-miller-myst-obduction-interview/ - Wikipedia, HyperCard: https://en.wikipedia.org/wiki/HyperCard --- # Why the Sega 32X Failed: The $159.99 Add-On Flop URL: https://404memoryfound.com/posts/what-happened-to-sega-32x-add-on.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-13 Updated: 2026-09-10 Topics: Gaming, Hardware, Business Blunders **Summary:** The Sega 32X was a $159.99 add-on that turned a Genesis into a 32-bit machine, released in North America on November 21, 1994. It sold an estimated 665,000 units, got a library of 40 games, and was discontinued in 1996 after Sega launched the Saturn six months into its life. Sega no longer supports it, and a used 32X runs about $200 in 2026. **Key facts:** - Launched: November 21, 1994 in North America, $159.99 - Units sold: About 665,000 by the end of 1994; 800,000 produced - Game library: 40 games, six of them needing a Sega CD too - Status: Discontinued in 1996, last units cleared at $19.95 - Price today: About $199.50 loose, $299 boxed (PriceCharting, September 2026) ## What was the Sega 32X, and what did it cost in 1994? The 32X was a black plastic wedge that sat in the cartridge slot of a Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html) Genesis and added two 32-bit SH-2 processors running at 23 MHz, plus a video chip that could put 32,768 colors on screen. It reached North American stores on November 21, 1994 at $159.99, with no game in the box. Europe got it in November 1994 at 169.99 pounds, and Japan followed on December 3, 1994 at 16,800 yen. Brazil got it in March 1995. Sega had shown the hardware at the Consumer Electronics Show in June 1994 and pitched it as the cheap door into 32-bit gaming. Inside the company it was Project Mars, and it began life as a standalone console before Sega of America research chief Joe Miller and his team argued it should bolt onto the Genesis instead. Setting one up was a project. The 32X needed its own power supply, a connection cable to the Genesis, and an extra conversion cable if you owned the first Genesis model. A Sega CD (https://404memoryfound.com/posts/what-happened-to-sega-cd-add-on.html) owner who wanted everything ended up with three boxes, two power bricks and a tangle of cables behind the TV. ## How many 32X units did Sega actually sell? Retailers wanted it at first. More than 1,000,000 units were ordered, Sega shipped roughly 600,000 by January 1995, and close to 500,000 sold in the United States by Christmas 1994. Then it stopped. The most widely cited estimate puts 32X sales at about 665,000 units by the end of 1994. Sega produced 800,000 and moved the leftovers at steep discounts. A full year later the number had barely moved. Asked by Next Generation in January 1996 how many he had sold, Sega of America president Tom Kalinske answered: "We'll be at around 700,000 by December. It's not bad." For scale, Kalinske told the same magazine in March 1995 that Sega and Nintendo together expected to sell through 9.5 million 16-bit machines in 1994. The 32X was the upgrade path Sega offered those owners, and by his own count it never cleared 700,000. ## Who pushed the 32X, Sega of America or Sega of Japan? Both, at different moments, which is the whole problem. At the Winter Consumer Electronics Show in January 1994, Sega president Hayao Nakayama called Joe Miller's Las Vegas hotel suite and pressed for a fast answer to the Atari (https://404memoryfound.com/posts/what-happened-to-atari-jaguar-64-bit-console.html) Jaguar. What the American team came back with was the 32X. In public, Kalinske sold it on price. "No matter how great Saturn is, or PlayStation is, or Ultra 64 is, we will outsell them by an enormous amount with 32X," he told Next Generation in March 1995, predicting the add-on would stay big for three or four years. In private the plan was smaller. "It really was designed to be an interim piece and to prolong the life of the 16-bit platform," Kalinske told the site Sega-16 in 2006, adding the lesson he took from it: "Don't ever expect an add-on device to be as important as a true, new platform." Nakayama's view won anyway. Sega of Japan cut support for the Genesis family to concentrate on the Saturn, so the add-on spent its short life competing with its own company. Kalinske's position in that 2006 interview was that he would rather have waited: "I felt that we were rushing Saturn. We didn't have the software right." ## Which of the 40 32X games were actually good? The library finished at 40 titles, six of which also required a Sega CD. Sega had promised 12 games at launch and another 50 during 1995 from outside publishers. Six were there on day one. The ones people still name: - Doom , which Kalinske named in 1995 as his favorite game on the system because it shipped with its sound intact, though later critics knocked the port for missing levels. - Virtua Racing Deluxe , the launch title reviewers treated as the honest showcase for the extra processors. - Star Wars Arcade , another launch cartridge and a straight arcade port. - Knuckles' Chaotix , the 32X-exclusive Sonic spinoff that GamesRadar later called the best game on the machine. - Kolibri , a shooter starring a hummingbird, and the closest the system came to a cult classic. - Virtua Fighter , which arrived in late 1995, after the Saturn had already shipped its own version. The other end of the list was worse than the good end was good. Launch fighter Cosmic Carnage played so poorly that reporters made jokes about it, and a large share of the catalog was Genesis ports with more colors, including NFL Quarterback Club and World Series Baseball '95. ## How did the Saturn's early launch finish the 32X? On May 11, 1995, under six months after the add-on went on sale, Sega put the Sega Saturn (https://404memoryfound.com/posts/what-happened-to-sega-saturn-console.html) into 1,800 stores across the United States and Canada, four months ahead of its own schedule. It cost $399 on its own, or $449 with Virtua Fighter, with five other titles available. Next Generation's summary of what that meant for the older machine, written at the end of 1995, was four words long: "32X? Forget it." The price collapsed. In September 1995 the 32X dropped to $99. That December, Next Generation gave the system one star out of five, and when the magazine asked Kalinske whether people who bought a 32X a year earlier felt Sega had delivered on its promises, his answer was a single word: "No." Sega discontinued the 32X in 1996 and cleared the remaining stock at $19.95, about 12 percent of the launch price. The Saturn did not save the company either, and by 2001 Sega was out of the hardware business, a slide the Dreamcast (https://404memoryfound.com/posts/why-sega-stopped-making-consoles-dreamcast.html) ended rather than caused. ## What does a Sega 32X cost second-hand in 2026? More than it did new. PriceCharting, which tracks completed sales, listed the 32X unit at $199.50 loose and $299.00 complete in box when checked in September 2026. A sealed unit was tracking above $600. That is roughly ten times the $19.95 clearance price of 1996, and it buys you a peripheral, not a console. A working 32X still needs a Genesis, the link cable, the A/V cable and its own power supply, and the cables go missing more often than the hardware does. Where to find one today: Stone Age Gamer keeps the boring but necessary parts in stock, including the 32X link connector cable at $11.99 and the A/V cable that feeds a Genesis Model 2, Sega CD and 32X at $7.99, in its Sega 32X section (https://stoneagegamer.com/sega/32x/). Sort out the cables before you chase the unit itself. ## Frequently Asked Questions ### How much did the Sega 32X cost when it launched? The Sega 32X launched in North America on November 21, 1994 at $159.99, with no pack-in game. It fell to $99 in September 1995, and Sega cleared the last units at $19.95 before discontinuing the add-on in 1996. ### How many games were released for the Sega 32X? The Sega 32X library ended at 40 games, six of which needed a Sega CD attached as well. Sega had promised 12 titles for the November 1994 launch and delivered six, among them Doom, Virtua Racing Deluxe and Star Wars Arcade. ### Is the Sega 32X better than the Sega Saturn? No. The Saturn, which reached US stores on May 11, 1995 at $399, was a full 32-bit console, while the 32X was a $159.99 add-on that used slower versions of the same SH-2 chips. Sega itself said the 32X could not handle a port of Virtua Fighter 2, which ran on the Saturn. **Sources:** - Next Generation, Issue 3, March 1995: Tom Kalinske interview: https://archive.org/details/NEXT_Generation_03 - Next Generation, Issue 13, January 1996: Kalinske interview and the 1995 year in review: https://archive.org/details/NEXT_Generation_13 - Sega-16: Interview with Tom Kalinske, July 11, 2006: https://www.sega-16.com/2006/07/interview-tom-kalinske/ - Wikipedia: 32X: https://en.wikipedia.org/wiki/32X - PriceCharting: Sega 32X unit prices: https://www.pricecharting.com/game/sega-32x/sega-32x-unit --- # The First Banner Ad of 1994: What AT&T Really Paid URL: https://404memoryfound.com/posts/first-banner-ad-hotwired-1994-history.html Author: Dana Reyes (404 Memory Found) Published: 2026-05-13 Updated: 2026-09-18 Topics: Internet Culture, Then vs Now **Summary:** The first banner ad appeared on HotWired on October 27, 1994, one of about a dozen that launched with the site, and AT&T paid roughly $30,000 for a 12 week sponsorship. Reports put its click-through rate near 44 percent, a level no display ad has touched since. Banner advertising grew into the $81.6 billion US display business that IAB and PwC counted for 2025, where the average click rate now sits under half a percent. **Key facts:** - Launched: October 27, 1994, on HotWired - Advertiser: AT&T, using its "You Will" tagline - Price: About $30,000 for a 12 week sponsorship - Click rate then and now: Reported 44 percent in 1994, 0.46 percent on the Google Display Network in 2025 - Status today: US display advertising worth $81.6 billion in 2025 (IAB and PwC) ## Who actually built the AT&T banner? The banner came out of Modem Media, the Connecticut agency AT&T used for its early interactive work. Creative director Joe McCambley led it, and Craig Kanarick, who later co-founded Razorfish, handled the build. The credits on thefirstbannerad.com, the project site the team published on October 27, 2014, also list Otto Timmons of TANGENT Design on the creative team, Brent Hood and Dave Evant on production, and Bill Clausen as the AT&T client partner. Nobody on the job had made an internet ad before, because there were none to copy. They were designing for Netscape Navigator (https://404memoryfound.com/posts/how-netscape-lost-the-browser-war.html) and Mosaic on small screens, with no color standard, no agreed size, and no way to know where on the page the thing would land. That same project site is blunt about the mythology it helped build. "No, it wasn't the 'first' as there were a handful of other ads that ran on various sections of hotwired.com," the 2014 page says. ## What did the banner say, and where did the click land? The copy was one line, "Have you ever clicked your mouse right HERE?", with an arrow pointing to the answer, "YOU WILL." It borrowed the tagline from AT&T's "You Will" television campaign, which was already promising videoconferencing and tablets on network TV. Clicking it did not open a product page or a form. It opened a hand built directory of virtual museum tours, including the Louvre, the Andy Warhol Museum and the Library of Congress, as the Internet History Podcast oral history recorded in 2014. Accounts of the size disagree. The team designed to roughly 468 by 60 pixels, Digiday's 2017 oral history describes the rendered banner as 476 by 56, and the Internet History Podcast puts it at 546 by 56 once browsers added their own borders. What mattered was the deal underneath. AT&T was buying a placement, not clicks, and HotWired made no promises about traffic. ## Was the 44 percent click-through rate real? Close to it, yes, though nobody audited the number the way a 2026 campaign is audited. HotWired sold the AT&T placement as a 12 week sponsorship for about $30,000, roughly $10,000 a month, priced against print magazine page rates, and the banner stayed live for around ten months. The people in the room remembered a range rather than one figure. Rick Boyce, who ran ad sales at HotWired, described "click-through rates as high as 50%, maybe even higher" in the Internet History Podcast oral history in 2014. The 44 percent figure is the one that stuck in the retrospectives. Andrew Anker, HotWired's first chief executive, gave the sharper benchmark for the era in the same 2014 oral history: "if someone's click-through rates fell below 2%, we knew the banner wasn't good." A 2 percent floor in 1995 is more than four times what an average display ad returns in 2026. ## How fast did banners spread across the web? Inside a year, everywhere. CNET ran its first banner in June 1995, Lycos (https://404memoryfound.com/posts/what-happened-to-lycos-search-engine.html) and Excite (https://404memoryfound.com/posts/what-happened-to-excite-search-engine-google.html) launched advertising through the spring and summer of that year, and Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) started selling ads in November 1995, according to the 2014 Internet History Podcast oral history. About a dozen advertisers had shared HotWired's launch day on October 27, 1994, among them MCI, Volvo, Club Med, Sprint, IBM and Zima. By 1996 the banner was simply how a website paid for itself. The people who made it did not see that coming. "I didn't think it was the beginning of a giant industry," Jonathan Nelson, co-founder of the agency Organic, told Digiday in 2017. The fight inside Wired had been louder than the launch. "People told us if you put ads online, the internet would throw up on us," Wired co-founder Louis Rossetto told Digiday in 2017. ## What is a banner click worth in 2026? Display advertising did not die. IAB and PwC put US display revenue at $81.6 billion for full year 2025, up 9.8 percent and worth 27.7 percent of all digital media spend, inside a $294.6 billion internet advertising market, in the report published in April 2026. What changed is what the money buys. "This revenue growth reflects a market that has reoriented around performance channels," IAB chief executive David Cohen said in that 2026 release. Search took $114.2 billion of the 2025 total and video took $78 billion. The click itself is now a rounding error. Focus Digital's December 2025 benchmark study put the average Google Display Network click-through rate at 0.46 percent, ranging from 1.08 percent in real estate to 0.22 percent in B2B services. At 0.46 percent, matching the 44 percent of 1994 would take roughly 96 times the impressions. ## Why the 1994 number can never come back That rate measured novelty, not intent. HotWired's readers in October 1994 were people who had gone to real trouble to find a website at all, and a rectangle that answered a mouse click was itself the thing being demonstrated. Three forces then ground the number down. Volume, as pages filled with banners and readers learned to skip the top 60 pixels. Measurement, as advertisers moved from paying for placements to paying for clicks and conversions. And the auction, which pulled budgets toward search results and later toward feeds, where the user had already said what they wanted. Ad blocking finished the job on a large share of desktop screens, and the dot com crash of 2000 and 2001 wiped out the companies that had bought banners hardest, Pets.com (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html) among them. The AT&T banner still works as a demonstration. It just cannot be run again. ## Frequently Asked Questions ### What was the first banner ad on the internet? The first banner ads ran on HotWired, Wired magazine's website, on October 27, 1994. The best remembered is AT&T's, which read "Have you ever clicked your mouse right HERE?" and led to a page of virtual museum tours. About a dozen advertisers launched alongside it that day. ### How much did AT&T pay for the 1994 HotWired banner ad? AT&T bought a 12 week sponsorship on HotWired for about $30,000, roughly $10,000 a month, and the banner stayed live for around ten months into 1995. The price was set against print magazine page rates, because no online rate card existed in 1994. ### What is the average banner ad click-through rate today? Focus Digital's December 2025 study put the average Google Display Network click-through rate at 0.46 percent, against the roughly 44 percent reported for the 1994 AT&T banner on HotWired. US display advertising was still worth $81.6 billion in 2025, according to IAB and PwC. **Sources:** - Internet History Podcast, An Oral History of the Web's First Banner Ads (2014): https://www.internethistorypodcast.com/banner/ - Digiday, An oral history of the first banner ad (2017): https://digiday.com/media/history-of-the-banner-ad/ - thefirstbannerad.com, the AT&T You Will banner project site (2014): http://thefirstbannerad.com/story.html - IAB and PwC, Internet Advertising Revenue Report, Full Year 2025 (April 2026): https://www.iab.com/news/digital-ad-revenue-climbs-to-nearly-300b-as-iab-celebrates-30-year-anniversary/ - Focus Digital, Average CTR for Display Ads benchmarks (December 2025): https://focus-digital.co/average-display-ad-ctr-2025-benchmarks/ --- # The Game Genie Beat Nintendo. Who Owns It Now? URL: https://404memoryfound.com/posts/what-happened-to-game-genie-galoob-cheat-device.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-12 Updated: 2026-09-18 Topics: Gaming, Hardware **Summary:** The Game Genie is discontinued and has been since the mid 1990s, when cartridge consoles gave way to disc systems. Galoob, the toy company that sold it in the United States, was bought by Hasbro in 1998, so the name sits inside Hasbro today. What survives is the 1992 Ninth Circuit ruling that made home game modification legal, and a second-hand NES unit that costs about $22. **Key facts:** - Launched: 1990, designed by Codemasters, sold in the US by Galoob - US price: About $60 at launch in 1990 - Units sold: About 2.5 million NES units through 1993 - Court outcome: Galoob won in 1992 and collected Nintendo's $15 million bond - Status today: Discontinued; Galoob owned by Hasbro since 1998; loose units about $22 ## How did a plastic sleeve rewrite a NES game? The Game Genie was an L-shaped plastic pass-through cartridge. A game cartridge slotted into the top of it, the whole stack went into the console, and a code screen loaded before the game did. The player typed in six or eight character codes and the game booted with more lives, more speed, or a level select the developers never shipped. Nothing on the cartridge changed. The device sat on the data line between the cartridge and the processor and swapped single bytes as the chip asked for them. The Ninth Circuit put it plainly in 1992: "The Game Genie does not alter the data that is stored in the game cartridge. Its effects are temporary." Codemasters, a British developer, built the hardware. Galoob, a toy company founded in 1957 and based in South San Francisco, took the United States rights and put the device in American toy aisles in 1990 at about $60, a figure Mental Floss cites for the launch. ## Why did Nintendo take a cheat device to court in 1990? Nintendo of America ran the NES (https://404memoryfound.com/posts/how-nintendo-nes-saved-video-games-1983-crash.html) as a closed system. It licensed every cartridge, capped what publishers could ship, and used a lockout chip to keep unapproved hardware out. A $60 sleeve that let an eleven year old skip the hard part was a direct challenge to that control, and Nintendo treated it as one. The company argued that a Game Genie session produced a derivative work based on its copyrighted games, and that selling the device made Galoob a contributory infringer. Howard Lincoln, then Nintendo's senior vice president, said in 1990 that the company could not "stand by while the essence of our business" was put at risk by the product, in remarks Mental Floss later quoted. Galoob and Codemasters had already gone to court first, asking a federal judge to declare that the device infringed nothing. Nintendo won a preliminary injunction anyway, and the Game Genie stayed off United States shelves for roughly a year while the case ran. ## Was the Game Genie a derivative work? That was the whole case. A copyright owner controls derivative versions of its work, and Nintendo said every altered game was one. The Ninth Circuit disagreed on May 21, 1992, in Lewis Galoob Toys, Inc. v. Nintendo of America, Inc., 964 F.2d 965. A derivative work has to incorporate the original in some concrete or permanent form, and a Game Genie display incorporated nothing. It intercepted bytes in transit and left the cartridge untouched, so pulling the device out left the game exactly as Nintendo shipped it. The panel added that the device "is useless by itself, it can only enhance, and cannot duplicate or recast, a Nintendo game's output." Even if an altered display had counted as a derivative work, the court held that a child changing a game at home for no commercial purpose was fair use, and that Nintendo had shown no market it was losing. The 1992 opinion also carried a line hardware companies have leaned on ever since: "Technology often advances by improvement rather than replacement." ## What happened to Nintendo's $15 million bond? Preliminary injunctions are not free. To stop the sales, Nintendo had to post a bond that would cover Galoob's losses if the injunction turned out to be wrong. It was raised over the course of the fight to $15 million. The injunction did turn out to be wrong. After Galoob won at trial, the district court ran a lost sales hearing and found the year off the shelves had cost the company at least $15,138,048 in profits, more than the bond covered. It awarded Galoob the entire $15 million, plus interest and the cost of restarting the line. Nintendo appealed that too and lost again on February 17, 1994, in Nintendo of America, Inc. v. Lewis Galoob Toys, Inc., 16 F.3d 1032. The rule the panel applied, as the published summary of the 1994 decision states, is that "a party wrongfully enjoined is presumptively entitled to recover damages up to the bond amount." Mental Floss puts the money Nintendo handed over in 1994 at $16.1 million. Few companies have paid eight figures for the privilege of winning a preliminary injunction. ## Why did Galoob stop making them? The product did not fail. Galoob sold about 2.5 million of the NES version through 1993 by Mental Floss's count, and millions more of the Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html) and Super Nintendo versions after that. Versions for the Sega Genesis (https://404memoryfound.com/posts/sega-genesis-games-how-to-play-2026.html) and the Game Gear followed. What ended the line was the hardware generation, not the courtroom. Every Game Genie was built around a specific cartridge slot, and by the middle of the 1990s the money had moved to the PlayStation and the Sega (https://404memoryfound.com/posts/what-happened-to-sega-saturn-console.html) Saturn, which read discs. There was no cartridge for the device to sit in front of. Action Replay and GameShark carried the cheat market onto disc consoles with memory card tricks and boot discs, and the Game Genie name stopped appearing on new hardware. Galoob lasted a few more years on its own. Hasbro bought the company in 1998 for $220 million, and Galoob has been a Hasbro brand name since, turning up on die-cast collectibles rather than on anything that plugs into a console. ## What does a Game Genie cost in 2026? A loose NES Game Genie ran about $22 on PriceCharting in September 2026. A complete boxed copy sat near $59 and a sealed one near $92. Graded sealed examples are their own market, with one VGA 85 copy recorded at $813. The Super Nintendo, Game Boy and Sega versions trade in the same band, and none of them are scarce. Galoob made millions, and most of those units are still in American basements. Two things to check before buying. A manual alone runs roughly $8 to $20 on PriceCharting, so a listing that includes the code book is priced higher for a reason. And the device needs a clean card edge connection on both ends, which means a healthy cartridge slot in the console as well as in the Game Genie. Where to find one today: a retro specialist sells tested accessories with a return policy, which is worth the premium over an untested auction lot at these prices. Stone Age Gamer (https://stoneagegamer.com/nintendo/nes/accessories/) stocks NES accessories. ## Frequently Asked Questions ### Who owns the Game Genie name now? Hasbro. Hasbro bought Lewis Galoob Toys, the company that sold the Game Genie in the United States, in 1998 for $220 million, and Galoob has been a Hasbro brand name ever since. No Game Genie hardware has been built in that lineage since the 1990s, and later cheat products sold under a similar name were not the Galoob and Codemasters device. ### Is it legal to use a Game Genie in the United States? Yes. In Lewis Galoob Toys, Inc. v. Nintendo of America, Inc., 964 F.2d 965, the Ninth Circuit held on May 21, 1992 that changing a game's display at home with a Game Genie created no derivative work and was fair use. That 1992 ruling still stands and is cited whenever a console maker objects to third-party hardware. ### How much is an NES Game Genie worth in 2026? An NES Game Genie was worth about $22 loose on PriceCharting in September 2026, about $59 complete in its box, and about $92 sealed. Prices have stayed flat because Galoob sold roughly 2.5 million NES units through 1993, so supply is deep. **Sources:** - Lewis Galoob Toys, Inc. v. Nintendo of America, Inc., 964 F.2d 965 (9th Cir. 1992): https://law.resource.org/pub/us/case/reporter/F2/964/964.F2d.965.91-16205.html - Nintendo of America, Inc. v. Lewis Galoob Toys, Inc., 16 F.3d 1032 (9th Cir. 1994), case brief: https://www.studicata.com/case-briefs/case/nintendo-of-america-v-lewis-galoob-toys - Mental Floss, When the Game Genie Made Beating Nintendo Games Easy: https://www.mentalfloss.com/article/655393/when-game-genie-made-beating-nintendo-games-easy - Galoob, company history (Wikipedia): https://en.wikipedia.org/wiki/Galoob - PriceCharting, Game Genie (NES) price guide: https://www.pricecharting.com/game/nes/game-genie --- # What Happened to Geocaching, the GPS Treasure Hunt That Started With a Bucket and a Can of Beans URL: https://404memoryfound.com/posts/what-happened-to-geocaching-gps-treasure-hunt.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-12 Topics: Hardware, Internet Culture At midnight on May 1, 2000, the United States government did something that should have made the front page of every newspaper and instead made almost no news at all. It turned off a setting on its global positioning satellites called Selective Availability. In the time it took the satellites to receive the order and apply it, the typical accuracy of every civilian GPS receiver on Earth jumped from about 100 meters to under 20 meters, roughly a ten-fold improvement. A device that previously could only tell you, vaguely, what neighborhood you were standing in could now point to a specific tree in that neighborhood. The next day, a computer consultant in Beavercreek, Oregon named Dave Ulmer drove to a wooded clearing off the side of the road, dropped a black plastic bucket into a depression in the ground, put a logbook and some small trinkets inside it, covered it with branches, and posted the latitude and longitude coordinates to an internet newsgroup with the subject line "GPS Stash." Within three days, two people had driven out to Oregon to find it. Within ten years, what Ulmer had invented would have over five million participants and would become the first true mass-market activity to require a satellite signal, an internet connection, and a willingness to walk into the woods looking for things. The activity was called geocaching, eventually. And the story of how it grew, how it survived the smartphone, and what it tells us about the early consumer internet is one of the more under-told case studies in user-generated platform design. ## The policy change that started everything To understand geocaching, you have to understand that for most of the 1990s, the United States military deliberately broke its own technology for non-military users. GPS, the Global Positioning System, was built by the US Department of Defense. By the early 1990s the satellite constellation was operational and civilians could buy receivers that talked to it. The catch was that the government did not want enemy militaries piggybacking on American satellites to guide weapons or troops. So the Pentagon introduced something called Selective Availability, which intentionally degraded the civilian-accessible portion of the signal. The military got the precise version. Everyone else got noise added to their location data, deliberately, by design. In practice, civilian GPS receivers had a typical accuracy of about 100 meters horizontally. That is fine for an airplane finding an airport. It is useless for a hiker trying to navigate a trail or a driver trying to find a specific address. The civilian market for GPS had been a slow burn through the 1990s for exactly this reason. Nobody could justify paying for a device that could only tell them roughly where on a football field they were. On May 1, 2000, President Bill Clinton issued a statement announcing that Selective Availability would be discontinued, effective immediately. The Pentagon, after years of internal debate, had concluded that the technology to degrade civilian signals could be replicated cheaply and that the strategic value was now lower than the economic value of an unleashed civilian GPS market. The downgrade went away overnight. Civilian accuracy went from useless to genuinely useful. "The decision to discontinue SA is the latest measure in an ongoing effort to make GPS more responsive to civil and commercial users worldwide," the official White House statement read. The word "geocaching" did not appear in the press release. The activity did not yet exist. The press coverage was modest. Most consumer media did not run the story at all. The people who did notice, immediately, were the small but obsessive community of GPS hobbyists who had been hanging out on Usenet newsgroups like sci.geo.satellite-nav. ## Dave Ulmer hides a bucket in the woods Dave Ulmer was one of those hobbyists. He had been running test routes in Oregon to measure the new accuracy of his GPS receiver, and on May 3, 2000, he decided to test the social possibilities of the new system. He stashed a five-gallon black plastic bucket near a forest road in Beavercreek. Into the bucket he put a logbook and pencil, two CD-ROMs of DeLorme Topo USA mapping software, a George of the Jungle VHS tape, a book, four dollars, a cassette recorder, a slingshot, and a can of beans. He posted the coordinates to the newsgroup along with what is now treated as the founding principle of the whole activity: "Take some stuff, leave some stuff. Record it all in the logbook." Within 24 hours, a Vancouver, Washington reader named Mike Teague had driven to Beavercreek, found the bucket, signed the log, and traded items. Teague then started a website to track other people doing the same thing. Hobbyists in California, Illinois, and Kansas hid their own caches within a week. Teague's site listed them all. For about four months, that was the entire community. A few dozen people, mostly engineers and outdoor types, scattered across the United States, hiding plastic boxes in the woods and trading the coordinates on a single ad-hoc website run by one guy in his spare time. That is when Jeremy Irish entered the story. ## How a Seattle web developer built the platform Jeremy Irish was a 28-year-old web developer at a Seattle interactive agency called Sunrise Identity. In July 2000, a friend forwarded him a link to Mike Teague's GPS Stash Hunt website. Irish read it, immediately bought a Garmin eTrex, drove out to find a cache near his Seattle home, and got hooked in the way that people get hooked on things that combine a treasure hunt, a hike, and a community of strangers who are all in on the same secret. Irish offered to take over the website. Teague, who was running it alone and getting overwhelmed by traffic, agreed. In September 2000, Irish registered the domain Geocaching.com and launched the rebranded site on September 2, 2000. At launch, there were 75 known geocaches in the world. This was a brand decision that mattered. "Stash hunt" sounded faintly illegal. "Geocaching" sounded scientific. Irish later said the name had to be searchable and had to not freak out parents. Irish co-founded a company called Groundspeak with Bryan Roth and Elias Alvord to operate the site. There was no revenue model for the first few years. Irish kept his day job. The site's initial server costs were funded by selling 144 geocaching t-shirts to early community members. Listings grew through word of mouth on outdoor forums and through coverage in early 2000s tech magazines that were hungry for stories about novel uses of the new accurate GPS. ## The device that made it work The unsung hero of geocaching's first decade was the Garmin eTrex. Garmin had been making professional GPS units for marine and aviation use since 1989 (the company was founded in Lenexa, Kansas in 1989 by Gary Burrell and Min Kao (the company name is a portmanteau of their first names) and later moved to nearby Olathe in 1996). In late 1999 and into 2000 they launched the eTrex, a 5.3-ounce handheld receiver priced at about $120 to $150 depending on the configuration. The eTrex was the right device at the right moment. It was cheap enough to be impulse-bought by hobbyists. It ran for 22 hours on two AA batteries. It was waterproof enough to survive Oregon rain. It had a high-contrast LCD that worked in direct sunlight. And it could store hundreds of waypoints, which was exactly what a geocacher needed to do. The dependence on the eTrex is the kind of detail that gets missed in retellings of geocaching's history. Without a sub-$200 handheld GPS that worked reliably in the woods, geocaching would have stayed inside the hobbyist niche forever. The platform side of the activity, hosted at Geocaching.com, was completely conventional 2000-era web infrastructure. The hardware side was the actual enabling technology. Geocaching grew at the speed of the eTrex sales curve, and the eTrex sales curve grew at the speed of word-of-mouth among the people Geocaching.com brought into the hobby. The two systems reinforced each other. Garmin did not sponsor geocaching, did not advertise to geocachers specifically, and never publicly tied its sales numbers to the activity. But the company's revenue from consumer-grade handhelds went from a rounding error in 1999 to a meaningful product line by 2003, and geocachers were a measurable chunk of that growth. ## The platform that grew under it Geocaching.com is interesting as a piece of platform design because it does almost nothing on its own. The platform does not generate caches. The platform does not place caches in the world. The platform does not verify caches exist or that they have not been muggled (geocaching slang for "discovered by a non-participant who removes or destroys it"). All of that work is done by users. What the platform does is run the database, the search interface, the user accounts, the maps, and the logging system. A cache hider creates a listing with coordinates and a description. A reviewer (a volunteer) approves it. The cache appears on the public map. Finders log their visits, upload photos, trade items, and mark whether the cache was found, not found, or needs maintenance. The cache owner is notified of every log and is expected to manage the cache's lifecycle in the real world. This is a remarkably hands-off design for a 2000-era website. Most internet hobby communities of that era had centralized moderators and a lot of top-down editorial work. Geocaching.com pushed almost all of that responsibility onto the users themselves. Volunteer reviewers in each region handled cache approvals. Cache owners handled their own maintenance. Finders handled their own logs and disputes. The platform mostly just held the data and ran the search. This is essentially the model that Wikipedia would popularize a few months later (Wikipedia launched in January 2001) and that platforms like Yelp, Reddit, and TripAdvisor would build entire businesses on. Geocaching.com was doing community-curated, user-generated, peer-moderated content in 2000, with a paid subscription tier introduced in 2001, and the whole thing was running out of a Seattle office that fit under a dozen employees for years. ## The money question For the first year, Groundspeak operated as a hobby. By 2001, server costs and traffic were getting unsustainable. Irish and his co-founders introduced Premium Membership, a $30-per-year subscription that gave paying users access to additional features (advanced search, instant notification of new caches in a region, certain "members-only" cache types). The basic experience stayed free. This freemium structure was unusual in 2001. Most consumer websites of the era were either ad-supported portals or paid-only services. Charging a fraction of users for premium features while keeping the core product free was the kind of model that would become standard in mobile apps a decade later. Groundspeak got there early, partly because there was no good way to monetize a non-commercial outdoor activity with banner ads. Groundspeak never raised venture capital. Irish said publicly, multiple times, that the team was not interested in growth-at-all-costs and preferred to remain private and independent. The company stayed in Seattle, in a brick building in Fremont, with maybe 80 employees at its peak. Geocaching.com became, for a long stretch, the largest content-driven hobby community on the internet that nobody in venture capital had ever paid for a piece of. ## The smartphone problem The transition to smartphones around 2008 to 2010 was, on paper, the moment geocaching should have collapsed. Smartphones killed dedicated MP3 players, dedicated cameras, dedicated GPS units, and dedicated handheld games. The Garmin eTrex was exactly the kind of product whose entire job was now a free app on the device in your pocket. Groundspeak's response was the official Geocaching app, released for iPhone in 2008 and Android in 2010. The app cost $10, which by the standards of 2010 app pricing was on the high end. It worked well. Geocaching.com did not lose its community. It just migrated. What is interesting is that smartphone GPS killed dedicated GPS hardware for almost every market segment except the one geocaching actually used. Smartphones are excellent for navigation when you have a cell signal and a battery and good weather. They are mediocre to bad for the conditions that actual geocaching often requires: deep woods, no cell signal, all-day battery, rain, freezing temperatures, dropped on rocks. The Garmin eTrex line survived the smartphone era specifically because of the segment of geocachers who needed the durability. The result was a strange hybrid. New geocachers came in through the smartphone app and treated it as an urban scavenger hunt. Experienced geocachers kept their handheld receivers for the harder, remoter caches. The platform supported both modes simultaneously, and the database kept growing. ## What the data looks like today As of the current Geocaching.com public statistics, there are over three million active geocaches hidden in 191 countries. There are more than 200,000 active caches in the United States. The site has logged over a billion individual "found it" logs across all caches in its history. Cache types have expanded well beyond the original plastic bucket model: traditional, multi-cache, mystery, EarthCache, Letterbox Hybrid, virtual, Wherigo, and a few more. The activity has changed substantially since 2000. There are now Adventure Labs, which use GPS to send participants to specific spots without requiring a physical container. There is a thriving secondary economy in Geocoins and Trackable items, which are coins or tags with unique numbers that travel from cache to cache and are logged as they move. There are official events and competitions. There is a global ranking system. There is even an annual film festival of geocaching videos. What has not changed is the basic loop. Somebody hides something. Somebody else uses coordinates to find it. Both of them sign a logbook. The platform records that it happened. The trust required between strangers, for the whole thing to work, has not been re-engineered in 25 years. It is the same trust that Dave Ulmer asked for when he posted the original GPS Stash coordinates to a Usenet newsgroup. ## Why geocaching matters as a piece of internet history Geocaching is small compared to the platforms that came after it. Three million active caches and a few million active participants is not big-tech scale. But the activity matters out of proportion to its size, for three reasons. First, it is one of the first popular consumer activities that required both the open internet and the open GPS system to exist simultaneously. The Department of Defense unlocked the satellites on May 2, 2000. Dave Ulmer published coordinates to Usenet on May 3. Mike Teague found the bucket within a day. The lag time between a major government policy change and a viable consumer internet activity built on top of it was effectively zero. That speed is not normal. It happened because the right hobbyist community was already there, waiting, with the receivers in hand. Second, Geocaching.com was an early proof that a small independent platform could build and sustain a global community without venture funding, advertising revenue, or a public exit. Groundspeak has been operating Geocaching.com for over 25 years on a freemium subscription model. Most platforms of that era either got acquired, pivoted, raised mega-rounds, or died. Geocaching.com just kept running. Third, geocaching demonstrated that user-generated, peer-moderated platforms could work for activities that have a physical, real-world component. Wikipedia proved it for text. Geocaching proved it for objects in the actual world. That model is now everywhere. AirBnB is essentially "Geocaching.com for spare bedrooms with money attached." OpenStreetMap is essentially "Geocaching.com for roads and buildings." iNaturalist is essentially "Geocaching.com for living things." The pattern works because Dave Ulmer's original loop works: somebody puts something in the world, somebody else finds it, both events get logged, the platform holds the data. The activity that emerged from a single bucket in the Oregon woods is now a model that runs an enormous fraction of the modern consumer internet. Whether the people who built those bigger platforms knew they were copying from a 2000-era hobbyist site is a different question. The pattern was there. Geocaching demonstrated it first. ## Frequently Asked Questions ### When did geocaching officially start? Dave Ulmer hid the first cache, then called a "GPS Stash," on May 3, 2000, near Beavercreek, Oregon. He posted the coordinates to the sci.geo.satellite-nav Usenet newsgroup that day. The activity was renamed "geocaching" in September 2000 when Jeremy Irish registered the domain Geocaching.com. ### What was Selective Availability and why did it matter? Selective Availability was a feature of the United States military's GPS satellites that intentionally degraded the accuracy of civilian signals to about 100 meters. President Bill Clinton announced on May 1, 2000 that Selective Availability would be discontinued at midnight that day, immediately improving civilian GPS accuracy by roughly a factor of ten. Geocaching emerged within 24 hours of the change because hobbyists had been waiting for civilian-grade GPS to become precise enough to make the activity possible. ### Who owns Geocaching.com? Geocaching.com is operated by Groundspeak, Inc., a privately held company based in Seattle, Washington. Groundspeak was founded in 2000 by Jeremy Irish, Bryan Roth, and Elias Alvord. The company has never raised institutional venture funding and has remained independently owned and operated. ### How does Groundspeak make money? Geocaching.com offers a free tier and a Premium Membership tier, currently around $30 per year. Premium members get expanded search tools, member-only cache types, instant notifications, and additional features. The company also sells merchandise, trackable items like Geocoins, and event licenses. ### What is the Garmin eTrex and why is it associated with geocaching? The eTrex is a line of handheld GPS receivers launched by Garmin in late 1999, with broader availability in 2000. The original model was small, ran 22 hours on two AA batteries, was weatherproof, and cost roughly $120. It became the default early-2000s device for the new geocaching community because it was the first GPS unit that combined consumer-grade pricing with field-grade durability. ### Did smartphones kill geocaching? No, though they came close to killing dedicated GPS hardware overall. Geocaching.com released official mobile apps for iPhone in 2008 and Android in 2010, which brought a new generation of casual urban geocachers into the activity. Experienced geocachers continued to use dedicated handhelds for harder, remoter caches where smartphones suffer from battery, signal, and durability limitations. The hybrid mode has kept the community growing across both device types. ### How many caches and players are there today? Geocaching.com lists more than three million active geocaches in 191 countries. Over a billion individual "found it" logs have been recorded across the history of the platform. Groundspeak does not publish exact active-user counts, but the community is widely estimated in the low single-digit millions of regular participants worldwide. ### What was in the original cache that Dave Ulmer hid? The original five-gallon bucket cache included a logbook and pencil, two CD-ROMs of DeLorme Topo USA mapping software, a George of the Jungle VHS tape, a book, four dollars, a cassette recorder, a slingshot, and a can of beans. The "can of beans" became a meme inside the geocaching community and is still referenced in modern Beavercreek tribute caches placed near the original hide location. --- # Why the Power Glove Failed but Never Really Died URL: https://404memoryfound.com/posts/what-happened-to-nintendo-power-glove.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-12 Updated: 2026-09-18 Topics: Gaming, Hardware **Summary:** The Power Glove was a motion controller Mattel released for the NES in October 1989 at about $100, and it was off store shelves by 1990 because its ultrasonic tracking was too imprecise to improve any NES game. Nintendo never designed it, the marketing firm Abrams/Gentile Entertainment licensed it, and only two games ever used its extra moves. In 2026 it is a collector item and a maker platform, with loose units near $100 and boxed ones above $200. **Key facts:** - Launched: October 1989, Mattel in the US and PAX in Japan - US price: About $100, near $200 in 2020 dollars per ACMI - Units sold: 1.3 million worldwide per co-creator Chris Gentile, 600,000 of them in Japan - Discontinued: 1990, after roughly one year on shelves - Status today: Collector item and maker platform, about $100 loose and $216 boxed on PriceCharting in 2026 ## How did a $10,000 lab glove become a $100 toy? The Power Glove descends from the VPL DataGlove, a research instrument sold to labs and simulation programs in the late 1980s. Its core idea came from Thomas G. Zimmerman, who filed a patent on a bend sensor for a glove on September 29, 1982. It was granted on September 17, 1985 as US 4,542,291, "Optical Flex Sensor," and the abstract describes electronics that "detect a combination of direct light rays and reflected rays when the flexible tube is bent." Zimmerman built the first version himself. "I built a glove with an optical sensor that could pick up finger bends," he told Mental Floss in 2017. In the same oral history, VPL co-founder Jaron Lanier put the research version's price plainly: "The glove went for about $10,000." Nintendo licensed the consumer rights to Abrams/Gentile Entertainment, a New York marketing firm, and never designed the product itself. Mattel manufactured it for the United States and PAX handled Japan. The engineering brief was brutal. AGE co-founder Chris Gentile told Mental Floss the job "became about taking a $10,000 device and turning it into $26 worth of materials," and that "we had about nine months to get it done." ## How was the Power Glove supposed to work? Two systems ran at once. Ultrasonic speakers on the glove pinged receivers mounted around the television, and the console worked out where your hand was by triangulating those pings. Separately, bend sensors ran along four fingers, the pinky excluded, so the glove could tell a fist from an open hand. The cost cutting landed on both halves. The lab glove's fiber optic sensors were swapped for cheap resistive strips, and the fine positional resolution that made the DataGlove useful in simulation work did not survive the trip to a $100 retail price. What reached kids in October 1989 was a plastic forearm brace with a directional pad, A and B buttons, and a numeric keypad used to punch in program codes, one per NES game, that mapped arm gestures onto ordinary button presses. Mattel engineer Will Novak told Mental Floss in 2017 that the ultrasonic triangulation genuinely worked, but that calibrating it was hard enough that no kid was ever going to bother. A controller needing a setup ritual before every session is a controller that loses to the one sitting next to it. ## Did The Wizard sell more gloves than the ads? The Power Glove had a feature film working as its commercial. Universal released The Wizard on December 15, 1989, ten days before Christmas, with Nintendo hardware in nearly every act. The rival gamer Lucas Barton opens a steel briefcase, pulls on the glove, and delivers the line that outlived the product: "I love the Power Glove. It's so bad!" Novak remembered the scene landing exactly as intended. "The kids were playing Rad Racer with it. 'Oh, it's so bad,'" he told Mental Floss. The film gave a peripheral almost nobody had used a demonstration that looked effortless, on a screen, to an audience of children with two weeks left to ask for it. The trade press was already primed. In 1989, Design News ran a cover story that opened by telling readers to "Throw away your joysticks, kids," a line the Australian Centre for the Moving Image later quoted as the high-water mark of the hype. Nintendo had spent the decade rebuilding an industry that the NES pulled out of the 1983 crash (https://404memoryfound.com/posts/how-nintendo-nes-saved-video-games-1983-crash.html), and in 1989 almost anything with its logo near it moved. ## What went wrong once kids got it home? Novak said Nintendo's own research showed a typical play session ran 90 to 120 minutes. "With the Glove, your arm would get tired after 15 or 20 minutes," he told Mental Floss. Holding an arm out at a television is exercise, and nobody had tested for that before the tooling was cut. The software was worse. Only two titles, Super Glove Ball and Bad Street Brawler, were built around the glove, and neither shipped in the box. Both played fine with a normal controller, so the glove added nothing you could not already do. Caesar Filori of Nintendo gave Mental Floss the one sentence version: "There was no game experience that became better for having used the Glove." Mattel executive Hall summed up the home experience for Mental Floss: "Your arm getting tired, you have no idea where the center is, there's no 3D. So you shove it into a closet." Mattel pulled the Power Glove from shelves in 1990, after roughly a year. It joined a short list of Nintendo era hardware that sold on promise and died on use, a pattern that repeated six years later with Nintendo's own Virtual Boy (https://404memoryfound.com/posts/what-happened-to-virtual-boy-nintendo.html). Nobody cancelled the Power Glove in a press release. It simply stopped being reordered. ## How many Power Gloves were actually sold? The numbers depend on who is counting. Chris Gentile told Mental Floss that the debut at CES produced "700,000 orders," with Toys R Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html) and Kmart (https://404memoryfound.com/posts/is-kmart-still-open-stores-left.html) each committing to 100,000 units before a single consumer had touched one. On lifetime sales, Gentile's figure is the one from inside the company: "We sold a total of 1.3 million Gloves, including in Japan." His brother John Gentile put the Japanese share at 600,000 units, sold under the PAX brand rather than Mattel's. The Australian Centre for the Moving Image gives a higher estimate, roughly 2 million units and about $88 million in revenue. Either way, the Power Glove was not a sales failure. It was a usage failure. The gloves moved at retail on the strength of a movie and a magazine cover, then stopped moving once the first wave of owners told their friends. That gap between units shipped and units used is why the accessory reads as a flop despite seven figures of sales. ## What is a Power Glove worth in 2026? PriceCharting tracks a loose NES Power Glove at $99.94 in September 2026, with a sale logged at $111.20 on August 15, 2026. Complete in box sits at $216.45, sealed examples run to $641, and a graded new one sold for $3,493.05. Condition of the foam and the ultrasonic receiver bar drives most of the spread, because both degrade in storage. The more interesting market is the one that cuts them open. Hackaday has covered a decade of rebuilds: Dillon Markey converted a glove into a stop motion animation controller in 2015, Nolan Moore rebuilt one as a universal wireless HID that flew an AR Drone in 2016, Alessio Cosenza shipped a Bluetooth nine axis version called Power Glove Ultra in 2017, and Teague Labs bolted one to an HTC Vive the same year. The shell is comfortable, distinctive and cheap, which is exactly what a maker wants around modern sensors. Where to find one today: working loose units turn up in the $100 to $150 range and boxed ones above $200, so expect to pay more than a Wii and its motion controllers (https://404memoryfound.com/posts/what-happened-to-nintendo-wii.html) for something that does far less. Retro specialists such as Stone Age Gamer's NES accessories shop (https://stoneagegamer.com/nintendo/nes/accessories/) are the safer route than an untested auction listing, since the failure points are internal. ## Frequently Asked Questions ### Does the Power Glove still work on an NES today? A Power Glove from 1989 will still plug into an original NES and register button presses, since the wired controller portion is simple and durable. The ultrasonic motion tracking is the part that fails, because the receiver bar and the glove's sensors drift with age and the system needed careful calibration even when new. Most 2026 owners use them as display pieces or donor hardware rather than controllers. ### How many games used the Power Glove? Only two NES games were designed around the Power Glove: Super Glove Ball and Bad Street Brawler, both from 1990. Neither was packaged with the glove, and both were fully playable with a standard NES controller, so the glove-specific moves were optional. Every other NES game had to be driven through numeric program codes that translated arm motion into ordinary button presses. ### Who invented the Power Glove? The underlying bend sensing technology came from Thomas G. Zimmerman, whose optical flex sensor patent was filed in 1982 and granted in 1985, and from the VPL DataGlove he developed with Jaron Lanier. The consumer version was designed by Abrams/Gentile Entertainment under a Nintendo license and manufactured by Mattel for the 1989 US launch. Nintendo itself did not design the Power Glove. **Sources:** - An Oral History of Nintendo's Power Glove, Mental Floss (2017): https://www.mentalfloss.com/article/91939/losing-their-grip-oral-history-nintendos-power-glove - US Patent 4,542,291, Optical Flex Sensor, Thomas G. Zimmerman: https://patents.google.com/patent/US4542291A/en - The promise of the Nintendo Power Glove, Australian Centre for the Moving Image: https://www.acmi.net.au/stories-and-ideas/nintendo-nes-power-glove/ - NES Power Glove price history, PriceCharting: https://www.pricecharting.com/game/nes/power-glove - Power Glove project archive, Hackaday: https://hackaday.com/tag/power-glove/ --- # theGlobe.com Today: The 606% IPO That Left a Shell URL: https://404memoryfound.com/posts/what-happened-to-theglobe-com-dot-com-ipo.html Author: Dana Reyes (404 Memory Found) Published: 2026-05-12 Updated: 2026-09-18 Topics: Internet Culture, Business Blunders, Money & Tech **Summary:** theGlobe.com is not a website anymore. The community site went dark in August 2001, and the company behind it, theglobe.com, inc., is now a shell with no employees, no operations and $15,115 in cash as of June 30, 2026. Its November 13, 1998 IPO closed 606 percent above the $9 offer price, a first-day record no American IPO has broken since. **Key facts:** - Launched: May 1, 1995, by Cornell students Stephan Paternot and Todd Krizelman - IPO: November 13, 1998, priced at $9 a share by Bear Stearns - First-day close: $63.50, about 606 percent above the offer, still the record - Community site: Shut down in August 2001, months after the Nasdaq delisting - Status today: Shell company, ticker TGLO, no employees, $15,115 in cash (June 2026) ## Who owns theGlobe.com now, and what is left? theglobe.com, inc. still exists as a Delaware corporation and still files with the Securities and Exchange Commission. There is nothing inside it. Delfin Midstream, a liquefied natural gas developer, has held about 70.9 percent of the stock since a purchase agreement signed on December 20, 2017. The company reported no employees as of March 23, 2026. Its only officer, Frederick Jones, holds the president, chief executive and chief financial officer titles at once and takes no pay for any of them. The balance sheet is the real answer. On June 30, 2026 the company held $15,115 in cash, its only asset, and it had lost $123,504 over the first six months of that year. Its own filing puts the risk in one line: "We prefer to avoid filing for protection under the U.S. Bankruptcy Code." The stock still changes hands over the counter as TGLO, with 441,480,473 shares outstanding as of August 5, 2026. It has not traded on Nasdaq since April 2001. ## How did two Cornell students build a 1995 social network? theglobe.com, inc. was incorporated on May 1, 1995 and started operating the same day, according to its own filings. Stephan Paternot and Todd Krizelman, both undergraduates at Cornell University, built it so that anyone with a browser could claim a homepage, join a themed club and talk to strangers. The product went head to head with GeoCities (https://404memoryfound.com/posts/history-of-geocities-websites.html), which Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) bought in 1999. theGlobe.com gave members free pages through a tool called uPublish, topic rooms called globeClubs, chat and message boards, plus small business hosting and a games network. At the scale of the time, it worked. The company's Form 10-K for 1999 reported 3.6 million registered members in the United States and abroad, 4.7 million unique visitors in December 1999 and a 7.2 percent reach of the internet as measured by Media Metrix. About 220 people worked there at the end of that year. Advertising paid for all of it. Banner and sponsorship sales produced roughly $15.0 million of the $18.6 million the company booked in 1999, about 80 percent of revenue. ## Why did a $9 stock close at $63.50 on day one? Bear Stearns priced the offering at $9 a share on November 13, 1998, after the deal had slipped twice in October. The company sold 3.1 million shares, a thin slice of its stock, into a market that had decided internet shares were scarce. The first trade printed far above the offer. CBS News reported an opening price of $67 that day, while later accounts put the first print at $87. The rest is not in dispute. The stock touched $97 and closed at $63.50. That close sat about 606 percent above the offer price, the largest first-day gain an American IPO had ever produced. CBS News rounded it to 605 percent in its story that evening. Nothing has beaten it in the 28 years since. The business underneath was thin. theGlobe.com had booked $1.2 million of revenue against $5.8 million of losses in the first half of 1998. Tom Taulli of Silicon Investor told CBS News that day, "We're kind of in a mania here. It doesn't make sense." ## What did the vinyl pants clip cost theGlobe.com? In 1999 a CNN crew followed Paternot to a Manhattan nightclub and filmed him dancing on a table in shiny vinyl pants. On camera he said, "Got the girl. Got the money. Now I'm ready to live a disgusting, frivolous life." The clip traveled the way things traveled before social media. It ran on television, it got quoted in print, and it followed Paternot for decades under a nickname he never shook, the CEO in the plastic pants. The segment did not cause the collapse, and treating it as the cause is the lazy version of the story. What it did was put a face on a suspicion investors were already forming in 1999, that the people running these companies had no clear idea what the money was for. Paternot later wrote a memoir about the run, A Very Public Offering, and was still arguing with the caricature years afterward. "No longer were we the only idiots in town," he wrote of the wave of internet offerings that followed his, in a line Salon quoted in 2001. ## Why did 3.6 million members not pay the bills? Community sites in 1999 sold banner ads, and banner ads were about to get very cheap. theGlobe.com's revenue grew from $5.5 million in 1998 to $18.6 million in 1999. Its net loss grew faster, from $16.0 million to $49.6 million, which works out to about $2.67 of loss for every dollar of revenue. There was no second revenue line worth the name, and no protection from anyone who wanted to copy the idea. The 1999 annual report said so in its own risk language, warning that "Barriers to entry are relatively insubstantial" in the business the company had chosen. When online ad rates fell apart in 2000 and 2001, nothing was left to fall back on. Paternot and Krizelman were pushed out in 2000, Nasdaq delisted the stock in April 2001, and the community site itself was switched off that August. The arithmetic was not unique to theGlobe.com. Pets.com (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html) and Webvan (https://404memoryfound.com/posts/what-happened-to-webvan-grocery-delivery.html) died of the same thing, spending heavily to buy customers who were never going to be worth what they cost. ## Where are Stephan Paternot and Todd Krizelman now? Todd Krizelman moved to the German media group Bertelsmann after theGlobe.com and worked with its publishing businesses. In 2007 he co-founded MediaRadar, an advertising intelligence company he still runs as chief executive. He told TheWrap in 2024 that he "went in with my eyes open" the second time around. Stephan Paternot was worth roughly $97 million on paper at the close of the first trading day, by Salon's 2001 account. He borrowed against that stake, put money into the delivery startup UrbanFetch and lost about a million dollars of his own when it folded. He has spent the years since in film and technology ventures. The corporation they started outlived both of their roles in it, on paper only. It sold its last operating business, the .travel registry manager Tralliance, on September 29, 2008, and has had no operations since. What is left is a ticker, a Delaware charter and a set of quarterly filings paid for by a natural gas company. That is the entire estate of the biggest first-day IPO in American history. ## Frequently Asked Questions ### Is theGlobe.com still in business in 2026? No. theglobe.com, inc. is a shell company with no employees and no operating business, and it has been one since it sold the Tralliance registry business on September 29, 2008. Its June 30, 2026 filing reported $15,115 in cash and warned that it may not continue as a going concern beyond the next twelve months. ### Was theGlobe.com's IPO really the biggest first-day pop ever? Yes, and it still is. theGlobe.com priced at $9 a share on November 13, 1998 and closed its first session at $63.50, a gain of about 606 percent. No American IPO has posted a larger first-day rise in the years since. ### Why did theGlobe.com shut down? theGlobe.com sold banner advertising to fund a free community site, and it lost $49.6 million on $18.6 million of revenue in 1999 alone. When online ad rates collapsed, Nasdaq delisted the stock in April 2001 and the community site was turned off that August. **Sources:** - theglobe.com, inc., Form 10-K for fiscal year 2025, SEC EDGAR: https://www.sec.gov/Archives/edgar/data/1066684/000110465926038019/tglo-20251231x10k.htm - theglobe.com, inc., Form 10-Q for the quarter ended June 30, 2026, SEC EDGAR: https://www.sec.gov/Archives/edgar/data/1066684/000110465926096764/tglo-20260630x10q.htm - theglobe.com, inc., Form 10-K for fiscal year 1999, SEC EDGAR: https://www.sec.gov/Archives/edgar/data/0001066684/000089534500000218/0000895345-00-000218.txt - Internet IPO Theglobe.com Soars, CBS News, November 13, 1998: https://www.cbsnews.com/news/internet-ipo-theglobecom-soars/ - Dumb, dumber and theglobe.com, Salon, August 22, 2001: https://www.salon.com/2001/08/22/paternot/ --- # What Happened to the Atari Lynx, the Color Handheld That Lost to Game Boy URL: https://404memoryfound.com/posts/what-happened-to-atari-lynx-handheld.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-11 Topics: Hardware, Gaming, Then vs Now Picture this: it's the fall of 1989. You're standing in the handheld aisle at Toys R Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html), and there are exactly two things you have never seen before in your life. On one peg, in a tiny cardboard box, there's the Nintendo Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html). Gray, the size of a Walkman, four shades of green, $89.99. Next to it, taking up about three times the shelf space, is a thing called the Atari (https://404memoryfound.com/posts/who-owns-atari-now.html) Lynx. It's the size of a paperback novel, has a backlit color screen, costs $179.95, and somebody at Atari has decided to advertise it as the future of portable gaming. Reader, I have to tell you something. The Atari Lynx was, by every spec sheet you could throw at it, the better machine. It was not even close. The Game Boy looked like a calculator. The Lynx looked like the year 2000. And the Lynx got destroyed. This is the story of a handheld that was years ahead of its time, was built by ex-Amiga engineers in a shed at a struggling software company, was launched by an Atari that was already losing the plot, and was buried by a $90 brick that played Tetris. If you were there, in the GameStops and Electronics Boutiques and Babbage's of the early '90s, you saw the whole thing happen in slow motion. The Lynx deserved better. Pull up a chair. ## How a Bankrupt Software Company Built the Most Advanced Handheld of the 1980s To understand the Lynx you have to go back to 1986, and a place called Epyx. If that name doesn't ring a bell for you, that's fine. Epyx made California Games, Summer Games, Winter Games, the Olympic-style sports titles you played on your Commodore (https://404memoryfound.com/posts/what-happened-to-commodore-computers.html) 64 in the back room of somebody's house. They were a software company, not a hardware company, and by the mid-'80s they were starting to feel the squeeze of the post-crash console market. So Epyx hired two guys nobody outside of Silicon Valley had heard of: RJ Mical and Dave Needle. These two were not normal engineers. They had just come off building the Commodore Amiga, which is to say they had built one of the most beautiful, most ahead-of-its-time computers ever made. Epyx wanted them to design a portable game system. Mical and Needle started in 1986. They had a working prototype by 1987. The internal codename was Handy. Here's what Handy was: a custom 8-bit 65SC02 CPU running at 4 MHz, paired with two custom chips Mical and Needle named Mikey and Suzy. Mikey did sound and video. Suzy was the wild one, a 16-bit blitter chip that handled sprite scaling, distortion, and rotation in hardware. The screen was a 3.5-inch backlit color LCD with a 4,096-color palette, displaying 16 colors on screen at once from that palette. You could rotate the whole machine, and the screen would flip with it, so left-handed players could just turn the whole thing around and play. You could daisy-chain up to eight units together with a cable called ComLynx for multiplayer. In 1987, in a backpack, this was insane. Nintendo's R&D2 team in Kyoto was working on the Game Boy at the same time, and Gunpei Yokoi's design philosophy was the polar opposite. Yokoi famously preached "lateral thinking with withered technology," which is a Japanese way of saying use the cheap stuff that already works. The Game Boy ran a Sharp LR35902 at 4.19 MHz, had a 160x144 monochrome dot matrix LCD, no backlight, and was designed to sip AA batteries. The Lynx ate AA batteries. Six of them. For about four to five hours. That single design choice, that one tradeoff, is the whole movie. But we're getting ahead of ourselves. ## How the Handy Became the Atari Lynx By 1988, Epyx was in trouble. They had a finished portable game system in their hands and no money to manufacture it. They started shopping it. Nintendo, who was secretly working on the Game Boy, took a look and said no. Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html), who would later launch the Game Gear in 1990, took a look and said no. A few other companies passed. The buyer that finally said yes was Atari Corporation, which by 1988 was a different beast than the Atari that ruled your living room in 1982. After the video game crash of 1983, the original Atari had been split up and sold off. Jack Tramiel, the legendary and famously combative founder of Commodore, bought the consumer division in 1984 and renamed it Atari Corp. By 1988, Atari Corp was making the Atari ST line of computers and the 7800 console, and they were getting absolutely lapped by Nintendo in the home market. Atari and Epyx struck a deal. Atari would handle manufacturing, distribution, and marketing. Epyx would handle the software development. By the end of 1989, Epyx had filed for Chapter 11 bankruptcy anyway, and Atari essentially owned the whole project. They renamed it the Atari Lynx and brought it to market in September 1989. Price: $179.95. Pack-in game: California Games, the Epyx classic, ported beautifully. I want you to understand the timing here, because it matters. The Game Boy launched in Japan in April 1989 and in North America on July 31, 1989. The Lynx launched in September 1989. They were almost the same Christmas. Two completely different visions of what a portable game console should be, sitting on the same shelf, six weeks apart. The Lynx was twice the price of the Game Boy, three times the size, ate six batteries instead of four, and lasted half as long. It also had color, hardware sprite scaling, and looked like a piece of Star Trek equipment. Both of these were true at the same time. ## Christmas 1989, and the First Sign of Trouble Atari had a problem at launch, and it was not the product. It was that they could not make enough of them. Christmas 1989 was supposed to be the Lynx's coming-out party. Instead, what showed up was a national supply shortage. Stores got tiny allocations. Most people who walked into Toys R Us looking for a Lynx walked out with a Game Boy, because that's what was actually in stock. Nintendo, by contrast, was Nintendo. They had been manufacturing electronics and toys at industrial scale since the late 1800s. They had the Game Boy stacked in pyramids in every retail aisle in America. They had Tetris bundled in every box. And here's where it gets interesting. By the end of 1990, Atari had moved roughly 500,000 Lynx units in the US. That's a real number. That is also a number that gets dwarfed by Game Boy units shipped in just the first six months of 1990. By late 1991, total Lynx sales were estimated at around 800,000. The Game Boy, in the same window, was selling that many units in good months. The math was already against the Lynx and we were only two Christmases in. ## The Lynx II, and the Doomed Comeback In July 1991, Atari did the thing every struggling console maker eventually does. They put out a revision. The Lynx II. The Lynx II was actually pretty good. They shrunk the chassis, added rubberized hand grips on the back, swapped in a clearer backlit screen, added a stereo headphone jack to replace the original's mono output, and put in a power-save mode that turned the screen off when you weren't using it. They also dropped the launch price to $99.99, which is the price the original Lynx probably should have been all along. For about ten minutes, this looked like it might work. The smaller form factor felt better in your hands. The new screen looked cleaner. The price was finally in Game Boy territory. Magazine reviewers gave it polite second looks and said things like "Atari has finally made the Lynx the system it should have been." And then, in April 1991, Sega launched the Game Gear in North America for $149.99, also color, also backlit, also chewing through batteries, also starring Sonic eventually. Now the Lynx wasn't even the only color handheld on the shelf. It was the third option, behind the Nintendo juggernaut and the Sega challenger. By 1995, total combined sales of the Lynx and the Game Gear were under 7 million units. The Game Boy, by itself, in the same period, sold over 16 million. And that's before you count the Game Boy Pocket, the Game Boy Color, and eventually the Advance, all of which would extend that platform into the 2000s. ## The Games Were Actually Good Here's the part that hurts. The Lynx had genuinely great games. Not enough of them, not advertised loudly enough, but they existed and they were special. Blue Lightning, an after-burner-style flight combat game, was the launch demo most stores looped on the kiosk and it was the closest thing to a polygonal arcade experience anybody had seen on a portable. Klax on the Lynx is, depending on who you ask, the definitive port of Klax. Todd's Adventures in Slime World, Chip's Challenge, Lemmings, Rampage, Toki, Xenophobe, all of these landed on the Lynx. The library got to about 75 official commercial releases over the lifetime of the system. For comparison, the Game Boy library had cleared 1,000 titles by the late '90s. So even when the Lynx had a great game, you had eight Game Boy games to choose from in the same slot at Funcoland. The shelf space told the whole story. The other thing that hurts: ComLynx multiplayer was, for its time, beyond anything else available. You could plug eight Lynx units together with cables and play head-to-head. In 1990. On a portable. Kids playing Game Boy multiplayer needed a Game Link cable and a friend, and they got two-player. Lynx kids could in theory get a school cafeteria full of Slime World running at lunchtime. The catch was that you needed eight Lynx units in the same room, which, in a world where the Lynx was outsold ten-to-one, meant nobody actually did this in practice. Beautiful idea, no install base. ## Why It Lost: A Plain English Postmortem People love to argue about why the Lynx lost. Some of the answers are technical, some of them are about marketing, and some of them are about Nintendo just being Nintendo. Here is what actually killed it, in order of how much it mattered: 1. Battery life. Game Boy: 15 to 30 hours on four AAs. Lynx: 4 to 5 hours on six AAs. If you are 11 years old at summer camp with no outlet, this difference is the entire ballgame. Color was cool. Battery anxiety was death. 2. Price. $179.95 versus $89.99 in 1989. This is roughly $475 versus $235 in today's money. For the same kid in the same store with the same allowance, only one of those two prices was a real option. 3. Software pipeline. Nintendo had Tetris, Mario, Zelda, Pokémon, Final Fantasy Legend, Kirby, the entire third-party world following them onto the platform. Atari had a few hundred third-party developers and a frequently rotating internal team. The Lynx never got a single killer app on the level of Tetris or Pokémon, and Tetris alone moved more Game Boys than the entire Lynx library moved Lynxes. 4. Atari itself. By 1991, Atari Corp was bleeding money, fighting Nintendo in a federal antitrust lawsuit, getting ready to launch the Jaguar (https://404memoryfound.com/posts/what-happened-to-atari-jaguar-64-bit-console.html) console (spoiler: that didn't go well either), and giving the Lynx less and less marketing oxygen each year. By 1994, the Lynx was effectively abandoned. By 1996, Atari Corp itself was being merged into JT Storage and would soon disappear as an active company. 5. Distribution. Nintendo had relationships with every toy and electronics retailer in North America, and they enforced exclusivity terms that made carrying competitors difficult. Atari had Atari's relationships, which is to say less leverage and less shelf space. Pick whichever one you like. The honest answer is that all five mattered, and they compounded. ## The Long Tail: Lynx After the Lynx Officially, the Lynx died around 1995. Atari Corp stopped active marketing. The last commercially released cartridge in the original era was around 1995 to 1996. The system was quietly retired with about 75 official releases and a couple million units in homes worldwide. But here's the thing about the Lynx. It refused to fully die. The homebrew community for the Lynx is one of the most active in retro gaming. Songbird Productions, run by Carl Forhan since the late '90s, has been releasing brand-new commercial Lynx games into the 2000s and 2010s. Atari Age and the Lynx homebrew scene have produced new cartridges, new demos, and entire new game releases decades after Atari left the room. There are Lynx tournaments at retro game conventions. There are people who still gather with their ComLynx cables and play Battlewheels in 2026. If you have ever been in a room of those people, you understand. They are not letting it go. They are not going to. And in 2019, on the Lynx's 30th anniversary, RJ Mical and Dave Needle did a series of interviews and conference appearances reflecting on what they had built. Sadly, Dave Needle passed away in 2016, but Mical has spoken extensively about the Lynx and the Amiga at conventions including the Game Developers Conference, and he has a real, justified pride about the thing he built. He should. They invented the modern color portable. The world just wasn't ready to pay for it. Every time you flip open a Switch and play in handheld mode, every time you boot up a Steam Deck on a flight, you are using something that the Atari Lynx prototyped in 1989. Color screen. Backlit. Networked multiplayer. Hardware sprite scaling. Battery anxiety. All of it. ## What the Lynx Tells Us About Hardware Wars Look, the Lynx vs Game Boy story gets told two ways. The first way is "Atari mismanaged a great product." That is partly true. The second way is "the better technology doesn't always win." That is also partly true. But neither of those is quite right on its own. The real lesson is that hardware is a system, not a spec sheet. You can have the better display and lose because of the worse battery. You can have the better software and lose because the price is wrong. You can have the better idea and lose because your distributor can't get the product on the shelf in October. Nintendo didn't beat Atari in 1989 because the Game Boy was a better console. They beat Atari because the Game Boy was a better product: cheaper, longer-lasting, more available, with better games coming, sold by a company that knew what to do with all of that. Atari shipped the future and forgot to also ship the present. The Lynx was a portable Star Trek display that needed to live in 1989, when 1989 wanted Tetris on a green screen for under a hundred bucks. ## Frequently Asked Questions About the Atari Lynx When was the Atari Lynx released? The original Atari Lynx launched in North America in September 1989, with a retail price of $179.95. The Lynx II revision arrived in July 1991 at $99.99. The European launch followed in 1990. How many Atari Lynx units were sold? Atari never released official lifetime sales figures. Industry estimates put US sales at roughly 500,000 units by the end of 1990 and around 800,000 by late 1991. Combined lifetime sales of the Lynx and Sega Game Gear together were estimated at under 7 million by 1995, against more than 16 million Game Boys in the same window. Who designed the Atari Lynx? The Lynx was designed by RJ Mical and Dave Needle, both of whom previously designed the Commodore Amiga. They started the project at Epyx in 1986 under the codename Handy, and it was sold to Atari Corporation when Epyx ran into financial trouble. Why did the Atari Lynx fail? It came down to four big things: a much higher price than the Game Boy ($179.95 vs $89.99 at launch), much shorter battery life (about 4 to 5 hours on six AAs vs 15 to 30 hours on four), a smaller and weaker software library, and Atari's inability to match Nintendo on manufacturing scale, retail distribution, and marketing. Was the Atari Lynx 16-bit? Mostly marketing. The CPU was an 8-bit 65SC02 running at 4 MHz. The Suzy graphics co-processor handled some graphics operations in 16-bit chunks via its custom blitter, and Atari leaned on that to call the system 16-bit in advertising. By modern standards it sat between the 8-bit Game Boy and the true 16-bit consoles like the SNES and Genesis. What was the Lynx's pack-in game? California Games, ported by Epyx, came in the box with the original 1989 Lynx. It featured surfing, BMX, halfpipe, and footbag events, and it was widely considered one of the best showcases of the Lynx's color hardware. Are people still making games for the Atari Lynx? Yes. Songbird Productions, founded by Carl Forhan in the late '90s, has continued to release new commercial Lynx cartridges into the 2010s and beyond, and the broader homebrew community on AtariAge has produced demos, new games, and re-releases for decades. Is the Atari Lynx worth collecting today? Working original Lynx and Lynx II units in clean condition typically sell in the $100 to $300 range as of recent years, depending on condition and bundled games. Boxed launch titles and rarer late-life releases like Battlewheels can run higher. The screens often need recapping or rubber dome replacement to stay usable, which is its own little community of repair tutorials. What other handhelds were on the market at the same time? The Nintendo Game Boy launched in 1989 at $89.99. The Sega Game Gear launched in 1990 in Japan and 1991 in North America at around $149.99. The NEC TurboExpress, a true portable TurboGrafx-16, launched in 1990 at $249.99 and is its own forgotten chapter. The Lynx was technically the first color handheld of that generation. The Lynx deserved better. It just had the misfortune of being right at the wrong time, in the wrong company's hands, against the most disciplined consumer electronics outfit on the planet. Pour one out, plug six AAs in, and load up California Games. The screen still pops. --- # What Happened to Excite, the Search Engine That Turned Down Google for $750,000 URL: https://404memoryfound.com/posts/what-happened-to-excite-search-engine-google.html Author: Dana Reyes (404 Memory Found) Published: 2026-05-11 Topics: Internet Culture, Business Blunders, Software & Apps In January 1999, a search engine called Excite had a stock price of around $128 a share, a market value of roughly $35 billion, and a CEO named George Bell who had just turned down the chance to buy a 4-person company called Google for $750,000. That sentence contains the entire decade of the late dot-com era. The number that's wild is not the $750,000. The number that's wild is the $35 billion. This is the story of Excite. Not the punchline version, where George Bell is the guy who passed on Google. The actual version, where six Stanford friends in a Cupertino apartment built one of the original web search engines, took it public in the same month as Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) and Lycos (https://404memoryfound.com/posts/what-happened-to-lycos-search-engine.html), merged it with the largest broadband company in America, and then watched the entire $35 billion structure collapse on top of itself in less than three years. Excite is the cleanest case study we have for what the late 1990s actually were. Smart people, real technology, real users, decisions that made sense in the moment, and a market that punished every one of those decisions in retrospect. Let's walk through it. ## Six Stanford Kids and a Company Called Architext In June 1993, six undergraduates at Stanford incorporated a company called Architext Software. Their names were Graham Spencer, Joe Kraus, Mark VanHaren, Ryan McIntyre, Ben Lutch, and Martin Reinfried. The original idea, before there was even much of a public World Wide Web to search, was to build software that could search and analyze large databases of text using statistical word relationships rather than just exact keyword matching. Architext was funded with a $15,000 loan from Joe Kraus's father. The team set up shop in Joe Kraus's parents' garage in Los Altos and then in a rented house in Cupertino. The Wired-era cliche of "founders in a garage" was, in this case, literal. By 1994, the technology landscape had changed under their feet. Mosaic had launched in early 1993. Netscape (https://404memoryfound.com/posts/is-netscape-still-around.html) was about to. The web was no longer a niche academic protocol. It was the next mass medium. The Architext founders pivoted from selling enterprise text-search to building a search engine for this new web. They renamed the product, and eventually the company, Excite. Vinod Khosla at Kleiner Perkins Caufield and Byers led an early venture round, putting Kleiner alongside Institutional Venture Partners on the cap table. By the time Excite went public, Kleiner Perkins owned roughly 25% of the company. Here's the thing. Excite's actual technology was good. Their search engine used statistical concept extraction, the idea that the word "doctor" and the word "physician" should return similar results because they are conceptually similar, even if a user types one and a document contains the other. That sounds like nothing now. In 1995 it was a real differentiator. Excite was widely considered one of the technically more sophisticated search engines of its era. ## October 1995: Excite Launches Excite launched as a public website in October 1995. By then there were already several search engines on the web. Yahoo, founded by Jerry Yang and David Filo at Stanford, had launched in early 1994 as a hand-curated directory and was the dominant front door to the web. WebCrawler, Lycos, Infoseek, and AltaVista (https://404memoryfound.com/posts/what-happened-to-altavista-search-engine.html) were either already live or about to be. The interesting fact is that, in late 1995, nobody knew which of these would win. They were all roughly the same size, all roughly the same speed, all funded by the same handful of VCs, and all losing money. The model was: get users, sell ads, become the front door. Excite's pitch had two layers. One was the technology, the concept search. The other was the editorial wrapper, news and content channels around the search box, plus user-customization with a feature called My Excite. The company very deliberately moved away from being just a search engine and toward being a "portal." That word, "portal," ran the late '90s. Yahoo was a portal. Excite was a portal. AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html) was a portal. Microsoft built MSN as a portal. The thesis was that web users would pick one home page and largely live there, the way TV viewers pick one channel. ## The April 1996 IPO Triple-Header In April 1996, three search engine companies went public on the Nasdaq in the same month: Yahoo, Lycos, and Excite. Yahoo's first-day pop is the one history remembers, but Excite's IPO worked too. The company priced shares at $17, raised about $34 million, and within months had a market value in the hundreds of millions. This was a year when "we are losing money but we have user growth" was a complete answer. The market loved it. By late 1996 Excite was acquiring smaller players, including Magellan and McKinley's Internet directory, and consolidating itself as one of the top three or four destinations on the web by traffic. And here's where the Google moment lives. Pull back from the headline. ## The 1999 Google Pitch (And Why George Bell Wasn't Crazy) In late 1998 and into early 1999, Larry Page and Sergey Brin were two PhD students at Stanford running a search engine called Google out of a Menlo Park garage they were renting from Susan Wojcicki. They had built something called PageRank that ranked web results by the link structure of the web itself, treating each link as a vote. It was much better than what was on the market. They knew it was much better. They were also broke and exhausted, and they wanted to get back to their PhDs. So they tried to sell Google. The asking price went through Vinod Khosla as an intermediary. The pitch landed in front of George Bell, who was the CEO of Excite at the time. The number on the table was first $1 million, then negotiated down to $750,000. And George Bell said no. The version of this story that gets retold makes Bell sound stupid. He wasn't. Years later, in interviews including with the Internet History Podcast, Bell laid out his reasoning. The deal-breaker, he said, was a non-negotiable condition that Page and Brin attached to the sale. They wanted Excite to rip out its existing search technology and replace it entirely with Google's. Excite at that moment had hundreds of engineers whose jobs depended on the Excite search stack. Bell had a board, a quarterly earnings cadence, and a culture built around that technology. Saying yes to Google's terms wasn't just a $750,000 line item, it was an internal political event that would gut his engineering organization. And here's the thing the Excite story always misses. Better search was, in 1999, a worse business model than worse search. Excite made money by keeping you on Excite. Showing you a banner ad next to your results, then a news feed, then a sports score, then a horoscope, then more ads. The point was the portal. A search engine that sent you off the page faster, the way Google was designed to, was actively bad for the page-view economics every public search company was trading on. Bell was running a media company that happened to have a search box, not a search company that happened to have ads. He was looking at Google through that lens. From inside that lens, Google's offer was "pay us $750,000 and we will optimize for the metric you are not measured on." Bell was wrong, in the way that people are wrong who can describe their world accurately and just don't notice the world is changing under them. He saw the choice exactly as it was framed, and the framing itself was about to be obsolete. The lesson of Excite vs Google is not that George Bell was a bad CEO. The lesson is that he was a perfectly competent CEO of a business model that was 18 months from being dead. ## The Other 1999 Decision: The @Home Merger While the Google offer was on the table, Excite was negotiating something far bigger. In January 1999, Excite agreed to merge with @Home Network (https://404memoryfound.com/posts/what-happened-to-excite-at-home-broadband.html), the country's largest broadband cable internet provider, in a deal valued at $6.7 billion in stock. The combined company became Excite@Home. George Bell became CEO of the combined entity. Tom Jermoluk, who had been running @Home, stayed on as chairman. The thinking behind the merger was elegant on paper. @Home had pipes. Excite had eyeballs. Cable broadband was about to change the home internet experience permanently. If you owned both the pipe into someone's house and the home page they saw when they opened their browser, you owned the most valuable real estate of the next decade. AT&T, which had a controlling stake in @Home through its acquisition of TCI in 1998, was a strategic backer. The market loved it. Excite@Home traded as high as around $128 per share in the first quarter of 1999, with a peak market cap somewhere around $35 billion. For context, the original Excite IPO three years earlier had priced at $17 a share and raised $34 million total. The implied wealth creation between 1996 and early 1999 was, on paper, more than 100x. And then it didn't work. Almost none of it worked. ## Why Excite@Home Fell Apart Three things went wrong, in roughly this order. First, the integration didn't deliver synergies. @Home's broadband customers were not particularly more likely to use Excite than any other portal, and the bundling of "Excite as your default home page" felt to many users like the kind of intrusion that made AOL annoying in 1997. The product fit was real on a slide deck and weak in practice. Second, AT&T started behaving like a controlling shareholder rather than a partner. By 2000, AT&T had increased its ownership stake and begun negotiating directly with cable partners to take broadband subscribers off the @Home network when contracts expired. This was, from AT&T's view, a normal commercial move. From Excite@Home's view, the company that owned a controlling chunk of them was simultaneously planning to walk away with their largest customer base. It made the long-term enterprise value of Excite@Home very hard for the public market to underwrite. Third, the dot-com bubble burst. The Nasdaq peaked in March 2000. By the second half of 2000, every internet advertising business was watching its CPMs collapse. Excite was a portal that monetized through banner ads. When the ad market dried up, the entire revenue side of Excite's business stalled. Combined with the fixed costs of running @Home's broadband infrastructure, the company's cash burn became unsustainable. The stock cratered. From the $128 peak in early 1999, Excite@Home was trading around $15 by September 2000, when George Bell stepped down as CEO. By late 2001, it was trading under $1. On October 1, 2001, Excite@Home filed for Chapter 11 bankruptcy in the Northern District of California. In the messy unwind that followed, AT&T took over service for many of the broadband subscribers, and the Excite portal was sold off to iWon for a fraction of its former value, then changed hands several more times in subsequent years. There is, technically, still an excite.com today. It is a thin portal page kept alive by a series of small operators. It is not the company. ## What Excite Got Right (Yes, Really) It's tempting to write the whole story as a parade of mistakes, capped by the Google decision. That's not quite fair. Look at what Excite actually pulled off. They took a research-stage search technology out of Stanford and turned it into one of the four most-visited destinations on the web within three years. They went public successfully in the most contested IPO month of the decade. They built a personalization product, My Excite, that anticipated every "your home page" feature now built into Google News, Apple News, and Yahoo's portal. They were one of the first major web companies to sign content distribution deals with newspaper chains and wire services. They acquired Magellan, McKinley, MatchLogic, and a handful of other operators in moves that, at the time, looked smart. And they correctly identified that broadband was going to reshape the consumer internet. They were just early on the timing, paired with the wrong partner, and exposed to the wrong commercial structure with AT&T. The strategic logic was not crazy. The execution and the macro environment were unkind. ## The Three Stories Inside the Excite Story If you zoom out, Excite is actually three stories layered on top of each other. Story one is about the founders. Six Stanford kids, $15,000 from a parent, a garage, a real piece of technology, an IPO three years later. By any reasonable measure, this is the dream version of a startup. Joe Kraus has spent the post-Excite years as an entrepreneur and investor, including building JotSpot (acquired by Google in 2006) and serving as a general partner at GV. The founders did fine. The technology is what got buried. Story two is about the late-1990s portal model. Pageviews, banner ads, sticky home pages, vertical content channels, a CEO from the magazine industry. (Bell came to Excite from Times Mirror Magazines.) This model dominated the late '90s web and was, in retrospect, almost exactly wrong. Google didn't win because it had better technology, although it did. It won because it had a better economic model. Search ads on intent versus banner ads on attention. The portal era ended. The search era started. Excite was on the wrong side of that line. Story three is about cable broadband. AT&T spent the late '90s buying up cable systems through the TCI acquisition and trying to use those pipes for high-speed internet via @Home. The integration of internet and cable would eventually happen, but not on @Home's terms and not on Excite@Home's timeline. Comcast and Time Warner Cable and Cox would build their own broadband businesses through the 2000s, with their own portals and their own back-end deals. The era of "one company owns the pipe and the home page" never arrived in the form Excite@Home bet on. It arrived in a different form, and Google was waiting at the front door of every browser by then. ## The Real Punchline Here's the part that doesn't get told enough. The reason George Bell's $750,000 decision feels so absurd in retrospect is not that he failed to see the future. It's that the future arrived faster than anyone, including the founders of Google, expected. In 1999, Page and Brin were trying to sell Google because they could not figure out how to make money from search. In 1999. The AdWords product, the thing that turned Google into a money geyser, did not launch until October 2000. The keyword auction model that became AdWords' big leap, sometimes called the Overture model after the company GoTo.com became, didn't get adopted in its modern form until 2002. The Google IPO was August 2004. If George Bell had bought Google for $750,000 in 1999, he would have inherited a tiny company with no revenue model, told his board he had paid for it by ripping out his own engineering team, and watched the dot-com bubble burst eighteen months later. The Excite@Home merger would have happened anyway. The $35 billion would have been built on the same broken structure. Whether his small Google acquisition would have survived inside that wreckage, the way Hotmail survived inside Microsoft, is genuinely unclear. The convenient story is that one decision changed everything. The truer story is that Excite's whole model was on a clock, and by the time the alarm went off, everyone in the building was already at their desks, doing the work, and no one heard it ring. ## Frequently Asked Questions About Excite When did Excite start and when did it end? Excite was founded as Architext Software in June 1993 by six Stanford undergraduates. It launched as a public website in October 1995. Its parent company Excite@Home filed for Chapter 11 bankruptcy on October 1, 2001. A small successor portal at excite.com still exists today under different ownership, but the original company is long gone. Who founded Excite? Graham Spencer, Joe Kraus, Mark VanHaren, Ryan McIntyre, Ben Lutch, and Martin Reinfried, all then-students at Stanford University. Joe Kraus is the co-founder most often associated with the company in interviews, and his account in Jessica Livingston's book Founders at Work is the canonical first-person source on the early years. Did Excite really turn down buying Google for $750,000? Yes. In 1999, Larry Page and Sergey Brin, with Vinod Khosla acting as intermediary, offered to sell Google to Excite for $1 million, then negotiated down to $750,000. Excite CEO George Bell turned it down. Bell has said in subsequent interviews, including with the Internet History Podcast in 2014, that the main sticking point was Page and Brin's demand that Excite replace its search stack with Google's, which would have required gutting Excite's existing engineering organization. Was Excite ever the largest search engine? No. Yahoo was the dominant front door to the web through the late '90s, with Excite usually ranking in the top three to top five depending on the metric. By traffic, Excite was a major player but never the leader. What was Excite@Home? The combined entity formed in 1999 when Excite merged with @Home Network, the largest broadband cable internet provider in the United States at the time, in a stock deal valued at about $6.7 billion. The combined company peaked at a market cap of roughly $35 billion in early 1999 and filed for bankruptcy in October 2001. Why did Excite fail? A combination of factors. The portal-and-banner-ad business model was being undermined by Google's intent-based search ads. The @Home merger created strategic and operational tension with AT&T as a controlling shareholder. The dot-com crash in 2000 collapsed the online ad market that Excite depended on. And competing search engines, particularly Google, simply offered a better core product as the web grew larger and noisier. Does Excite still exist? A site at excite.com is still online as of today, but it has been sold and resold multiple times since 2001, most recently operating as a thin portal and webmail service. It has no continuity of staff, technology, or strategy with the original company. What happened to the founders? All six founders moved on after the bubble. Joe Kraus founded JotSpot, which was acquired by Google in 2006, and later joined Google Ventures (now GV). Graham Spencer also joined Google. Several of the others moved into investing, smaller startups, or stepped away from the industry. What's the lesson of Excite? Picking the right business model matters more than picking the right technology. Excite had real engineers, real users, real revenue, and a real strategy. It also had a model that was about to be replaced. Better technology can lose to better economics, especially when the better economics show up while you are negotiating a $6.7 billion merger to defend the older model. --- # Why the TurboGrafx-16 Won Japan and Lost America URL: https://404memoryfound.com/posts/what-happened-to-turbografx-16-nec-forgotten-console.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-11 Updated: 2026-09-18 Topics: Gaming, Hardware, Business Blunders **Summary:** The TurboGrafx-16 launched in Japan as the PC Engine in October 1987 and became a real rival to the Famicom, then reached the United States in August 1989 two weeks behind the Sega Genesis and never recovered. NEC reported only 750,000 US consoles sold by March 1991, against 18 million Genesis units, and Turbo Technologies dropped the platform in May 1994. In 2026 Konami owns the library, the 99.99 dollar TurboGrafx-16 Mini from 2020 is the last official hardware, and an original console runs about 207 dollars loose. **Key facts:** - Launched: October 30, 1987 in Japan as the PC Engine; August 1989 in the United States - Launch price: 199.99 dollars for the console, 399.99 dollars for the TurboGrafx-CD add-on in November 1989 - Units sold: 10 million in Japan by 1995; 750,000 in the United States by March 1991 - Owner today: Konami, which merged Hudson Soft into its games arm in March 2012 - Status today: Discontinued in North America in May 1994. Konami's 99.99 dollar TurboGrafx-16 Mini from 2020 is the only official hardware; original consoles sell for about 207 dollars loose ## How did a small Japanese console outsell the Famicom? Hudson Soft designed the machine and NEC built it, and it went on sale in Japan on October 30, 1987 as the PC Engine. It measured 14 by 14 by 3.5 centimeters, which still makes it the smallest major home console ever released. It arrived two years before anything comparable reached American stores, and Japanese publishers backed it early. By 1989 NEC had sold more than 1.2 million consoles in Japan along with over 80,000 CD-ROM units. The shoot em up library did most of the work. Hudson and Irem kept the release schedule full while Nintendo's Famicom was aging, and for stretches of 1988 and 1989 the PC Engine was the system Japanese retailers could not keep in stock. By 1995 NEC had sold 10 million PC Engine units in Japan, plus 1.92 million CD-ROM and Duo units by March 1996. Measured only against its home market, the platform was a success. ## What did American buyers see on the shelf in 1989? NEC put the redesigned console into New York City and Los Angeles test markets in late August 1989, two weeks after Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html) of America test-marketed the Sega Genesis (https://404memoryfound.com/posts/sega-genesis-games-how-to-play-2026.html) in the same two cities. Sega had barely changed its Japanese hardware. NEC had spent months restyling its own. The retail price was 199.99 dollars, though Game Developer's 25th anniversary account of the launch puts it at 189 dollars. Either way the box came with Keith Courage in Alpha Zones, a Hudson platformer nobody in the United States had heard of, while the Genesis shipped with a port of the arcade hit Altered Beast. NEC's American operation in Chicago built 750,000 units on the assumption that demand would follow. It did not. Ken Wirt, who ran NEC's TurboGrafx unit, told Game Developer in 2014 that Sega "outspent us probably 4-to-1 in marketing, or even higher". The people running the launch were consumer electronics staff, not games people. Carol Balkcom, who was on that launch team, put it plainly to the same outlet in 2014: "They came from electronics sales and marketing, but not from the game business." ## Was the TurboGrafx-CD the first CD game system? It was. The CD-ROM2 add-on went on sale in Japan on December 4, 1988, which made the PC Engine the first video game console to use CD-ROM as a storage format. It sold 60,000 units in its first five months. The American version, the TurboGrafx-CD, followed in November 1989 at 399.99 dollars with no game in the box. A console and drive together cost about 600 dollars in 1989 money, at a moment when most American parents had never seen a CD-ROM drive of any kind. NEC was three years ahead of the Sega CD (https://404memoryfound.com/posts/what-happened-to-sega-cd-add-on.html), which reached the United States in 1992, and four years ahead of the 3DO (https://404memoryfound.com/posts/what-happened-to-3do-700-dollar-console.html). Being first bought NEC very little. By March 1991 it had moved 500,000 CD-ROM units worldwide, most of them in Japan. ## When did NEC give up on the United States? By March 1991 NEC reported 750,000 TurboGrafx-16 consoles sold in the United States, a number that matched its original production run almost exactly. Sega eventually sold 18 to 18.5 million Genesis units in the same market by June 1997, and Nintendo sold 34 million NES consoles across the Americas. In mid-1992 NEC and Hudson Soft handed North American management to a joint venture called Turbo Technologies, Inc., which launched the TurboDuo with the CD drive built in. It arrived into a market already split between the Genesis and the Super NES. In May 1994 Turbo Technologies announced it was dropping support for the Duo. It kept repairing existing units and let independent companies publish the last few games. That was the end of NEC as a console maker in the United States, five years after the launch. John Greiner, Hudson Soft's first American employee, described the expectations that preceded all of it to Game Developer in 2014: "the hype had a life of its own". ## Who owns the TurboGrafx-16 library in 2026? Konami does. It bought into Hudson Soft in 2001, completed a full takeover in January 2011, and merged Hudson into Konami Digital Entertainment on March 1, 2012. Bonk, Star Soldier, Adventure Island and the rest of the Hudson catalog went with it. Konami used that catalog once, properly. It announced the TurboGrafx-16 Mini in July 2019 at 99.99 dollars, shipped it in Japan on March 19, 2020, and released it in North America on May 22, 2020 after the pandemic shut the Chinese factory building it. The international model carries 57 games, the Japanese model 58. NEC itself left the business long ago. Its follow-up, the PC-FX, launched only in Japan and failed, and the company has not made a game console since. In 2026 the TurboGrafx-16 is a Konami asset, not an NEC one, and Konami has announced no successor to the Mini. ## What does a TurboGrafx-16 cost second-hand today? PriceCharting tracks a loose TurboGrafx-16 console at about 207 dollars in 2026, 325 dollars complete in box, and 821 dollars sealed. The TurboDuo is the collector's item: roughly 558 dollars loose and 1,075 dollars complete. Games are cheaper than the hardware, because the American library was small and the good ones were printed in quantity. Bonk's Adventure runs about 49 dollars loose, Blazing Lazers about 56 dollars, Ninja Spirit about 59 dollars and Military Madness about 43 dollars. The exception is Magical Chase, a late 1993 shooter with a tiny print run, at roughly 3,955 dollars loose. Where to find one today: Stone Age Gamer (https://stoneagegamer.com/turbografx-16/) stocks TurboGrafx-16 consoles, controllers, HuCards and the flash carts that hold the whole library, which is the practical route if the goal is playing rather than collecting. Expect 200 to 350 dollars for a working console and a controller before you buy a single game. ## Frequently Asked Questions ### Is the TurboGrafx-16 still made in 2026? No. Turbo Technologies discontinued the TurboGrafx-16 and TurboDuo in North America in May 1994, and NEC has not built a game console since. The closest thing to new hardware is Konami's TurboGrafx-16 Mini, a 99.99 dollar plug-and-play unit released in North America on May 22, 2020, which is no longer in production. ### How many TurboGrafx-16 consoles were sold in the United States? NEC reported 750,000 TurboGrafx-16 consoles sold in the United States by March 1991, which was roughly the size of its original production run. For comparison, Sega sold 18 to 18.5 million Genesis consoles in the United States by June 1997 and Nintendo sold 34 million NES units across the Americas. In Japan the same machine, sold as the PC Engine, reached 10 million units by 1995. ### Who owns the TurboGrafx-16 games now? Konami owns them. It completed its takeover of Hudson Soft, the TurboGrafx-16's designer, in January 2011 and merged the studio into Konami Digital Entertainment on March 1, 2012, taking the Bonk and Star Soldier series with it. Konami used that library for the 2020 TurboGrafx-16 Mini and still controls any future re-release. **Sources:** - Game Developer: Stalled engine: The TurboGrafx-16 turns 25: https://www.gamedeveloper.com/business/stalled-engine-the-turbografx-16-turns-25 - Wikipedia: TurboGrafx-16: https://en.wikipedia.org/wiki/TurboGrafx-16 - Video Game Console Library: NEC TurboGrafx-16 / PC Engine: https://www.videogameconsolelibrary.com/console/nec-turbografx-16/ - Konami: TurboGrafx-16 Mini console release details: https://www.konami.com/games/us/en/topics/1312/ - PriceCharting: TurboGrafx-16 price guide: https://www.pricecharting.com/console/turbografx-16 --- # What Happened to PointCast and the 1990s Push Technology Boom URL: https://404memoryfound.com/posts/what-happened-to-pointcast-push-technology.html Author: Dana Reyes (404 Memory Found) Published: 2026-05-11 Topics: Internet Culture, Software & Apps, Business Blunders In March 1997, Wired magazine put a full image of a tombstone on its cover. The tombstone read "The Web." The cover story argued that the web was already obsolete, that browsing pages on demand was a clumsy way to consume information, and that the future of the internet was a technology called push. The headline product driving that argument was a Cupertino-based company called PointCast. Two years later, in May 1999, PointCast sold for about $7 million. It had turned down a reported $450 million acquisition offer from News Corporation in January 1997. That delta, from a billion-dollar covers-of-Wired moment to a $7 million fire sale, in roughly 28 months, is one of the cleanest case studies in tech history of how a real insight, wrapped in the wrong product at the wrong moment, can disintegrate. The interesting part is not that PointCast failed. The interesting part is what it got right. PointCast understood, before nearly anyone else, that pull-based browsing was an inefficient way for humans to keep up with information. They built the first mainstream attempt at solving that problem. They lost. And then, fifteen years later, everyone who lost to them found different ways to win on the same insight. ## What PointCast Actually Was PointCast Network was free software for Windows and Mac that ran in the background on your office or home PC. When your machine sat idle, the software activated as a screensaver. Instead of flying toasters or starfields, you got a customizable scroll of news headlines, stock quotes, weather, and sports scores, pulled from sources like CNN, Reuters, the Boston Globe, the New York Times, and the Los Angeles Times. The official launch was February 13, 1996. The company was based in Sunnyvale, California, and had been founded in 1992 by Christopher R. Hassett, originally under the name PED Software. By 1996 they had renamed the product PointCast Network and pivoted from CD-ROM-based publishing into internet-based delivery. The shift was the whole bet. Here is the technical idea. The web in 1996 worked one way: a user opened a browser, typed a URL, and the browser pulled the page from a server. Every visit required intent. If you wanted to know the closing price of Microsoft stock, you had to open Netscape (https://404memoryfound.com/posts/is-netscape-still-around.html), type in a URL, and wait for the page to load over a 28.8 kbps modem. PointCast inverted this. Instead of the user pulling content, the PointCast server pushed updates to the client software at scheduled intervals. The software stored those updates locally. When the user walked away from their desk, the screensaver kicked in and displayed the most recent content. The user did nothing. The content just arrived. This sounds boring today. In 1996 it felt like science fiction. ## The Cultural Moment To understand why PointCast briefly captured the imagination of the entire tech press, you have to remember what 1996 actually felt like. The web was four years old as a public phenomenon. Most people had encountered it for the first time within the previous 18 months. Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) had IPO'd in April 1996 and was the dominant directory of the internet, a hand-curated tree of links. Google did not exist yet. Search was bad. RSS would not be invented until 1999. The notion of an algorithmic feed of personalized content was not yet a notion. What PointCast offered, in this context, was the first product that let you specify what you cared about and then receive a stream of it without further effort. You picked which news categories you wanted. You picked which stocks to follow. The software went out, fetched updates from PointCast's servers, and put the result on your screen. For knowledge workers in 1996, this was meaningful. The alternative was opening five different websites every hour to check what had moved. PointCast collapsed that entire workflow into a passive ambient display. Adoption was rapid. By late 1996, PointCast claimed somewhere between 1 and 1.7 million active users depending on the source and the measurement method. For free software in 1996, with a 28.8 modem as the average connection speed, those numbers are genuinely impressive. The product was being used in newsrooms, trading floors, marketing departments, and law firms, places where staying current mattered and where IT had budget to install niche software. Wired magazine, in March 1997, ran the now-famous "Push!" cover. The thesis was that the pull-based browsing model was obsolete and that push would become the dominant paradigm of internet content delivery. PointCast was the lead example. The cover did not say it explicitly, but the message between the lines was that companies like Netscape and Yahoo, built on pull-based models, were sitting on a foundation that was about to crumble. In 1997, the consensus position among tech journalists was that the browser was a transitional technology and push would replace it. That consensus did not survive the next 18 months. ## The $450 Million Offer The Wired moment was preceded by a more concrete signal of mainstream interest. In December 1996, Microsoft announced a partnership with PointCast that would integrate push technology into Internet Explorer 4. The partnership was widely covered as validation. Microsoft, the dominant force in consumer software, was effectively saying that push was the future of the desktop internet experience. Around the same window, News Corporation, the global media conglomerate led by Rupert Murdoch, approached PointCast with an acquisition offer. The reported number was $450 million. PointCast turned it down. The exact reasoning depends on whose account you trust. Some accounts say Hassett and the board believed the company would be worth significantly more. Other accounts, including comments from News Corporation's James Murdoch in later interviews, point to internal hesitation at PointCast around revenue projections and execution. By March 1997, the offer was off the table. PointCast announced plans to go public instead. In hindsight, the timing of that decision is extraordinary. The Wired cover hit newsstands roughly the same month that News Corp walked away. PointCast was, by external measures, at peak cultural relevance. The IPO was filed in 1997. It would never close. ## The Bandwidth Problem Here is where the logic starts to break down. PointCast worked by having the client software check in with PointCast's servers at regular intervals, typically every few hours, to download fresh content. The downloaded content included not just text headlines but graphics, advertisements, and small animations. Each update was a few hundred kilobytes. Multiply that by tens or hundreds of users in a single company office, all pulling updates simultaneously over a shared T1 line, and the math gets ugly. By mid-1997, corporate IT departments started noticing that PointCast was consuming a disproportionate share of their internet bandwidth. The complaints were not theoretical. A 1.5 Mbps T1 line shared by 50 employees, half of whom had PointCast running, could see meaningful chunks of its capacity tied up by background pushes during the workday. Pages loaded slower. Email moved slower. Video, which was barely a thing in 1997, became almost impossible. The response from corporate IT was straightforward. PointCast was banned. By the end of 1997, many large companies had explicit policies prohibiting installation. The product's primary user base, knowledge workers in office environments, was systematically locked out. Imagine pouring a swimming pool through a garden hose. That was the corporate internet of 1997, and PointCast had decided to fill the entire pool ten times a day on top of whatever else was already running. The bandwidth problem was solvable. Send less data. Compress better. Update less frequently. But PointCast's business model required active, engaged users seeing fresh content and advertisements. Reducing the frequency of pushes meant reducing ad impressions, which meant reducing revenue. They were stuck between a technical problem and a revenue problem, and they could not optimize for one without breaking the other. ## The Ad Revenue Question PointCast was free to users. The revenue model was advertising. Brands could pay to have their ads appear in the screensaver rotation alongside the news headlines. Wells Fargo, GM, Visa, and other Fortune 500 advertisers signed on early. The pitch to advertisers was elegant on paper. You could not skip a PointCast ad the way you skipped a television ad. The ad was on screen while the user was away from their desk and again when they returned. The audience was educated, employed, and the kind of person who paid attention to news. CPMs in the early days were reportedly high. Two things went wrong with this revenue model in practice. First, the same bandwidth issues that made corporate IT depart-ments hostile also meant that ad rotations were not actually being seen as often as the pitch deck suggested. Second, advertisers in 1996 and 1997 were still figuring out how to measure online ad effectiveness at all. The metrics that would later make digital advertising a $500 billion industry, click-through rates tied to conversion data tied to attribution models, did not exist in mature form yet. By the time PointCast's advertisers started asking harder questions about return on investment, the user base was already eroding from the corporate bans. ## The IPO That Never Happened Internal turmoil compounded the external pressure that followed the News Corp collapse. Christopher Hassett, the founder and CEO, was replaced in October 1997 by David W. Dorman, a former executive at Sprint and Pacific Bell. Dorman was brought in to bring adult supervision and to navigate either an IPO or a sale. The IPO came first. PointCast filed in May 1998 with a reported valuation in the $250 million range, well below the $450 million News Corp offer it had refused 16 months earlier. The filing was withdrawn within two months. The public market was not buying. A failed IPO in 1998 was not, by itself, fatal. The dot-com bubble was still inflating. Capital was available. But the withdrawal was widely covered as a confidence event. The narrative shifted, fast. PointCast went from being the future of the internet to being a cautionary example almost overnight. By late 1998, PointCast was visibly in decline. Internet Explorer 4 had shipped with its push technology features. They went unused by most consumers. The Microsoft partnership, which was supposed to validate the model, had instead helped commodify it without driving meaningful PointCast adoption. A consortium-led acquisition deal valued around $100 million was discussed in late 1998 and ultimately did not close. Meanwhile, a completely different idea was taking shape in Mountain View. Sergey Brin and Larry Page had incorporated Google in September 1998. The product was a search engine. Pull, not push. Better than the alternatives by an embarrassing margin. The bet that won the internet was the exact opposite of the bet PointCast had made. ## The $7 Million Exit In May 1999, PointCast was sold to Launchpad Technologies, a San Diego company backed by Idealab. The reported price was approximately $7 million. Dorman had resigned as CEO in March 1999, and Phil Koen, the former COO, had stepped in to manage what was effectively a wind-down. $7 million versus $450 million is the number people remember. The decline is roughly 98 percent in a little over two years. It is one of the cleanest documented examples of how fast a tech valuation can collapse when the underlying thesis breaks. Launchpad rebranded the service and continued operating it for less than a year. The PointCast network was shut down in 2000. The technology was effectively dead by the time of the dot-com crash in March 2000, though most of the press coverage of the crash never bothered to mention it. PointCast had already been written off. ## What PointCast Got Right It is tempting to read the PointCast story as a simple failure narrative. Bad bet, bad management, bad ending. The more interesting reading is that PointCast was wrong about everything except the central insight, and the central insight turned out to be enormous. The insight: most information consumption should not require active retrieval. Users should be able to specify what they care about once, and then have a continuous stream of relevant content delivered to them. This is, almost word for word, the founding thesis of every algorithmic feed product that came after. Look at what replaced PointCast and you see the insight everywhere. Google News, launched in 2002, was a personalized news aggregator. RSS readers in the mid-2000s. Twitter, founded 2006, was a manually curated push feed. Facebook News Feed, introduced in 2006, became the dominant news distribution channel of the 2010s. Apple News, Flipboard, Pocket, Feedly, Substack, TikTok's For You feed. All of them are versions of the same idea PointCast tried to sell in 1996. What changed was the delivery mechanism. PointCast's failure was insisting on a heavy desktop client that fought corporate IT and burned through scarce bandwidth. The products that won used the browser or the phone, asked for less per session, and arrived after broadband and mobile networks had absorbed the bandwidth costs that killed PointCast. PointCast was right that pull was inefficient. They were just early by about a decade. ## The Screensaver Problem One smaller observation that gets lost in the larger narrative is that PointCast bet on the screensaver as the primary surface. This was a clever and self-defeating choice. Clever, because the screensaver was the one piece of screen real estate that nobody else was using. While the user was at their desk, that screen belonged to whatever app they were working in. The moment they walked away, the screen was idle. PointCast claimed that real estate and turned it into a content channel. Self-defeating, because the screensaver is exactly the part of the desktop that the user is, by definition, not looking at. If the user was looking at the screen, the screensaver would not be running. PointCast solved this partially by also offering a foreground app, but the foreground app was less compelling than the screensaver, and it competed for attention with everything else on the user's desktop. The successors to PointCast almost universally rejected this surface choice. The browser tab became the news delivery surface. Then the mobile home screen. Then the lock screen notification. Each successive generation of news delivery moved closer to where the user was actually looking, not further away. ## The Push Technology Hangover The collapse of PointCast was not just a single-company event. It was the failure of an entire category narrative, and for the rest of the 1990s, the term "push" became almost unmentionable in technology marketing. BackWeb, Marimba, Castanet, NewsEdge, BellCharles, and a dozen other companies had launched products under the push technology banner in 1996 and 1997. Most of them pivoted, downsized, or shut down by 1999. Marimba, founded in 1996 by Kim Polese and three colleagues from Sun Microsystems with a heavyweight Java pedigree, survived by repositioning entirely as an enterprise software distribution company, eventually being acquired by BMC Software in 2004 for around $239 million. The others quietly disappeared. Internet Explorer 4's push features, the result of the much-hyped Microsoft partnership, shipped to consumers, were ignored, and were quietly removed in later versions of the browser. The technology was called Active Channels. Most users never knew the feature existed. The push hangover was so severe that when notifications became a core part of smartphone operating systems in the early 2010s, almost nobody called them push technology even though that was exactly what they were. The category had a reputational scar that took a full product generation to fade. ## Why PointCast Matters Now Three things make the PointCast story worth revisiting today, even if you have never heard of the company before now. The first is that PointCast is the cleanest example in tech of a company that was right about the future and wrong about the present. Their thesis about the inefficiency of pull-based information retrieval was correct. Their product was a dead end. The lesson is that being directionally right does not protect you from being executionally wrong, and the distance between those two things can be measured in $443 million of evaporated valuation. The second is that the bandwidth story has a precise modern parallel. Every few years, a new category of internet product emerges that consumes more bandwidth than the existing infrastructure was sized for. Streaming video did this in the late 2000s. Cloud gaming attempted it in the early 2020s. Generative AI is doing it now in 2026, with model inference and data transfer chewing through enterprise networks in ways that IT departments are still figuring out how to manage. The PointCast pattern, of a useful product becoming a network parasite and being administratively blocked, recurs every cycle. The third is that the screensaver bet illustrates how easy it is to confuse available surface area with attention. PointCast assumed that because the screensaver was unused, it was therefore available for content delivery. The unused screensaver was not unused because nobody had thought of it. It was unused because the user was not looking. A lot of modern product strategy still confuses these two things. Empty real estate is not the same thing as attention. This is essentially what every notification overload critique twenty years later was about. You can fill an unused channel with content. That does not mean anyone is going to see it. ## What Happened to the People Christopher Hassett, the founder, went on to start several other companies, including a digital media company called Activate.net in the late 1990s. The post-PointCast trajectory was quieter, and he largely receded from the consumer tech press. David Dorman, the CEO brought in to navigate the exit, moved on to become the CEO of Concert, the BT and AT&T global venture, and later the CEO of AT&T itself from 2002 to 2005. The PointCast period is a small footnote in his executive resume. The PointCast brand, after Launchpad rebranded the service, became Infogate. Infogate operated for a couple more years before shutting down entirely. The domain pointcast.com floated through several owners and currently resolves to unrelated content. The pattern is familiar. The technology lost. The people moved on. The lessons quietly entered the bloodstream of the next generation of product builders, most of whom were too young at the time to remember why nobody talked about push technology in 2001. ## Frequently Asked Questions What was PointCast? PointCast Network was free desktop software that used push technology to deliver personalized news, stock quotes, weather, and advertisements to users' computers, typically appearing as a screensaver when the machine was idle. It launched on February 13, 1996 from Sunnyvale, California, and was the most famous example of the 1990s push technology category. Who founded PointCast? PointCast was founded by Christopher R. Hassett in 1992, originally as PED Software. The company pivoted from CD-ROM-based publishing into internet-based push delivery and rebranded as PointCast Network in the lead-up to the 1996 launch. Did PointCast really turn down $450 million? Yes. In January 1997, News Corporation reportedly offered approximately $450 million to acquire PointCast. The offer was either rejected or withdrawn after a few months of negotiation, depending on which account you trust, and was definitively off the table by March 1997. How much did PointCast eventually sell for? Approximately $7 million in May 1999. The buyer was Launchpad Technologies, a San Diego company backed by Idealab. The service was rebranded and shut down by 2000. Why did PointCast fail? Several reasons compounded. The product consumed disproportionate amounts of corporate internet bandwidth and was banned by IT departments at many large companies by 1997. The push technology category lost narrative momentum after the Wired "Push!" cover became seen as overhyped. A failed IPO in 1997 eroded confidence. Internal leadership transitions slowed product decisions. And the rise of search-based pull models like Google starting in 1998 reframed the entire question of how users should find information. Was PointCast really on the cover of Wired? The March 1997 issue of Wired magazine ran a cover story titled "Push! Kiss your browser goodbye," featuring an image of a tombstone for the web. PointCast was the most prominent product cited in the cover story, although the article covered the broader push technology category. What was Microsoft's relationship with PointCast? Microsoft announced a partnership with PointCast in December 1996 to integrate push technology features into Internet Explorer 4. The resulting feature in IE4 was called Active Channels. The integration shipped, was largely ignored by users, and was quietly removed in later versions of the browser. The partnership did not result in an acquisition. How is PointCast different from modern news apps? Conceptually, very similar. Practically, very different. PointCast required a heavyweight Windows or Mac client, ran as a screensaver, and pushed content over scheduled intervals using the corporate or home internet bandwidth of 1996. Modern news apps run on phones, occupy small surfaces in the foreground, and rely on broadband and mobile networks that did not exist at consumer scale during the PointCast era. The push concept survived. The PointCast delivery model did not. Did anyone else try push technology and succeed? Indirectly, yes. Smartphone notifications, RSS readers, algorithmic feeds on Twitter, Facebook, TikTok, and others all implement variants of the push idea. The category never recovered the "push" name itself, but the underlying logic of automatically delivering personalized content to a user without active retrieval is now the dominant pattern of consumer internet consumption. Is there anything left of PointCast today? No active operations. The pointcast.com domain has changed hands multiple times. The original software does not run on modern operating systems. Some historical screenshots and a few archived press releases from the 1996 to 1999 window survive on the Wayback Machine, which is the only place most of the original product still exists. If you ever wonder why your phone buzzes 150 times a day with notifications you never explicitly asked for, the historical answer goes back to a Sunnyvale company in 1996 that bet the entire future of the internet on a screensaver. They got the future right and the screensaver wrong, and the rest of the industry spent the next 25 years quietly proving both halves of that statement. --- # How the Nintendo NES Saved Video Games After the 1983 Crash URL: https://404memoryfound.com/posts/how-nintendo-nes-saved-video-games-1983-crash.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-10 Topics: Gaming, Hardware, Business Blunders Picture this. It is the fall of 1985. You are eight years old. You are walking through the toy aisle of a Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html) in suburban America with your mom, and you stop in front of a display you have never seen before. There is a sleek gray plastic box that does not look like the Atari (https://404memoryfound.com/posts/who-owns-atari-now.html) your cousin had. There is a robot called R.O.B. with glowing red eyes. There is a plastic gun shaped like a science fiction laser pistol. And there is a TV running a demo of a tiny Italian plumber jumping over a turtle to the catchiest 8-bit music you have ever heard. You stare at it for ten minutes. You drag your mom over. She squints, reads the price tag, sighs, and says, maybe for Christmas. That is the moment Nintendo had been planning for. Because two years earlier, in the fall of 1983, the entire American video game industry had collapsed in on itself. Stores were piling unsold cartridges into landfills in New Mexico. Atari, the company that had been synonymous with the word video game, was hemorrhaging hundreds of millions of dollars and getting carved up by Warner Communications. The phrase video game had become poison at retail. And here was a Japanese playing card company that nobody in middle America had heard of, walking into that wreckage and saying, we are going to sell you a console. In a recession. After the crash. With a robot. And the wild thing is, it worked. The Nintendo Entertainment System sold somewhere around 62 million units worldwide before it was done, with roughly 34 million of those in North America. It put Nintendo at the center of pop culture for the next decade. It minted Mario as the most recognizable character in entertainment. It built the rules every console maker still operates by. And it did all of this because of a series of decisions that, at the time, looked completely insane. ## The Crash That Killed American Video Games To understand why the NES was such a long shot, you have to understand how bad 1983 really was. Most people who were not there have a vague sense that there was some kind of video game crash, that E.T. on the Atari 2600 was bad, that there were buried cartridges in the desert. The reality was much worse than that. In 1982, the American home video game market was worth about $3.2 billion. Two years later, in 1984, it was worth roughly $100 million. That is not a downturn. That is a 97 percent collapse. It is the kind of number you usually only see in stock crashes or banking panics, and it happened to an industry that had only existed for a decade. The reasons were a perfect storm. Atari had let its 2600 platform get flooded with garbage. There were no licensing controls, no quality standards, no way for stores to know what was good. By 1982 there were dozens of third-party publishers, many of them brand new and incompetent, dumping unfinished games onto retail shelves. Pac-Man on the 2600 was a notorious mess. E.T., made in five and a half weeks to hit Christmas 1982, became the punchline. Activision was producing brilliant work, but for every Pitfall there were ten knock-offs gathering dust. Then the price war hit. Retailers were stuck with mountains of unsold inventory and started slashing prices. Cartridges that had launched at $35 were getting dumped at $5. Once that happens, you cannot un-happen it. Buyers learn that if they wait six months, the price will collapse, so they wait. New releases die at full price. Publishers stop investing. Stores stop ordering. The whole engine seizes. By 1984, retailers had decided the entire category was a fad that had run its course. Toys R Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html) slashed shelf space. Department stores moved consoles to clearance bins. The trade press was openly speculating that home video games were over for good, the same way pet rocks and CB radio had been over. This is the room Nintendo walked into. ## The Famicom: A Hit in Japan, a Risky Bet Abroad Nintendo's Japanese console, the Family Computer or Famicom, had launched on July 15, 1983, just as the American market was starting to fall apart. The Famicom was designed by a small team led by Masayuki Uemura, an engineer Nintendo president Hiroshi Yamauchi had personally poached from Sharp years earlier. Yamauchi told Uemura two things. He wanted hardware that nobody else could clone for at least a year. And he wanted a low retail price. Uemura's team got there by stripping the design to the studs. The Famicom used a custom Ricoh 2A03 chip based on the MOS 6502. It used a custom picture processing unit Nintendo had designed in-house. It launched at 14,800 yen with three games, two of them Donkey Kong titles and the third a port of Popeye. It sold extremely well in its first months. By 1984, the Famicom was the best-selling console in Japan and Nintendo had a logistical problem keeping up with demand. So Yamauchi looked west. The natural move would have been to license the hardware to an American partner who already had retail relationships. Atari was the obvious choice. There was a serious negotiation with Atari in 1983 to distribute the Famicom in the United States as an Atari product. The deal fell apart, partly over Coleco showing off an unauthorized port of Donkey Kong on its Adam computer at the Consumer Electronics Show, which Yamauchi took as a sign Atari could not be trusted to protect the IP. That left Nintendo with a brutal choice. Either give up on America entirely, or do the launch themselves through their fledgling subsidiary Nintendo of America, run by Yamauchi's son-in-law Minoru Arakawa out of a warehouse in the Seattle area. Arakawa picked the second option. He had no idea how hard it was about to be. ## Selling a Dead Industry to Skeptical Retailers This is the part that gets glossed over in most retellings, and it is the most important part of the whole story. Nintendo of America did not get to walk into a Toys R Us in 1985 and say, here is our new video game console. Nobody wanted a video game console. The category was poison. Buyers at the major chains had told their reps in plain English to stop pitching them. Arakawa's team had to do something almost magical. They had to convince retailers that the gray plastic box on the desk in front of them was not, in fact, a video game console. So they renamed everything. The Famicom became the Nintendo Entertainment System, with the word Entertainment doing all the heavy lifting. The cartridges became Game Paks. The console did not have a console-style top-loading cartridge slot, even though the Famicom did. Nintendo of America insisted on a redesigned chassis with a front-loading slot that pushed the cartridge in like a VCR tape, because VCRs were the hot new home electronics category and nobody hated VCRs. The R.O.B. peripheral, the Robotic Operating Buddy with the light-up sensor and the plastic arms that picked up gyros, was almost entirely a marketing prop. R.O.B. was designed to make the NES look like an interactive robot toy that happened to have games, not a video game system that happened to have a robot. There were only two real R.O.B. games released, Gyromite and Stack-Up, and they were not particularly fun. R.O.B. existed because R.O.B. could get you onto the toy aisle planogram. The NES Zapper, the gray and orange light gun, served the same purpose. It was a toy. Toys had a category. Toys were not dead in 1985. Nintendo of America launched the NES in a single test market first. October 18, 1985, in New York City, at a handful of select retailers. The deal Arakawa had to cut to get on those shelves is now legendary. Nintendo essentially guaranteed unsold inventory. They put their own people in stores to set up the displays. They paid for in-store demo kiosks. They handled returns. They turned a sales pitch into something closer to a consignment relationship, which retailers were willing to accept because the financial risk had moved entirely onto Nintendo's balance sheet. The New York test sold a healthy share of the units Nintendo had committed to the launch by the end of the holiday season. That was not a triumph. It was, however, a survival. Bigger chains agreed to take the system in 1986. The national rollout began in early 1986 in Los Angeles, and continued through additional cities over the year. By the end of 1986, the NES was in stores across the country. By the end of 1987, it had sold roughly 1.8 million units in North America. By 1988, it was selling millions more in a single year. And by 1990, almost a third of American households with a television had an NES under it. ## Super Mario Bros and the Pack-In That Changed Everything None of the above happens without one specific cartridge. Super Mario Bros was developed by a small team at Nintendo R and D4 led by Shigeru Miyamoto and Takashi Tezuka, with Toshihiro Nakago as the lead programmer. It launched in Japan on September 13, 1985, and was bundled with the NES in North America from the 1985 test launch onward, often paired with Duck Hunt on a single cartridge in the so-called Action Set bundle that became the dominant SKU in 1988. The bundle deal was unprecedented in the American market. Most consoles before the NES had not shipped with software at all. You bought the hardware and then you bought a separate cartridge. Nintendo's pack-in strategy meant that the second you got the system home, you had a transcendent piece of software waiting for you. Super Mario Bros was so far beyond what anyone had seen on home hardware that it was almost difficult to process. The screen scrolled smoothly to the right. The sprites were enormous and recognizable. The music, written by Koji Kondo, lodged itself in your head on the first playthrough. The level design rewarded experimentation in a way that felt new. There were secret pipes that took you to underground bonus rooms. There was a hidden warp zone. There was a final boss who was not actually the final boss, because the kingdom was always in another castle. The cartridge sold roughly 40 million copies as a pack-in over the lifetime of the NES, which made it the best-selling video game in history for two decades. It was finally surpassed by Wii Sports in 2009, which was also a pack-in. The pattern is not a coincidence. If you give people the most charming software possible the moment they unbox the hardware, they tell their friends. Word of mouth in 1986 was kids in elementary school cafeterias trading rumors about the warp zone in 1-2. ## The Lockout Chip and the Iron Grip on Quality Nintendo had watched what happened to Atari and decided, in detail, never to let it happen to them. The 1983 crash had been driven in large part by an avalanche of bad third-party games that Atari had no power to stop. Nintendo's response was a piece of silicon called the 10NES, a small lockout chip that lived inside every NES console and inside every licensed Game Pak. The chip in the cartridge had to handshake with the chip in the console for the game to boot. No handshake, no game. That single design decision rewrote the rules of the console business. To get the lockout chip, you had to be a Nintendo licensee. To be a Nintendo licensee, you had to agree to terms that, at the time, sounded outrageous. You had to manufacture your cartridges through Nintendo, paying the company a per-cartridge fee on top of the underlying production cost. You had to commit to minimum order quantities that put the inventory risk back on you, not Nintendo. You had to limit the number of titles you released per year, capped at five or so for most publishers. You could not release the same title on a competing platform within a window of two years. And every game had to pass through Nintendo of America's quality assurance process, which is where the famous gold Nintendo Seal of Quality came from. This is where it all gets interesting, because Nintendo's licensing terms were genuinely heavy-handed. They got Nintendo sued by Atari Games, which set up a subsidiary called Tengen specifically to crack the 10NES chip and ship unlicensed cartridges. Tengen reverse-engineered the lockout chip, lost a lengthy legal battle, and eventually had to stop manufacturing NES cartridges. Nintendo also drew the attention of regulators, settling a price-fixing investigation in 1991 that resulted in rebate coupons sent to NES owners. And here is the thing. The licensing terms were anti-competitive. They were also, in retrospect, exactly what the industry needed. Nintendo's quality bar kept retailers convinced that the NES library was worth shelf space. Stores knew that whatever Nintendo Seal cartridge they put on the rack had passed through a real review process. The seal of approval was, functionally, a brand promise to retailers as much as to consumers. That promise is what kept Toys R Us from pulling the rug out from under video games again. ## Nintendo Power and the Cool Patrol Nintendo of America was not just selling a console. It was building a media empire around it. Nintendo Power magazine launched in July 1988 as a free issue, then converted to a paid subscription. Within a couple of years it had over a million subscribers, making it one of the largest paid magazines in the United States, period. Not just gaming. The whole magazine industry. A children's product magazine was outselling adult-targeted lifestyle publications. Before Nintendo Power, there was the Nintendo Fun Club newsletter, which had been free since 1987. And before that, there was the Nintendo Game Counselor hotline, a phone line you could call to get tips on stuck games. Kids ran up phone bills calling that hotline. Parents got the bills, and you would hear stories about hundreds of dollars of phone charges from a single weekend of trying to beat Battletoads. Then came the Nintendo World Championships in 1990, a multi-city tournament that Nintendo turned into the gaming equivalent of a Coca-Cola commercial. There was The Wizard, a 1989 film starring Fred Savage that was essentially a 100-minute advertisement for Super Mario Bros 3 and the Power Glove (https://404memoryfound.com/posts/what-happened-to-nintendo-power-glove.html). The Power Glove itself was a colossal flop as a product, but a marketing win because it was so weird that it became iconic. Nintendo even ran a kids' show on Saturday mornings, The Super Mario Bros Super Show, which ran from 1989 to 1990 with live-action segments featuring Captain Lou Albano as Mario. The show was widely considered terrible. It also reinforced, in the brain of every American child watching cartoons in the bowl-of-cereal hour, that the Nintendo characters were celebrities. By 1990, almost a third of American households with a television also had an NES under it. Nintendo had captured kids in a way no electronics company had managed since the introduction of the transistor radio. ## The Library and the Killer Apps The NES library ended up at roughly 715 licensed games in North America by the time the system was officially discontinued. The hits during the system's run kept compounding on each other. The Legend of Zelda launched in North America in August 1987 and shipped with a battery-backed gold cartridge, the first console game in the United States to let you save your progress. Metroid arrived a few months later. Mike Tyson's Punch-Out came out in 1987, with the actual world heavyweight champion's name on the box. Tetris, in its NES form, launched in late 1989 after a famously messy licensing dispute. Final Fantasy and Dragon Quest opened the door for Japanese role-playing games on American shelves, even if Final Fantasy 1 sold only modestly its first year. Castlevania, Mega Man, Contra, Duck Tales, Battletoads, the licensed Disney games from Capcom, the Sega-poaching of Capcom and Konami onto the platform. The list is long because the system kept producing them. Super Mario Bros 3 is the one most people remember as the peak. It launched in North America in February 1990 and went on to sell well over 17 million copies, making it one of the highest-grossing pieces of entertainment of any kind from that period. By 1991, Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html) had launched the Genesis in the United States with the explicit intention of breaking Nintendo's grip. The console wars that followed are their own story, but the important point is that the Genesis only got to fight Nintendo because Nintendo had already rebuilt the entire industry from scratch. Sega did not have to convince Toys R Us that video games were a real category anymore. The NES had already done that work. ## The Slow Goodbye: 1993 to 2003 The Super Nintendo launched in North America in August 1991. From that point on, the NES was technically a legacy product, but Nintendo did not pull the plug. There was a redesigned NES-101 model released in 1993 at $49.99, a cute top-loading version that looked like a pillow with a controller port. The NES kept selling new games into 1994. Kirby's Adventure, Tetris 2, the U.S. release of Mega Man 6, the cartridge port of Wario's Woods. Nintendo of America formally discontinued sales of the NES in 1995. In Japan, Nintendo continued repairing and servicing Famicom units until 2007, when the company officially ended hardware servicing for the system. That tail is a hint of what made the NES different. Most consoles get cut off the second the next generation launches. Nintendo kept supporting a system at retail for years after a successor existed because, in many American households, the NES was still the only console. Sega had a more aggressive customer churn strategy. Nintendo had, in effect, a customer base. There were elementary school kids in 1992 who got an NES instead of a Super Nintendo because their older sibling already had one and the cartridges were cheaper. ## What the NES Actually Built The NES did not just save video games. It rewrote what a video game console even meant. Almost every rule in modern console business comes from decisions Nintendo made between 1985 and 1990. The pack-in game as a launch tool. The console as a media platform with magazines, hotlines, TV shows, and licensed merchandise. The licensing program with mandatory royalties and quality control. The seal of approval as a brand promise to retailers. The first-party flagship character used to anchor every generation. The exclusivity windows that locked publishers into platform commitments. The peripheral as a marketing device, even if the peripheral itself was a flop. The tournament as a customer engagement event. The deliberate cultivation of a kid-first audience as the long-term economic base. Sony's PlayStation, Microsoft's Xbox, even Nintendo's own Switch, all of them are still operating inside the framework Nintendo built in the wreckage of the 1983 crash. When Sony decided in 1994 that PlayStation games would be manufactured by Sony, with publishers paying a per-disc royalty and quality control going through Sony, that was an NES idea. When Microsoft decided in 2002 that Xbox Live would centralize multiplayer infrastructure rather than let publishers run their own servers, that was an NES idea about controlling the platform. When the Switch launches with a Mario or Zelda title every time, that is the pack-in lesson with three more decades of polish on it. The NES was the moment video games stopped being a fad and became infrastructure. The gray plastic box, the gold-sealed cartridges, the laser gun, the robot. All of it was a Trojan horse for a 40-year industry that has not had a serious crash since. ## Frequently Asked Questions ### How many NES units were sold in total? Nintendo's official lifetime figure for the Famicom and NES combined is approximately 61.91 million units worldwide. North America accounts for the largest share at around 34 million sales, Japan accounts for about 19 million, and the rest were sold in Europe, Australia, and other regions where the system launched in the late 1980s. ### Was the NES really styled to look like a VCR? Yes. Nintendo of America explicitly redesigned the Japanese Famicom into a front-loading, gray and black chassis to dodge the toxic video game label. Buyers at chains like Toys R Us were refusing to consider new game consoles after the 1983 crash, so Nintendo positioned the NES as a piece of consumer electronics in the same general category as VCRs, which were a hot new product. The Robotic Operating Buddy and the Zapper light gun were extensions of the same strategy, designed to push the system into the toy aisle while the entertainment system framing kept it eligible for electronics shelves too. ### What was the role of the 10NES lockout chip? The 10NES was a security chip Nintendo embedded in every NES console and licensed Game Pak. It performed a handshake at boot to verify that the cartridge was officially licensed. Without that handshake, the console would refuse to play the game. The chip enforced Nintendo's strict licensing program, which required publishers to pay royalties, manufacture cartridges through Nintendo, and pass quality reviews before getting the gold Nintendo Seal of Quality. The system was challenged in court by Atari Games through its Tengen subsidiary, which reverse-engineered the chip and ultimately lost the resulting copyright lawsuit in the early 1990s. ### What were the best-selling NES games? Super Mario Bros sits at the top with around 40 million copies sold, the vast majority bundled in as a pack-in. Super Mario Bros 3 sold roughly 17 million. Other major sellers include Super Mario Bros 2, Duck Hunt, Tetris, The Legend of Zelda, Zelda II, and Mike Tyson's Punch-Out. The library overall reached approximately 715 licensed games released in North America by the end of the system's commercial life. ### When was the NES officially discontinued? Nintendo of America stopped distributing new NES hardware in 1995. The system was kept alive much longer in Japan, where Nintendo continued repairing and servicing Famicom units into the late 2000s. By that point, the NES had been on the market in some form for two decades, an extremely rare lifespan for any piece of consumer electronics. ### Did the NES really save the video game industry? The American home console market collapsed from roughly $3.2 billion in 1982 to about $100 million in 1985, a 97 percent drop driven by oversaturation, low-quality software, and the resulting retail panic. Nintendo's launch strategy and licensing controls rebuilt the category from scratch over the second half of the 1980s. By 1990, the home console business was again worth several billion dollars in the United States alone, and almost every successful console business since has used a version of the playbook Nintendo wrote during the NES era. By any reasonable read, yes, the NES saved video games as a mainstream consumer product. --- # What Happened to Standalone GPS Devices: TomTom and the Smartphone Killer URL: https://404memoryfound.com/posts/what-happened-to-standalone-gps-tomtom-garmin-magellan.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-10 Topics: Hardware, Business Blunders, Then vs Now In 2008, the global market for personal navigation devices shipped roughly 42 million units. By 2014, it had collapsed to under 18 million. Within a single product cycle, an entire category of consumer electronics had quietly stopped being a category. The brands that had defined it, Garmin and TomTom and Magellan, were either pivoting violently into other businesses or laying off staff in waves. The product had not gotten worse. It had just gotten free. Most stories about the standalone GPS device get told as a one-line obituary. The smartphone killed it. That is true, but it leaves out the much more interesting question: why was the standalone GPS market so big in the first place, and why did it die that fast? The short answer is that GPS hardware spent twenty years as a closed, expensive, slow-evolving business that depended on three accidents of timing. When those accidents reversed, the market did not slow down. It vaporized. The longer answer is a story about a US Air Force satellite system, a Korean Air Lines tragedy, a 21-second instruction signed by Bill Clinton, and a Dutch software company that bet the firm on Christmas 2004. To understand what happened, you have to start with the technology itself. ## How a Cold War Satellite System Became a Consumer Product The Global Positioning System started as a US military program in the 1970s. The first operational satellites went up in the late 1970s and early 1980s. The full constellation, 24 satellites in medium Earth orbit, did not reach initial operational capability until 1993 and full operational capability until 1995. For most of the system's first decade, it was a strictly military asset. The pivot to civilian access happened because of a tragedy. On September 1, 1983, Korean Air Lines Flight 007, a Boeing 747 carrying 269 passengers, drifted off course on a flight from Anchorage to Seoul and was shot down by a Soviet Su-15 interceptor over the Sea of Japan. The plane had wandered into Soviet airspace because of a series of navigation errors made by the crew. President Reagan, in the wake of the incident, announced on September 16, 1983, that the GPS signal would be made available to civilian aircraft once the system was operational, with the goal of preventing similar navigation accidents. That announcement set up the entire civilian GPS industry, but with a catch. The signal civilians could access was deliberately degraded. The military maintained a feature called Selective Availability, or SA, which intentionally added timing errors to the public GPS signal to prevent hostile actors from using it for precision guidance. With SA on, civilian GPS receivers were typically accurate to within about 100 meters, which is fine for telling you which highway you are on but not fine for turn-by-turn directions on city streets. The first civilian GPS receivers were industrial tools. Magellan Systems Corporation, founded in 1986, shipped the NAV 1000 in 1989. It was a brick of a handheld receiver, weighed about a pound and a half, retailed for thousands of dollars, and could pull a position fix in roughly a minute. Customers were surveyors, fishermen, hikers with money, and military and aviation buyers. The category was a low-volume professional tool business through the early 1990s. Garmin entered the market in 1989, founded by two engineers named Gary Burrell and Min Kao. The company name is a portmanteau of their first names. Garmin's pitch was simple. They could build a GPS receiver that was smaller, cheaper, and more reliable than what Magellan and the other early entrants were shipping. Their first commercial product was the GPS 100, a marine receiver that debuted in 1990, followed by the GPS 100AVD aviation panel-mount unit in 1991. The company spent most of the 1990s building specialty receivers for aviation, marine navigation, and outdoor recreation, all premium professional segments where customers would pay hundreds to thousands of dollars per unit. ## The Decision That Cracked the Market Open The GPS industry stayed niche through most of the 1990s for a single reason. Selective Availability made the consumer use case marginal. You could not give meaningful turn-by-turn driving directions to a car when your position fix was a hundred meters off. You needed accuracy on the order of meters, not the length of a city block. That changed at midnight on May 1, 2000. President Bill Clinton signed an order ending Selective Availability for civilian GPS users. The decision was driven by several factors at once. The Federal Aviation Administration was lobbying for higher accuracy because they were starting to use GPS for civilian aviation. Differential GPS techniques had largely defeated SA in practice anyway, so the strategic value was eroding. And the consumer GPS market had finally grown large enough that the Pentagon could see the economic upside of making the signal more useful. Within hours of the order taking effect, civilian GPS accuracy improved by roughly a factor of ten. Receivers that had been showing 100-meter circles of probable position were suddenly showing 10-meter circles. That is the difference between knowing you are somewhere on this block and knowing you are in the right driveway. That single technical shift is the reason the standalone GPS market exists. Before May 1, 2000, you could not reasonably build a turn-by-turn car navigator for the consumer market. After May 1, 2000, you could. Garmin, Magellan, TomTom, and a long list of smaller competitors all built their consumer businesses on top of that one administrative decision. ## TomTom and the Christmas 2004 Bet TomTom is the company that turned consumer GPS from a niche aftermarket product into a mass consumer category. The company had been founded in Amsterdam in 1991 under the name Palmtop Software, originally building business applications for handheld devices like the Psion organizer. Through the 1990s, they pivoted slowly toward navigation software, releasing TomTom Navigator in 2002 as software you ran on a Palm (https://404memoryfound.com/posts/is-palm-still-a-company.html) or Pocket PC connected to a separate Bluetooth GPS receiver. The Navigator was a bet on the smartphone-shaped future. It also was not really working. PDAs were a niche product, the user experience of pairing a Bluetooth receiver was painful, and most consumers did not own the underlying hardware. The category was bottlenecked by the customer needing to assemble the system themselves. In 2004, TomTom made the opposite bet. Instead of selling navigation software for someone else's hardware, they would build their own dedicated navigation device with the software burned in. The TomTom Go launched in March 2004 at a price that worked out to roughly €800 in the home market, with regional pricing close to £499 in the United Kingdom. It was the first mass-market all-in-one GPS car navigator. You bought one box, mounted it to your windshield with a suction cup, and it just worked. There was no pairing, no separate handheld, no software install. It was a navigation appliance. The Go sold faster than TomTom could manufacture it. By Christmas 2004, the company was the talk of every consumer electronics buyer in Europe. The product hit American shelves in 2005. By 2007, TomTom revenue had grown from €192 million in 2003 to €1.74 billion. The company went public on the Amsterdam Stock Exchange in 2005, and at peak in 2007, its market capitalization was over €5 billion. Garmin's stock, riding the same wave, peaked at over $120 a share in October 2007 with a market cap above $20 billion. This is the moment everyone in the industry felt invincible. They had a product category that did not exist five years earlier and was now selling tens of millions of units a year at retail prices between $200 and $600. The hardware had healthy gross margins. The map data licensing was contained. The user base was growing globally. By every conventional metric, this was a great business. ## The Map Data Wars To understand what happened next, you have to understand the second pillar holding up the consumer GPS market. The hardware was a commodity. The proprietary moat was the map data. Two companies, Tele Atlas and Navteq, controlled essentially all of the high-quality, fully verified vector road map data for North America and Europe. Tele Atlas was a Belgian-Dutch company, founded in 1984, that had spent two decades building a digital road network database. Navteq was an American company, founded in 1985 as Karlin and Collins and renamed Navigation Technologies, that had done the same on the other side of the Atlantic. Both companies had armies of survey vehicles, often actual cars driving every road in their coverage area, plus deals with public agencies and aerial photography vendors. The data was extraordinarily expensive to produce and basically impossible to recreate from scratch on a competitive timeline. Every consumer GPS device licensed map data from one of these two vendors. TomTom favored Tele Atlas. Garmin favored Navteq. The map updates pushed to consumers cost real money, often $50 to $100 per annual update, and the underlying license fees the device makers paid to Tele Atlas and Navteq were a meaningful part of every device's cost structure. In 2007, the slow consolidation of this layer turned into a war. On July 23, 2007, TomTom announced an agreed €2 billion bid for Tele Atlas, planning to vertically integrate the map supplier into its navigation business. Just over two months later, on October 1, 2007, Nokia (https://404memoryfound.com/posts/does-nokia-still-make-phones.html) announced it would acquire Navteq for $8.1 billion, a stunning price for a company most consumers had never heard of. Garmin briefly counter-bid for Tele Atlas in late October, forcing TomTom to raise its offer. The bidding war eventually settled at roughly €2.9 billion, and TomTom completed the Tele Atlas acquisition in mid-2008. Within a year, the two companies that controlled the world's road map data had been bought by their two biggest customers, Nokia on the device side and TomTom on the navigation side. Garmin, having lost out on Navteq, was suddenly in a bad position. Their primary map vendor was now owned by a competitor in the smartphone business. They scrambled, eventually deepening a license deal with Navteq under Nokia, but the writing was already on the wall. The map data layer was no longer a neutral utility. It was being absorbed into vertically integrated platforms. ## Google Walks In The thing that actually killed the standalone GPS market was not a hardware competitor. It was a free software product from a company that had no traditional mapping business at all five years earlier. Google Maps launched on the web in February 2005. The interactive panning and zooming behavior, which feels obvious now, was a new pattern at the time, lifted from a small startup called Where 2 Technologies that Google had acquired in 2004. Within a year, Google Maps was the dominant web-based mapping service. Google launched a mobile version, Google Maps for Mobile, in late 2005, initially as a Java app for feature phones and BlackBerry (https://404memoryfound.com/posts/what-happened-to-blackberry-smartphone.html) devices. The category-killer move came on October 28, 2009, when Google announced free turn-by-turn navigation as part of Google Maps Navigation on Android 2.0. The feature shipped on the Motorola Droid, which launched on November 6, 2009. It included spoken turn-by-turn directions, traffic-aware routing, satellite imagery, and Street View, all built on top of Google's own map data, which the company had been quietly assembling since 2008 to reduce its dependence on Tele Atlas. The day Google Maps Navigation shipped, the share prices of Garmin and TomTom dropped roughly 16 percent and 21 percent respectively. The stock market understood instantly what had just happened. Free, smartphone-based, traffic-aware navigation was now a default feature of Android. Apple would later ship its own native maps app on iPhone in 2012. The dedicated GPS device was now competing with a feature that came pre-installed on the phone in your pocket, for free, with always-on cell data and live traffic information that no offline GPS unit could match. This is not a story about Google being smarter than Garmin and TomTom. It is a story about Google having a fundamentally different business model. Google did not need navigation to be profitable. Navigation was a feature that drove smartphone adoption, which drove search queries, which drove ad revenue. Google was effectively cross-subsidizing a product category that Garmin and TomTom had to charge $300 a unit for. There is no way to win that pricing fight when your competitor is treating your entire category as a marketing expense. ## The Collapse of the Mass-Market PND The collapse of the personal navigation device, or PND, market did not happen all at once. It happened over about five years and looked, from the outside, like a textbook product cycle decline. Global PND shipments peaked at roughly 42 million units in 2008. They dropped to about 33 million in 2010. By 2014, the global figure was under 18 million units a year, and most of those sales were to commercial vehicle fleets, rental car operators, and a small population of older consumers who preferred a dedicated device. TomTom's revenue followed the curve. The company peaked at €1.74 billion in 2007, fell to €1.27 billion by 2011, and continued declining steadily. The company started layoffs in 2010 and ran successive rounds of restructuring through the 2010s. The headcount fell from over 4,000 employees at peak to roughly 2,500 by the late 2010s. Magellan was the first to break. The brand had been bought by Thales in 2001 and then sold to MiTAC International in 2008, partly to integrate with MiTAC's own Mio navigation business. Magellan kept shipping consumer GPS devices through the 2010s, but the brand was hollowed out, and the parent company eventually wound down active development on consumer devices, retreating into outdoor handhelds and OEM contracts. The dedicated GPS device was now competing with a feature that came pre-installed on the phone in your pocket, for free, with always-on cell data and live traffic information that no offline GPS unit could match. ## Why Garmin Survived and TomTom Got Smaller Of the three companies, Garmin is the one that genuinely came out the other side intact, and the reason is structural. Garmin had spent the 1990s building specialty GPS hardware for aviation, marine, and outdoor recreation customers. The car navigator business was a profitable detour for the company in the 2000s, but it was never the core business. When the consumer market collapsed, Garmin still had its aviation and marine and outdoor lines, and it had spent enough on R and D in those segments that it had moats the smartphone could not breach. The smartphone was a bad fit for a Cessna cockpit. It was a worse fit for a sport-fishing boat or a bicycle handlebar in pouring rain. Garmin's specialty hardware was rugged, had vastly better battery life, and integrated with avionics and marine instruments in ways a phone never could. Garmin also leaned into the fitness wearables market, launching the Forerunner running watches in 2003 and steadily expanding through the 2010s into the broader smartwatch category. By 2020, Garmin's fitness segment alone was generating more revenue than the company's entire business had in the early 2000s. TomTom did not have those side businesses. The company had made a focused bet on consumer car navigation, and when the consumer car navigation market disappeared, there was no obvious adjacency to retreat into. The company has spent the last decade trying to pivot into B2B services, selling mapping data and ADAS components to automotive OEMs. The bet has been partly successful. TomTom is still listed in Amsterdam, still profitable in some years, and its automotive licensing business is real. But the company that hit a €5 billion market cap in 2007 is now valued at a small fraction of that. The lesson from Garmin and TomTom is not that one team made the right decision and the other made the wrong one. It is that they started with different exposure. Garmin had insurance against the consumer collapse because it had spent the 1990s in specialty markets. TomTom had no insurance, because TomTom had bet everything on the bull case for consumer car navigation in the years before Google Maps Navigation existed. ## The Hidden Inheritance: Why Your Car Still Has a GPS Antenna The standalone GPS as a consumer category is gone. The technology, however, is now everywhere. Every smartphone has a GPS receiver. Every modern car has built-in navigation, often supplied by Garmin or TomTom in the back end through automotive licensing deals. Every smartwatch worth buying has GPS. Every commercial fleet management system, every food delivery app, every ride-share platform, every fitness tracker, every drone, every kid's smartwatch with a parent-tracking feature, every photo with location metadata, every airline app showing your flight on a map. Every one of those is a GPS use case that did not exist as a consumer product when the NAV 1000 shipped at $3,000 in 1989. The category that died was not GPS. It was the dedicated, single-purpose GPS device sitting on a dashboard. That product existed in a narrow window, roughly 2000 to 2010, when the technology was good enough for consumer turn-by-turn navigation but not yet integrated into the phone everyone was already carrying. The window was always going to close. The companies that thrived through it were the ones that recognized, early enough, that the dashboard device was a means, not the end. The other lesson, the one about platforms, is harder. Tele Atlas and Navteq spent decades building map databases that looked like an unassailable moat. Two acquisitions closed in 2007 and 2008 worth a combined $11 billion locked those moats up under Nokia and TomTom. Within five years, both moats had been outflanked by Google, which built its own map dataset largely from Street View vehicles, satellite imagery, and crowdsourced corrections from billions of Android users. The thing that looked like the most defensible asset in the entire industry, proprietary high-resolution road data, turned out to be reproducible at scale by a company with a large enough mobile install base. This is essentially the same story Apple and Spotify lived through with the music industry (https://404memoryfound.com/posts/napster-destroyed-music.html)'s licensing catalogs, that Netflix lived through with cable's content catalogs, and that any number of platforms have now lived through with mapping. When the underlying data is reproducible, the company with the larger distribution wins. The standalone GPS market was the first consumer electronics category to learn that lesson. It will not be the last. ## Frequently Asked Questions ### What was Selective Availability and why did it matter? Selective Availability, or SA, was an intentional degradation of the public GPS signal that the US military maintained from the system's earliest operational years. With SA active, civilian GPS accuracy was generally limited to about 100 meters. President Bill Clinton ordered SA turned off effective May 1, 2000, which improved civilian accuracy by roughly a factor of ten and made consumer-grade turn-by-turn navigation viable for the first time. The standalone GPS device market essentially could not exist before that order. ### When did the personal navigation device market peak? Industry analyst figures put the global peak for personal navigation devices, or PNDs, at around 42 million units shipped in 2008. By 2014, that figure had fallen to under 18 million units a year, and most of the remaining volume was to commercial fleets, rental cars, and a smaller base of consumer holdouts. The peak-to-trough decline took roughly six years. ### Why did TomTom buy Tele Atlas? TomTom announced an agreed offer for Tele Atlas in July 2007 at €2 billion, eventually completing the acquisition in mid-2008 at roughly €2.9 billion after a bidding war with Garmin. The strategic logic was vertical integration. Map data was one of the most expensive recurring costs in TomTom's bill of materials, and owning the data supplier would protect TomTom's margins and lock out competitors. Months into the process, Nokia announced its own acquisition of Navteq, the other major Western road map vendor, for $8.1 billion. With Navteq about to be controlled by a smartphone competitor, TomTom's leadership believed it had no choice but to push the Tele Atlas deal through. ### What killed the standalone GPS device? The single biggest event was Google's launch of free, smartphone-based turn-by-turn navigation on Android 2.0, announced on October 28, 2009. The feature included spoken directions, traffic-aware routing, and full integration with Google Maps, all at no cost to the user. The day of the announcement, Garmin and TomTom shares both fell sharply. Within five years, the consumer market for dedicated GPS devices had been more than halved, and the trend continued through the 2010s as smartphone navigation matured. ### Is Garmin still in business? Yes, and Garmin is one of the rare consumer hardware companies that successfully escaped a collapsing core market. The company retreated from in-car consumer GPS during the 2010s and pivoted into specialty segments where smartphones are weak substitutes, including aviation, marine, outdoor handhelds, and especially fitness wearables and smartwatches. By the 2020s, Garmin's revenue and profit had grown well past their pre-collapse highs, with fitness wearables becoming the largest single business unit. Garmin trades on the New York Stock Exchange and is generally profitable. ### What happened to Magellan? Magellan, the company that shipped the first commercial handheld GPS in 1989, was acquired by Thales in 2001 and then sold to Taiwanese electronics manufacturer MiTAC International in 2008. MiTAC integrated Magellan with its existing Mio navigation business and continued to ship consumer GPS devices through the early 2010s, but the brand quietly faded from major retail channels. Magellan-branded products today are largely focused on outdoor handhelds and dash-cam style automotive accessories, a long way from the brand's late-1990s peak. --- # What Happened to TechTV and The Screen Savers Channel URL: https://404memoryfound.com/posts/what-happened-to-techtv-screen-savers-channel.html Author: Dana Reyes (404 Memory Found) Published: 2026-05-10 Topics: Internet Culture, Hardware, Music & Entertainment Picture this: 2002. You're 14, it's a Tuesday night, and you've staked out the basement TV. Your dad is watching the news upstairs, your mom is on the phone with your aunt, and you've got the cable box dialed to channel 174 or 211 or whatever obscure number TechTV happened to live on in your zip code. The intro hits: bouncy synth, neon graphics, and a guy in a bowling shirt named Patrick Norton walking around an industrial-looking studio in San Francisco. Leo Laporte is at his desk holding a hard drive like it's a small animal. This is The Screen Savers. And for about six years, this was the closest thing American cable had to a clubhouse for the kind of kid who knew what an IRQ was. If you're old enough to remember TechTV, you remember it the way you remember discovering a really good pizza place that closed before anybody else found out about it. You remember the regret. You also remember, vaguely, that some company you'd never heard of bought the channel, turned it into something else, and then a few years later that something else was also gone. This is the story of a 24-hour cable channel that should have become CNN for nerds and instead became a cautionary tale about what happens when the people running a network do not actually like the people watching it. ## From ZDTV to TechTV: How Ziff-Davis Started a Cable Channel By the late '90s, Ziff-Davis was the most powerful tech magazine company in the United States. PC Magazine. PC/Computing. MacUser. Computer Shopper, the legendary doorstop of a monthly that you could fracture a toe on if it slid off a desk. The kind of titles you'd see stacked at a Borders (https://404memoryfound.com/posts/who-owns-borders-now-bookstore.html) newsstand thicker than a phonebook. So when Ziff-Davis decided to launch a 24-hour cable channel called ZDTV in 1998, they had a built-in audience of subscribers who already trusted the brand. The channel went live on May 11, 1998, on a tiny handful of cable systems: Las Vegas, parts of Detroit, sections of Georgia near Atlanta, parts of Maine. It was a soft launch in the most literal sense. Most of the country could not actually find it on their cable lineup. But the people who could find it were already addicted by week three. ZDTV's pitch was simple. Imagine if all those tech magazines were a TV channel. Reviews of new gear. News about Microsoft and Apple. Programming help. Internet culture coverage. And, critically, a live nightly call-in show where two guys would walk a confused viewer through reformatting their Windows 98 machine over the phone, on the air, for two hours. That was the bet. And in November 1998, Paul Allen's investment company Vulcan Ventures bet $54 million on it, taking a 33 percent stake in the network. Allen was not a casual tech investor. He had co-founded Microsoft. He understood the audience. The press release did not bother dressing it up: this was a long play, and Allen wanted in. A year later, Vulcan came back for the rest. In November 1999, Allen's people moved to acquire the remaining two-thirds. The deal closed January 21, 2000, for $204.8 million. Ziff was out. Vulcan owned the channel outright. On August 21, 2000, ZDTV got the rebrand it probably should have launched with. The name became TechTV. New logo, new sets, new ad campaign. Same shows, same hosts, same vibe. Just a cleaner package. For about three years after that, TechTV ran the best lineup of programming about technology that any television network has ever attempted. ## The Screen Savers Was the Show You can argue about which show was the channel's flagship. Tech Live had the news. Call for Help had the largest reach. Extended Play, later renamed X-Play, owned video games. But for most viewers who actually remember TechTV, The Screen Savers was the show. It premiered the same day the channel launched, May 11, 1998. The original hosts were Leo Laporte and Kate Botello. Leo had spent the early '90s as a tech radio host in San Francisco. Kate was a former PR person and stage actor who could be funny on demand. They had the right chemistry: he was the patient explainer, she was the smart skeptic asking the questions that the at-home audience was actually thinking. The show was live. That part is important. In an era when basic cable had moved almost entirely to taped programming, The Screen Savers was running in real time, two hours a night, taking phone calls from viewers and answering them with whatever was on hand. Want to hear how to install RAM in your Compaq (https://404memoryfound.com/posts/compaq-lost-pc-business-hp-merger.html) Presario? Call now. Your Outlook Express is throwing an unrecoverable error and your inbox is gone? Hold for Leo. That structure made the show feel different from anything else on cable. It was technical, but it was warm. It moved fast, but it was patient with people who didn't know what they were doing. If your grandmother was crying because she'd installed BonziBuddy (https://404memoryfound.com/posts/what-happened-to-bonzibuddy-spyware.html) and now her browser opened to a different homepage every time, The Screen Savers would walk her through removing it. They cared. In April 2000, Botello left to co-host GameSpot TV, and Patrick Norton joined Leo as co-host. Norton came over from PC Magazine, where he'd been a writer and reviews editor. The Leo and Patrick era is the version of the show most people remember. Norton was loose and funny and willing to break things on camera. The two of them together could carry an entire 7 PM hour with nothing more than a new sound card and a screwdriver. Behind them rotated a whole bench of recurring contributors. Yoshi DeHerrera was the no-nonsense overclocking guy. Kevin Rose was the segment producer who became an on-air host and would later launch Digg (https://404memoryfound.com/posts/what-happened-to-digg-social-news.html). Sumi Das did the news. Megan Morrone covered games. Cat Schwartz, who had been on ZDTV's original Internet Tonight show, was a regular face. Together they made the network feel like a real workplace where smart people were having fun and you got to listen in. ## Call for Help, X-Play, and the Rest of the Lineup The Screen Savers was the cool show. Call for Help was the necessary one. Hosted by Leo Laporte solo for most of its run, Call for Help was an hour of pure tech support television. Viewers called in with broken PCs, and Leo would talk them down off the ledge. It was so practical that some people watched it the way other people watched the morning news. You'd come home, eat dinner, turn on TechTV, and watch Leo solve other people's problems while taking notes for whatever was going wrong with your own machine. It was the original IT helpdesk YouTube channel, except you had to watch it when it aired or you missed it. Then there was Extended Play, which premiered July 4, 1998 on the original ZDTV programming block. Adam Sessler hosted from the start. Extended Play was a weekly half-hour video game review show, and for years it was the most reliable place on television to actually find out whether a game was good or bad. Sessler was sharp, opinionated, and would spike scores he didn't believe in. On April 28, 2003, the show was rebranded as X-Play, and Morgan Webb joined Sessler as co-host. The Sessler and Webb era of X-Play is the one most gamers remember, and it is also the version that managed to survive the G4 transition the longest. Other shows had their fans. Tech Live was the daily news block, the channel's CNN equivalent. Internet Tonight covered web culture before web culture was a beat. Fresh Gear reviewed gadgets the way Top Gear reviewed cars. The Lab With Leo ran briefly. There was even an interview format called Big Thinkers that hosted people like Steve Wozniak. The thing that made all of this work was that everybody on the channel cared about the subject. They were not actors hired to read tech news. They were tech people who had stumbled into television. You could feel the difference. ## The Vulcan Years and What Made Them Special Under Paul Allen's Vulcan Inc., TechTV had real money behind it. The studio sat at 650 Townsend Street in San Francisco, a renovated industrial space that you can still see in old episodes if you look closely. Big windows, exposed brick, way too many monitors. The kind of place where Linux geeks could believably hang out. The channel was not always profitable, and it never had enormous reach. By 2003, TechTV was carried in roughly 43 million American homes. That sounds like a lot, but for context, ESPN was in around 88 million. CNN was in about 88 million. TechTV was niche. It was always going to be niche. That was actually fine. The viewership was small but rabid. Forums sprang up around every show. Leo Laporte had email lists with tens of thousands of subscribers. Kevin Rose was building the audience that would eventually become Digg. The audience was the kind of audience that bought magazines, built PCs, and convinced their families to upgrade to broadband. Advertisers like Intel, AMD, NVIDIA, and Logitech understood exactly who was watching. You could even see the lifestyle creep into the gear getting reviewed on screen. Patrick Norton would unbox a new Garmin eTrex GPS unit on Fresh Gear and walk you through using it. The Screen Savers would do segments comparing the original iPod (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html) against the Creative Nomad Jukebox. Years later, when the Microsoft Zune (https://404memoryfound.com/posts/why-microsoft-zune-actually-failed.html) launched in late 2006, the channel that had been TechTV was already gone, but the heirs to that audience, on G4 and the early podcast networks, gave the Zune the only fair coverage it ever got. This was the audience corporate America did not understand. And in early 2004, corporate America came for it. ## The G4 Merger and the Long Slow Death On March 25, 2004, Vulcan announced that it was selling TechTV to Comcast. Specifically, to Comcast's gaming-focused cable channel G4. The acquisition closed on May 10, 2004. The combined network launched on May 28, 2004, under the name G4techTV. This is where it all falls apart. G4 had launched in April 2002 as a 24-hour video game channel. Same general demographic as TechTV but a totally different vibe. G4 was younger, louder, more focused on consoles and pop culture. TechTV was older, more technical, more about computers and the internet. Cramming them together was a marketing decision that ignored basically everything that made either channel work. The original plan was that G4techTV would keep both audiences. In practice, the new owners wanted G4's audience and treated the TechTV side like an annoying inheritance. The studio was shut down. Production moved from San Francisco to Los Angeles, into G4's existing facility. Most of the TechTV staff were not invited to come along. The ones who were got told their shows would be reformatted for the new audience. "They wanted what we built but not the people who built it." That was the sentiment circulating among San Francisco staffers who watched their colleagues get cut in the spring and summer of 2004. The Screen Savers got moved to LA in late 2004. Leo Laporte, after months of trying to make the new arrangement work, was let go in 2004 once the merger was finalized. Patrick Norton left around the same time. Kevin Rose became one of the new hosts of the show, alongside Alex Albrecht. The first LA based episode aired September 7, 2004. The new Screen Savers was unrecognizable. Less tech, more pop culture, more skits, more young men yelling at each other. The legacy audience hated it. The new audience never showed up in the numbers the network needed. On March 18, 2005, The Screen Savers ended for good and was replaced by Attack of the Show, a magazine-format pop culture program with new hosts and a new tone. X-Play survived because video games were always going to live on a gaming channel. Sessler and Webb actually got more polished air time. But everything else from the TechTV era was gone within 12 months. On February 15, 2005, the network dropped the TechTV part of the name entirely. It was just G4 from then on. The brand TechTV, which had been built across two ownership changes and seven years of programming, was deleted from the channel logo and never came back. Less than a year. That's all it took. ## After TechTV: Leo, Kevin, Patrick, and the Podcast Generation Here's the part that almost no other cable channel obituary can claim. Almost everybody who got fired or pushed out in 2004 went on to do something more interesting than what they had been doing on TV. Leo Laporte launched the TWiT podcast network, originally short for This Week in Tech, in April 2005. It became one of the longest running podcast networks in the world. He recorded the first episodes essentially out of his cottage. By the late 2010s, TWiT was producing dozens of weekly shows out of a custom studio in Petaluma, California. Leo had basically rebuilt TechTV, except he owned it, and the audience came with him. Patrick Norton co-hosted DL.TV with Robert Heron, then Tekzilla on Revision3, then later collaborated with Leo on TWiT shows. He has been on YouTube, podcasts, and live streams ever since. Kevin Rose was a co-founder of Digg, which launched publicly on December 5, 2004, while still ostensibly hosting on G4. By 2008 Digg was being valued at well over $100 million and was widely seen as the front page of the internet, until Reddit ate its lunch a couple of years later. He also launched Pownce, Revision3, and a string of other ventures, eventually becoming a partner at Google Ventures. Not bad for the guy who used to demonstrate hard drives on a basement TV channel. Adam Sessler stayed at G4 for years before leaving in 2012. He has run gaming review channels and worked in the games industry ever since. Morgan Webb went on to host games coverage for various outlets and is now mostly active in voiceover and gaming streams. Cat Schwartz, Sumi Das, Yoshi DeHerrera, Megan Morrone, and most of the rest of the bench scattered across the early 2000s tech media world. Several of them landed at the same podcast networks the others built. The pattern is hard to miss. The audience was real. The talent was real. The only thing that did not survive the merger was the network itself. ## Why TechTV Still Matters Think about what this channel did, day after day, for six years. It taught a generation of people how to install RAM. It explained broadband. It walked viewers through their first home network. It demystified how computers actually worked at a moment when computers were going from being a tool a few people used at the office to being a thing that lived in everybody's house. There has not been anything like it since. Not on television. Not really online either. YouTube has thousands of tech channels, and a handful are excellent. But TechTV had something that YouTube does not. It was a place. A schedule. A nightly habit. You knew that at 7 PM Pacific, Leo and Patrick were going to be on, live, and they were going to take calls. You knew that on Friday night X-Play was going to review whatever just came out. You knew where to find the news at 5. The shape of a TV schedule made the channel feel inhabited in a way an algorithmic feed cannot. This is the part that hits when you watch old episodes today on the Internet Archive. Not the content. The pacing. The fact that this was real time television about technology, made by people who liked you and assumed you were smart, broadcast every night for six years until somebody decided that pop culture skits would draw a younger demographic. The Dreamcast deserved better. The Nokia 3310 deserved better. TechTV belongs on that same list. The channel was built for an audience that turned out to be much bigger than anyone running it could see, and by the time anyone realized it, the audience had taken its talent and gone to make podcasts. If you ever called in to The Screen Savers, you remember it. If you watched X-Play religiously every Friday, you remember the music sting and the way Sessler delivered a 1 out of 5 like a personal insult. If you had Call for Help on while you ate dinner, you remember the relief of hearing somebody on cable television tell you that, no, your hard drive crashing was not your fault. That's the thing about a great cable channel. You don't realize it was a city until they tear it down. ## Frequently Asked Questions When did TechTV launch and when did it shut down? ZDTV launched May 11, 1998, was rebranded TechTV on August 21, 2000, and was acquired by Comcast's G4 on May 10, 2004. The combined channel was called G4techTV until February 15, 2005, when the TechTV name was dropped entirely from the channel's branding. Who owned TechTV? ZDTV started under Ziff-Davis. Vulcan Ventures, Paul Allen's holding company, took a 33 percent stake in November 1998 and bought the rest on January 21, 2000, for $204.8 million. Vulcan sold the channel to Comcast's G4 in 2004. When did The Screen Savers end? The original San Francisco run of The Screen Savers ended in 2004 after the move to Los Angeles. The Los Angeles version, hosted by Kevin Rose and Alex Albrecht, ran from September 7, 2004 to March 18, 2005, when it was replaced by Attack of the Show. Leo Laporte revived the format as The New Screen Savers on TWiT from 2015 to 2018. Who were the hosts of The Screen Savers? The original hosts were Leo Laporte and Kate Botello, from May 1998 to April 2000. Patrick Norton replaced Botello in April 2000 and stayed through the network's closure. Kevin Rose and Alex Albrecht hosted the LA version from 2004 to 2005. Many other on-air contributors rotated through the show, including Yoshi DeHerrera, Sumi Das, Megan Morrone, and Cat Schwartz. What happened to Leo Laporte? Leo was let go in 2004 after the G4 merger. He launched the TWiT podcast network in April 2005, and it grew into one of the largest independent podcast networks in the world. He revived the Screen Savers format on TWiT from 2015 to 2018. Did G4 ever come back? G4 was effectively wound down for new programming by 2014, and Comcast removed the channel from its Xfinity systems on January 6, 2014. A revived G4 launched in 2021 as a streaming and limited cable channel, but it shut down again in 2022. Where can I watch old TechTV episodes today? The Internet Archive has saved many episodes of The Screen Savers, Call for Help, and X-Play. Fan uploads on YouTube cover most of what is left. There is no official streaming archive of the TechTV catalog. --- # What Happened to Wine.com and the Internet Wine Wars URL: https://404memoryfound.com/posts/what-happened-to-wine-com-internet-wine-wars.html Author: Dana Reyes (404 Memory Found) Published: 2026-05-10 Topics: Business Blunders, Money & Tech, Internet Culture In January 1995, on a Tuesday afternoon, a customer in California ordered a bottle of wine from a website. The site was called Virtual Vineyards. The wine was a bottle from a small Napa producer. The transaction was processed essentially by hand by the founders, Robert Olson and Peter Granoff, in a Los Altos office that doubled as a startup workspace. That sale, on January 24, 1995, is widely cited as the first wine ever sold on the U.S. internet. It was also the start of one of the most expensive misfires of the entire dot-com era. Over the next six years, three companies would burn through more than $200 million trying to sell wine online. They would acquire each other. They would merge. They would file for bankruptcy. By April 2001, two of the three brands were effectively dead, and the third, an Oregon startup called eVineyard, would walk away with the only thing that mattered: the wine.com domain name. This is the story of how three companies tried to do the same thing at the same time, why they all failed for the same reason, and how a quieter operator with less money ended up owning the category. ## The Original: Virtual Vineyards Goes Live Virtual Vineyards launched in 1995 in Los Altos, California. The company was founded by Robert Olson, who had recently left a job at Silicon Graphics where he had managed a marketing team building software for interactive television, and Peter Granoff, his brother in law and the 15th American to earn the title of Master Sommelier in 1991. They were joined by information architect Harry Max. The business model was novel for its time. Virtual Vineyards was not a wholesaler with a website. It was a digital wine merchant. Granoff hand selected small wineries that could not afford national distribution. The site offered tasting notes, food pairing suggestions, and Granoff's own assessments. It treated visitors like wine buyers, not browsers. The first sale, on January 24, 1995, was processed essentially by hand. By the end of 1995, the operation was small but real, with about a dozen wineries on the platform. Look, Virtual Vineyards was not solving an obvious consumer problem. Most American wine buyers in 1995 were perfectly happy walking into Trader Joe's. The company was solving a discovery problem. If you wanted to find a 200 case Pinot Noir from a Sonoma vineyard with no distribution east of San Jose, the internet was the only way that producer was ever going to reach you. That niche, small wineries finding distant buyers, turned out to be real. Virtual Vineyards grew steadily through 1996 and 1997. The company was profiled in Inc. magazine in June 1996 as a textbook example of a working internet retail business. It was, briefly, the proof of concept everybody was waiting for. Then the capital arrived, and everything changed. ## The Capital Arrives, and So Does the Competition The first sign that the wine category was about to get crowded came in May 1998, when an entrepreneur named Peter Sisson founded WineShopper.com. Sisson had a different theory of the case than Granoff and Olson. Where Virtual Vineyards focused on small producers and educated buyers, WineShopper.com would be the Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html) of wine. Vast inventory. Aggressive pricing. National scale. WineShopper.com raised $46 million from a roster that read like a Silicon Valley dream sheet: Kleiner Perkins Caufield and Byers, Amazon.com itself, and a handful of strategic investors who saw wine as a high margin category waiting for an Amazon style shakeup. That was problem number one for Virtual Vineyards. They had been in the market for three and a half years and had raised a fraction of that. In 1999, Virtual Vineyards used its own venture funding to buy the Wine.com domain from another startup that had failed to make a business out of it. Reports at the time pegged the price at over $10 million for the URL alone. The Virtual Vineyards site was renamed Wine.com. The brand they had built since 1995, gone in a press release. There was logic in the move. The wine.com domain was the most valuable URL in the category, and it was newly available because the original holders had run out of cash. Spending $10 million to lock in a category-defining URL was, in 1999 dollar terms, normal. Pets.com (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html) had done the same kind of math earlier the same year. Here's the thing. Renaming yourself after a URL works only if the URL becomes the thing buyers search for. In 2000, Americans were not yet searching for wine online in any serious volume. The category was being created, not captured. Meanwhile a third player entered the market. eVineyard launched in late 1999, headquartered in Portland, Oregon. The company took a different approach. eVineyard would not chase a national brand. It would focus on operational efficiency, regulatory compliance, and getting the boring stuff right. By the start of 2000, three companies were all competing to dominate online wine sales. Wine.com, formerly Virtual Vineyards, was the brand. WineShopper.com had the capital. eVineyard had the discipline. Two of them were about to disappear. ## The Three Tier System Was Always Going to Win Here is the part that nobody on television talked about in 1999. The reason wine is hard to sell online in the United States is not technical. It is legal. After Prohibition was repealed in 1933 by the 21st Amendment, every state was given the authority to regulate alcohol sales within its own borders. Most states adopted what is called the three tier system. Producers can sell only to licensed wholesalers. Wholesalers sell to licensed retailers. Retailers sell to consumers. The chain cannot be skipped. For an online wine retailer, the three tier system is a brick wall. If you are an online retailer in California and you want to ship a bottle of Cabernet to a customer in Pennsylvania, you have to be a licensed retailer in Pennsylvania, and Pennsylvania may require you to use a licensed wholesaler in Pennsylvania, and the customer may have to buy through a state run liquor store anyway. Multiply that across 50 states and the District of Columbia. Some states allowed direct shipping. Some states banned it outright. Some states allowed it for in state producers but not out of state retailers. A few states made it a felony to ship wine to a consumer. In 1999, the year all the wine dot coms were burning the most cash, only a handful of states allowed any meaningful direct to consumer wine shipping under reciprocal arrangements. The rest either banned it or required licensing layers that were economically unworkable for an out of state retailer. The pitch deck math assumed that the entire United States was the addressable market. The legal reality was that the actually shippable market was significantly smaller, and that footprint shifted state by state with every legislative session. This is essentially what TikTok shop sellers ran into 25 years later when they tried to ship CBD products and ammunition. Some categories are not federally regulated. They are state by state political artifacts. You cannot scale them at internet speeds because the legal infrastructure does not exist. The wine startups in 1999 either knew this and chose to ignore it, or did not understand it. Either way, the bill came due. ## The Merger That Made Things Worse By the second half of 2000, both Wine.com and WineShopper.com were burning cash faster than they could replace it. Their respective venture rounds had been raised in the dot com peak of 1999, and the public market windows that would have produced an IPO had slammed shut by April 2000. In November 2000, the two companies announced a merger. The combined entity would keep the Wine.com brand. WineShopper.com founder Peter Sisson framed the deal as bringing together complementary strengths. Wine.com CEO Bill Newlands estimated that the combined company would be able to serve up to 95 percent of the U.S. market by the end of 2001. That number was aspirational. The 95 percent assumed that state laws would cooperate. They did not. The merger closed in 2001. The combined company kept burning cash. Layoffs followed. By spring 2001, the new Wine.com was insolvent. In April 2001, eVineyard stepped in. The Oregon company acquired the Wine.com brand and assets out of bankruptcy. The terms were not disclosed publicly, but eVineyard's leadership made clear that they had not assumed Wine.com's $17 million in liabilities. eVineyard president Larry Gerhard told reporters at the time that they "didn't pay anywhere close to $17 million." Industry coverage described the actual purchase price as a small fraction of the more than $200 million collectively raised by Virtual Vineyards, WineShopper.com, and the original Wine.com. That was the entire dot com wine boom in one transaction. Hundreds of millions in venture funding, marketing campaigns, lobbying budgets, and customer acquisition costs, sold for less than the cost of a couple of Super Bowl ads. eVineyard kept the wine.com domain, the customer database, and the brand. The rest of the wreckage was abandoned. ## Why eVineyard Won The reason eVineyard survived comes down to a small number of choices. First, eVineyard had less capital and was forced to be more efficient. Wine.com and WineShopper.com had raised so much money that they could afford to staff up before they had figured out unit economics. eVineyard's investors gave them less rope, and as a result they hired slower, advertised more conservatively, and focused on volume per active customer rather than raw audience size. Second, eVineyard had built its operations around the legal patchwork from day one. Rather than treating the three tier system as friction to be overcome, the company treated it as the design constraint. eVineyard developed licensed warehouse partnerships in multiple states and routed orders to whichever facility could legally fulfill the customer's address. It was less glamorous than a coast to coast distribution model, but it was actually shippable. Third, eVineyard avoided the brand obsession. The company was not trying to become the Amazon of wine. It was trying to become a wine retailer that happened to be online. The difference shows up in how it spent money. Less on television, more on customer service. Less on URL acquisition, more on warehouse logistics. When eVineyard inherited the wine.com domain in 2001, it kept its operating philosophy and adopted the better URL. The result, two decades later, is a company that quietly does several hundred million dollars a year in revenue, depending on the year, and is still the largest online wine retailer in the country. "They didn't pay anywhere close to $17 million." That was eVineyard president Larry Gerhard, asked about the price his company paid to inherit the entire Wine.com brand and customer base in April 2001. There is a lesson here that every category teaches. The first mover does not always win. The best capitalized does not always win. The company that builds for the legal and operational reality of the category is the one that ends up still standing. ## What the Wine Wars Tell Us About Dot Com Era Investing Look at the trajectory. Virtual Vineyards, the most thoughtful and founder driven of the three, lost the brand it had spent four years building because the URL it bought had more equity than the name it had earned. WineShopper.com, the best capitalized, lost because it tried to scale a regulated category at unregulated speed. eVineyard, the least visible, won because it adapted to the actual market. The wine wars are a small case study, but they happen to map onto the entire 1999 to 2001 dot com period almost cleanly. Pets.com raised hundreds of millions and lost it all because the unit economics of shipping pet food never made sense. Webvan (https://404memoryfound.com/posts/what-happened-to-webvan-grocery-delivery.html) raised over $800 million between 1999 and 2001 and lost it all because the warehouse infrastructure required to do same day grocery delivery in 1999 simply did not exist outside of a handful of dense urban markets. Boo.com (https://404memoryfound.com/posts/what-happened-to-boo-com-fashion-dot-com.html) raised $135 million and burned it on television advertising for a fashion site that took 30 seconds to load on dial up. Wine.com fits the same shape. The capital was real. The product was real. The customer demand was real. The infrastructure to do the thing economically did not exist yet, and the regulatory environment refused to bend. The investors who burned the most money were the ones who assumed the internet would force markets to change. The companies that survived were the ones that built around the markets that already existed. That is essentially the same lesson that played out twenty years later with cannabis dot coms, ride share startups in regulated jurisdictions, and the first wave of crypto exchanges. The ones who treated the law as a problem to be overcome lost. The ones who treated the law as a design parameter survived. It is not a glamorous lesson. It is also the only lesson the dot com era reliably teaches. The flipside also holds. Whenever a venture round pours capital into a category before the operating model has been proven, the math will eventually catch up. WineShopper.com's $46 million round looked, on paper, like a reasonable bet on a clear category leader. The reality, by the time the books closed in 2001, was that the company had spent a meaningful portion of that capital fighting a regulatory environment that did not negotiate. The dollars were not wasted in any single decision. They were wasted across hundreds of small ones, each defensible at the time, none of which addressed the structural ceiling on how big the company could realistically grow. ## Where Wine.com Is Today Wine.com under eVineyard's ownership has been a quietly profitable operation for most of the last 25 years. The company is headquartered in San Francisco. It is the largest online wine retailer in the United States by revenue. It does not disclose annual sales publicly, but industry estimates have placed revenue in the low to mid hundreds of millions in strong years. The company expanded shipping coverage as state laws softened, particularly after the 2005 Supreme Court decision in Granholm v. Heald, which struck down state laws that allowed in state wineries to ship directly to consumers while banning out of state shipments. That decision opened up several previously closed states. The three tier system did not disappear, but the most extreme barriers came down. Wine.com today operates with a model that looks more like the original Virtual Vineyards thesis than the WineShopper.com one. It curates. It carries small producers. It writes tasting notes. It has a master sommelier on staff. It charges premium prices and competes on selection rather than discount. Peter Granoff, one of Virtual Vineyards' two founders, eventually returned to wine retail as a partner in Oxbow Wine Merchants in Napa. He continues to teach Master Sommelier candidates. Robert Olson moved on to other startups in the early 2000s. The original Wine.com brand survived. Just barely. The companies that built it did not. The contrast between the original Wine.com and the eVineyard inheritor is worth sitting with. The first version of Wine.com tried to win the category with funding, marketing, and a category defining URL. The second version won by working through the actual fulfillment problem, state by state, license by license, until the operation was something a wine buyer in Ohio could reliably use without thinking about the legal apparatus underneath. The customer never sees the three tier system. That is the entire trick. Most internet retail looks easy in retrospect. Most of it was hard at the time, and the hard parts were almost never the parts the founders pitched investors on. Wine.com is a small example, but it is a clean one. The lesson is not that internet retail does not work. It is that internet retail works only after somebody has done the unglamorous work of mapping the real world rules. ## Frequently Asked Questions Who actually invented online wine sales? Virtual Vineyards, founded in Los Altos, California by Robert Olson, Peter Granoff, and Harry Max, sold the first bottle of wine on the U.S. internet on January 24, 1995. The company was renamed Wine.com in 1999 after acquiring the URL. How much money did the wine dot com era burn through? Combined funding for Virtual Vineyards / Wine.com, WineShopper.com, and eVineyard's pre 2001 expansion exceeded $200 million in 1999 to 2001 dollars. WineShopper.com alone raised approximately $46 million from Kleiner Perkins, Amazon, and other investors. The Wine.com brand was eventually sold out of bankruptcy in April 2001 to eVineyard for an undisclosed amount widely reported as a small fraction of total invested capital. What was the three tier system and why did it matter? The three tier system is the post Prohibition U.S. structure that requires alcohol producers to sell to wholesalers, who sell to retailers, who sell to consumers. States set their own rules within that framework. In 1999, only a small minority of states allowed any meaningful direct to consumer wine shipping. The result was that the addressable market for online wine retailers was significantly smaller than the dot com era pitch decks assumed. How is Wine.com doing today? Wine.com, owned by what was originally eVineyard, is the largest online wine retailer in the United States. Industry estimates have placed annual revenue in the low to mid hundreds of millions. The company is privately held and does not disclose detailed financials. Did the 2005 Granholm v. Heald decision fix the legal problems? Partially. The Supreme Court's ruling in Granholm v. Heald struck down state laws that discriminated between in state and out of state wineries on direct shipping. It did not eliminate the three tier system. State by state restrictions still apply, though most states allow some form of direct to consumer wine shipping today. What happened to Peter Sisson and the WineShopper.com team? Peter Sisson moved on after the merger and has worked across various technology and consumer ventures. WineShopper.com's brand effectively ceased to exist after the 2001 merger and bankruptcy. Who founded WineShopper.com? WineShopper.com was founded in May 1998 by Peter Sisson. It raised about $46 million from Kleiner Perkins Caufield and Byers, Amazon.com, and other strategic investors before merging with Wine.com in 2000 and 2001. --- # What Happened to the Nokia 3310, the Phone That Refused to Die URL: https://404memoryfound.com/posts/what-happened-to-nokia-3310.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-08 Topics: Hardware, Then vs Now Picture this. It is the autumn of 2000, and you are standing in a Cingular Wireless kiosk at the mall, holding a Nokia (https://404memoryfound.com/posts/does-nokia-still-make-phones.html) 3310 in your palm for the first time. The thing is hilariously small by year 2000 standards. It feels solid in a way that feels expensive but also somehow casual. The screen is monochrome, the keypad has these chunky tactile buttons that click when you press them, and the back is some swappable plastic shell in a color you actually picked out. A salesperson is telling you the battery lasts a week. You are pretty sure he is exaggerating, but he is not. Nothing about this device looks like the future. Everything about this device feels indestructible. Twenty-six years later, that exact same phone is still a meme. Still in drawers. Still in headlines whenever someone drops one and it survives a fall that would have shattered an iPhone into glittering pieces. The Nokia 3310 sold 126 million units, became the punchline of an internet joke that refuses to die, and got resurrected from the grave in 2017 because too many people would not let it go. This is the story of how a small Finnish company built a phone so good it accidentally turned into a cultural symbol, and then watched that exact symbol outlive the company itself. ## Finland in 2000: When Nokia Owned the World To understand the Nokia 3310, you have to understand what Nokia actually was in the year 2000. This was not some plucky underdog. This was the absolute peak of the company. Nokia held about 30 percent of the global mobile phone market that year, almost twice the share of its nearest competitor Motorola. At its peak, Nokia represented roughly 4 percent of Finland's entire GDP, 21 percent of total Finnish exports, and 70 percent of the market capitalization of the Helsinki stock exchange. Let that sink in for a second. One company. One small Northern European country. The kind of dominance you usually only see in textbook chapters about Standard Oil. And the wild part? They were not famous for any one phone. They were famous for making phones that just worked. Reliable. Practical. The phone in your dad's coat pocket was a Nokia. The phone your aunt finally caved and bought was a Nokia. The phone every European kid was begging their parents for during high school was a Nokia. They had the supply chain locked down, the carrier relationships dialed, the hardware quality tuned to a level nobody else could match. The 3310 was supposed to be a follow-up. Nothing more. The Nokia 3210 had launched in 1999 and absolutely crushed it, selling something like 160 million units over its lifetime. The 3310 was meant to be the iterative refresh. A little smaller, a little better screen, a little more polish, swappable covers, a slightly different vibe. Nokia announced it on September 1, 2000, and shipped it to retail in the fourth quarter. Retail price across Western Europe was somewhere in the 129 to 159 euro range, which translated to roughly $120 to $150 in American money at the time. Cheap enough to be an upgrade. Expensive enough to feel real. And here is the thing that was different about it: it felt good in your hand. That sounds like marketing fluff but it is not. The 3310 was 113 grams. It fit in a pocket. The keypad had this satisfying click that you could feel through your thumb. The corners were rounded just enough to not jab into your thigh when you sat down. Whoever designed this phone was thinking about how you actually carried it, not just how it looked on a shelf. ## The Indestructibility Was Not a Marketing Claim Here is where the legend starts. The Nokia 3310 was not advertised as indestructible. Nokia did not run ads claiming it could survive a fall from a building. The whole "indestructible Nokia" thing was a folk legend that grew up over a decade because too many people had the same experience. They dropped the phone. The phone did not break. They dropped it again. The phone still did not break. They put it through the laundry, sat on it, ran it over with a bicycle, threw it across the room in frustration over a lost game of Snake. And the phone just kept working. The reason for this is not magic. It is just good engineering and a complete absence of fragile components. The 3310 had no glass screen. The display was a tiny 84 by 48 pixel monochrome panel buried under a thick plastic window. There was no touchscreen to crack. No camera lens to scratch. The internal components were soldered to a stiff motherboard that fit inside a chassis with almost no air gaps, which meant nothing rattled loose when you dropped it. The plastic shell was thick. The battery was a brick. The whole device was basically a small block of polymer with some electronics in the middle. Compare that to a modern phone, which is essentially a sandwich of two glass sheets glued together with a slab of aluminum in the middle and a battery designed to bend if you breathe on it wrong. The 3310 had nothing to break because Nokia engineers had built it for people who would drop it, not for people who would baby it. They expected real human use, including the kind of real human use where you fling it at a wall. ## Snake II: The Mobile Game Before Mobile Gaming If you owned a 3310, you played Snake II. There is no version of this story where you did not. The phone shipped with four games, technically. Pairs II was a memory match thing. Space Impact was a side-scrolling shooter. Bantumi was a digital version of an African strategy game most Americans had never heard of. But Snake II was the one. Snake II was the reason classes ran long and meetings dragged and people missed bus stops. The original Snake had been on the Nokia 6110 since 1997. Snake II added more features. You had walls you could pass through. You had different mazes. You had bonus items that gave you extra points. You had four difficulty levels. The graphics were the same chunky black-on-grey blocks, but the gameplay was tighter. And because the 3310's screen had slightly better resolution than its predecessors, the game felt smoother. If you were a teenager in 2001 and you got a 3310, you spent the first two weeks of ownership trying to beat your high score on Snake II. Then you spent the next year refusing to give your phone to anyone else because you knew they would mess up your save. What Snake II actually did, beyond being addictive, was prove that mobile gaming was a real thing people wanted. This was years before the App Store. Years before mobile gaming was a market category anyone took seriously. Snake II was a game that came on your phone for free, and millions of people played it for hundreds of hours. The lesson Nokia could have taken from this was that mobile gaming was about to be massive. The lesson Nokia actually took was that they should keep including Snake II on phones. Which is, honestly, an extremely Nokia move. ## Xpress-on Covers and the Customization Thing One of the genuinely smart moves with the 3310 was the Xpress-on covers. The phone was designed so the front and back plastic shells could be swapped out by the user. You popped the back off to access the battery, and you could pop the front off too. This meant Nokia could sell you a new color shell for $20 instead of a new phone for $200, and it meant third-party manufacturers could sell you covers in literally any design imaginable. And they did. By 2002, you could walk into a phone shop or a mall kiosk and buy a 3310 cover with the Tasmanian Devil on it, or a chrome metallic finish, or a faux-wood texture, or your favorite soccer team's logo, or a pattern of cartoon flames. It was the original phone case industry, and it was massive. People who could not afford a new phone could still feel like they had a new phone every few months. Teens who wanted their phone to feel like theirs could mod it without voiding the warranty. The customization layer turned the 3310 from a device into something more like a fashion accessory you happened to make calls on. This was also genuinely innovative for the time. Most consumer electronics in 2000 were sealed black bricks designed to look the same forever. The 3310 said the opposite: this is yours, change it, make it weird. The lesson got lost over the next decade as smartphones moved toward sealed unibody designs, and we are only now getting back to anything close to that level of personalization through cases and skins. ## The Battery Life Was the Actual Superpower This is the part that hits hardest if you grew up on phones. The Nokia 3310 had a 900 mAh removable nickel metal hydride battery. Nokia rated it for up to 260 hours of standby and around 4.5 hours of talk time. In real-world use, where most people barely talked on the phone and mostly just had it sitting in their pocket waiting for a call, you got something like five to seven days of normal use between charges. Five days. Seven days. On a 900 mAh battery from 2000. Compare that to a modern flagship smartphone, which has a 4,000 to 5,000 mAh battery, a chip designed in a 3 nanometer process, hardware-accelerated AI, and dies after about a day of moderate use. The 3310 was not running a multi-tasking operating system. It was not powering a 6.7 inch OLED display at 120 hertz. It was running a stripped-down proprietary Nokia OS called Series 20, which existed mostly to make calls, send SMS messages, run a calculator, and play Snake. The whole point was efficiency. The chip used almost no power. The display used almost no power. The radio modem was the most power-hungry part of the phone, and it was idle 99 percent of the time. What this meant practically was that you forgot you needed to charge your phone. You charged it on Sunday night. You did not think about it again until the following Friday or Saturday when the low battery beep started. There was no anxiety about whether your phone would last the day. There was no carrying a charger to a coffee shop. There was no battery percentage living rent-free in your brain. The 3310 freed you from a problem we did not even realize we had until smartphones gave it back to us. ## SMS and the Great Unanticipated Revolution SMS messaging existed before the Nokia 3310. It had been part of the GSM standard since 1992. But it was the 3310 that made it a mass-culture phenomenon, because the 3310 was the phone everyone in your circle had. Once everyone has the same device, the network effect kicks in. And once you can fire off a 160-character message to anyone you know for the cost of a few cents, the way humans communicate fundamentally changes. The 3310 had a few SMS innovations that mattered more than people gave them credit for. It supported predictive text input via T9, which made typing roughly twice as fast as multi-tap. It supported long messages by automatically chaining multiple SMS together up to a 459-character limit. It had a "chat" view that grouped messages from the same contact into a back-and-forth conversation, which is wild to think about because that interface is now the default everywhere and the 3310 was one of the first phones to do it. You sent texts. Lots of texts. Coordinating where to meet. Asking your crush what they were doing. Confirming pizza orders. Telling your mom you were on the bus. Forwarding terrible jokes. The 3310 turned text messaging from a niche carrier feature into the default mode of communication for millions of people. The whole emotional shape of what mobile communication would become, the awkward silences, the typing-bubble anxiety, the carefully worded one-line message you sent to someone you liked, all of it can be traced back to a generation of people learning to type with their thumbs on a Nokia 3310 keypad. ## The Decline: Color Screens, Cameras, and the End of an Era So what killed the Nokia 3310? It was not one thing. It was the slow, inevitable march of feature creep that happens to every successful consumer product. Once color screens became affordable, customers wanted color screens. Once camera phones became viable, customers wanted cameras. Once polyphonic ringtones replaced monophonic ringtones, the 3310's tinny 35-tone library started to sound dated. The 3310 was a single-purpose device built when single-purpose was acceptable. The market wanted multi-purpose. Nokia did what any company would do. They iterated. The 3310 was succeeded by the Nokia 3410 in 2002, which added a color-capable display and a primitive Java runtime for downloadable games. Then came the 3510 with GPRS support for early mobile internet. Then the 3530 with a color screen. Then a parade of variants: 3315, 3320, 3330, 3350, 3360, 3390, 3395. Each one slightly different, each one targeting a slightly different market, each one trying to be a 3310 with one more feature. None of them caught on the way the original did. Nokia officially discontinued the 3310 in 2005 after roughly five years on the market, which is an absolutely insane lifespan for a consumer electronics product. By comparison, most phones today have a market life of about 18 months before they are end-of-lifed. And here is the bigger thing. By 2007, when Apple announced the iPhone, Nokia was still the largest phone maker on Earth. They had something like 50 percent of the smartphone market. They had the supply chain, the carrier deals, the brand recognition, the engineering talent. And they completely failed to see what was coming. Nokia executives reportedly looked at the iPhone and said it was a niche product that would not appeal to mainstream users. Five years later, Nokia's mobile phone division was sold to Microsoft for $7.2 billion. By 2017, Nokia was no longer in the consumer phone business at all. ## The 2017 Resurrection Here is the part nobody saw coming. In 2016, a Finnish company called HMD Global, founded by ex-Nokia executives, acquired the rights to use the Nokia brand on phones. Their first big move was not some flagship Android device. It was a deliberate, almost theatrical resurrection of the Nokia 3310. They announced it at Mobile World Congress in Barcelona on February 26, 2017, and the response was wild. Tech journalists who had not written about Nokia in a decade suddenly had something to talk about. The internet flooded with nostalgia takes. The phone became a story before it had even shipped. The new 3310 was not really the same phone. It had a 2.4 inch color screen. It had a 2 megapixel rear camera. It ran a feature phone OS called Series 30+ instead of the old Series 20. It had FM radio, an MP3 player, Bluetooth 3.0, and an Opera Mini browser that could pretend to load web pages on a 2G network. It also had Snake. Of course it had Snake. The whole point was the snake. The launch was also an old-school marketing dream. Nostalgia journalism, Twitter takes, a wave of "remember when phones were like this" articles. The phone went on sale in May 2017 in Europe and Asia, sold out in many markets within weeks, and HMD Global later released a 3G variant on October 29, 2017, to address the original 2G-only design. Nokia branded feature phones, which had basically died as a category, suddenly had a market again. HMD sold around 13.5 million Nokia feature phones in the third quarter of 2017 alone, with the 3310 driving most of the press attention. What this revival proved was something that had been hiding in plain sight. There is a real, persistent, possibly growing audience of people who do not want a smartphone. People who want to make calls and send texts and not get pulled into a notifications hellscape every time they sit down. The 3310 reissue gave that audience a product, and the product worked because the brand carried so much accumulated trust that you could essentially put a "made with feeling" sticker on a piece of plastic and people would buy it for what it represented. ## Why the Nokia 3310 Still Matters in 2026 Here is what is interesting about the 3310 now, looking back. It was not a great phone in any modern sense. The screen was tiny and monochrome. The keypad was slow. The internet connection, when it eventually got one in later models, was barely functional. You could not do almost anything with it that you can do with a $200 Android phone today. By every objective metric, the 3310 was worse than what came after. And yet. People still talk about it. People still buy them on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html) for $40 to $80 in working condition. There is an entire genre of internet meme dedicated to its toughness. The phone gets brought up in conversations about minimalism, about digital detox, about the way smartphones have changed our brains for the worse. The 3310 has become shorthand for a different relationship with technology. One where the device served you instead of the other way around. That is the part Nokia accidentally got right. They built a phone for people who needed a phone, not a lifestyle. They built it tough because life is tough. They made the battery last a week because nobody likes charging. They put Snake on it because waiting in line is boring. Every design decision was about reducing friction in your day, not adding new things to your day. And then somewhere in the years that followed, the entire industry forgot that was the point. The Nokia 3310 still matters because it is a museum piece for a road not taken. ## Frequently Asked Questions ### When was the Nokia 3310 released and how many did it sell? The Nokia 3310 was announced on September 1, 2000, and went on sale in the fourth quarter of that year. Over its production run from 2000 to 2005, it sold roughly 126 million units worldwide, making it one of the best-selling mobile phones in history. ### Was the Nokia 3310 actually indestructible? Not literally, but its reputation is mostly earned. The 3310 had no glass screen, no camera lens, a thick plastic shell, and very few internal components that could rattle loose. It survived drops, washing machines, and general abuse far better than the modern smartphones that replaced it. Nokia never marketed it as indestructible. The legend grew organically because so many users had the same experience. ### How long did the Nokia 3310 battery last? The original 3310 used a 900 mAh nickel metal hydride battery. Nokia rated it for up to 260 hours of standby and approximately 4.5 hours of talk time. In typical real-world use, most owners got five to seven days between charges, which is roughly five to seven times longer than a modern flagship smartphone. ### What games came on the Nokia 3310? The 3310 shipped with four built-in games: Snake II, Pairs II, Space Impact, and Bantumi. Snake II was by far the most popular and became a cultural touchstone for an entire generation of phone owners. It was an upgraded version of the original Snake that had launched on the Nokia 6110 in 1997. ### Why did the Nokia 3310 get discontinued? The 3310 was officially discontinued in 2005, after a roughly five-year market life that is almost unheard of for a consumer phone. The decline was driven by the rise of color screens, camera phones, polyphonic ringtones, and early mobile internet. Customers wanted devices with more features, and Nokia moved on to successors like the 3410, 3510, and 3530, none of which captured the same cultural moment. ### What was the 2017 Nokia 3310 reissue? In February 2017, HMD Global, a Finnish company that had licensed the Nokia phone brand, announced a modernized version of the 3310 at Mobile World Congress in Barcelona. The reissued phone had a 2.4 inch color screen, a 2 megapixel camera, an FM radio, an Opera Mini browser, and a redesigned version of Snake. It launched in May 2017 in Europe and Asia and sold out quickly, proving that there was a real market for simple, durable feature phones in the smartphone era. ### Can you still buy a Nokia 3310 in 2026? The 2017 HMD Global reissue and its 3G variant from later that year are still available through some retailers and on the secondary market. Original 2000 to 2005 era 3310 units are also widely available on eBay and similar marketplaces, typically priced between $40 and $80 in working condition, with collector-grade examples going for more. --- # What Happened to NCSA Mosaic, the Browser That Built the World Wide Web URL: https://404memoryfound.com/posts/what-happened-to-ncsa-mosaic-browser.html Author: Dana Reyes (404 Memory Found) Published: 2026-05-08 Topics: Internet Culture, Software & Apps In January 1993, a 22-year-old computer science student at the University of Illinois at Urbana-Champaign was earning $6.85 an hour writing code in a beige basement office. Six months later, the software he and a colleague had built would be the most-downloaded program on the early internet. Three years later, it would be at the center of the largest IPO Wall Street had ever seen for a company with no profits. And by the year 2000, almost nobody was using it anymore. The software was called Mosaic, and the student was Marc Andreessen. The story of NCSA Mosaic is the story of how a research project at a public university became, briefly, the center of gravity for the entire internet. It is also the story of how that center moved fast, fragmented faster, and ended up reshaping the next thirty years of how humans interact with information. Most people remember the browser wars of the late 1990s. Far fewer remember the browser before the war started. Mosaic was that browser. It is the missing chapter in the standard internet history. ## The World Before Mosaic To understand what Mosaic actually did, you have to understand how strange the internet was before it. By 1992, the internet was about a decade old as a public network. There was email. There were Usenet newsgroups, which were essentially threaded message boards. There was FTP for moving files around, Gopher for navigating menus of documents, and a handful of niche protocols nobody outside academia had ever touched. Tim Berners-Lee, working at the European particle physics lab CERN, had proposed a system he called the World Wide Web in 1989. By late 1990, he had built the first server and the first browser. The browser was named WorldWideWeb, ran only on his NeXT workstation, and was essentially text-only. By 1992, his system had grown into a small constellation of academic servers and a handful of browsers built by other researchers, most notably one called ViolaWWW. None of them had reached anything like a popular audience. The web in 1992 was a graduate-student tool. It was useful for sharing physics papers and lab documentation. It was not useful for almost anyone else. And the reason was not just that it was unfamiliar. It was that the available browsers were either text-only or so primitive in their visual design that they did nothing to suggest the web could be a medium for normal humans. ## Inside NCSA: The Government-Funded Sandbox The National Center for Supercomputing Applications was founded at the University of Illinois at Urbana-Champaign in 1986, funded primarily by the National Science Foundation. The mission was straightforward: make supercomputers useful. NSF had decided to fund several supercomputing centers around the country, and one of them needed to figure out how to put a friendly interface on these enormous, expensive machines so that researchers in other fields could actually use them. That mandate, "build interfaces that make hard things accessible," is what made NCSA the right place for Mosaic. The center had a culture of building software that real people could use. It had access to high-end Unix workstations. It had a steady stream of undergraduate and graduate computer science students looking for something interesting to work on. And in 1992, two of those students, Marc Andreessen and Eric Bina, started building a graphical browser for the World Wide Web. Andreessen was the project lead and the user-facing voice. Bina was the systems engineer who actually figured out how to make the rendering work. They started on the X Window System, which was the standard graphical environment on Unix workstations of the era. They called the early version xmosaic. Bina did most of the heavy lifting on the parsing engine and the rendering pipeline. Andreessen drove the user interface design and, crucially, the public communication. He was the one who would post on mailing lists, respond to user feedback, and push for features. Andreessen, at the time, was a junior earning $6.85 an hour as an NCSA programmer. He spent more time at the lab than in class. The work fit a pattern that would later become familiar in tech: a young, talented programmer who barely sleeps, working on something his bosses do not fully understand, building infrastructure that will eventually outpace the institution that paid for it. ## January 1993: The Release Mosaic version 0.5 was released for the X Window System on January 23, 1993. It was free. It was downloadable from NCSA's anonymous FTP server. And it was good enough that it started spreading immediately. The first month, NCSA recorded around 5,000 downloads. By March, that number was higher. By April, when version 1.0 was released, the team had ports underway for Mac and Windows. By the end of 1993, ports for both platforms had shipped, which was the part that really mattered. The Unix workstation market was small. The Mac and Windows market was enormous. Once Mosaic existed on the operating systems regular people actually used, the web stopped being a tool for graduate students and started becoming something else. Mosaic 1.0 was free, easy to install, and did something none of its predecessors had done: it made the web feel like a place. The graphics were small. The layouts were primitive. But you could click a link and arrive somewhere new, and that was new. What Mosaic did differently was a combination of small decisions that, taken together, mattered enormously. The interface was point-and-click in a way that earlier browsers were not. The "back" button was prominent and reliable. URLs were visible in an address bar at the top of the window. Bookmarks were a first-class feature. There was a forms interface that worked well enough to support the first interactive web pages. And there was the IMG tag. ## The IMG Tag: A Decision That Reshaped the Web On February 25, 1993, Marc Andreessen sent a message to the WWW-Talk mailing list announcing a new HTML element he and Bina had been working on. The element was called IMG, and it allowed Mosaic to display images inline with text on a web page. Earlier browsers had handled images by opening them in a separate viewer window. Mosaic showed them right where they belonged, embedded inside the document. Tim Berners-Lee was, by all accounts, hesitant about the proposal. He preferred a more general approach using extensions of existing tags rather than a new element. He told Andreessen, in effect, that he did not want to add new HTML tags until the specification had gone through a more formal review process. Andreessen's reply was not really a reply. By the time he had emailed the mailing list, Bina had already implemented IMG. The release shipped with it. The genie was out of the bottle. This is one of the most consequential design decisions in the history of the web. IMG turned web pages from documents into visual media. It made the web something a graphic designer could care about. It made personal home pages possible, because people wanted to share photos. It made news sites work, because layouts needed images. It made advertising work, because banner ads needed images. The entire visual language of the modern internet, from product photography to memes to TikTok thumbnails, traces back to a 22-year-old student deciding that Tim Berners-Lee's caution was not going to slow him down. Andreessen has been criticized for the unilateral move. Berners-Lee was right that more careful design would have produced cleaner standards. But the alternative was waiting, and waiting was the one thing nobody on the web in 1993 could afford to do. The next year saw an explosion of websites that would have been impossible without inline images. Mosaic shipped. The web grew. The standards process caught up later, when IMG was incorporated into HTML 2.0 in 1995. ## The Numbers: From Thousands to Millions By the end of 1993, downloads of Mosaic from NCSA's servers had passed one million. By mid-1994, they had passed several million. The web itself had grown from a few hundred servers in early 1993 to tens of thousands by the end of 1994. Mosaic was responsible for almost all of the growth in non-academic web traffic during that period. Other browsers existed, but Mosaic was, by a wide margin, the one most people used. This made NCSA, briefly, the most important software shop in the consumer internet. It also made NCSA's position increasingly awkward. The center was a research institution. It was funded by NSF. It was not built to support millions of users running its software on Mac and Windows machines around the world. The team behind Mosaic, originally a few students and staff engineers, was overwhelmed. They needed support, documentation, bug fixes, marketing. The university's bureaucracy was not designed to provide any of that. The other thing happening in parallel was that NCSA was claiming credit for Mosaic in ways the team itself found alienating. Press releases featured center directors rather than the engineers who had built the software. Marketing materials emphasized the institution. Andreessen, in particular, felt sidelined. By late 1993, he had graduated and moved to Silicon Valley, looking for what to do next. He did not have to look long. ## Jim Clark and the Birth of Netscape Jim Clark was, at the time, the founder of Silicon Graphics, a workstation company that had peaked and was looking for new opportunities. He had heard about Mosaic. In late 1993 and early 1994, he started reaching out to people in the Mosaic community. He met Andreessen in February 1994 at a coffee shop in Palo Alto. The meeting ran long. Within weeks, they had agreed to start a company. The company was incorporated on April 4, 1994, originally under the name Mosaic Communications Corporation. Clark put up $4 million of his own money. Kleiner Perkins joined as a venture investor. The plan was straightforward: build a better Mosaic, sell it to companies that needed to put themselves on the web, and capture the wave that was already building. The University of Illinois objected to the name. Mosaic was their trademark. After several months of negotiation, Mosaic Communications Corporation became Netscape (https://404memoryfound.com/posts/is-netscape-still-around.html) Communications Corporation on November 14, 1994. The product, originally called Mosaic Netscape, became Netscape Navigator (https://404memoryfound.com/posts/how-netscape-lost-the-browser-war.html). The first public version, Mosaic Netscape 0.9, was released on October 13, 1994. It was, to put it mildly, a hit. Within four months of release, Netscape Navigator had captured roughly three quarters of the browser market. The original Mosaic team, plus several engineers Clark had hired from elsewhere, was building features at a pace NCSA could not match. Netscape Navigator added cookies, Java support, JavaScript, frames, plugins, and a user interface that felt years ahead of anything else. By 1995, Mosaic was the project that started the browser. Netscape was the browser that everyone actually used. ## The Spyglass Deal: How Mosaic Became Internet Explorer Here is where the story gets interesting from a business perspective. NCSA, recognizing that it could not commercialize Mosaic itself, licensed the technology and trademarks to a small company called Spyglass, Inc. in 1994. Spyglass was supposed to be the official commercial path for Mosaic. They built their own browser, sometimes called Spyglass Mosaic, that they marketed to enterprise customers. It was, by most accounts, fine. Not as good as Netscape, but solid. In January 1995, Microsoft licensed Spyglass Mosaic. The deal, on paper, was structured as a $2 million upfront payment plus a per-copy royalty on every copy of the browser Microsoft sold. The browser became Internet Explorer 1.0, released in August 1995 as part of the Microsoft Plus! pack for Windows 95. The catch was that Microsoft never sold Internet Explorer. They bundled it with Windows. Free. Every copy of Windows came with a copy of Internet Explorer at no additional charge. Microsoft argued that since they were not technically selling the browser, the per-copy royalty did not apply, and they only owed Spyglass the minimum quarterly fee. Spyglass, watching their licensing revenue evaporate as Microsoft used Mosaic-derived code to compete with the very browsers their customers were buying, threatened a contractual audit in late 1996. On January 22, 1997, Microsoft settled with Spyglass for $8 million. Spyglass got a check. Microsoft got the browser. The original code, ultimately derived from work Andreessen and Bina had done at NCSA on a $6.85-an-hour student wage, became the foundation of the browser that would dominate the consumer internet for the next decade. Look, the lesson here is one of the cleanest case studies in the dangers of mismatched leverage in software licensing. Spyglass was a small company with a license that assumed a normal business model. Microsoft was a giant company that did not need to follow a normal business model because their leverage came from Windows distribution. Spyglass found out, the hard way, that licensing revenue against a per-copy royalty is meaningless if the licensee can choose to give the product away. ## The Slow Death of NCSA Mosaic NCSA officially discontinued development of Mosaic in January 1997. The team that had built it was long gone. The market had moved entirely to Netscape and Internet Explorer. The original codebase had been licensed to Spyglass, modified into Internet Explorer, and was about to enter the most consequential antitrust battle in the history of the software industry. Mosaic itself, the program, had become a historical artifact within four years of its initial release. NCSA still maintains a Mosaic archive page on its website. The original FTP server is long gone, but copies of the code float around in the usual places. Running Mosaic in 2026 is technically possible, although you have to either find an old machine, run an emulator, or compile the source against modern libraries. The web it was designed to render is also, in a sense, gone. Modern websites assume CSS, JavaScript, video, fonts, layout engines, and security models that Mosaic was not built to handle. What survives, and what matters, is the design. The basic visual model of the browser, the address bar at the top, the back button on the left, the bookmarks menu, the forward button next to back, all of it traces in a direct line back to Mosaic. Open Chrome or Safari or Edge in 2026, look at the chrome of the window, and you are looking at decisions Marc Andreessen and Eric Bina made in 1992 in a basement office in Illinois. ## The Bigger Lesson: Distribution, Not Code The real question is what Mosaic teaches us about how technology actually wins. The answer is not what most engineers want to hear. Mosaic was not the first browser. It was not the most technically sophisticated browser. The IMG tag was a hack that Tim Berners-Lee, the actual inventor of the web, did not endorse. The user interface was crude even by 1993 standards. There were better-engineered alternatives in the Unix world. What Mosaic had was distribution. It was free. It worked on the platforms people used. It was easy to install. It was promoted by an organization, NCSA, that had access to media coverage and academic networks. It launched at exactly the moment when the underlying internet had reached enough size to support a mass-market application but had not yet acquired one. The combination was unrepeatable. This is essentially what happened ten years later with Firefox, which beat Internet Explorer not on technical merit but on distribution and timing. It is what happened with Chrome, which beat Firefox by being shipped to every Google search user. It is what happens with TikTok's algorithm, which beat YouTube's algorithm not because TikTok understood human attention better but because TikTok had the right hooks at the right moment in the right form factor. The Mosaic story is the original draft of a pattern that has played out over and over since. The other lesson is about ownership. NCSA built Mosaic and lost it. Andreessen built Netscape and lost most of the upside to Microsoft, although he did become extremely wealthy in the IPO. Spyglass licensed the technology and got pennies on the dollar. Microsoft, which contributed the least to the original innovation, captured the most value because it controlled the distribution channel that mattered. That is not a moral lesson. It is just how software economics work when distribution matters more than invention. ## Why Mosaic Still Matters The reason to revisit Mosaic in 2026 is not nostalgia. It is that the patterns it established are still running the modern internet. Inline images, the back button, bookmarks, the address bar, the URL as a first-class concept, the idea that any document on the web can link to any other document, all of it was either invented or popularized by Mosaic. The engineers who built it did so quickly, without permission, in a research lab that did not fully understand what they were doing. The institution that funded the work failed to capture any of the upside. The student who led the project went on to become one of the most powerful venture capitalists in the world. If you are trying to understand why the internet looks the way it does, why browsers feel the way they feel, why the web grew so fast and then fragmented so quickly, the answer is not in the polished history that gets told about Netscape and Microsoft. It is in the eight months between January and August 1993 when a browser built by two students at a public university quietly became the front door to the entire World Wide Web. ## Frequently Asked Questions ### When was NCSA Mosaic released? The first public version, Mosaic 0.5 for the X Window System on Unix, was released on January 23, 1993. Version 1.0 followed on April 21, 1993. Ports for Macintosh and Microsoft Windows were released later in 1993, which is when adoption among non-academic users began to explode. ### Who created NCSA Mosaic? Mosaic was created at the National Center for Supercomputing Applications at the University of Illinois at Urbana-Champaign. The lead developers were Marc Andreessen, who drove the project and the user interface, and Eric Bina, who handled most of the rendering and parsing engine work. They were assisted by a small team of NCSA programmers, including Aleks Totic, Mike McCool, and Chris Wilson. ### What made Mosaic different from earlier web browsers? Mosaic introduced inline images via the IMG tag, ran on the platforms most people used (Mac and Windows in addition to Unix), had a polished point-and-click interface, and was free. Earlier browsers were either text-only, available only on niche platforms, or built primarily for academic users. Mosaic made the web feel like a consumer product. ### What is the relationship between Mosaic and Netscape? Marc Andreessen, who had led the Mosaic project at NCSA, co-founded Mosaic Communications Corporation with Jim Clark on April 4, 1994. The company was renamed Netscape Communications Corporation on November 14, 1994 after the University of Illinois objected to the use of the Mosaic name. Netscape Navigator, the company's flagship product, was a from-scratch rebuild rather than a fork of NCSA Mosaic. ### What is the relationship between Mosaic and Internet Explorer? NCSA licensed Mosaic technology to Spyglass, Inc. in 1994. Microsoft licensed Spyglass Mosaic in January 1995 for $2 million plus a per-copy royalty, then bundled the resulting browser, Internet Explorer, with Windows for free. Spyglass sued, and Microsoft settled for $8 million on January 22, 1997. The original Internet Explorer codebase was directly derived from Spyglass Mosaic, which itself was derived from Mosaic. ### When was NCSA Mosaic discontinued? NCSA officially discontinued Mosaic in January 1997. By that point, the browser market had moved entirely to Netscape Navigator and Internet Explorer. The original team had moved on, mostly to Netscape, and the market for a research-grade browser from a university lab had effectively disappeared. ### Can you still run Mosaic today? Yes, with effort. The original source code is still available, and various enthusiasts have ported Mosaic to modern operating systems. NCSA also maintains an archive page about the project. The practical problem is that the modern web assumes CSS, JavaScript, modern security protocols, and many other features that Mosaic does not support, so most contemporary websites will either fail to render or look very strange in the original browser. --- # What Happened to the Sega CD, the Add-On That Promised Gaming's Future URL: https://404memoryfound.com/posts/what-happened-to-sega-cd-add-on.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-07 Topics: Gaming, Hardware Picture this. It is the fall of 1992. You are 11 years old, walking into a Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html) at the Northridge Fashion Center, and there is a glass display case at the back of the electronics department with a black plastic slab the size of a paperback novel sitting next to a Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html) Genesis. The sticker says $299. Your dad whistles, the way dads do when they see a price they have no intention of paying. The TV above the case is playing a game called Sewer Shark, which appears to be actual video footage of an actor screaming at you from inside what looks like a sewer. You have never seen anything like it. Neither has anyone else, because this is the Sega CD, and Sega has spent the last year telling you that this little black box is the future of video games. It was not the future. It was, depending on who you ask, a brilliant misfire, an expensive footnote, or one of the most fascinating disasters in console history. By 1996 it was gone, less than four years after it landed. Sega never released official lifetime sales numbers, but the most commonly cited figure is around 2.24 million units worldwide. For context, the Genesis it bolted onto sold somewhere north of 30 million. If you owned one, you remember. The weight of it. The whir of the disc spinning up. The agonizing load times. Lunar: Silver Star on a Saturday afternoon in your friend's basement, snow falling outside, that incredible animated intro playing while you both tried to figure out what was happening. The Sega CD deserved better than the way it ended. It also, honestly, kind of brought the ending on itself. ## Why Sega Built a CD Drive in the First Place To understand why the Sega CD even exists, you have to remember where Sega was in 1990. The Genesis had launched in North America in August 1989 and was actually winning. By the end of 1991 Sega's bench was deeper than it had any right to be. Sonic the Hedgehog dropped in June of that year and turned a hardware company into a brand. Sega of America, run by Tom Kalinske, was running circles around Nintendo of America with marketing campaigns that practically dared kids to admit they still played SNES. Genesis Does What Nintendon't was on every TV during commercial breaks for The Simpsons. And here is the thing. When you are winning, the temptation is always the same. Push harder. Bigger. Get there first. Sega wanted to leapfrog Nintendo into the next generation before Nintendo even noticed the floor was moving. CD-ROM was the obvious target. Cartridges in 1991 maxed out around 8 megabits, which sounds adorable now. A CD held roughly 650 megabytes of data. That is not a typo. The math is genuinely insane. You could fit hundreds of cartridges' worth of audio, art, and video onto a single disc. Engineers at Sega had been quietly working on a CD add-on for the Mega Drive (the Japanese name for the Genesis) for several years. The pitch internally was simple. Add-on, not new console. Bolt it to the existing Genesis install base. Use the disc capacity for huge games, full motion video, CD-quality audio. Get a multi-year head start on whatever Nintendo and Sony eventually came up with. That last part is its own story, by the way. In 1988, Nintendo had partnered with Sony to develop a CD-ROM add-on for the Super NES. That partnership famously collapsed at CES 1991, which freed Sony to eventually release a console called the PlayStation. Sega's executives were watching all of that play out in real time, and the message they took from it was: get to CD first or die trying. ## December 1991 in Japan, October 1992 in America The Mega-CD launched in Japan on December 12, 1991, priced at 49,800 yen. North American gamers had to wait nearly a year. The Sega CD hit US shelves on October 15, 1992, at $299, which is around $660 in 2026 dollars (https://404memoryfound.com/posts/90s-tech-inflation-calculator.html), and that price did not include the Genesis you needed to plug it into. Sega only had about 50,000 launch units available because of production problems. They sold past 200,000 by the end of 1992 and 300,000 by July 1993. The packaging strategy was classic early-90s ambition. The original Sega CD was a tray-loading unit that sat under your Genesis like the bottom layer of a cake. There was a little lever you flipped to slide the tray out. The whole thing felt like science fiction in your hands and looked like a piece of professional A/V equipment in your TV cabinet, which is basically what your parents thought it was. The launch lineup was where the cracks started showing. The pack-in game was Sewer Shark, developed by a studio called Digital Pictures. Sewer Shark was a full motion video on-rails shooter, which is a fancy way of saying you watched grainy live-action footage of a guy yelling at you while you pressed buttons that occasionally fired a gun at things. It was, and this is being charitable, not actually fun. But it looked like nothing else, and that was the point. Other launch and early titles included Black Hole Assault, Cobra Command, and a bunch of Genesis ports with redbook audio soundtracks. The killer app, the thing that was supposed to sell hardware, never quite arrived during the launch window. Sonic CD would not show up until November 1993 in North America. By then a lot of people had already made up their minds. ## The Full Motion Video Promise Here is where the Sega CD's whole identity comes from. Full motion video. FMV. The idea that instead of pixel-art sprites bouncing around 2D backgrounds, you could play games made out of actual filmed footage of actual humans pretending to be in a game. It is hard to explain to someone who did not see it in 1992 how revolutionary this looked. Genesis games were beautiful in their own way, but they were obviously cartoons. The Sega CD let developers cut to footage of a real actor's face on your TV, and your brain registered it as a different category of entertainment entirely. For about six months, kids in arcades and Toys R Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html) aisles all over America stood and stared at FMV demo loops with their jaws somewhere around their sneakers. And here is where it all falls apart. Because the FMV that the Sega CD could actually display was a tiny, postage-stamp-sized window of compressed, blocky, deeply unimpressive video. The hardware had to decompress motion JPEG-style frames in real time, and the Genesis architecture it was tethered to could only display so many colors at once. The result, in practice, was that most FMV games looked like a smudged VHS tape playing inside a small television inside your big television. The illusion held in commercials. It usually fell apart in your living room. Digital Pictures became the main FMV studio for the platform, and they produced a string of titles that, depending on your point of view, are either fascinating cultural artifacts or some of the worst video games ever sold for money. Sewer Shark, Ground Zero Texas, Make My Video Marky Mark and the Funky Bunch edition (yes, that was a real product), and the one that became infamous: Night Trap. ## Night Trap and the Senate Hearings If you remember one thing about the Sega CD that is not the price, it is probably Night Trap. And if you do not remember Night Trap, the short version is this. It was an FMV game where you, the player, monitored security cameras in a house full of college girls being attacked by hooded figures called Augers. Your job was to trap the Augers using house security mechanisms. It was filmed in 1987 for an unreleased Hasbro VHS-based console concept, then dusted off and shipped on the Sega CD in 1992. It was campy, weirdly tame for what it was being accused of, and it became the centerpiece of a national panic over video game violence. On December 9, 1993, the United States Senate held a committee hearing led by Senators Joe Lieberman of Connecticut and Herb Kohl of Wisconsin, focused on violence in video games. The two main exhibits were Mortal Kombat and Night Trap. Lieberman publicly described Night Trap as promoting violence and sexual aggression against women. Toys R Us and Kay-Bee Toys pulled the game from shelves in December 1993. Sega stopped producing copies in January 1994. The hearings did two things at once. First, they put a target on the Sega CD's back at exactly the moment Sega needed parents to be excited about the platform, not afraid of it. Second, and this is the part that actually mattered long term, the threat of federal regulation pushed the industry to create its own ratings board. The Entertainment Software Rating Board (ESRB) launched in 1994, and that is the same ESRB whose rating is on the back of every game console title sold in North America today. So in a strange way, the most embarrassing game on the Sega CD is also the reason your local GameStop (https://404memoryfound.com/posts/is-gamestop-still-in-business.html) has age ratings on its shelves. ## The Tower of Power and the Beginning of the End By late 1993 it was already clear that something was off. Sega's executives knew Sony was working on the PlayStation. They knew the Saturn was coming. They needed to keep Genesis owners spending money in the meantime. The answer, somehow, was to layer more hardware on top of the existing stack. Sega redesigned the Sega CD in 1993 as a top-loading unit (the Sega CD model 2), made smaller and cheaper at around $229 retail. That part was actually a good idea. Then in November 1994, Sega launched the Sega 32X (https://404memoryfound.com/posts/what-happened-to-sega-32x-add-on.html), a separate add-on cartridge that plugged into the top of your Genesis and added 32-bit graphics processing. If you owned a Genesis, a Sega CD, and a 32X, you stacked them together into a teetering plastic tower with three power supplies, multiple cables, and an aesthetic that can only be described as cyberpunk garage sale. People called it the Tower of Power. Sega did not, officially, but the nickname stuck. The math on this stack was brutal. By the holiday season of 1994, a Sega fan had been asked to spend roughly $89 on a Genesis, plus $230 on a Sega CD, plus $159 on a 32X, plus games for each, in less than three years. And the Sega Saturn launch was already on the calendar. Tom Kalinske, the Sega of America CEO who had built so much of the company's American success, has said in interviews over the years that he was opposed to the 32X internally. He believed Sega should skip the add-on and put everything behind the Saturn. He lost that argument. Whether that decision alone killed the company is debatable, but it definitely killed the trust between Sega and its hardcore fans, the people who had spent their parents' money on every previous Sega product. ## The Hidden Gems This is the part that always hurts a little. Because for all the FMV nonsense and the corporate chaos, the Sega CD's library, especially in its second and third year, contained some genuinely excellent games. Forget the marketing for a second. There is Sonic CD, released in November 1993 in North America, which is one of the best 2D Sonic games ever made and contains, depending on which version you played, one of the most iconic anime intro sequences in console history. The opening cinematic was produced with the involvement of Toei Animation, the studio behind Dragon Ball Z, with the actual animation work handled by Studio Junio. That kind of production budget was unprecedented for a console game in 1993. There is Lunar: The Silver Star, a JRPG by Game Arts published in North America in 1993 by Working Designs, with a localization that fans still talk about as one of the best of the 16-bit era. There is Lunar: Eternal Blue, the sequel, in 1994. There is Snatcher, a cyberpunk noir adventure designed by Hideo Kojima, the same Kojima who would later make Metal Gear Solid. The Sega CD got the only English release of the original Snatcher for years, which is why used copies of that game now go for genuinely deranged amounts of money on eBay. There is Popful Mail, Eternal Champions: Challenge from the Dark Side, Final Fight CD, the criminally underrated Heart of the Alien, and the Sega Classics Arcade Collection. There is the much-loved Earthworm Jim Special Edition with extra levels and CD-quality audio. If you only judged the Sega CD by its hidden gems, you would walk away thinking it was a brilliant, niche, beloved platform. The problem is that nobody at the time bought hardware based on hidden gems. They bought based on what was on the front of the box at Toys R Us. And what was on the front of the box was Sewer Shark. ## Why It Actually Failed People love to say the Sega CD failed because of the games. Or because of the price. Or because of FMV. The truth is that all of those mattered, but the bigger story is structural. The Sega CD failed because it was an answer to a question Sega had asked itself, not a question consumers had asked. Consumers in 1992 were asking: I have a Genesis, what is the next great game I can put in it? Sega answered: a $299 hardware upgrade plus an entirely new game library. That is a much harder pitch. It also failed because Sega could not decide what the Sega CD was actually for. Was it a multimedia device? A way to play CD-quality music? A platform for interactive movies? A storage upgrade for traditional games? The marketing tried to be all of those things at once, which meant most parents at Sears could not articulate what the box on the shelf actually did. And it failed because the platform's lifespan got cannibalized. Sega CD games started coming out in late 1992. The Sega Saturn launched in Japan in November 1994 and in North America in May 1995. That gave the Sega CD roughly two and a half years before its parent company started openly competing with it. Compare that to the Genesis, which had years of headroom before the SNES showed up in some markets and was supported well into the mid-90s. The Sega CD never had time to grow up. ## The Long Tail and the Quiet Death By early 1995, Sega had effectively stopped advertising the Sega CD. Resources moved to the Saturn. The North American release window for major Sega CD titles wound down in 1995, with the last few stragglers limping out in 1996. There was no funeral. There was no Sega press release announcing the end. The platform just stopped getting games, then stopped getting shelf space, then stopped existing. The retail picture turned ugly fast. By the spring of 1995 you could walk into a Toys R Us and buy a Sega CD for $129, then $99, then bargain bin clearance. Software prices collapsed in parallel. The fire sale created a generation of kids and college students who bought the platform after it was already dead, played 30 weird games on it across two summers, and walked away with weirdly fond memories of a system they had only ever experienced as a clearance item. That bargain bin chapter is honestly part of why the Sega CD has the cult status it has now. People who could not afford one in 1992 picked one up in 1996 for less than dinner at Olive Garden. It became the platform you discovered, not the platform you bought. ## The Legacy Here is the strange thing. The Sega CD lost. Commercially, financially, in the public memory, in every metric that matters. And yet most of what console gaming became in the years after, in some real sense, started here. Optical media as the standard format for console games? PlayStation went all in on CDs in 1995, and within a generation cartridges were a memory. CD-quality soundtracks in games? Sega CD pioneered redbook audio in console gaming, and now nobody ships a console game without a real recorded score. Live action footage in games? FMV cutscenes never went away, they just got better, and the cinematic style of modern games like Death Stranding or the Yakuza series owes more to Sewer Shark than anyone wants to admit. Even the hardware modularity question never really died. PlayStation 2's hard drive add-on, Xbox 360's Kinect, the Wii's MotionPlus, every weird peripheral that ever shipped after 1995 is in some way a descendant of Sega's belief that you could just keep stapling things to a console. The Sega CD also, in a way that nobody at Sega meant or wanted, helped create the modern video game ratings system. Without Night Trap and the Senate hearings, the ESRB might not exist in the form it does, and the politics around game content might look very different. ## What It Felt Like None of this is what you actually remember about the Sega CD if you owned one. What you remember is the weight of the unit, the way the disc tray felt heavier than a CD player tray for some reason, the click of locking it onto your Genesis. The first time you saw the Sega CD bios screen with that floating logo and the orchestral score that played on bootup. You remember the load times, which were genuinely awful, and you remember not caring because there was something on the screen that felt like the future. You remember Lunar's opening cinematic playing in your friend's basement on a 19-inch Magnavox tube TV with the volume turned down so his mom would not yell. You remember pretending Night Trap was scary even though it was, in retrospect, more campy than scary. You remember reading GamePro previews of FMV games that turned out to be terrible but sounded amazing. The Sega CD is a story about ambition out-running execution, and a company so addicted to being first that it forgot to ask whether being first was actually winning. It is also, weirdly, a story about hope. Every Sega CD owner I have ever talked to remembers the moment they thought, watching that disc spin up for the first time, that they were holding the future. They were not. But they were not totally wrong either. ## FAQ When did the Sega CD launch in North America? The Sega CD launched in North America on October 15, 1992, at a retail price of $299. The Japanese version, called the Mega-CD, had launched in December 1991. The European launch followed in April 1993. How many Sega CD units were sold? Sega never released official lifetime sales figures, but the most commonly cited number is approximately 2.24 million units worldwide. By the end of 1992, North American sales had passed 200,000 units, and by July 1993 they had passed 300,000. Why was Night Trap so controversial? Night Trap was an FMV game shipped on the Sega CD that depicted hooded figures attacking young women in a house. It became one of the main exhibits at the December 9, 1993 United States Senate hearings on video game violence, led by Senators Joe Lieberman and Herb Kohl. Toys R Us and Kay-Bee pulled it from shelves in December 1993, and Sega stopped producing it in January 1994. The hearings ultimately led to the creation of the Entertainment Software Rating Board (ESRB) in 1994. What was the Tower of Power? The Tower of Power was the unofficial nickname for the full Sega hardware stack: a Genesis with a Sega CD attached underneath and a Sega 32X plugged into the cartridge slot on top. By the end of 1994, a fan who owned every piece had spent close to $480 in hardware before buying any games. What are the best Sega CD games? The most enduringly praised titles include Sonic CD, Lunar: The Silver Star, Lunar: Eternal Blue, Snatcher by Hideo Kojima, Popful Mail, Final Fight CD, and Earthworm Jim Special Edition. The general FMV catalog from Digital Pictures has aged poorly, but the platform's RPG and action library holds up surprisingly well. When did Sega discontinue the Sega CD? Sega began phasing out support for the Sega CD in early 1995, redirecting marketing and development resources toward the Sega Saturn. The last few first-party releases trickled out in 1995 and 1996, but there was no formal discontinuation announcement. The platform simply stopped getting new releases and shelf space. Was the Sega CD a failure? Commercially, yes. It sold a fraction of what the Genesis sold, never achieved its sales targets, and contributed to the collapse of trust between Sega and its core audience. Culturally and historically, it is more complicated. The Sega CD pioneered CD-ROM gaming on consoles, popularized redbook audio soundtracks, hosted some genuinely beloved RPGs and action games, and indirectly helped create the modern video game ratings system. It lost, but it left fingerprints on almost everything that came after. --- # What Happened to Prodigy, the Online Service That Beat AOL to America URL: https://404memoryfound.com/posts/what-happened-to-prodigy-online-service.html Author: Dana Reyes (404 Memory Found) Published: 2026-05-07 Topics: Internet Culture, Business Blunders By the end of 1990, Prodigy had 465,000 paying subscribers, the second-largest commercial online service in the United States. AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html), then transitioning out of its Quantum Computer Services days, was still much smaller, with subscriber numbers well under six figures. CompuServe (https://404memoryfound.com/posts/what-happened-to-compuserve-online-service.html), the elder statesman of the field, was sitting on about 600,000 and watching nervously. By 1993, Prodigy had passed CompuServe and was the largest online service in the country. By 1996, IBM and Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html), the two corporate parents who had spent more than a billion dollars building it, sold the entire operation for $200 million. Carlos Slim, the Mexican telecom billionaire, came in as a minority investor on the buyer's side. By 2001, the Prodigy brand had effectively been folded into SBC Communications and quietly retired. Today most people under 40 have never heard the name. The story of how that happened is more interesting than "AOL won." It is a story about what happens when two of the most powerful corporations in America build a product around what they think users should want, then refuse to listen when users tell them something different. ## Trintex: A Joint Venture That Should Not Have Existed Prodigy was incorporated on February 13, 1984 under the name Trintex. The three founding partners were CBS, IBM, and Sears, Roebuck and Company. Each contributed something the other two did not have. CBS brought media and content expertise. IBM brought hardware, networking, and engineering. Sears brought distribution, retail, and a 100-year-old relationship with American households. On paper this was a dream team. In practice it was a corporate Frankenstein. Three CEOs from three different industries trying to agree on what an online service for ordinary Americans should look like in a world where most ordinary Americans had never used a modem. CBS lost interest first. In 1986, CBS CEO Tom Wyman was selling off non-core assets, and an experimental data network for home computers fit cleanly into that bucket. CBS exited, and the venture continued with IBM and Sears as co-owners. The remaining two partners had a clear shared vision, even if it took them another four years to figure out how to ship it. The pitch internally was simple. Online services like CompuServe were aimed at engineers, hobbyists, and corporate users. They were expensive, charged by the hour, and required you to read a manual. Prodigy would be different. It would be aimed at ordinary households. Flat monthly fee. Polished graphical interface. Pre-packaged content from real publishers. And, critically, advertising. The Sears half of the partnership saw an obvious opportunity: an entire shopping channel, but built into a computer. ## The 1988 Soft Launch and the 1990 National Push Prodigy went live regionally in 1988 in three test markets: Atlanta, Hartford, and San Francisco. The pricing was $9.95 per month, flat rate. For families used to CompuServe's per-minute charges, the difference was almost difficult to comprehend. You could log on, browse for an hour, and not have to mentally calculate how much you were spending. The national launch came on September 6, 1990, supported by a marketing campaign developed by the ad agency J. Walter Thompson and its direct response sister company JWT Direct. IBM and Sears bundled Prodigy with new IBM PS/1 and PS/2 home computers, with Hayes modems, and with a long list of clones. If you bought a computer in a Sears in 1990, there was a meaningful chance Prodigy was already in the box. Within months, Prodigy had crossed half a million subscribers. By 1993, it was the largest online service in the United States. It was, by every measure of mainstream consumer reach, winning. ## The Interface That Made Sense to Sears Executives The Prodigy interface is one of the most fascinating UX artifacts of the early online era. It was built around a custom graphical client, written by IBM engineers, that ran on top of a proprietary protocol. The screen was divided into a content frame on top and an advertising banner along the bottom. The whole experience felt less like browsing and more like flipping through a digital magazine. Look at it now and the design choices read like a corporate strategy document made visible. The content frame was where Prodigy editors curated stock quotes, news headlines, sports scores, weather, encyclopedia entries, kids' content from Sesame Street, and shopping links. The bottom banner was where Sears, Coldwell Banker, JCPenney, and other partners sold things to you. The interface looked like television because the people who designed it thought of the online world as television with shopping attached. This is essentially what Pinterest's home feed and Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html)'s product recommendations would do twenty years later, just with worse graphics and a $9.95 monthly subscription. Prodigy was, in a real sense, the first major attempt to build the thing that mainstream consumer internet would eventually become. The problem was that Prodigy thought users wanted to consume what Prodigy curated. The users wanted something else. They wanted to talk to each other. ## The Bulletin Board Surprise This is the part of the Prodigy story that the founders never planned for. The service had bulletin boards, message areas where users could post and reply to each other. They were buried in the interface, treated as a small feature among many. The Sears and IBM executives saw the boards as a sideline, not a destination. The users had other ideas. By 1991, Prodigy's bulletin boards were generating volumes of messages and engagement that the company had never modeled for. People were not logging on to read curated stock quotes. They were logging on to argue about politics, to discuss parenting, to swap recipes, to flirt, to vent. The boards were where the actual product lived. The shopping mall the founders had built was almost empty. The town square next door was packed. This created a problem the company did not know how to solve. Their per-user economics had been built around a flat fee plus advertising revenue. They had assumed the average user would log on, click through some content, look at some ads, and leave. Bulletin board addicts did not behave that way. They stayed for hours. They posted hundreds of messages. They cost Prodigy real money in network usage and storage. And here the logic of the business starts to break down. ## The 1991 Email Charge Disaster In January 1991, Prodigy modified its pricing. Subscribers would still pay $9.95 a month, but they would now be limited to 30 free email messages per month. Each additional email would cost 25 cents. The reasoning, from the company's perspective, was straightforward. Email was being used at volumes nobody had projected. The infrastructure was straining. The shopping advertising model was not throwing off enough revenue to subsidize unlimited communication. Charge for email past 30 messages, recover some of the costs, and steer users toward the parts of the service that actually paid the bills. From the user's perspective this looked, accurately, like a betrayal. People had subscribed to Prodigy under the belief that the flat $9.95 fee covered email. Suddenly it did not. Some subscribers organized in protest, formed user groups, posted angry messages on the bulletin boards, and began contacting Prodigy advertisers directly to complain. Prodigy's response was to suspend the accounts of users coordinating the protest, which made the original story much worse. The company eventually backed down on the per-email fee, but the damage was real. The narrative stuck. Prodigy was the online service that charged you to read your own mail. ## The Censorship Problem The other long-running issue, which started early and never fully resolved, was content moderation. Prodigy moderated user-generated content with a heavy hand. The company employed editors who could pull posts they considered obscene, libelous, or commercially harmful. Bulletin board posts about competitors of Sears, for example, sometimes vanished. Posts critical of Prodigy itself sometimes vanished too. Prodigy framed this as family-friendly stewardship. Users framed it differently. The phrase that circulated on the bulletin boards in the early 90s was that Prodigy was a "censored service for the censored mind." The company had a mailing address that protesters could send physical mail to. Some did, in volume. The censorship policy also created an unexpected legal precedent. In a 1995 New York case, Stratton Oakmont v. Prodigy Services, a court ruled that because Prodigy actively moderated content, it could be held legally liable as a publisher for user-posted defamatory statements, much like a newspaper. The ruling was eventually superseded by Section 230 of the Communications Decency Act in 1996, which essentially flipped the logic. But for a brief window, Prodigy's editorial choices had made it the test case for how online services would be regulated for decades. That is a remarkable amount of internet legal history for a service most people now barely remember. ## Banner Ads Before the Web Had Banner Ads Here is something that does not get said often enough. Prodigy ran banner ads on its home interface in 1990. Four years before the web's first famous banner ad, the AT&T ad on HotWired (https://404memoryfound.com/posts/first-banner-ad-hotwired-1994-history.html) in October 1994, Prodigy was already showing on-screen promotional graphics at the bottom of every user's screen, integrated into the platform interface. By any reasonable definition, Prodigy invented the format that would later become the dominant economic engine of the entire commercial internet. The catch was that Prodigy's banner ads were not particularly effective, partly because click-through was not really a concept yet, partly because the ads were mostly for Sears products, and partly because users were already irritated by Prodigy and disinclined to engage with anything the platform promoted. The lesson here is one that internet history has a hard time articulating. Being first is not the same as being right. Prodigy got the format. AOL and the open web got the model. ## The Internet Arrives The thing that ultimately killed Prodigy was not AOL. It was the open internet, and Prodigy's inability to figure out how to compete with something it did not control. Through 1993 and 1994, the World Wide Web went from an experimental academic protocol to a consumer phenomenon. NCSA Mosaic, released in 1993, made the web visual. Netscape Navigator (https://404memoryfound.com/posts/how-netscape-lost-the-browser-war.html), released in late 1994, made it usable for people who had never touched a command line. Internet service providers like NetCom, Pipeline, and a wave of regional dial-up companies started selling raw internet access for $19.95 a month, often with no minutes cap, no advertising, and no censorship. Prodigy's business model was not built for this. The whole point of Prodigy was the curated experience, the controlled interface, the ad-supported wall around the content. Once consumers could just dial into the open internet and go anywhere, the walled garden looked like a smaller, less interesting version of the thing right next door. Prodigy launched a web browser feature in early 1995, but it was bolted onto the existing client and felt like exactly that. AOL did not have a much better product, but AOL had marketing, distribution, and a willingness to flood every magazine and mailbox in America with free trial CDs. Between 1994 and 1996, AOL went from a competitor Prodigy did not fully take seriously to the company that owned the consumer online category. ## The 1996 Sale By 1996, Prodigy's subscriber growth had stalled around 2 million. The numbers do not look terrible at first glance, but they were the wrong numbers. AOL had passed Prodigy in subscribers in 1994 and was growing at a pace Prodigy could not match. IBM and Sears had now spent more than a billion dollars combined on the venture and were not seeing a credible path to profitability. In 1996, IBM and Sears sold their stakes for $200 million to a buyer group led by the former founders of Boston Technology, operating as International Wireless. Carlos Slim, then the principal owner of the Mexican telecom giant Telmex, came in as a minority investor. The new ownership group renamed the company Prodigy Communications Corporation, restructured it around traditional internet service provision, and tried to position it as a premium ISP rather than a walled-garden online service. The strategy had some merit. Prodigy's brand recognition with US consumers was still high, and the new ownership team genuinely understood telecom. The company went public on the NASDAQ in 1999 at the height of dot-com enthusiasm, and at one point had a market capitalization in the billions. None of the traditional online service features survived the transition. The famous Prodigy interface, the banner ads, the curated content, the bulletin boards, all of it was effectively gone. ## The SBC Acquisition and the Quiet Ending In 2001, the regional Bell company SBC Communications, which would later become AT&T, acquired Prodigy. By then Prodigy was a mid-tier ISP with around 3 million subscribers, competing in a market full of mid-tier ISPs. SBC paid in stock and rolled the operations into its broader internet services division. The Prodigy brand limped along inside SBC for a few years. The Prodigy.net domain stayed live. Some long-time subscribers continued to receive email at prodigy.net addresses well into the late 2000s. There was no formal funeral. The brand simply faded. The story of Prodigy ends not with a collapse but with a slow administrative absorption, the way most failed corporate experiments end in America. ## What Prodigy Got Right Look at what Prodigy was actually trying to build, and the modern internet looks suspiciously similar to the version Prodigy sketched out in 1990. Curated content delivered to a consumer through a graphical interface, supported by advertising and integrated shopping, with social features layered on top. That is essentially Facebook. It is also Amazon. It is also the homepage of every major news site. The walled garden Prodigy built has become the dominant architecture of the consumer web, just under different ownership and with much better front-end design. Prodigy got the on-screen advertising format right. It got the flat-fee subscription model right. It got the idea that ordinary households would pay every month for a polished consumer service right. It got the importance of bulletin boards, eventually, even if the company never fully embraced them. What it missed was openness. Users wanted to talk to each other, and Prodigy treated that as a cost center rather than a product. Users wanted to publish their own content, and Prodigy treated that as a moderation problem. Users wanted to roam beyond the walls, and Prodigy could not bring itself to let them. ## What Prodigy Got Wrong The clearest lesson from the Prodigy story is that Sears and IBM were not stupid. They were just looking at the wrong question. They were trying to figure out how to extend the existing world (catalog shopping, broadcast advertising, branded content) into the online era. They were not trying to figure out what the online era would actually be. The companies that won the next decade, AOL, Yahoo, then Google, then Facebook, were not better engineers than IBM. They were not better marketers than Sears. What they had was a willingness to follow user behavior wherever it led, even when that meant abandoning the original product plan. Prodigy could not do that. Its founders had spent too much money on a vision that was already calcifying by the time the service launched. The other lesson is about ownership structure. Prodigy was always slow because it had two parents with different priorities, and later three layers of corporate oversight. Decisions that AOL could make in a week took Prodigy three months. By the time Prodigy responded to a competitor's move, the competitor had already moved twice more. That is what corporate joint ventures look like when the underlying market is moving faster than the partners can agree on a meeting time. ## The Cultural Memory Prodigy lives on in the memories of people who logged in for the first time as teenagers in the early 1990s and remember the strange textured beige of the welcome screen, the loading sound of the IBM client, the chunky pixelated graphics of the bulletin board interface, the distinctive feel of using something that was clearly trying very hard to be a magazine. It also lives on in retrocomputing communities, in archived screenshots, in academic papers about the prehistory of the consumer internet. The original client software has been preserved by hobbyists, and you can run a working Prodigy emulator on a modern Windows machine if you know where to look. The bulletin board content, mostly, is gone. Most of the content was hosted on IBM mainframes that were decommissioned years ago. The clearest place to find Prodigy's fingerprint, though, is on the modern internet itself. Every time a website serves you curated content alongside a banner ad and a recommendation widget, every time a social network moderates user content according to its own policies, every time a flat-fee subscription service tries to figure out how to handle power users who use the product more than the average customer, the ghost of Prodigy is somewhere in the room. ## FAQ When did Prodigy launch? Prodigy was incorporated as Trintex on February 13, 1984. The service launched regionally in 1988 in Atlanta, Hartford, and San Francisco, then nationally on September 6, 1990. Who owned Prodigy? Prodigy was originally a joint venture of CBS, IBM, and Sears, Roebuck and Company. CBS exited in 1986. IBM and Sears jointly owned the service through 1996, when they sold their stakes to a buyer group led by International Wireless and including Carlos Slim as a minority investor. SBC Communications acquired Prodigy in 2001. How many subscribers did Prodigy have? Prodigy crossed 465,000 subscribers by 1990 and reached approximately 2 million subscribers at its peak in the mid-1990s. After its 1996 transition to a traditional ISP, it grew further to around 3 million subscribers by the time of the SBC acquisition. What was the controversy with Prodigy email? In January 1991, Prodigy capped free email at 30 messages per month and charged 25 cents for each additional message. Subscribers organized in protest, and the company eventually rescinded the policy. The episode permanently damaged the company's reputation with its most engaged users. Did Prodigy invent banner ads? Prodigy ran banner-style on-screen advertising at the bottom of its interface from its national launch in 1990, four years before the famous AT&T web banner ad on HotWired in October 1994. By any reasonable definition of the format, Prodigy was the first major commercial online service to integrate persistent on-screen banner advertising into the user experience. Why did Prodigy fail? Prodigy failed because its walled-garden model could not adapt to the open internet. The service was built around curated content, controlled interface, and integrated shopping. When the web arrived in 1993 and 1994, users gained the ability to access unlimited content without curation, advertising, or content restrictions. AOL adapted faster, marketed more aggressively, and absorbed many of Prodigy's would-be subscribers. Internal corporate politics between IBM and Sears slowed Prodigy's response further. Is Prodigy still around? The Prodigy brand was effectively retired after SBC Communications absorbed it in the early 2000s. Some long-time users continued to receive email at prodigy.net addresses for several years afterward, but the service no longer exists as a meaningful product. SBC later became part of AT&T. --- # What Happened to the 3DO, the $700 Console That Bombed URL: https://404memoryfound.com/posts/what-happened-to-3do-700-dollar-console.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-07 Topics: Gaming, Hardware, Business Blunders Picture this. It is October 1993. You are wandering through the electronics section at Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html) on a Saturday afternoon with your dad, and you stop dead in front of a glass display case. Behind the glass is a sleek black console you have never seen before. It looks like something out of RoboCop. The price tag taped to the front says $699.99. You read it twice, just to make sure your eyes are working. Your dad reads it. You both look at each other. That is more than a month of rent in a lot of American cities in 1993. That is a used car. That is roughly two and a half Super Nintendos with a stack of games on top. And it is the price of a single console called the 3DO Interactive Multiplayer. If you were a kid in the early 1990s, you probably saw a 3DO at some point. Maybe at a Babbage's. Maybe at an Electronics Boutique. Maybe in a magazine ad with the tagline about being twice as fast as anything else. And then one day, almost overnight, it was gone. Pulled from shelves. Quietly forgotten. By 1996, the company was out of the hardware business entirely. The 3DO is one of the strangest stories in console gaming history. It was supposed to be the VHS of video games, a single open standard that everyone would build for and that would crush Nintendo and Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html). Instead, it became a cautionary tale about pricing, partnerships, and what happens when you bring a Mercedes to a Honda Civic fight. So what happened? ## How the 3DO Got Made: Trip Hawkins Bet His Career To understand the 3DO, you have to understand Trip Hawkins. By 1991, Hawkins was already a legend in gaming. He founded Electronic Arts in 1982, built it into one of the most respected publishers in the business, and made a lot of people very rich along the way. He was the guy who pushed the idea that game developers should be treated like rock stars and movie directors. He was also restless. By the early 1990s, Hawkins had decided that the future of gaming was not in cartridges. It was in CD-ROMs, full-motion video, and a single open hardware standard that anyone could build. So in 1991, Hawkins left EA and founded a new company called The 3DO Company. The pitch was wild. Instead of building hardware themselves like Nintendo or Sega, 3DO would design the console specifications and license them to manufacturers. Panasonic, Goldstar, Sanyo, and others would actually build and sell the boxes. 3DO would collect a small royalty on every game sold, similar to how Philips and Sony collected royalties on CD technology. The vision was that the 3DO platform would become the VHS of video games. One open standard. Many manufacturers competing on price and features. A massive shared library of titles. Game over for Sega and Nintendo. It was a genuinely smart idea. On paper. Hawkins assembled an all-star roster of partners. Time Warner, Matsushita (which owned Panasonic), MCA, AT&T, and Electronic Arts itself all bought in. The press loved it. Major tech magazines wrote glowing previews. Investors threw money at the company. The 3DO Company went public in 1993 with a valuation that briefly made Trip Hawkins one of the most-watched executives in tech. And then the actual product launched. ## The Launch: October 4, 1993 The Panasonic FZ-1 R.E.A.L. 3DO Interactive Multiplayer hit American shelves on October 4, 1993. The launch price was $699.99. To put that in context, in October 1993 you could buy a Sega Genesis with Sonic 2 for under $100. You could buy a Super Nintendo with Mario Kart for around $150. You could buy a fully working PC with a 486 processor for about $1,500. The 3DO was sitting in a price bracket that nobody had ever asked for. It was too expensive to be a toy and not powerful enough to be a workstation. It was the most awkward shelf placement in the history of consumer electronics. And here is the thing. The hardware itself was actually impressive. The 3DO had a 32-bit ARM60 CPU, two custom video co-processors, a double-speed CD-ROM drive, and 2 MB of RAM with another 1 MB of video RAM. In 1993, that was a serious machine. It could render 3D polygons that would have made the Super Nintendo cry. It could play full-motion video at a quality that made you stop and watch. The launch lineup included Crash N Burn, a futuristic combat racing game that genuinely looked like nothing else on the market. For about ten minutes after you turned it on, it really did feel like the future. The problem was the price. $699.99 in October 1993 dollars works out to roughly $1,500 today, adjusted for inflation. Imagine walking into a Best Buy in 2026 and seeing a new console with a price tag of $1,500. Now imagine that on the same shelf, there is a different console for $250 that has Mario on it. That is what the 3DO was up against. ## Why the Price Was So High: The Open Standard Trap Here is where the strategy started to eat itself. Sega and Nintendo could afford to sell consoles at a loss because they made the money back on game sales. Sony eventually used the same playbook with the PlayStation. The hardware was a Trojan horse for the software business. Sell the box cheap, take the loss, lock in the customer, and rake in the cash on $50 game cartridges for the next five years. The 3DO Company could not do that. They did not own the manufacturing. Panasonic, Goldstar, and Sanyo were the ones building the actual hardware, and those companies needed to make a profit on every unit sold. They were not going to eat $200 per console out of the goodness of their hearts. So the price had to cover the manufacturing cost plus the manufacturer's margin plus the 3DO Company's royalty plus retailer markup. By the time you added it all up, $699.99 was actually a slim-margin number for everyone involved. This is the part where it all falls apart. The brilliant open-standard strategy that was supposed to be the 3DO's biggest advantage was actually its biggest weakness. Hawkins thought he was building the VHS of gaming. What he had actually built was a console that was structurally incapable of being cheap. And in 1993, with a Genesis on every kid's Christmas list and a Super Nintendo on the rest of them, cheap was the only thing that mattered. Trip Hawkins said in later interviews that the licensing model was the single biggest mistake of his career. He believed for years that an open standard could win the way VHS won. Looking back, he admitted that for a console to succeed, one company had to control hardware, software, and marketing as a single integrated effort. By the time he understood that, it was too late. ## The Software Problem: A Library of B-Tier Games Even if you could swallow the $699.99 price tag, the next problem was games. And this is where the 3DO really stumbled. The console had a CD-ROM drive, which meant it could store dramatically more data than a Super Nintendo cartridge. That should have been an advantage. Instead, it became a liability. Developers in 1993 did not really know what to do with that much storage. So they filled it with full-motion video. If you bought a 3DO in 1993 or 1994, you ended up playing a lot of games that were basically interactive movies with grainy live-action footage. Night Trap. Plumbers Don't Wear Ties. Slam City with Scottie Pippen. There were good games on the system, sure. Road Rash on 3DO is genuinely a classic. Crash N Burn is fun. Star Control 2 was a beautiful port. But for every memorable title, there were ten games that felt like glorified PowerPoint presentations with bad acting. The library never reached critical mass. Compare that to what was on Super Nintendo in the same window. Super Mario All-Stars came out in 1993. Donkey Kong Country dropped in November 1994 and sold over 9 million copies. Final Fantasy VI launched in 1994. Chrono Trigger was right around the corner. Sega Genesis had Sonic 3, Streets of Rage 3, Ecco the Dolphin, Phantasy Star IV. Nintendo and Sega were shipping all-time classics one after another. The 3DO was shipping a Putt-Putt Saves the Zoo CD-ROM and asking for $699.99. ## The Saturn and PlayStation Show Up: 1994 to 1995 And then the real killers arrived. The Sega Saturn (https://404memoryfound.com/posts/what-happened-to-sega-saturn-console.html) launched in Japan in November 1994 and in North America in May 1995. The Sony PlayStation launched in Japan in December 1994 and in North America in September 1995. Both consoles offered comparable or better 3D graphics performance than the 3DO. The Saturn launched at $399. The PlayStation launched at $299. And both companies had something the 3DO never had: a deep pipeline of exclusive games from established studios. The PlayStation in particular was a knockout punch. Sony had been quietly courting third-party developers with developer-friendly tools and reasonable royalty rates. Within a year of launch, the PlayStation had Ridge Racer, WipEout, Tekken, and a roadmap that included Resident Evil, Tomb Raider, Final Fantasy VII, and Metal Gear Solid. The 3DO had nothing comparable. It had Gex. The price gap is what really killed it, though. By the time the PlayStation hit American shelves in September 1995, the 3DO had been forced to drop its price to around $399. But the Panasonic FZ-1 was a year and a half old, the library was thin, and consumer awareness was already shifting toward Sony. Why would you spend $399 on a 3DO when you could spend $299 on a PlayStation with Ridge Racer in the box? ## The M2 Project: A Sequel That Never Shipped The 3DO Company saw the writing on the wall in 1994 and started developing a successor console called the M2. On paper, the M2 was incredible. It was supposed to deliver next-generation 3D graphics, a CD-ROM drive that was either an upgrade module for the existing 3DO or a standalone console, and performance that would have crushed both the Saturn and the early PlayStation. Magazines printed previews. Developers got dev kits. Hype was building. And then, in late 1995, Matsushita bought the M2 rights from the 3DO Company for around $100 million. The 3DO Company took the cash and got out of the hardware business entirely. Matsushita sat on the M2 technology for a few years, used pieces of it in arcade boards and industrial products, and eventually let the project die. The console version of the M2 never shipped. To this day, prototype M2 dev units are some of the most valuable pieces of vaporware in gaming history. You know what happened next? The 3DO Company pivoted. With the hardware business sold off, Trip Hawkins reinvented The 3DO Company as a third-party publisher. They put out games like Army Men, High Heat Major League Baseball, BattleTanx, and Might and Magic VII. Some of those titles sold reasonably well. None of them were enough to save the company. By 2003, the 3DO Company filed for Chapter 11 bankruptcy. Trip Hawkins paid roughly $405,000 at auction for the rights to a portfolio of old brands and patents. The Army Men franchise went to Take-Two. Might and Magic went to Ubisoft. High Heat Baseball ended up at Microsoft. The rest got scattered to the wind. ## How Many 3DOs Actually Sold? Estimates vary, but most sources put global sales of the 3DO platform at around 2 million units across all manufacturers over its entire lifespan. For comparison, the Sony PlayStation sold over 100 million units. The Super Nintendo sold roughly 49 million. Even the Sega Saturn, which is often called a commercial failure, sold somewhere around 9 million units. The 3DO was outsold by the Saturn by more than four to one. It was outsold by the PlayStation by fifty to one. And here is the wild part. Despite all of that, the 3DO had its moment. Time magazine named it the 1993 Product of the Year. The hardware really was ahead of its time. The branding was sleek. The marketing was aggressive. There were genuine moments where it felt like the 3DO might actually become the open standard Hawkins promised. It just kept losing on the only number that mattered, which was the number on the price tag. ## What the 3DO Got Right It is easy to dunk on the 3DO in retrospect, but the console got some things right that the industry eventually copied. It pushed CD-ROM as the standard storage format for home consoles, which Sony and Sega both eventually adopted. It treated the console as a multimedia device that could play movies, music, and games, which is now exactly how the PlayStation 5 and Xbox Series X market themselves. It used a 32-bit ARM-based architecture at a time when most consoles were still on custom 16-bit chipsets, anticipating where the entire industry was headed. And the open-standard idea was not crazy. It was just early. Twenty years later, Android would basically prove that an open hardware ecosystem with multiple manufacturers all building to a shared software platform could absolutely beat a vertically integrated competitor. Hawkins was not wrong about the model. He was wrong about the timing, the price, and the size of the developer community needed to make it work. ## The Collector's Market Today Today, the 3DO has a small but devoted collector following. A working Panasonic FZ-1 in good condition typically sells for $200 to $400 on the used market, with boxed units in mint condition fetching considerably more. Loose games range from a few dollars for the FMV junk titles up to several hundred dollars for rare imports and limited releases. Some of the system's exclusive games, particularly imports, command serious money. A complete-in-box copy of certain Japanese 3DO releases can cost more than the console did at launch, adjusted for inflation. If you want to actually play 3DO games today, emulation is your friend. There are reasonable 3DO emulators available, and the entire library is small enough that you can sample everything worth sampling in a weekend. The good games hold up better than you might expect. The bad games are an extremely specific kind of bad that only existed in that 1993 to 1995 window when developers were still figuring out what to do with CD-ROM technology. ## The Real Lesson The 3DO is the console industry's reminder that hardware specs do not sell consoles. Price sells consoles. Software sells consoles. Brand recognition sells consoles. The 3DO had impressive hardware, and that hardware was almost completely irrelevant to its commercial fate. Trip Hawkins built a technically excellent product that died because it cost too much, shipped without enough good games, and tried to compete with companies that were willing to lose money on every box to win the long game. If you owned a 3DO in 1993, you were either a journalist who got a review unit, a kid whose parents had genuinely lost their minds, or a serious gaming enthusiast who saved for two years and treated the console like a religious artifact. Most people just saw the price tag, walked past the display case at Sears, and went home with a Sega Genesis. The 3DO deserved better than that, in some ways. But the market does not care what something deserves. It only cares what something costs. ## Frequently Asked Questions ### When was the 3DO released and how much did it cost? The 3DO Interactive Multiplayer launched in North America on October 4, 1993, with the Panasonic FZ-1 model. The launch price was $699.99 in the United States. By 1994 the price had dropped to around $499.99, and by 1995 the second-generation Panasonic FZ-10 was selling for closer to $399.99. None of those price points were competitive with the Sega Genesis, Super Nintendo, or eventually the Sony PlayStation. ### Who made the 3DO console? The 3DO Interactive Multiplayer was a hardware standard designed by The 3DO Company, founded by Electronic Arts founder Trip Hawkins in 1991. Unlike most consoles, the 3DO was manufactured by multiple licensees rather than the platform owner. The primary manufacturers were Panasonic (which made the FZ-1 and FZ-10), Goldstar (now LG), and Sanyo. Creative Labs (https://404memoryfound.com/posts/what-happened-to-creative-labs-sound-blaster.html) also produced a 3DO-compatible expansion card for PCs called the 3DO Blaster. ### Why did the 3DO fail? The 3DO failed for a combination of reasons. The launch price of $699.99 was far too high for a mass-market gaming console. The licensing model meant manufacturers had to make a profit on every unit, which prevented the kind of below-cost pricing that Sega, Nintendo, and Sony used. The game library never reached critical mass and relied too heavily on shovelware full-motion video titles. And in 1995, the Sony PlayStation arrived with comparable hardware, a stronger software pipeline, and a launch price of $299, which effectively ended the 3DO's competitive viability. ### How many 3DO consoles were sold? Most industry estimates put total worldwide sales of the 3DO platform across all manufacturers at around 2 million units over its entire commercial lifespan from October 1993 to its discontinuation in 1996. By comparison, the Sega Saturn sold roughly 9 million units, the Nintendo 64 sold around 33 million, and the Sony PlayStation sold over 100 million units. ### What happened to The 3DO Company after the console failed? After exiting the hardware business in 1995 and 1996, the 3DO Company sold the rights to its M2 successor console technology to Matsushita (Panasonic) for a reported $100 million. The 3DO Company then pivoted into being a third-party game publisher, releasing franchises like Army Men, High Heat Major League Baseball, and Might and Magic. The company filed for Chapter 11 bankruptcy on May 28, 2003. Its game franchises were auctioned off to publishers including Microsoft, Take-Two Interactive, Namco, and Ubisoft. ### Are 3DO games worth playing today? The 3DO library has a handful of genuine standouts that are worth tracking down through emulation or original hardware. Road Rash, Crash N Burn, Need for Speed (the original 3DO version was the first release of the franchise), Star Control 2: The Ur-Quan Masters, and Gex are all considered classics or near-classics. The full-motion video titles from the era have aged poorly and are mostly of historical interest. The total game library is small enough, with most counts placing the worldwide total in the 250 to 300 range, that a dedicated collector can sample most of the best of it in a few weekends. --- # What Happened to Microsoft Bob, the Friendly Interface That Failed URL: https://404memoryfound.com/posts/what-happened-to-microsoft-bob-friendly-interface.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-05-07 Topics: Software & Apps, Business Blunders, Then vs Now Microsoft Bob shipped on March 11, 1995, retailed for $99, and sold an estimated 58,000 copies during its entire commercial life. For context, that is fewer copies than a moderately successful Steam indie game sells in its first weekend today. Bob was discontinued by early 1996. PC World later named it the seventh worst tech product of all time. Time magazine put it on its list of the 50 worst inventions in human history, alongside hydrogen blimps and asbestos. None of which would matter much, except that Microsoft Bob was not a side project from a forgotten division. It was a flagship consumer release. It was personally championed by Bill Gates. It had a marketing budget that included a Murphy Brown crossover, a Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html) tie-in, and a media blitz that declared March 31, 1995, to be Microsoft Bob Day. And it was managed by a young woman named Melinda French who, three months earlier, had married the CEO of Microsoft. So how does a product like that ship, fail, and get pulled from shelves in under twelve months? The short answer is that Microsoft Bob got almost everything wrong about how regular humans actually use computers. The longer answer is more interesting, because it gets at a question that the entire tech industry was asking in 1995, and that we are still arguing about thirty years later: what does a friendly computer actually look like? ## The Problem Microsoft Was Trying to Solve In 1994, Microsoft had a real concern, and it was not a bad one. The personal computer was about to go mainstream in a way it never had before. Windows 95 was coming. Internet access was becoming a household line item. PCs were dropping below $2,000 for the first time. Tens of millions of new users were going to buy a computer in the next three years, and most of them had never touched a keyboard outside of a typing class. The interface those users would encounter was, frankly, hostile. Windows 3.1 expected you to understand file paths, drive letters, and the Program Manager. The default look was gray on gray. Error messages used jargon. Even the user-friendly applications like WordPerfect (https://404memoryfound.com/posts/what-happened-to-wordperfect-word-processor.html) or Lotus 1-2-3 (https://404memoryfound.com/posts/what-happened-to-lotus-1-2-3-spreadsheet.html) had toolbars that looked like the cockpit of a 747. Microsoft's own internal research showed that new computer buyers were spending hours on simple tasks, getting frustrated, and returning their PCs. The question Microsoft was trying to answer was a good one. What if a computer interface was actually friendly? What if instead of file folders and drive letters, you saw something familiar, like a house with rooms? What if the computer talked to you in plain English and helped you instead of throwing error codes? These were legitimate UX questions in 1994. Apple was asking them. IBM was asking them. The entire industry was looking for the answer that Microsoft Bob ended up trying to give. ## The Pitch: A Cartoon House with Talking Pets The product that shipped in March 1995 was bizarre, and it is worth describing it in detail because the description is the criticism. When you launched Microsoft Bob, you did not see a desktop. You saw a cartoon house. The house had rooms. The rooms had furniture. To do anything, you clicked on the furniture. Clicking on the calendar gave you a calendar app. Clicking on the writing pad gave you a word processor. Clicking on the checkbook gave you a checkbook program. You were not alone in the house. You were accompanied by a cartoon assistant. There were a dozen of them to choose from, including a yellow dog named Rover, a parrot named Scuzz, a turtle, a cat, a dragon, and a rat. The assistants spoke to you in cartoon speech bubbles. They offered help. They asked if you wanted to redecorate. They had personalities, opinions, and what the marketing team called social presence. The home environment came pre-loaded with eight applications: a letter writer, an address book, a calendar, a household manager, a financial guide, a checkbook, an email program, and a basic geography game. They were not bad applications, technically. They were just trapped inside a metaphor that nobody asked for, and that made every task slower than the equivalent task in regular Windows. ## The People Behind Bob The Microsoft Bob team was led by program manager Karen Fries, who had been pushing the concept of a social interface inside Microsoft Research for years. Fries had run prototype usability tests with new computer users and observed that they responded warmly to animated guides. In one famous test, a cartoon duck walked test subjects through a software demo, and Fries watched the participants relax, smile, and engage in a way they did not with traditional interfaces. That observation became the seed of Bob. The marketing manager assigned to the project was Melinda French, who had joined Microsoft in 1987 as the company's first female MBA hire. She married Bill Gates on January 1, 1994, and took on the Bob marketing role shortly after, while keeping her name as French during the product cycle (she would later go by Melinda Gates publicly, then return to Melinda French Gates after her divorce). French has spoken publicly about Bob being a formative failure for her, the kind of project that taught her how to recognize when an idea was not going to work, even when senior leadership wanted it to. And then there was the technical work, which is where Microsoft Bob accidentally produced its single most consequential cultural artifact. While testing the Rover character, Microsoft graphic designer Vincent Connare noticed that Rover's speech bubbles displayed text in Times New Roman. Connare, who had a background in comic books, thought it looked absurd to have a cartoon dog speaking in a serif typeface designed for newspapers. He sat down and designed a casual, hand-drawn-feeling typeface to replace it. He called the typeface Comic Sans. Comic Sans did not actually ship with Microsoft Bob. The font was finished too late. But Connare was at Microsoft, and the typeface ended up bundled with Microsoft Plus! for Windows 95 and then Windows 95 itself. Twenty years later, Comic Sans would become the most ridiculed and most widely recognized typeface in computing history. It exists because of a cartoon dog in a piece of failed software that you have probably never seen. ## What Bob Got Wrong Look. The fundamental problem with Microsoft Bob was not that it was friendly. The problem was that the friendliness slowed everything down without making anything easier. The metaphor of a house with rooms sounded intuitive in a focus group. In actual use, it was extra friction. Consider a normal task: you want to write a letter. In Windows 3.1, you double-click the WordPerfect or Word icon and you start typing. Two clicks, maybe three. In Microsoft Bob, you opened the application, navigated to your house, walked into the den, found the writing desk, clicked on the writing desk, watched a cartoon character pop up, dismissed the cartoon character, and then started typing. The interface was added cognitive load wrapped in a smile. Users tolerated it for about an hour, then started looking for the exit. The hardware requirements were also brutal for 1995. Microsoft Bob required at least 8 MB of RAM, a relatively recent CPU, and Windows 3.1 or Windows 95. In March 1995, the average home PC had 4 to 8 MB of RAM. Most users did not have the memory to run Bob smoothly. The product that was supposed to be approachable for new computer users was actively crashing on the machines those users were most likely to own. The reviews were brutal. The New York Times described the cartoon house as the work of, and this is a real direct quote, an aesthetically challenged sixth-grader. The Washington Post called the home environment sterile and lifeless. Computing magazines picked the product apart. The user community, such as it was, found that Bob was simultaneously condescending toward beginners and useless to anyone with even minor computer literacy. ## Sales Numbers and the Quiet Pull Here is the thing about Bob that does not get reported often enough. Microsoft did not actually fail to sell Bob because the product was insulted in the press. They failed to sell it because nobody wanted it. According to PC Data, the leading retail sales tracker at the time, total Microsoft Bob sales from launch in March 1995 to discontinuation came to roughly 58,000 copies. Microsoft had been internally projecting sales in the millions, comparable to Microsoft Works or Encarta (https://404memoryfound.com/posts/what-happened-to-microsoft-encarta.html), both of which routinely moved millions of units per year. Microsoft pulled Bob from active production in early 1996, less than a year after launch. There was no big announcement. No press release. The product just stopped getting shelf space. By the time the second version that had been planned was due to ship, the project had been quietly euthanized. The internal team was reassigned. Karen Fries continued at Microsoft on related research work. Melinda French moved on to other projects, including her eventual transition into philanthropy and her current work at Pivotal Ventures. ## Where the Bob DNA Actually Survived Bob did not totally die. Pieces of it migrated into other Microsoft products, sometimes in ways that became famous on their own. The most direct descendant was Office Assistant, which shipped with Office 97. The animated paper clip named Clippy (https://404memoryfound.com/posts/what-happened-to-clippy-microsoft.html), the Einstein character, the cat, the dog, all of those animated guides were direct descendants of the Microsoft Bob assistant concept. Clippy was Karen Fries' work refined into a more constrained form. Inside Office, where the user already knew what they were trying to do, the animated guide was less intrusive. It was still annoying enough to become an internet joke for the next twenty-five years, but it sold tens of millions of copies attached to Office, which is more than Bob ever managed on its own. Rover the dog showed up again in Windows XP, released in October 2001. The XP search companion feature, the cute animated yellow dog that walked across the screen while you searched your file system, was Rover from Microsoft Bob. Microsoft had quietly preserved the character and brought him back when the search interface needed a friendly face. Rover lived in XP for years, was disabled by most users immediately, and was finally killed off in Windows Vista in 2007. Comic Sans, of course, became its own cultural object. The typeface that exists because Microsoft Bob needed a cartoon dog to look right is now used on everything from elementary school flyers to the original 2012 Higgs boson discovery announcement at CERN, which the physics community is still arguing about. ## Why Bob Failed and What It Tells Us About UX Here is the thing. The conventional story about Microsoft Bob is that it failed because it was condescending. That is partially true, but it is not the whole picture. Bob failed because Microsoft was solving a problem that was about to solve itself. Bob shipped in March 1995. Windows 95 shipped in August of the same year. Windows 95 introduced the Start menu, the taskbar, the Recycle Bin, and a coherent file management system that made the Program Manager from Windows 3.1 feel ancient overnight. The interface improvements in Windows 95 made the friendliness of Bob largely unnecessary. The actual desktop became approachable enough that the cartoon-house alternative looked silly by comparison. This is a recurring pattern in user interface history. A product solves a problem in an elaborate, theatrical way, and then a year later the underlying platform improves enough that the problem evaporates. AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html) solved the problem of getting on the internet through walled garden navigation, and then web browsers got good enough that the walled garden became a constraint instead of a feature. Real Networks solved the problem of streaming media on slow connections through proprietary players, and then web standards caught up and the proprietary player was unnecessary. Bob solved the problem of unfriendly Windows, and then Windows became friendly without him. The deeper UX lesson is one Don Norman has been repeating for decades. The most usable interface is usually the most invisible one. Bob made the interface visible, animated, talkative, and decorated. Windows 95 made the interface invisible enough that you forgot you were using it. Invisible won. ## The Revisionist View: Was Bob Actually That Bad? Here is a take that has gained traction in the last few years among UX researchers and tech historians. Microsoft Bob was not a bad product. It was a product that shipped at the wrong moment, on hardware that could not run it well, with a metaphor that was too literal for the use case it was trying to serve. None of those problems are inherent to the underlying ideas. They are execution failures, and execution failures get romanticized into cosmic disasters when the company is large enough that everyone watches. Look at the actual user research that came out of the Bob project. Karen Fries' team documented something real. New computer users in 1994 and 1995 did respond positively to friendly, animated guides. They did get less anxious. They did complete tasks faster in certain contexts. The problem was that the version of friendliness Bob shipped was too maximalist. It tried to replace the entire interface, when what users actually wanted was a friendly assistant that lived alongside the standard interface and could be dismissed at will. That is exactly what Office Assistant did two years later, and what every modern AI chatbot does today. The pattern that worked is: keep the standard interface, layer a friendly assistant on top, make the assistant easy to ignore, and let the user decide how much help they want. Bob got the first two parts right. It got the third and fourth parts catastrophically wrong. You could not ignore Bob. You had to live inside the cartoon house. There was no exit door labeled just give me the desktop. The other revisionist point is that Bob's commercial failure was partially a marketing positioning problem. The product was sold to existing PC owners as a replacement shell. The actual target audience, which was first-time computer buyers, was almost never reached because they bought their machines pre-configured at retailers and never went to the software store to look for a friendlier interface. The 58,000 copies Bob sold were almost entirely to existing Windows users curious about the experiment. The new users Bob was designed for never even saw it. ## The Modern Echoes of Microsoft Bob Look at any modern AI assistant, smart home interface, or onboarding flow, and you will find traces of Bob. The friendly avatar that introduces itself when you open a new app. The chatbot that asks if you need help. The wizard interface that walks you through setup. The character with a name and a personality that exists to make your software feel less like software. All of this is Microsoft Bob with thirty years of refinement. This is essentially what Apple's Siri does today, what Amazon's Alexa does today, what Microsoft's own Copilot does today. The pendulum swung back. Bob was wrong about almost everything except the fundamental insight that human users respond to interfaces that feel like collaborators rather than tools. That insight just needed faster computers, better natural language, and the patience to make the assistant useful instead of intrusive. Karen Fries, the program manager who pushed Bob into existence, spent the rest of her career working on assistant technologies at Microsoft. The work she started in 1995 looks a lot more reasonable in 2026 than it did in 1995. She was right about the destination. She was just twenty-five years too early on the timeline. ## Frequently Asked Questions ### When was Microsoft Bob released and what did it cost? Microsoft Bob was released on March 11, 1995. The retail price was $99 in the United States. The product required Windows 3.1 or Windows 95 and at least 8 MB of RAM, which was a demanding requirement for a typical home PC in early 1995. ### Who was on the Microsoft Bob team? The Microsoft Bob program management team was led by Karen Fries, who had researched the social interface concept inside Microsoft Research before launching the product. The marketing manager was Melinda French, who had married Bill Gates on January 1, 1994. Vincent Connare, a Microsoft graphic designer, designed the Comic Sans typeface during the Bob project, although the font itself shipped with Microsoft Plus! and Windows 95 rather than with Bob. ### How many copies of Microsoft Bob were sold? According to PC Data, the leading retail sales tracker at the time, Microsoft Bob sold approximately 58,000 copies between its launch in March 1995 and its discontinuation in early 1996. Microsoft had been internally projecting sales comparable to Microsoft Works or Encarta, which typically moved millions of units per year. Bob was discontinued less than twelve months after launch. ### Why did Microsoft Bob fail? Microsoft Bob failed for several connected reasons. The cartoon house metaphor added cognitive load and slowed every task compared to standard Windows applications. The hardware requirements were too steep for the average 1995 home PC. The product was widely panned in reviews from outlets including The New York Times, The Washington Post, and major computing magazines. And critically, Windows 95 launched in August 1995 with significant interface improvements (the Start menu, the taskbar, the Recycle Bin) that made the underlying problem Bob was trying to solve largely obsolete within months of Bob's release. ### What is the connection between Microsoft Bob and Comic Sans? Microsoft graphic designer Vincent Connare designed Comic Sans during the Microsoft Bob development cycle. He was working on the Rover character and noticed that the cartoon dog's dialogue appeared in Times New Roman, which Connare found inappropriate for a casual cartoon character. He designed a hand-drawn, casual typeface to replace it. The font was not ready in time to ship with Microsoft Bob, but it was bundled with Microsoft Plus! for Windows 95 in August 1995 and later with Windows 95 itself, which is how it became one of the most widely distributed typefaces in computing history. ### Did anything from Microsoft Bob survive into later Microsoft products? Yes. The most direct descendant was the Office Assistant feature in Microsoft Office 97, including the famous Clippy character, which was a refined version of the Bob assistant concept. Rover the cartoon dog from Bob returned in Windows XP in 2001 as the Search Companion in Windows Explorer, where he assisted users with file searches until being removed in Windows Vista in 2007. Comic Sans remains in active use across Microsoft products. The broader concept of a friendly conversational assistant interface has continued to evolve through products like Cortana and Microsoft Copilot. --- # What Happened to the Sony Discman, the CD Player That Changed Music URL: https://404memoryfound.com/posts/what-happened-to-sony-discman-portable-cd-player.html Author: Marcus Vale (404 Memory Found) Published: 2026-05-06 Topics: Hardware, Music & Entertainment Picture this. It is 1996, and you are sitting in the back seat of your parents' minivan on a road trip to your aunt's house three hours away. You have got a Sony Discman balanced on your knee, a burnt CD of songs you recorded off the radio using Windows Sound Recorder, and a pair of those terrible foam-pad headphones that came in the box. Every time your dad hits a pothole, the music skips. You press the disc down with your thumb through the closed lid, hoping that somehow applying pressure will keep the laser on track. It does not work. It never works. But you do it anyway, because this is the closest thing to freedom you have ever felt. If you grew up in the 1990s, you know exactly what I am talking about. The Sony Discman was not just a portable CD player. It was a rite of passage. It was the device that sat between the cassette Walkman and the iPod (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html), bridging two completely different eras of how we listened to music. And for about fifteen years, it was everywhere. School buses, gym locker rooms, airport terminals, and every single car that did not have a CD player built into the dashboard. The Discman was the soundtrack device of an entire generation. And then it vanished. Not with a dramatic death or a corporate scandal. It just quietly stopped being relevant. So what happened? ## The Birth of the Discman: November 1984 The story of the Discman starts with a guy named Kozo Ohsone, the same Sony engineer who had led the development of the original Walkman. Ohsone looked at the compact disc, which Sony had co-developed with Philips and launched commercially in 1982, and asked a simple question: can we make this portable? The answer, at first, was basically "not really." CD players in 1982 were large, expensive, and extremely sensitive to vibration. The idea of carrying one around in your bag seemed borderline absurd. But Sony did it anyway. In November 1984, they released the D-50, also known as the D-5 in some markets. It was the world's first portable CD player. It cost 49,800 yen in Japan, which translated to about $350 in the United States. For context, that is roughly $1,050 in today's money. You were paying a thousand bucks, adjusted for inflation, to carry around a device that skipped if you looked at it wrong. And people bought it. Not in massive numbers at first, but enough that Sony knew they were onto something. The D-50 was not exactly pocket-sized. It was about the dimensions of four CD jewel cases stacked on top of each other, and it weighed just over a pound. The battery life was rough. The skip problem was real from day one. But here is the thing: it played compact discs. Away from your home stereo. In 1984, that was genuinely revolutionary. CDs themselves were still a new format, and most people were still buying cassettes. The idea that you could take this shiny, futuristic silver disc and play it on the bus was the kind of thing that made you the coolest person in your friend group. ## The 1990s: When Everyone Had One The Discman did not really become a mass-market product until the early 1990s. Through the late 1980s, Sony kept iterating on the design, making it smaller, lighter, and cheaper. By 1990, you could get a basic portable CD player for under $200. By the mid-90s, the price had dropped to around $50 to $100 for a decent model, and that is when the floodgates opened. And it was not just Sony anymore. Panasonic, Aiwa, Philips, JVC, and a dozen other brands were making portable CD players. The market exploded. By the mid-1990s, portable CD players were outselling portable cassette players for the first time. The Walkman, Sony's own legendary product, was being cannibalized by its younger sibling. Which is kind of insane when you think about it. Sony was essentially competing with itself, and the CD player was winning. The 1990s Discman experience was defined by a few universal truths. First, the skip problem. Every single person who owned a portable CD player has a skip story. Walking too fast? Skip. Jogging? Forget about it. Riding a bike? You might as well have been listening to a remix. The mechanical reality of a portable CD player is that a tiny laser has to read data off a spinning disc, and any vibration disrupts that process. It was a fundamental engineering challenge that Sony and every other manufacturer struggled with for years. Second, the accessories. The car cassette adapter was maybe the most important peripheral of the 1990s. If your car had a tape deck but no CD player, you could buy this little gadget that looked like a cassette tape with a wire coming out of it, plug it into your Discman's headphone jack, and suddenly your car stereo was playing CDs. It was janky, the audio quality was mediocre at best, and the wire was always getting caught on the gear shift. But it worked. And for millions of people, it was the bridge between old car technology and new music technology. Third, the CD binder. Nobody carried around jewel cases. You had a zippered nylon binder, usually black, that held anywhere from 24 to 128 discs in those clear plastic sleeves. Your entire music collection, organized however you wanted. Mine was organized by genre, then alphabetically, which in retrospect was way too much effort for a thirteen-year-old. But that binder was sacred. Losing your CD binder was a genuine tragedy. People mourned those things. ## The Anti-Skip Revolution: ESP and G-Protection Sony knew the skipping problem was killing them. People loved the idea of portable CD music but hated the reality of it. So in the early 1990s, they started working on electronic skip protection, which they branded as ESP. The concept was clever: instead of playing music directly from the disc in real time, the player would read ahead and store several seconds of audio in a RAM buffer. If the laser got knocked off track by a bump or vibration, the player could keep playing from the buffer while the laser found its place again. The first generation of ESP, which showed up around 1992 to 1994, gave you about three seconds of buffer. Three seconds. That sounds pathetic now, but at the time it was a revelation. Three seconds of bump protection meant that a casual walk would not cause skips anymore. It did not help much on a bumpy bus ride, but for everyday carrying, it was a game changer. By the mid-1990s, the buffer had grown to ten seconds, then twenty, then forty. Sony's D-777 model from 1995 featured a ten-second anti-skip buffer, and it was considered a premium feature. The RAM required to store that much uncompressed CD audio was expensive. We are talking about an era when a desktop computer might have 4 to 8 megabytes of RAM total, and Sony was cramming 1.5 megabytes of buffer RAM into a portable music player just to solve the skip problem. By the late 1990s and early 2000s, anti-skip technology had gotten legitimately good. Sony introduced G-Protection, their advanced version of ESP, and buffer times climbed to 45 seconds, then 60, then 120 for MP3-CD players. The skip problem that had defined the Discman experience for a decade was, for all practical purposes, solved. You could jog with a late-model Discman and it would keep playing. The irony is that by the time they finally fixed the biggest complaint about portable CD players, the format was already on its way out. ## The Name Game: Discman to CD Walkman Here is a detail that most people do not know. Sony actually retired the "Discman" brand name in 2000. After that, all of their portable CD players were marketed under the "CD Walkman" name, folding them back into the Walkman family. The reasoning made sense from a branding perspective. "Walkman" was one of the most recognized brand names on the planet, and Sony wanted to consolidate everything portable under that umbrella. But culturally, nobody cared. Everyone still called them Discmans. The name had become generic, like Kleenex or Band-Aid. You did not say "hand me my CD Walkman." You said "where's my Discman?" And that is how most people still remember them. ## The Peak: 1997 to 2001 The golden age of the Discman, if you want to pin it down, was roughly 1997 to 2001. This was the period when portable CD players were at their cheapest, most reliable, and most ubiquitous. The skip protection was good enough for daily use. The prices had dropped to a point where a basic model cost $30 to $40 at Walmart. And CDs themselves were at the absolute peak of their commercial dominance. In 1999, the U.S. music industry (https://404memoryfound.com/posts/napster-destroyed-music.html) hit its all-time peak for CD revenue, with CD sales generating roughly $13 billion of the industry's record $14.6 billion total. The format had never been bigger, and it would never be this big again. Think about what the world looked like during peak Discman era. Napster (https://404memoryfound.com/posts/who-owns-napster-now.html) had just launched but had not yet destroyed the music industry's business model. The iPod did not exist yet. Streaming was a fantasy. If you wanted to listen to music away from your home stereo or car, you had two choices: a portable cassette player or a portable CD player. And by 1999, the choice was obvious. CD audio quality was vastly superior to cassette. CDs did not degrade with repeated plays. You could skip to any track instantly instead of fast-forwarding through tape. The Discman had won. Every high school hallway in America during this period had the same scene. Kids with Discman players clipped to their belts or stuffed in backpack front pockets, headphone cables running up under their shirts to those earbuds that hooked over your ears. Teachers confiscating them during class. The battery compartment always slightly loose because you had opened it so many times to swap in fresh AAs. The lid hinge getting wobbly after a year of use. These were shared experiences for an entire generation. ## The Fall: How the iPod Changed Everything On October 23, 2001, Apple introduced the iPod. And just like that, the Discman was living on borrowed time. The iPod did not kill the Discman overnight. The first iPod cost $399, only worked with Macs, and held 5 gigabytes of music, which was about 1,000 songs. For most people, that was not an immediate upgrade from a $40 Discman and a binder full of CDs. But the iPod represented something that the Discman could never match: the entire concept of carrying your whole music library in your pocket. No discs. No skipping. No bulk. No CD binder. Just a smooth white rectangle that held everything. The transition took a few years. From 2001 to about 2005, portable CD players and MP3 players coexisted. Some manufacturers tried to split the difference with MP3-CD players, devices that could read discs burned with MP3 files, fitting 10 to 12 albums on a single CD-R. That bought the format a little extra time. But once the iPod Mini launched in 2004 at $249, and especially once the iPod Nano hit in September 2005 starting at $199 for the 2GB model, the math stopped working for portable CD players. Why carry a bulky disc player and a binder of CDs when you could carry a device the size of a credit card that held your entire collection? Sony, to their credit, saw this coming. They had launched their own digital music players, including the ill-fated Network Walkman line that used Sony's proprietary ATRAC format instead of MP3, which was a disaster of a decision. But even Sony could not save the Discman from the march of technology. Portable CD player sales declined steadily through the mid-2000s, and by 2010, Sony had quietly discontinued their CD Walkman line entirely. ## Then vs Now: The Discman's Unexpected Afterlife Here is the part of the story that nobody predicted. The Discman, and portable CD players in general, have experienced a genuine nostalgia-driven revival. Starting around 2020, vintage Discman models began showing up on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html) and in thrift stores at premium prices. A mint-condition Sony D-50 from 1984 can sell for $500 or more. Even common 1990s models that originally cost $40 are going for $80 to $150 if they are in good working condition. Part of this is the broader vinyl-style nostalgia trend. Just as vinyl records came back as a physical, tactile alternative to streaming, CDs are experiencing their own smaller-scale comeback. CD sales in the U.S. actually increased in 2021 and 2022 after years of decline, driven largely by younger buyers who never owned CDs the first time around. There is something appealing about the physical ritual: opening the jewel case, reading the liner notes, placing the disc in the tray. It is the same impulse that drives vinyl collectors, just with a different format. But here is the honest truth. The Discman is not coming back in any meaningful commercial sense. It served its purpose brilliantly for about fifteen years, bridging the gap between analog portable audio and digital portable audio. It gave an entire generation the experience of personal, portable, high-quality music. And then technology moved on. The skipping, the bulk, the limited capacity, the battery drain: these were not problems to be solved. They were fundamental limitations of the format. The Discman did not fail. It was simply surpassed by something better. And if you were there for it, if you spent your teenage years pressing down on that lid and praying the skip protection would hold through one more song, you know that no amount of streaming convenience will ever fully replace the feeling. There was something about music that required effort. Music that you had to carry, protect, organize, and occasionally smack on the side to get working again. That relationship with your music collection, physical and fragile and personal, is something that a Spotify playlist will never replicate. The Discman gave us that. And for that, it deserves to be remembered. ## Frequently Asked Questions ### When was the first Sony Discman released? The first Sony Discman, the D-50 (also called the D-5 in some markets), was released in November 1984. It retailed for approximately $350 in the United States, which is about $1,050 adjusted for inflation. It was the world's first portable CD player and was developed under the leadership of Kozo Ohsone, who had also led the development of the original Sony Walkman. ### Why did portable CD players skip so much? Portable CD players work by using a laser to read data off a spinning disc. Any vibration or sudden movement can cause the laser to lose its tracking position on the disc, resulting in an audible skip or interruption. Sony and other manufacturers eventually developed electronic skip protection (ESP), which buffered several seconds of audio in RAM so the music could continue playing while the laser recovered. Early ESP systems provided about three seconds of protection, while later models offered 40 to 120 seconds. ### When did Sony stop making portable CD players? Sony discontinued their portable CD player line (branded as "CD Walkman" after 2000) around 2010 to 2011. The decline was driven by the rise of digital music players, particularly Apple's iPod, which launched in October 2001 and offered a fundamentally superior portable music experience by eliminating the need for physical discs entirely. ### Are old Discman players worth anything today? Yes, vintage Sony Discman models have become collectible. The original 1984 D-50 in good working condition can sell for $500 or more. Common 1990s models typically sell for $80 to $150 depending on condition. The value is driven by nostalgia collectors and the broader revival of interest in physical media formats. ### What replaced the Discman? The Discman was replaced by digital music players, most notably the Apple iPod (launched October 2001). MP3 players could store hundreds or thousands of songs without physical discs, eliminating the skipping problem and dramatically reducing the size and weight of portable music devices. Later, smartphones with music playback capabilities and streaming services like Spotify replaced dedicated music players entirely. --- # What Happened to Xanga, the Blog Platform Before Blogging Was Cool URL: https://404memoryfound.com/posts/what-happened-to-xanga-blogging-platform.html Author: Dana Reyes (404 Memory Found) Published: 2026-05-06 Topics: Internet Culture, Software & Apps In 2005, Xanga had roughly 27 million registered user accounts. For a platform most people over 30 have never heard of, that number tends to surprise. For context, Facebook, which had launched just one year earlier, had about 5.5 million users at the same point. MySpace (https://404memoryfound.com/posts/rise-and-fall-of-myspace.html) had more, sure. But Xanga was not trying to be MySpace. Xanga was something quieter, stranger, and in many ways more personal. It was where a generation of teenagers learned to write online, to publish their thoughts for strangers, and to discover that the internet could be a diary that talked back. Then it disappeared. Not in a dramatic flameout like Friendster (https://404memoryfound.com/posts/what-happened-to-friendster-social-network.html), or a slow public decline like MySpace. Xanga just sort of evaporated, one server at a time, until the platform that had hosted millions of teenage confessions, song lyrics, and late-night emotional posts simply stopped existing in any meaningful way. The real story of what happened to Xanga is more interesting than "Facebook killed it." It involves a $1 million federal fine, a business model that never quite worked, a last-ditch crowdfunding campaign, and a fundamental misunderstanding of what users actually wanted. ## How Xanga Started: Reviews, Not Blogs Xanga launched on April 4, 1999. The founders were Marc Ginsburg and John Hiler, two entrepreneurs based in New York City. The original concept had nothing to do with blogging. Xanga started as a platform for sharing book and music reviews. The name itself reportedly derived from a combination of cultural references, though its exact etymology has been debated. The important thing is what it was not: it was not designed as a personal diary platform. That part came later, and it came from the users, not the founders. The pivot happened around 2000 when Xanga added weblog functionality to user profiles. This was a significant decision, though at the time it probably seemed like a minor feature addition. In November 2000, Xanga officially introduced blogs as a core feature. Users could now publish timestamped entries, and other users could leave comments. This sounds unremarkable now, but in 2000, the concept of a "blog" was still niche enough that most internet users had never encountered one. The term "weblog" had only been coined by Jorn Barger in December 1997, and shortened to "blog" by Peter Merholz in 1999. Xanga was riding a wave that barely existed yet. Which brings us to the real question: why did Xanga succeed where dozens of other early blogging platforms struggled? ## The Secret: Social Features Before "Social Media" Existed The answer is deceptively simple. Xanga understood, earlier than almost anyone else, that writing online was a social activity. A blog post published into a vacuum is just a text file on a server. A blog post published into a community, with comments, friend lists, and the ability to subscribe to other people's updates, is something entirely different. It is a conversation. Xanga built social infrastructure around blogging before the phrase "social media" had been invented. Users had "blogrings," which were essentially groups organized around shared interests: a blogring for fans of a particular band, a blogring for students at a specific high school, a blogring for people who liked a certain anime. Users could subscribe to each other's blogs and see updates in a feed. They could leave "eProps," which were basically an early version of the Like button, years before Facebook made that concept universal. The platform also offered extensive customization. Users could modify their page layouts with custom HTML and CSS, choosing backgrounds, fonts, colors, and embedded music players. This was the era of auto-playing songs on personal web pages, and Xanga was ground zero for it. If you visited a Xanga page in 2004 and were not immediately blasted with a Dashboard Confessional track, something was wrong. The demographic was specific and intense. Xanga's core user base was American teenagers, roughly 14 to 22 years old, with a heavy concentration in the emo, punk, and alternative music scenes. The posts were confessional: breakups, friendships, arguments with parents, song lyrics posted without context, cryptic messages aimed at specific people who would definitely know it was about them. Xanga was where a generation practiced the art of public vulnerability, for better and worse. ## Growth and the Revenue Problem By 2005, Xanga had grown to an estimated 25 to 27 million registered accounts. The platform was adding users rapidly, particularly among high school and college students. The growth was real. The revenue was not. Xanga's business model was advertising-supported, supplemented by a premium subscription tier called Xanga Premium. For a few dollars a month, premium users got additional customization options, more storage space, and the removal of ads from their pages. The problem was that Xanga's core demographic, teenagers, had very little disposable income and almost zero interest in paying for something that felt like it should be free. Premium conversion rates were low. And the advertising revenue was constrained by the nature of the content: brands were not exactly lining up to place ads next to teenagers' emotional diary entries about their ex-boyfriends. This is a pattern that repeats throughout the history of social platforms aimed at young users. The audience is large and engaged, but the monetization is difficult because the content environment is unpredictable and the users have no money. Xanga was living this problem years before Tumblr (https://404memoryfound.com/posts/what-happened-to-tumblr-yahoo-billion-dollar-mistake.html), Vine, and Snapchat would encounter the exact same challenge. ## The FTC Fine: $1 Million and a Warning Sign In September 2006, the Federal Trade Commission hit Xanga with a $1 million civil penalty for violating the Children's Online Privacy Protection Act, commonly known as COPPA. At the time, it was the largest COPPA fine ever imposed. The FTC's complaint was straightforward: Xanga's terms of service stated that users had to be 13 or older to create an account. But the platform allowed visitors to create accounts even when they entered a birthdate indicating they were under 13. According to the FTC, Xanga had created approximately 1.7 million accounts for users who had explicitly identified themselves as children under 13, collecting personal information from those users without parental consent. The fine itself was significant, but the real damage was reputational and operational. Xanga was a relatively small company, and a million-dollar penalty hit hard. More importantly, the FTC case highlighted a structural problem: Xanga's growth had been driven partly by very young users, and the platform had been negligent about age verification. The case forced Xanga to implement stricter age controls, which reduced new signups at precisely the moment the platform needed growth to attract advertising revenue. Look at the timing. September 2006. Facebook had opened registration to the general public on September 26, 2006, just weeks after the Xanga fine was announced. The platform that was about to consume the entire social internet was throwing its doors open at the exact moment Xanga was dealing with a federal enforcement action. The coincidence was not causal, but it was devastating. ## The Facebook Effect: Death by a Thousand Features Facebook did not kill Xanga in one blow. It killed Xanga by being better at every single thing Xanga did, while also doing a hundred things Xanga could not. Xanga let you blog. Facebook let you post status updates (shorter, easier, lower friction). Xanga had blogrings. Facebook had groups (more discoverable, better moderated). Xanga had eProps. Facebook had the Like button (simpler, more addictive). Xanga let you customize your page. Facebook gave everyone a clean, consistent layout (which turned out to be what most people actually preferred). Xanga had friend lists. Facebook had a social graph that connected you to people from every part of your life: school, work, family, casual acquaintances. The key difference was friction. Writing a Xanga post required effort. You had to think of something to say, write it out in paragraph form, maybe add some HTML formatting, and publish it. A Facebook status update could be six words. "Math test was brutal today lol." The bar for participation dropped dramatically, and with it, the average post shifted from confessional essay to casual life update. For most users, that trade was worth making. Xanga's traffic metrics tell the story. According to multiple internet analytics services, Xanga's user engagement began declining significantly in 2007 and never recovered. The users who had been the platform's core, teenagers who were now becoming young adults, were migrating to Facebook, and they were not coming back. By 2009, Xanga's relevance in the broader internet conversation had effectively ended. ## The Crowdfunding Hail Mary: 2013 By 2013, Xanga was functionally dying. The platform could no longer sustain operations under its existing model. Server costs, development, and maintenance were exceeding revenue from the dwindling user base. In May 2013, CEO John Hiler announced that Xanga would shut down unless the community could raise $60,000 through a crowdfunding campaign by July 15. The plan was to migrate Xanga to a WordPress-based infrastructure, which would reduce hosting costs. The new platform, called "Xanga 2.0," would require users to pay an annual subscription fee of $48 to maintain their blogs. Free hosting, the model that had attracted tens of millions of users, would be eliminated entirely. Here is the thing. The campaign technically succeeded. Xanga 2.0 launched on September 3, 2013, with user accounts migrated to new servers. The platform offered free downloadable archives for approximately 2 million legacy blogs, allowing users to preserve their old content. On paper, it was a rescue mission that worked. In practice, it was an ending dressed up as a new beginning. Charging $48 per year for a blogging platform in 2013, when WordPress.com, Tumblr, and Medium all offered free alternatives, was asking for loyalty that the remaining user base simply did not have. The transition to Xanga 2.0 was less a relaunch and more a quiet hospice admission. The platform limped along, but it never regained any meaningful user base or cultural relevance. ## Why Xanga Actually Mattered The easy narrative is that Xanga was just another failed social platform from the early internet. That framing misses the point entirely. Xanga was, for millions of people, their first experience with publishing. Not commenting on someone else's work. Not sharing a link. Actually writing something original and putting it out into the world where strangers could read it. That is a fundamentally different experience from posting a photo on Instagram or typing a tweet. Xanga posts had paragraphs. They had narrative structure, even when the narrative was "my crush talked to me in the hallway today and I do not know what to do with this information." The platform taught a generation that their thoughts had value, that writing was a skill worth practicing, and that an audience, however small, was findable. The confessional nature of Xanga also foreshadowed something that would define social media for the next two decades: the tension between authenticity and performance. Xanga users wrote raw, unfiltered posts about their real emotions, often under pseudonyms, in spaces that felt semi-private. That dynamic, the feeling of writing honestly for a small audience that understands you, is essentially what made early Tumblr appealing, what drives private Instagram accounts today, and what keeps platforms like BeReal trying to capture some version of "real" online sharing. The business lesson is equally important. Xanga had the users. It had the engagement. It had the cultural moment. What it did not have was a sustainable revenue model, a defensible competitive position, or the willingness to evolve its core product before competitors made it irrelevant. The same story would repeat with Friendster, with Vine, with dozens of other platforms that confused user passion for business durability. ## Where Xanga Stands Today As of 2026, Xanga exists in name only. The domain is still registered, and some version of the Xanga 2.0 infrastructure technically persists, but the platform has no meaningful active user base and no cultural relevance. The millions of blog posts that users wrote between 1999 and 2013, the confessional essays, the song lyric posts, the late-night emotional dispatches, are largely gone. Some users downloaded their archives during the 2013 migration. Most did not. The content, like so much of the early web, has been lost. That loss matters more than it might seem. Xanga was not just a platform. It was a record of how a generation of young people thought, felt, and expressed themselves during a specific moment in internet history. The posts were messy, emotional, and often embarrassing in retrospect. They were also authentic in a way that modern social media, with its performative polish and algorithmic optimization, rarely achieves. When Xanga disappeared, it took that record with it. Which is, in the end, the most Xanga thing that could have happened. A platform built on teenage feelings, temporary by nature and permanent only in memory. ## Frequently Asked Questions ### When was Xanga created? Xanga launched on April 4, 1999, founded by Marc Ginsburg and John Hiler in New York City. It originally started as a platform for sharing book and music reviews before adding blog functionality in November 2000. ### How many users did Xanga have at its peak? At its peak around 2005 to 2006, Xanga had an estimated 25 to 27 million registered user accounts. The platform was particularly popular among American teenagers and young adults. ### Why did Xanga get fined by the FTC? In September 2006, the FTC fined Xanga $1 million for violating the Children's Online Privacy Protection Act (COPPA). The platform had created approximately 1.7 million accounts for users who indicated they were under 13 years old, collecting their personal information without parental consent. At the time, it was the largest COPPA fine ever imposed. ### Is Xanga still online? Xanga technically relaunched as "Xanga 2.0" in September 2013 after a crowdfunding campaign, but it has no meaningful active user base. The platform required a $48 annual subscription fee, which drove away most remaining users. The vast majority of original Xanga blog content from 1999 to 2013 has been lost. ### What killed Xanga? Xanga's decline was caused by a combination of factors: the rise of Facebook (which opened to the general public in September 2006), the $1 million FTC COPPA fine that same month, an advertising-dependent revenue model that never generated sufficient income from its young user base, and the broader shift from long-form blogging to short-form social media updates. By 2009, most of Xanga's core users had migrated to Facebook. --- # What Happened to StarCraft, the Game That Invented Esports URL: https://404memoryfound.com/posts/what-happened-to-starcraft-birth-of-esports.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-15 Topics: Gaming, Internet Culture Picture this. It's a Saturday afternoon in 1998, and you've just torn the shrinkwrap off a jewel case with a cover so good it should've been a movie poster. Three alien races. A tagline that promised galactic warfare. And a system requirements sticker on the back that made you pray your family PC could handle it. The game was StarCraft. And within two years of its release, it would do something nobody at Blizzard Entertainment, or anywhere else in the gaming industry, could have predicted. It wouldn't just become the best-selling PC game of its year. It would create an entirely new form of professional competition, turn a generation of South Korean teenagers into millionaire celebrities, and lay the foundation for what we now call esports, an industry worth billions. This is the story of how a real-time strategy game built in Irvine, California, accidentally rewired an entire country's culture and invented a new kind of sport. ## The Game That Almost Looked Like Warcraft in Space Blizzard started working on StarCraft in 1995, right after wrapping up Warcraft II: Tides of Darkness. The original plan was straightforward: take the Warcraft II engine, reskin it with a sci-fi aesthetic, and ship it. Simple enough. When they showed the first build at E3 1996, the reaction was brutal. Journalists called it "Warcraft in Space," and not as a compliment. The game looked derivative, felt clunky, and generated the kind of polite disappointment that kills hype faster than a bad review. And here's where Blizzard did something that most studios wouldn't have the guts to do. They scrapped almost everything and started over. The engine was rebuilt from the ground up. The three playable factions, Terran, Zerg, and Protoss, were redesigned to play in fundamentally different ways rather than being mirror images of each other with different skins. This was a radical design choice for real-time strategy games at the time. Most RTS games gave you factions that were basically cosmetic variations. StarCraft made each race feel like an entirely different game. By the time Blizzard showed the overhauled version in early 1997, the reception flipped completely. The game looked gorgeous for its time, the asymmetric balance was unlike anything in the genre, and the hype machine was back on. On March 31, 1998, StarCraft shipped. It sold 1.5 million copies worldwide in its first year, including 746,365 copies in the United States alone, making it the best-selling PC game of 1998. The expansion pack, Brood War, followed in late 1998 and refined the balance even further. But the real story wasn't happening in American living rooms. It was happening ten thousand miles away, in PC rooms packed shoulder to shoulder across South Korea. ## The PC Bang Phenomenon and South Korea's Perfect Storm To understand how StarCraft became the biggest thing in South Korean culture, you have to understand what was happening in the country in the late 1990s. The 1997 Asian financial crisis had devastated the Korean economy. Unemployment surged among young people. The government, looking for an economic recovery strategy, made a massive bet on broadband infrastructure. By the early 2000s, South Korea had some of the fastest and cheapest internet access in the world. At the same time, a new type of business was exploding across the country: the PC bang. These were internet cafes, but calling them cafes doesn't really capture it. They were more like gaming arcades rebuilt for the internet age. Rows and rows of high-spec PCs with fast connections, open late, priced cheap enough that teenagers could afford to spend hours there after school. By 2001, there were roughly 23,000 PC bangs across South Korea. StarCraft and Brood War landed in this environment like a match in dry grass. The game was perfect for PC bangs: it was competitive, it was social, it rewarded skill in a way that was immediately visible to spectators, and it ran on modest hardware. Korean teenagers didn't just play StarCraft. They lived it. And because Battle.net, Blizzard's free online service, let anyone play anyone, the competition got intense fast. Within a year or two of Brood War's release, the best players in South Korea were becoming local celebrities. People would gather around screens in PC bangs to watch elite matches. And someone, inevitably, realized there was money in this. ## When Gaming Became Television In 2000, two things happened that turned StarCraft from a popular game into a professional sport. First, the Korea e-Sports Association, or KeSPA, was established with backing from the South Korean government's Ministry of Culture, Sports and Tourism. This wasn't some grassroots fan organization. This was a government-backed body designed to professionalize competitive gaming. Think about that for a second. In 2000, the South Korean government was officially recognizing gaming as a sport. The rest of the world wouldn't catch up for over a decade. Second, two cable television channels began broadcasting StarCraft matches: OnGameNet, known as OGN, and MBC Game. OGN launched the OnGameNet Starleague, or OSL, which became the premier individual StarCraft tournament in the world. MBC Game followed with the MBC Starleague, or MSL. These weren't small operations. The production quality was legitimate: multiple camera angles, professional commentators (called casters), instant replays, player cameras showing their faces during tense moments, and live studio audiences. The prize pools were real too. Top tournaments offered hundreds of thousands of dollars. And the players, many of them barely out of high school, became genuine celebrities. They signed endorsement deals. They appeared on talk shows. Fans made signs and screamed their names. The StarCraft Proleague finals in 2005 drew an estimated 100,000 or more spectators to Gwangalli Beach in Busan, South Korea, one of the largest live audiences for any gaming event in history at that point. ## The Emperor and the Golden Age If StarCraft esports had a Michael Jordan, it was Lim Yo-hwan, better known by his Battle.net handle: BoxeR. Born in 1980, BoxeR started competing professionally in 1999 and quickly earned the nickname "The Emperor" for his dominance of the Terran faction. Between 2001 and 2002, he won two OnGameNet Starleague titles and two World Cyber Games gold medals. He held the number one ranking in KeSPA's official standings for a consecutive 17 months. BoxeR wasn't just good at StarCraft. He was entertaining. He pioneered aggressive, creative strategies that made spectators lose their minds. He would do things that other players considered reckless or impossible, and he'd make them work through sheer mechanical skill and tactical instinct. In 2002, he founded Team Orion, which later became SK Telecom T1 in 2004, one of the most legendary esports organizations in history. Yes, that SK Telecom T1, the same organization that would later dominate League of Legends with Faker. But BoxeR wasn't alone. The Brood War era produced an incredible roster of legendary players. Lee Young Ho, known as Flash, is widely considered the greatest Brood War player of all time, earning the title "God of StarCraft" for his near-perfect play. Lee Jae Dong, known as Jaedong, was Flash's eternal rival and arguably the greatest Zerg player in history. Hong Jin Ho, known as YellOw, became one of the most popular players despite being famous for his heartbreaking losses in finals. These players weren't niche internet figures. Flash reportedly earned over $500,000 per year at his peak through tournament winnings, salary, and endorsements. Korean Air, SK Telecom, Samsung, and KT all sponsored professional StarCraft teams. Major Korean corporations were investing real money in competitive gaming years before the rest of the world took esports seriously. ## Why It Worked: The Accidental Brilliance of StarCraft's Design Here's the thing that people who never played Brood War competitively might not understand: the game was almost perfectly balanced despite being absurdly complex. Three completely different races, hundreds of unit interactions, dozens of viable strategies, and yet at the highest level, no single race dominated for long. The win rates between the three matchups hovered remarkably close to 50/50 over the years. Some of this was intentional. Blizzard patched the game periodically for balance. But a lot of it was emergent. Players kept discovering new strategies, new build orders, new ways to exploit the mechanics. The metagame evolved constantly. A strategy that dominated one season would get countered the next. And because the game rewarded both strategic thinking and raw mechanical speed (top players performed over 300 actions per minute), there was always room for a new prodigy to emerge and shake things up. The game was also incredibly fun to watch, even if you didn't fully understand what was happening. Big army clashes were visually dramatic. Clutch moments, like a perfectly timed defensive hold or a devastating surprise attack, produced genuine emotional reactions from spectators. The Korean casters, famous for their rapid-fire commentary and excited screaming, turned matches into theater. And then there was the map design. Brood War maps weren't randomly generated. They were hand-crafted, and the competitive map pool rotated regularly. Each map had different characteristics: some favored aggressive play, others rewarded defensive strategies, some had tricky terrain that enabled creative flanking maneuvers. The map pool rotation meant the meta was always shifting, preventing any single strategy from dominating indefinitely. Players who could adapt across map types rose to the top. Specialists got exposed. The community that formed around competitive Brood War was also remarkably analytical. Korean fans dissected games with a level of detail that would make sports statisticians proud. Build order timings were documented to the second. Win rate statistics were tracked across every matchup on every map. Post-game analysis shows on OGN featured casters breaking down key decision points frame by frame. This wasn't just entertainment. It was a whole analytical culture that grew up around a computer game, years before "data analytics in esports" became a buzzword in the West. ## The Business of Brood War: When Corporations Came Calling Here's something that blows my mind every time I think about it. By 2004, professional StarCraft in South Korea wasn't some scrappy grassroots thing anymore. It was a full-blown corporate ecosystem. The biggest Korean conglomerates, companies like Samsung, SK Telecom, KT Corporation, and CJ Group, were sponsoring professional teams. These weren't token sponsorships either. Players received salaries, team houses (shared apartments where the entire roster lived and practiced together), coaching staff, and access to training facilities. The team house model was particularly fascinating and, honestly, a little intense. Young players, many of them 16 or 17 years old, would move into a shared apartment with their teammates. They'd practice for 10 to 14 hours a day, every day. There was a structured schedule: practice blocks, strategy review, physical exercise (some teams added this later), and very little free time. It was closer to how Olympic athletes train than how most people imagined professional gamers spending their days. And the competition for spots was brutal. KeSPA ran a draft system similar to professional sports. Teams had rosters with limited slots. Players who couldn't perform got cut. There was a constant stream of hungry teenagers trying to break into the scene, which meant established players could never get comfortable. The pressure was immense, and burnout was a real issue. Several top players retired in their early twenties, not because they'd lost their skills, but because they'd been grinding at that intensity since they were teenagers. The economics made sense for the corporations involved. A corporate-sponsored esports team provided brand visibility with South Korea's most coveted demographic: young, tech-savvy consumers. A single Proleague match on OGN or MBC Game could reach hundreds of thousands of viewers. For companies like SK Telecom, which was competing fiercely with KT in the telecommunications market, having the best StarCraft team was a legitimate marketing strategy. When SKT T1 won a championship, that logo was all over the broadcast. You can't buy that kind of sustained engagement with traditional advertising. ## The Culture: StarCraft as National Identity It's hard for people outside South Korea to fully grasp how embedded StarCraft was in the country's cultural fabric during the early 2000s. This wasn't a subculture. This was mainstream entertainment. Pro gamers appeared on variety shows. BoxeR's autobiography sold well. When Lee "Flash" Young Ho was at the height of his dominance, he was recognized on the street the way soccer stars are recognized in Brazil or basketball players in the United States. The game also became a shared reference point for an entire generation. If you grew up in South Korea in the late 1990s or early 2000s, you played StarCraft. You had opinions about whether Terran was overpowered. You had a favorite pro player. You'd skipped class to watch a crucial Starleague match at a PC bang, or at least knew someone who had. It was the social glue for a generation of young Korean men in particular, a common language that crossed regional and class boundaries. There's a reason Blizzard chose South Korea for the StarCraft: Remastered launch event in 2017. They knew where the game's spiritual home was. And when they held the event at a venue in Seoul, the response confirmed it. Fans lined up for hours. Former pros came out for exhibition matches. The whole thing had the energy of a reunion concert for a beloved band. Except the band was a 19-year-old video game. ## StarCraft II and the Complicated Sequel In 2010, Blizzard released StarCraft II: Wings of Liberty, twelve years after the original. It was one of the most anticipated sequels in gaming history. The game sold 1.5 million copies in its first 48 hours and received near-universal critical acclaim. Two expansion packs followed: Heart of the Swarm in 2013 and Legacy of the Void in 2015. StarCraft II was, by most measures, a great game. But it never quite replicated the cultural phenomenon of Brood War in South Korea. Part of this was timing. By 2010, League of Legends was beginning its own meteoric rise, and the MOBA genre was pulling players and viewers away from real-time strategy. Part of it was the game itself. Many Brood War purists felt that StarCraft II simplified too many mechanics, reducing the skill ceiling that had made the original so compelling to watch at the highest level. The transition from Brood War to StarCraft II was messy and political. KeSPA and Blizzard fought over broadcasting rights and tournament control. For a period, Brood War and StarCraft II ran parallel competitive scenes, splitting the player base and the audience. Eventually, KeSPA transitioned its leagues to StarCraft II, but by then the damage was done. The viewership numbers never matched the Brood War golden age. In 2017, Blizzard released StarCraft: Remastered, a graphical overhaul of the original game and Brood War that kept the gameplay identical but updated the visuals and audio. It was a love letter to the competitive community that had kept the original game alive for nearly two decades. And in South Korea, Brood War tournaments continue to this day, with Afreeca TV hosting leagues that draw tens of thousands of viewers. ## The Legacy: What StarCraft Built It's hard to overstate what StarCraft, and specifically its South Korean competitive scene, did for the broader world of esports. Before StarCraft, the idea of professional gaming existed only in fragments: some Quake (https://404memoryfound.com/posts/what-happened-to-quake-online-fps.html) tournaments here, a few Street Fighter competitions there. StarCraft in South Korea created the template that every major esport would eventually follow: professional teams sponsored by major corporations, salaried players, televised matches with professional production, government recognition, and a fan culture that rivaled traditional sports. When Riot Games built League of Legends into a global esports phenomenon in the 2010s, they were following a blueprint that StarCraft had written a decade earlier. When the Overwatch League, Call of Duty League, and Valorant Champions Tour adopted franchise models with city-based teams, they were iterating on structures that Korean StarCraft leagues had pioneered. When ESPN started broadcasting esports on television, they were doing what OGN and MBC Game had been doing since 2000. Even the concept of the esports player as a celebrity, someone with fans, endorsement deals, and a public persona, was forged in the StarCraft scene. BoxeR was famous in Korea before most Western gamers had even heard the word "esports." StarCraft sold over 11 million copies across its lifetime. But its real impact can't be measured in sales numbers. It proved that competitive gaming could be a spectator sport. That it could generate real revenue. That players could become professionals. That an entire industry could be built around watching other people play video games. Every time you tune into a Twitch stream or watch a League of Legends Worlds final with millions of concurrent viewers, you're watching something that started with a real-time strategy game on a CD-ROM and a million teenagers in South Korean PC bangs who just wanted to see if they could be the best. And honestly? For a game that started as "Warcraft in Space," that's a pretty incredible legacy. ## Frequently Asked Questions Is StarCraft still played competitively? Yes. While StarCraft II's official competitive scene has slowed significantly, Brood War continues to have an active professional scene in South Korea. Afreeca TV hosts regular Brood War tournaments that attract dedicated viewership. The community remains passionate, even if the audience is smaller than during the golden age. How many copies did StarCraft sell? StarCraft and Brood War sold over 11 million copies combined worldwide. It was the best-selling PC game of 1998, moving 1.5 million copies in its first year alone. Why was StarCraft so popular in South Korea specifically? A combination of factors: South Korea's massive investment in broadband infrastructure after the 1997 financial crisis, the explosion of PC bangs (internet gaming cafes), government support through organizations like KeSPA, and the game's inherent qualities as a spectator-friendly competitive experience all converged at the right moment. Who is considered the greatest StarCraft player of all time? Lee Young Ho, known as Flash, is widely regarded as the greatest Brood War player in history, earning the nickname "God of StarCraft." In StarCraft II, opinions vary, but many consider players like Jung "Mvp" Jong Hyun and Maru among the greatest. Can you still play StarCraft online? Yes. In 2017, Blizzard released StarCraft: Remastered, which updated the original game's graphics while keeping gameplay identical. Blizzard also made the original StarCraft and Brood War free to download. Both versions run on Battle.net and have active online communities. --- # What Happened to Kodak, the Company That Invented the Digital Camera Then Chose to Ignore It URL: https://404memoryfound.com/posts/what-happened-to-kodak-digital-camera-bankruptcy.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-15 Topics: Hardware, Business Blunders In 1975, a 24-year-old Kodak engineer named Steve Sasson built the world's first digital camera. It was the size of a toaster, weighed about eight pounds, recorded black-and-white images at a resolution of 0.01 megapixels, and stored them on a cassette tape. The process of capturing a single photograph took 23 seconds. Sasson brought his invention to Kodak's management. Their response, as he later recalled, was a version of: "That's cute, but don't tell anyone about it." The company that had built its empire on the phrase "Kodak moment" had just been handed the future of photography. And it decided the future could wait. Twenty-seven years later, Kodak filed for bankruptcy. ## The Company That Made Photography Democratic George Eastman didn't invent photography. But he made it accessible to everyone, which mattered more. In 1888, Eastman launched the original Kodak camera with a slogan that captured its entire value proposition in seven words: "You press the button, we do the rest." Before Eastman, photography required specialized knowledge, bulky equipment, and a darkroom. After Eastman, it required $25 and a functioning index finger. The business model was elegant in a way that would later be compared to razor-and-blades economics. The camera was the entry point. The real money came from film, processing, and printing. Every photograph taken was a consumable transaction. Every birthday party, every vacation, every first day of school generated revenue for Kodak. And because film needed to be developed, Kodak controlled both ends of the value chain: the capture and the output. By the mid-20th century, Kodak was one of the most recognizable brands on the planet. The company's yellow-and-red logo was ubiquitous, appearing in camera shops, drugstores, and tourist destinations worldwide. "Kodak moment" entered the English language as shorthand for a memory worth preserving. The company wasn't just selling film. It was selling the idea that your life's important moments deserved to be captured, and that Kodak was the way to do it. ## The Numbers at the Peak Look at the numbers and you start to understand why Kodak's leadership had a hard time imagining the end. In 1976, Kodak controlled approximately 90% of film sales and 85% of camera sales in the United States. Global employment peaked at 145,300 workers in 1988. Annual revenue hit $16 billion in 1996. The company's market capitalization reached $31 billion in 1997. These weren't the numbers of a company in trouble. These were the numbers of a company that had won. Kodak's dominance in film photography was so complete that it functioned almost like a utility. If you took pictures, you almost certainly used Kodak products somewhere in the process. The company's Rochester, New York headquarters was the economic engine of the entire region, and Kodak was routinely listed among the most valuable brands in America. The problem, of course, is that dominance in a declining market is still decline. And by the time Kodak's leadership fully acknowledged what was happening, the market had already moved. ## 1975: The Camera That Should Have Changed Everything Steve Sasson's digital camera prototype was, by any reasonable definition, a breakthrough. He had proven that you could capture an image electronically, without film, without chemicals, without a darkroom. The resolution was terrible. The process was slow. The storage was primitive. But the concept was sound, and concepts are what matter when you're looking at a 15-to-20-year technology horizon. Sasson and his colleague Robert Hills even built a more advanced prototype in 1989 that produced images closer to what consumers would eventually expect from digital photography. They showed it internally. The response from Kodak's film division was predictable. As Sasson later told The New York Times, management's reaction could be summarized as: "Why would anyone ever want to look at their pictures on a television set?" Here's the thing. The people asking that question weren't stupid. They were rational actors protecting a phenomenally profitable business. Kodak's film and processing operations had profit margins estimated between 60% and 70%. Digital photography, by contrast, eliminated the consumable. No film to sell. No processing fees. No prints unless the customer chose to make them. From a pure business perspective, digital photography was a threat to the most profitable part of Kodak's business model. This is the innovator's dilemma in its purest form. Clayton Christensen would later use Kodak as one of his defining case studies. The company that had the most to gain from leading the digital transition was also the company that had the most to lose from it. And Kodak chose to protect what it had rather than build what was coming. ## The Slow Decline Nobody Wanted to See The strange thing about Kodak's decline is that it wasn't sudden. It was a slow, visible erosion that took roughly 15 years to reach the point of no return. And during that entire period, Kodak made moves that suggested awareness of the digital threat, just never with the conviction required to actually pivot the business. In 1991, Kodak developed the Kodak DCS 100, one of the first commercially available digital cameras, in partnership with Nikon. It cost between $20,000 and $25,000 depending on configuration, targeting professional photojournalists rather than consumers. In 1995, Kodak released the DC40, one of the first consumer-oriented digital cameras, at a price point accessible to early-adopter consumers. The company invested heavily in digital imaging research throughout the 1990s and held thousands of patents related to digital photography technology. So Kodak wasn't ignoring digital. It was hedging. The strategy seemed to be: participate in digital just enough to stay relevant, but never commit so aggressively that it cannibalized the film business. On paper, this looked like prudent corporate strategy. In practice, it meant that Kodak was always two steps behind the companies that had nothing to lose. Canon, Nikon, Sony, and eventually Samsung and the smartphone manufacturers had no legacy film business to protect. They could go all-in on digital without worrying about destroying an existing revenue stream. Kodak could not. Or rather, Kodak chose not to, which amounts to the same thing. Consider the timeline. In 1997, when Kodak's market cap hit $31 billion, digital cameras were still expensive toys that produced mediocre images. A reasonable executive could look at the numbers and conclude that digital was years, maybe decades, away from threatening the core business. But technology doesn't move in straight lines. It moves in S-curves. By 2003, digital camera sales surpassed film camera sales globally. By 2006, Nikon had stopped producing film cameras entirely. The transition that seemed distant in 1997 was effectively complete by 2008. Kodak's annual reports from this period make for fascinating reading. The company acknowledged the digital transition explicitly. It discussed its digital strategy at length. It reported digital revenue alongside film revenue. And yet, year after year, the investment balance tilted toward protecting the existing film business rather than accelerating the digital one. The words said "digital transformation." The capital allocation said "maybe next year." ## The Final Decade By the early 2000s, the numbers were turning ugly. Digital camera sales surpassed film camera sales in 2003. Kodak's revenue began declining. The company started cutting jobs, closing factories, and restructuring repeatedly. In 2004, Kodak announced it would stop selling traditional film cameras in North America and Europe. In 2005, the company reported its first annual loss in over a decade. CEO after CEO cycled through, each with a new strategy. Antonio Perez, who took over in 2005, tried to transform Kodak into a digital printing company, leveraging the company's expertise in imaging science for commercial and consumer inkjet printing. The logic was defensible. The execution was too late and too underfunded. Meanwhile, the technology that would truly kill Kodak's remaining digital camera business was already in people's pockets. When Apple launched the iPhone in 2007, it didn't just disrupt the phone industry. It began the process of making standalone digital cameras unnecessary for most consumers. Why carry a separate camera when your phone took pictures that were good enough to share instantly on social media? Kodak's digital camera business, which had never been its strongest division anyway, collapsed alongside the rest of the consumer camera market. By 2010, Kodak was bleeding money. The company tried to monetize its vast patent portfolio, which included foundational digital imaging patents. It filed licensing lawsuits against Apple, Samsung, and others, essentially trying to extract revenue from the technology it had helped pioneer but failed to commercialize effectively. ## Bankruptcy and After On January 19, 2012, the Eastman Kodak Company filed for Chapter 11 bankruptcy protection. The filing listed assets of $5.1 billion and debts of $6.8 billion. A company that had once employed 145,300 people and generated $16 billion in annual revenue was now insolvent. Kodak emerged from bankruptcy in September 2013 as a much smaller company focused on commercial printing and imaging for business customers. The consumer photography business that had defined the brand for over a century was gone. The workforce had been reduced to a fraction of its former size. Rochester, which had depended on Kodak for generations, was left to reinvent itself. The Kodak brand still exists today. The company produces commercial printing solutions, advanced materials, and chemicals. In 2020, Kodak briefly made headlines when the Trump administration announced a $765 million loan to help the company produce pharmaceutical ingredients, though the deal was later put on hold amid insider trading allegations. The company's stock briefly surged from around $2 to over $60 before crashing back down. But the Kodak that most people remember, the one that was synonymous with photography itself, is gone. And the tragedy of it is that the company literally held the technology that would replace its core business 37 years before that replacement actually happened. ## The Fuji Problem and the Antitrust Distraction While Kodak was wrestling with the digital question internally, it was also fighting a very public battle against Fujifilm on the analog front. Throughout the 1990s, Kodak accused Fuji of unfair trade practices in Japan, where Fuji allegedly blocked Kodak from accessing distribution channels. In 1995, Kodak filed a petition with the U.S. Trade Representative under Section 301 of the Trade Act, escalating the dispute to the World Trade Organization. The WTO ruled against Kodak in 1998, finding insufficient evidence that Japan's market was systematically closed to foreign competition. The ruling was a blow to Kodak's management, which had invested significant time, money, and executive attention into the trade dispute. But the larger problem wasn't the ruling itself. It was the distraction. During the years that Kodak was focused on beating Fuji in the film market, the film market itself was shrinking. Fuji, notably, managed the digital transition somewhat better than Kodak, diversifying into healthcare imaging, optical films for LCD screens, and cosmetics (leveraging its expertise in collagen chemistry from film manufacturing). By 2012, while Kodak was filing for bankruptcy, Fujifilm had revenue exceeding $21 billion and remained profitable. Same industry, same disruption, very different outcomes. The comparison is instructive. Fujifilm's leadership made the painful decision to invest aggressively in non-film businesses even while film was still profitable. They accepted that the core business was dying and looked for adjacent markets where their chemical and optical expertise could be applied. Kodak's leadership, by contrast, kept trying to slow the decline of film while half-heartedly investing in digital. The difference wasn't intelligence or foresight. It was willingness to act on what both companies could clearly see coming. ## The Patent Portfolio: Kodak's Final Card One of the more poignant chapters in Kodak's decline was its attempt to monetize its intellectual property. Over decades of research and development, Kodak had accumulated a massive portfolio of patents related to digital imaging. More than 1,000 digital imaging patents covered technologies that were fundamental to how digital cameras, smartphones, and image processing systems worked. Starting in the mid-2000s, Kodak began filing patent infringement lawsuits against major technology companies, including Apple, Samsung, and LG. The company also tried to license its patents directly. In some cases, Kodak won settlements. Samsung paid $550 million to settle patent disputes with Kodak in 2010. LG settled for $414 million in 2009. But these were one-time windfalls, not a sustainable business model. And the irony was thick: Kodak was now making money from the very technology it had refused to fully embrace when it had the chance. The company that had invented core digital imaging technology decades earlier was reduced to suing the companies that had actually turned that technology into products people wanted to buy. During the bankruptcy proceedings, Kodak sold a portfolio of 1,100 digital imaging patents to a consortium of buyers including Apple, Google, Microsoft, Samsung, and others for approximately $525 million. The patents that might have been the foundation of a digital imaging empire were instead liquidated to fund the restructuring. ## Rochester: The City Kodak Left Behind You can't tell the Kodak story without talking about Rochester, New York. For most of the 20th century, Kodak was Rochester, and Rochester was Kodak. The company was the largest employer in the region. It funded hospitals, parks, universities, and cultural institutions. George Eastman himself had donated over $100 million (in early 1900s dollars) to institutions including the University of Rochester, MIT, and what became the Eastman School of Music. When Kodak began its decline, Rochester felt it in layers. First came the layoffs, wave after wave through the 2000s. Then came the real estate impact as former Kodak employees left the region. Then the secondary effects: restaurants that served Kodak workers closed, small businesses that depended on Kodak supply contracts folded, and the tax base that funded public services eroded. Rochester has worked hard to reinvent itself in the years since, investing in healthcare, education, optics research, and technology startups. The University of Rochester's optics program, partly Kodak's legacy, has become a hub for photonics research. But the city is still smaller and poorer than it was during the Kodak years, and the memory of what the company's decline cost the community remains vivid. The story of Kodak's bankruptcy also has an important footnote about timing. When Kodak filed in January 2012, the smartphone camera revolution was still accelerating. Instagram had launched in October 2010 and had already reached 15 million users by the end of 2011. The idea that every person would carry a high-quality camera in their pocket at all times, that photographs would be shared instantly with hundreds or thousands of people, that the "Kodak moment" would be replaced by the Instagram post: all of this was becoming reality while Kodak was negotiating with creditors. George Eastman's original insight, that photography should be simple and accessible, had been realized more completely than he ever imagined. The camera was now built into the device people already carried everywhere. The "you press the button, we do the rest" promise had evolved into tap the screen and share with the world. Eastman would have understood the appeal immediately. His company just couldn't find a way to be part of it. ## The Real Lesson It's tempting to frame Kodak's story as simple corporate stupidity. The company that invented the digital camera and then refused to use it. But that framing is too clean. Kodak's leadership understood the digital threat. They invested in digital technology. They held the patents. They released digital products. The problem wasn't ignorance. It was incentive structure. Every quarter that Kodak sold film and processing services, it generated enormous profits. Every quarter that it invested aggressively in digital, it accelerated the destruction of those profits. The stock market, the board, and the shareholders all rewarded short-term profitability. No CEO wanted to be the one who voluntarily cratered Kodak's earnings by cannibalizing the film business, even if doing so was the only viable long-term survival strategy. This is essentially what Netflix did when it abandoned its profitable DVD-by-mail business to go all-in on streaming. Reed Hastings took the short-term hit, the stock cratered, customers revolted, and the company was mocked. But it survived. Kodak's leadership never made that leap. Not because they couldn't see the cliff. Because jumping off it voluntarily felt worse than being pushed. The Kodak story is, at its core, a story about how rational, intelligent people can collectively make decisions that lead to disaster when the incentive structures reward preservation over transformation. Every incumbent industry facing technological disruption should study it. Most of them probably won't act on the lessons, for exactly the same reasons Kodak didn't. ## Frequently Asked Questions Did Kodak really invent the digital camera? Yes. In 1975, Kodak engineer Steve Sasson built the first working digital camera prototype. It was the size of a toaster, captured images at 0.01 megapixels, and stored them on a cassette tape. Kodak held numerous foundational patents in digital imaging technology. When did Kodak file for bankruptcy? Kodak filed for Chapter 11 bankruptcy protection on January 19, 2012. The company emerged from bankruptcy in September 2013 as a significantly smaller company focused on commercial printing and imaging. Does Kodak still exist? Yes, but as a very different company. Today's Kodak focuses on commercial printing, advanced materials, and chemicals. The consumer photography business that defined the brand for over a century no longer exists in any meaningful form. How many people did Kodak employ at its peak? Kodak's global employment peaked at approximately 145,300 workers in 1988. By the time of its bankruptcy in 2012, the workforce had been reduced dramatically through years of layoffs and restructuring. What was Kodak's peak revenue? Kodak's annual revenue peaked at approximately $16 billion in 1996. Its market capitalization reached $31 billion in 1997. By the time of its bankruptcy filing in 2012, the company listed assets of $5.1 billion against debts of $6.8 billion. Could Kodak have survived if it had embraced digital earlier? Possibly, though it would have required a fundamentally different business model. The core challenge was that digital photography eliminated the consumable (film and processing) that generated most of Kodak's profits. Successfully transitioning would have meant accepting years of lower margins while building a competitive digital business, something Kodak's corporate structure and shareholder expectations made extremely difficult. --- # What Happened to the Virtual Boy, Nintendo's Most Embarrassing Failure URL: https://404memoryfound.com/posts/what-happened-to-virtual-boy-nintendo.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-14 Topics: Gaming, Hardware You're standing in the electronics section of a big-box retailer in the summer of 1995. The shelves are packed with the usual Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html) cartridges, Nintendo magazines with crease-worn spines, bright yellow price tags on everything. And then you see it. This weird red thing sitting on a stand, like some kind of office equipment that escaped from a failed startup. It's got weight to it when you pick it up, heavier than you'd expect for something called "Virtual Boy." It's got this thick headset aesthetic, something between a View-Master you'd use as a kid and a pair of ski goggles designed by someone who'd never actually seen ski goggles. You look around to see if other people are noticing what you're noticing. Some are. Most aren't. One kid is trying to put it on without asking permission, which is exactly what you wanted to do but were too self-conscious to attempt. This is the story of how Nintendo took a legendary engineer, gave him a revolutionary technology, and created something so bewildering that it would basically disappear from gaming history in less than a year. This is also the story of how you could walk into that electronics section, see the future of gaming, completely not understand what it was, and walk right back out. ## When Nintendo Handed Gunpei Yokoi an Impossible Dream Let's talk about Gunpei Yokoi for a second because this whole story falls apart if you don't understand who this person was and what he meant to Nintendo. Yokoi wasn't just some engineer. He was the guy who looked at the concept of the Game Boy, looked at everything everyone was saying about how it needed a color screen and more processing power and flashy graphics, and then went ahead and made it anyway, knowing exactly what he was doing. He created the D-pad. He created Game & Watch, which was basically just blinking LED lights that somehow managed to be one of the most addictive entertainment products ever made. He understood something about Nintendo that a lot of people still don't understand: the innovation isn't always about raw power. Sometimes the innovation is about understanding what people actually want to hold in their hands and what will actually keep them entertained. By 1995, Yokoi had already proven this philosophy about thirty different ways. He'd made billions of dollars for Nintendo through restraint and cleverness rather than brute force. And then he got interested in 3D technology. Specifically, he got interested in parallax barriers and red LED displays, which is kind of insane when you think about it, because that's such a specific, weird technical corner to decide is your future. But Yokoi believed in it. He believed that you could create a 3D gaming experience without needing an expensive LCD screen or some kind of special glasses. You could do it with red lights. Red LEDs that create a stereoscopic 3D effect through something called parallax, which is basically when your two eyes see slightly different images at slightly different times, and your brain interprets that as depth. This was actually brilliant technology. Like, genuinely clever. Which makes what happened next so much worse. ## August 14, 1995: The Day That Would Haunt Nintendo The Virtual Boy released in North America on August 14, 1995, for $179.95, which was actually a pretty reasonable price for what Nintendo was claiming to deliver. What they were claiming to deliver was the future. "Experience a remarkable leap in gaming entertainment," the marketing copy promised. This is the language of transformation. This is the language of revolution. And here's where it gets interesting, because what Nintendo was actually delivering was a headset that you had to hold with a stand. You couldn't play it in bed. You couldn't play it on the school bus. You could basically only play it sitting at a table in front of the stand while you held this weird binocular thing up to your face. Think about that for a second. Think about the absolute insanity of that design choice in 1995. The Game Boy, Yokoi's previous masterpiece, was literally defined by portability. You could play it anywhere. It was the whole point. It was the entire philosophy. And now here's the Virtual Boy, marketed as a handheld system, except it isn't handheld at all. It requires furniture. It requires a stand. It's a portable system that needs to be installed before you can use it, which is kind of insane. And the 3D effect, which was supposed to be the whole reason for existence? Yeah, it worked, technically. You'd put the thing up to your face and you'd see this red monochromatic 3D environment, and yeah, it was three-dimensional. It also gave most people headaches within about twenty minutes. It caused eye strain. The packaging came with warnings about limiting play sessions to thirty minutes. A thirty-minute limit on a gaming console. A built-in instruction that your product is only tolerable for half an hour at a time. If that doesn't tell you everything you need to know about the Virtual Boy, I don't know what would. ## The Library That Never Was Here's the problem with hardware: it doesn't matter how clever your technology is if you don't have software worth playing. And the Virtual Boy had a software library that would generously be described as "sparse." In total, across the entire lifespan of the system, Nintendo released exactly twenty-two games. Twenty-two. The Nintendo Entertainment System had hundreds. The Game Boy had dozens right at launch. The Virtual Boy had, essentially, nothing. The best game on the system, by most accounts, was Wario Land: Virtual Boy. Which is funny because even in 1995, when Wario was still relatively young as a character, the fact that your flagship title was a Wario game instead of a Mario game said something about the console's positioning. This wasn't the main event. This was a spinoff. In some regions, Mario Tennis shipped as the pack-in title, which was fine but also felt like an attempt to create legitimacy through familiar branding rather than through actual innovation. But here's the thing about the Virtual Boy's software library: it wasn't small because Nintendo couldn't develop games. It was small because developers saw where this was headed almost immediately. You know what it takes to get people excited about buying a console? You need a pipeline of games. You need developers pumped up about the technology. You need momentum. The Virtual Boy had none of that. Even Nintendo's own internal teams seemed confused about what this thing was supposed to be. You could feel it in the software. There was nothing that felt essential, nothing that felt like it could only exist on the Virtual Boy. Everything felt like a demo. Everything felt like a proof of concept. And in the consumer market, proof of concept doesn't translate to sales. ## The Red Monochrome Headache That Wouldn't Go Away Let's talk about the actual experience of using the Virtual Boy, because the technical specifications don't tell the story. The 3D effect looked wild. It was impressive in the moment. You'd see these layers of depth, these game worlds that seemed to have actual space in them, and yeah, it was something you'd never seen on a Game Boy before. But that impression lasted about fifteen minutes. After that, you'd start to notice the eye strain. The red monochrome display would create this persistent afterimage in your vision when you looked away. The weight of holding the device up to your face would start to matter. Your neck would get tired. Your hands would cramp. The thirty-minute warning on the packaging started to feel less like a recommendation and more like a threat. And the red, specifically. This was a weird choice. Historically, red gets used in displays because it's the easiest color to produce at high brightness, but that doesn't mean red is the color you want to stare directly into for any length of time. There's something almost aggressive about it, something that feels wrong for extended play sessions. You know that feeling when you've been looking at something red and bright for too long and your eyes feel kind of raw? That was the Virtual Boy experience. That was what Nintendo asked gamers to accept as the future. The controller was actually pretty innovative, with those dual D-pads and the way it was designed to work with the system's specific input needs. The problem was that innovation in one area doesn't matter when the entire experience is fundamentally flawed. It's like designing an amazing seat for a car with no engine. Sure, the seat is great, but you're not going anywhere. ## The Collapse That Happened in Slow Motion The Virtual Boy launched in August of 1995. By early 1996, it was clear that something had gone catastrophically wrong. The system had sold roughly 770,000 units worldwide, which sounds like a real number until you realize that the Game Boy was already in its seventh year and had sold millions upon millions more. The Virtual Boy was supposed to be the future. Instead, it was becoming a clearance item. Nintendo discontinued it in 1996, basically admitting defeat in a medium that they'd previously owned completely. The numbers don't fully capture the humiliation, though. This wasn't just a failed product. This was Nintendo's biggest hardware flop at the time, and coming from a company that had basically defined home and portable gaming for the previous decade, it felt like a massive crack in the foundation. There were jokes in the gaming press. Gamers had opinions, which is to say that gamers had made jokes and moved on. Within a year, the Virtual Boy had become a punchline. Within two years, it was basically forgotten. People who'd bought one early tried to resell them and found almost no market. It became one of those things that showed up at yard sales, the weird tech that nobody wanted, the thing you'd pick up sometimes just to remind yourself that it had actually existed. ## What This Meant for Gunpei Yokoi And here's where the story gets genuinely sad. Gunpei Yokoi, this legendary engineer who'd basically created Nintendo's entire handheld dominance, who'd invented the D-pad and Game & Watch and the Game Boy, who'd been right about so many things for so long, was now associated with the company's most spectacular hardware failure. He left Nintendo after the Virtual Boy's failure. He didn't stay to fight for it or to iterate on it. He just left. And then on October 4, 1997, he died in a car accident, which is just the kind of coda that makes this whole story feel even more tragic than it already was. The Virtual Boy stands as this weird monument to what happens when innovation, technology, and market understanding don't align. Yokoi was brilliant at understanding what people wanted to hold in their hands. The Virtual Boy proved that even brilliant people can miss the mark, especially when they fall in love with a specific technology without asking if that technology actually solves any problems that people need solved. ## The Legacy That Almost Isn't There Here's what's interesting about the Virtual Boy now, in hindsight. The technology wasn't actually bad. The parallax barrier 3D effect is still used in some displays. The engineering was clever. The design philosophy of trying to create 3D without expensive LCD screens or cumbersome glasses was genuinely forward-thinking. But none of that matters because the product itself was fundamentally unpleasant to use. You can have the best technology in the world, but if holding it makes your eyes hurt and your neck tired and your hands cramp, people aren't going to want it. This is kind of a lesson that the tech industry needs to relearn about once a decade, which is hilarious and also sort of depressing. The Virtual Boy disappeared from stores. It disappeared from warehouses. It disappeared from collective memory so thoroughly that most people born after 1995 have never actually seen one, let alone used one. When it shows up now, it's usually in the context of "weird Nintendo failures" or "gaming history's biggest flops." There are collectors who seek them out, who appreciate them as technical oddities, who admire what Yokoi was trying to do even if the execution was catastrophic. But that's a niche appreciation at best. What the Virtual Boy actually proved was that Nintendo was human. They could fail. They could miscalculate. They could believe too much in a technology and not enough in the actual user experience. And in some ways, that's reassuring. It means that even the companies that seem to know everything about what people want are basically just guessing, like the rest of us. They just have bigger budgets and more to lose when the guess is wrong. ## FAQ ## Why was the Virtual Boy red and not some other color? The red color came directly from the technology. The display used red LED technology to create the 3D effect, so the entire visual output was, by necessity, red. You couldn't have made it blue or green or any other color without completely changing the underlying technology, which would have defeated the whole purpose. The red was the point. It was also, as it turned out, not a great color to stare into for extended periods. ## Could modern technology make a Virtual Boy that actually works? Technically, sure. Modern LCD screens and processing power could create a 3D handheld experience that doesn't cause headaches and doesn't require a stand. In fact, the Nintendo 3DS did exactly that, using a similar parallax barrier technology but with a color display and without the physical restrictions. But the Virtual Boy didn't fail because the technology was impossible. It failed because the execution was wrong. ## Did any games actually take good advantage of the 3D? Some tried. Wario Land did a decent job of using the depth perception for puzzle-solving and navigation. Mario's Tennis used the 3D to give you a sense of where the ball was in space. But the problem was that thirty-minute play session limit kind of prevented any game from developing complex mechanics that really required the 3D. You never got a full experience because the hardware itself prevented that full experience. ## Why did Nintendo think this would work as a portable system? Marketing strategy. The Game Boy had proven that "portable" was a magic word in gaming. Developers and consumers alike associated portability with Nintendo's success. So even though the Virtual Boy required a stand and a thirty-minute play limit and would give you a headache, Nintendo still called it a handheld system. It was kind of insane, but it also allowed them to position this weird lab experiment as part of the Game Boy legacy. ## Is the Virtual Boy worth anything now? Depending on condition and whether it includes original packaging, a Virtual Boy can go for anywhere from $200 to $800 or more on the used market. The games are also collectible, especially the rarer titles. But this isn't because anyone wants to actually use the system. This is purely nostalgia and hardware collecting. --- # What Happened to the Segway, the Invention That Was Supposed to Change the World URL: https://404memoryfound.com/posts/what-happened-to-segway-personal-transporter.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-14 Topics: Hardware, Business Blunders In 2001, venture capitalists predicted the Segway Personal Transporter would generate a billion dollars in sales within a year. By 2020, when production ended, it had sold roughly 140,000 units total across its entire lifetime. That's about 0.014% of the prediction. The Segway is a fascinating case study in how a genuinely innovative product can fail not because the technology is flawed, but because the assumptions about the world it was entering turned out to be wrong. ## The Setup Dean Kamen invented things. Serious things. His company DEKA Research had already created the iBOT, a wheelchair that could climb stairs and balance on two wheels. Kamen wasn't some garage tinkerer playing with hobby projects. He had credibility. So when Kamen began working on a new personal transportation device in the late 1990s, people paid attention. The project was codenamed "Ginger" internally, which should tell you something about how seriously he took the secrecy. The code name alone generated speculation and hype before anyone outside DEKA had actually seen the thing. Here's the thing: the technology itself was genuinely impressive. The Segway used gyroscopic sensors and accelerometers to maintain balance. You stood on a platform with two wheels and leaned forward or backward to move. It was intuitive in a way that defied the mechanical complexity underneath. From an engineering standpoint, Kamen solved a real problem: how to make self-balancing personal transportation actually work. The hype machine ran at full capacity. Steve Jobs allegedly told Kamen it would be "bigger than the PC." John Doerr, the legendary KPCB venture capitalist, invested in Segway and predicted billion-dollar revenues by 2002. Major newspapers speculated about what the mysterious device could be. Some guessed it would be a new form of wheelchair. Others thought it might be a motorcycle replacement. The ambiguity was intentional, and it worked perfectly. On December 3, 2001, Kamen unveiled the Segway on Good Morning America. America was primed. Investors were ready. The product was real, it worked, and it was unlike anything most people had seen before. Then it tried to revolutionize transportation at $4,950 per unit, which in 2001 dollars was a significant amount of money. For context, you could buy a used car for that price. You could buy a decent used motorcycle. You were choosing to spend this money on a standing platform with two wheels. ## The Assumptions That Broke The people backing Segway made several logical assumptions. None of them were obviously wrong in December 2001. Assumption one: cities want to reduce cars. Urban planners and transit advocates had been talking about this forever. Congestion was getting worse. Parking was a nightmare. If you could get more people moving without individual cars, that would solve a lot of problems. A Segway could be that solution. You could navigate urban streets faster than walking, without the carbon footprint of a car. Assumption two: people who want faster personal transportation will buy whatever works best. If the Segway actually delivered on its promise, people would adopt it. Economics would favor efficiency. Forget bicycles and walking. Segway was the better mousetrap. Assumption three: the regulatory environment would adapt. Cities would designate paths for Segways. Infrastructure would follow the product. The technology was so obviously useful that governments would clear the way for deployment. This is where the logic starts to break down. ## The Problem with Being First Cities didn't embrace Segways. They banned them. Or more precisely, they weren't sure what to do with them, so they defaulted to no. Sidewalks? Too crowded. Segways pose a pedestrian safety risk, or at least that was the concern. A person moving at 12 miles per hour toward you on a crowded sidewalk doesn't feel that different from a bicycle, except the Segway rider can't necessarily dismount as quickly. Cities looked at Segways and saw liability. Bike lanes? The device wasn't a bicycle, so it didn't quite fit the regulations written for bicycles. Were Segways vehicles? Were they pedestrian devices? The regulatory ambiguity was paralyzing. Different cities made different calls. San Francisco banned them from sidewalks. Other cities allowed them in certain areas. None of this was consolidated or predictable. Here's the thing: Kamen invented a product that didn't have a legal category. The infrastructure was written for cars and pedestrians. Segways lived in the uncomfortable middle. This wasn't a technology problem. This was a regulatory problem. And regulatory problems move slowly. The first two years still looked okay on paper. Segway sold about 30,000 units by 2003, which wasn't the 50,000 to 100,000 that year-one projections had hoped for, but it was movement. The company could still claim it was building momentum. Early adopters existed. Some tour operators in cities like Boston started using Segways. There was a niche. But a niche isn't a market transformation. And markets have a way of telling you when you've overestimated their size. ## The Long Decline What happened next wasn't a dramatic crash. It was slower than that. More depressing. The company went through multiple ownership changes as enthusiasm gradually evaporated. Jimi Heselden, a British businessman who had made his fortune in defense barriers, bought Segway in 2009. He seemed like a natural fit, someone who believed in the product's vision. Then on September 26, 2010, Heselden died in a Segway accident in Yorkshire, England. He was riding the device near a cliff edge on his estate and fell into the River Wharfe. The irony was brutal. The company's owner died riding its own product. It was the kind of detail that tends to stick in people's minds and overshadow whatever else you're trying to say about your company's potential. By 2015, Ninebot, a Chinese company that manufactured electric scooters, acquired Segway in April of that year. This was a different kind of pivot. Ninebot wasn't interested in the Segway as a standalone transportation revolution. They were interested in the brand and the technology, but also in pivoting toward products with broader appeal. Electric scooters. Go-karts. Things people actually wanted to ride. The real question is whether Segway's failure was inevitable or whether different decisions could have changed the trajectory. Kamen created something genuinely innovative. The problem wasn't the engineering. The problem was that cities weren't ready, regulators weren't set up to accommodate it, and the market revealed that most people don't actually want to stand on a platform while moving through urban space. They want to sit. They want to pedal. They want something that fits into an existing category. ## Why This Matters The Segway story shows up in every discussion about innovation for a reason. It's taught in business schools as an example of good technology meeting bad market assumptions. But the more important lesson is something else: being first isn't the same as being right about what people want. Look at what happened with electric scooters two decades later. The scooter companies, particularly Lime and Bird, faced many of the same regulatory questions that Segway did. But they approached the problem differently. They were willing to operate in a legal gray area while cities figured out what to do. They accepted lower profit margins to move volume. They weren't trying to charge $5,000 per unit. They were trying to become ubiquitous, even if individual scooters didn't make much money. This is essentially what Uber and Lyft did with ride-sharing. They launched in cities where it was technically illegal, operated anyway, got shut down, lobbied, and eventually forced regulatory adaptation. It's a more aggressive strategy than Kamen's, and it worked. That's not to say Kamen was wrong to respect regulatory boundaries. It's to say that respecting boundaries and working within them, when those boundaries don't actually exist yet, can slow you down enough that you never reach critical mass. ## The End Segway PT production ended on July 15, 2020. The device had been in production for nearly two decades. Total lifetime sales were estimated around 140,000 units. That's genuinely not nothing. There were tour operators and police departments and tourism attractions that relied on the Segway for nearly twenty years. It found a niche and occupied it reasonably well. But it didn't revolutionize transportation. It didn't change how cities moved. It didn't become the next generation of personal mobility. The Segway brand still exists, but under Ninebot's ownership it's become something else entirely. Electric scooters. Ride-on machines. Products that slot into markets that already exist rather than trying to create new ones. Kamen himself has moved on to other projects. He's been working on water purification and medical devices through DEKA. He's still inventing. He still respects the challenge of building things that actually work, even if they don't always find the market he imagines they will. ## FAQ ## Why did the Segway fail if the technology was so good? The technology wasn't the problem. The issues were regulatory, infrastructural, and market-based. Cities banned or restricted Segways on sidewalks because they weren't sure how to categorize them legally. They didn't fit existing transportation categories. Meanwhile, the $4,950 price point meant the addressable market was much smaller than venture capitalists had predicted. People who wanted personal urban mobility could get a motorcycle, a scooter, or a bicycle for less money. ## Could the Segway have succeeded with different marketing or pricing? Possibly, but it would have required accepting a much smaller market. If Segway had positioned itself from the start as a niche product for tourists and specialized use cases, rather than trying to be the future of urban transportation, expectations would have aligned with reality. Alternatively, a lower price point might have expanded the potential market, but that would have required different manufacturing and investment strategies. The company was built around the assumption of massive adoption, and when that didn't materialize, the economics stopped working. ## What did cities eventually do about regulating personal mobility devices? Most cities eventually created categories for electric scooters, bikes, and similar devices, but this took years. By the time regulatory frameworks actually existed, Segway had already lost momentum. Electric scooters came later and benefited from more lenient regulatory environments, partly because regulators had learned from earlier devices like the Segway. ## Is the Segway still being manufactured anywhere? Segway PT production ended in July 2020, but the Segway brand continues under Ninebot. They make electric scooters, go-karts, and other devices. If you want to buy a new Segway product today, you're buying something very different from the original Personal Transporter. The brand survived, but the original product line didn't. ## What would happen if the Segway launched today? It would likely face the same regulatory questions, but with more established precedent. Cities now have frameworks for electric scooters and bikes. A Segway-like device might fit into those categories more easily. That said, the fundamental market question remains: do people want this form of transportation enough to pay for it? The fact that electric scooters dominate the personal mobility space, despite being cheaper and more familiar, suggests that a new Segway wouldn't automatically succeed. --- # What Happened to the Game Boy, Nintendo's Unbreakable Handheld URL: https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-13 Topics: Gaming, Hardware Picture this: it's Christmas morning, 1989. You're sitting on the carpet in your living room, surrounded by torn wrapping paper, and you're holding this chunky gray rectangle that weighs about as much as a can of soup. The screen is a weird greenish tint. There's no backlight. The graphics look like something from five years ago. And you could not care less, because Tetris is falling from the top of that little screen and you are completely, hopelessly, irreversibly hooked. That gray brick was the Nintendo Game Boy, and it changed everything about how we thought about video games. Not because it was the most powerful. Not because it had the best screen. Not because it looked cool. It changed everything because one man at Nintendo understood something that the rest of the industry didn't: the best technology doesn't always win. Sometimes the best idea wins. And the Game Boy was, from top to bottom, one of the best ideas anyone in gaming ever had. Today, the original Game Boy is a collector's item. You can find them at swap meets, retro gaming stores, and your parents' attic. The line was eventually discontinued in the 2000s after spawning the Game Boy Pocket, Game Boy Color, and Game Boy Advance. But the story of how this thing came to exist, how it crushed technologically superior competitors, how it nearly died and then got resurrected by a bunch of fictional pocket monsters, that story is one of the wildest rides in gaming history. ## The Man Who Built a Game Boy Before There Was a Game Boy You can't talk about the Game Boy without talking about Gunpei Yokoi, and you can't talk about Gunpei Yokoi without understanding his philosophy. Yokoi was a maintenance man at Nintendo in the 1960s. Literally. He maintained the machines on the assembly line. One day, the president of Nintendo, Hiroshi Yamauchi, saw Yokoi playing with an extendable arm toy he'd built in his spare time. Yamauchi told him to turn it into a product. That toy became the Ultra Hand, and it sold over a million units. Yokoi was promoted out of maintenance and into product development, and Nintendo's trajectory changed forever. Yokoi developed a design philosophy he called "lateral thinking with withered technology." The idea was deceptively simple: don't chase cutting-edge components. Instead, use technology that's mature, cheap, well-understood, and abundant. Then find a creative new use for it. When Sharp and Casio were in a price war over digital calculator components in the early 1980s, Yokoi looked at all those cheap LCD screens and semiconductors flooding the market and thought: what if we made games with these? The result was the Game & Watch series, which sold 43.4 million units between 1980 and 1991. The Game Boy was the next logical step. Yamauchi wanted a portable console that could play interchangeable cartridges, something beyond the single-game limitations of Game & Watch. Yokoi and his R&D1 team got to work. And the decisions they made during development would define the entire handheld gaming market for the next decade. ## The Decision That Won the War Before It Started Here's where the story gets interesting. By the late 1980s, LCD technology had advanced to the point where color screens were possible in handheld devices. Yokoi knew this. His engineers knew this. They could have put a color screen in the Game Boy. They chose not to. Yokoi's reasoning was pure "withered technology" logic. A color screen in 1989 would drain batteries in a few hours. It would make the unit more expensive. It would generate more heat. And the display quality of affordable color LCDs at the time was honestly not great. So Yokoi went with a monochrome, green-tinted dot matrix screen. It wasn't pretty. It wasn't flashy. But it sipped power. The Game Boy ran for 30 or more hours on four AA batteries. Thirty hours. Think about that for a second. The other key decision was the processor. The Game Boy used a custom 8-bit Sharp LR35902 CPU running at 4.19 MHz. For context, this was roughly comparable to the processing power of the original NES, a home console from 1983. In 1989, this was ancient. But it was cheap, reliable, and Yokoi's team knew exactly how to squeeze every drop of performance out of it. The price tag? $89.95 at launch in the United States on July 31, 1989. And it came bundled with a game. Not just any game. The game. ## Tetris: The Deal That Changed Everything The story of how Tetris ended up bundled with the Game Boy is, frankly, the kind of thing that sounds like it was made up for a movie. (They did make a movie about it, actually. Apple TV+ released it in 2023.) Tetris was created by Alexey Pajitnov, a Soviet computer scientist, in 1985. By the late 1980s, the rights to the game were tangled in a web of licensing deals, sub-licenses, and Cold War politics. Multiple companies claimed they owned the rights to different versions on different platforms. It was a mess. Enter Henk Rogers, a Dutch-born game developer living in Japan. Rogers had discovered Tetris at the Consumer Electronics Show in January 1988 and immediately understood its potential. He secured the Japanese console rights and brought the game to Nintendo's attention. When he learned that the handheld rights were still up for grabs, he flew to Moscow. In the middle of winter. To negotiate directly with ELORG, the Soviet government's software export agency. Rogers got the deal done. He secured the handheld rights to Tetris for Nintendo. And then he said something to Minoru Arakawa, the president of Nintendo of America, that might be the single most important piece of marketing advice in gaming history: "If you want to sell Game Boys to little boys, bundle it with Mario. If you want to sell Game Boys to everyone, bundle it with Tetris." Nintendo bundled it with Tetris. The results were immediate and staggering. In Japan, the entire initial stock of 300,000 units sold out in two weeks. In the United States, one million units were sold within the first few weeks. The Game Boy didn't just launch well. It launched like a rocket. ## The Competition That Should Have Won (But Didn't) Here's where a lot of people get confused about the Game Boy's story. On paper, it should have been destroyed by its competitors. The Atari (https://404memoryfound.com/posts/what-happened-to-atari-lynx-handheld.html) Lynx launched just two months after the Game Boy, in September 1989, with a color screen, a more powerful processor, and the ability to link up to 18 units for multiplayer. It cost $179.99. The Sega (https://404memoryfound.com/posts/what-happened-to-sega-game-gear-handheld.html) Game Gear arrived in 1990 in Japan and 1991 in North America, also with a full-color backlit screen, at $149.99. Both of these handhelds were objectively more impressive from a technical standpoint. The Lynx's screen was gorgeous. The Game Gear was basically a portable Sega Master System. If you put them side by side with a Game Boy, you'd pick one of the competitors every time. The Game Boy looked like a calculator next to them. But here's what Atari and Sega didn't account for. The Lynx ate six AA batteries in four to five hours. The Game Gear was even worse, burning through six AA batteries in three to five hours. Parents buying Christmas gifts in 1989 and 1990 were doing the math. A Game Boy ran for 30 hours on four AAs. A Game Gear ran for maybe four hours on six. That's not a minor difference. That's the difference between a toy you can take on a road trip and a toy that dies before you reach the state line. And then there was the game library. Nintendo had relationships with nearly every major third-party developer. The Game Boy launched with titles like Super Mario Land, Tetris, Alleyway, and Baseball. Within two years, it had hundreds of games. The Lynx and Game Gear had decent libraries, but nothing close. The Game Boy was where the games were, and in gaming, that's all that matters. The Atari Lynx sold roughly 3 million units before being discontinued. The Sega Game Gear sold about 10.6 million. The Game Boy? Over 118 million when you include the Game Boy Color. It wasn't even close. ## The Mid-90s Slump and the Pocket That Fixed It By 1995, the Game Boy was starting to show its age. The original hardware was six years old. The SNES and Sega Genesis were dominating home gaming with 16-bit graphics, and the Game Boy's chunky 8-bit look felt increasingly dated. Sales were slowing down. Nintendo was focused on developing the Nintendo 64. The Game Boy wasn't dead, but it was definitely fading. Nintendo's answer was the Game Boy Pocket, released in Japan in July 1996 and in North America in September 1996. It was smaller, lighter, had a sharper screen (still monochrome, but with a true gray-and-white display instead of the green tint), and ran on just two AAA batteries. It was, in every way, a refinement rather than a reinvention. Same processor, same game compatibility, just a sleeker package. The Pocket helped, but what really saved the Game Boy was something nobody at Nintendo could have predicted. Something that started as a weird little project by a guy obsessed with collecting bugs as a kid. ## Pokemon: The Miracle That Gave the Game Boy a Second Life Satoshi Tajiri grew up in a suburb of Tokyo that, during his childhood, was still semi-rural. He loved catching insects and tadpoles. As the area urbanized and the fields disappeared, he wanted to give other kids the experience of collecting creatures. He pitched his idea to Nintendo in the early 1990s. It took six years to develop. The project nearly bankrupted his company, Game Freak. Multiple developers quit during production. Pokemon Red and Green launched in Japan on February 27, 1996. They were released for the Game Boy. Not the upcoming Nintendo 64. Not some new handheld. The Game Boy, a seven-year-old piece of hardware that most people in the industry had written off. The games were slow to catch on at first. Initial reviews were modest. But word of mouth spread, and by the end of 1996, Red, Green, and Blue (an updated version of Green) had sold 1.04 million copies in Japan. In 1997, that number exploded to 3.65 million, making Pokemon the best-selling game in the country, beating Final Fantasy VII. When Pokemon Red and Blue launched in North America on September 28, 1998, it was a phenomenon. Two hundred thousand copies sold in the first two weeks. By the end of the year, four million units were gone. The games single-handedly revived Game Boy hardware sales. Stores couldn't keep Game Boys in stock. In 1998, a device that was almost a decade old was suddenly the hottest thing in gaming again. The timing was perfect for the Game Boy Color, which launched in Japan in October 1998 and in North America in November. It was backward-compatible with every original Game Boy game. Pokemon Gold and Silver, released in 1999 in Japan and 2000 worldwide, would go on to sell 23 million copies. The Game Boy line wasn't just alive. It was thriving. ## Game Boy Advance and the End of an Era The Game Boy Advance launched on March 21, 2001, in Japan and June 11 in North America. It was the first true generational leap for the Game Boy line: a 32-bit ARM7 processor, a larger screen, and graphics roughly comparable to the SNES. It was still backward-compatible with Game Boy and Game Boy Color cartridges, which was huge. Your entire library carried over. The GBA sold 81.51 million units worldwide across its various models, including the GBA SP (the clamshell redesign that finally added a backlit screen in 2003) and the Game Boy Micro (a tiny, oddly beautiful device released in 2005 that nobody bought). But the writing was on the wall. The Nintendo DS launched in November 2004, with its dual screens and touch input, and it was clear that the Game Boy name was being retired. Nintendo never officially announced the death of the Game Boy line. There was no press release. The last Game Boy Advance games trickled out in 2007 and 2008, and production of GBA hardware ceased around the same time. The Game Boy just quietly stepped aside, its job done. ## The Legacy That Won't Quit The original Game Boy is one of the most important consumer electronics products ever made. That's not hyperbole. Before the Game Boy, portable gaming was a novelty. After the Game Boy, it was an industry. The entire lineage of handheld gaming, from the PSP (https://404memoryfound.com/posts/what-happened-to-sony-psp-handheld.html) to the Nintendo DS to the Switch to the Steam Deck, traces back to that chunky gray rectangle. Gunpei Yokoi didn't live to see the full extent of what he'd created. He left Nintendo in 1996 after the commercial failure of the Virtual Boy (https://404memoryfound.com/posts/what-happened-to-virtual-boy-nintendo.html), a 3D headset that was ahead of its time but executed poorly. On October 4, 1997, Yokoi died in a car accident on a highway in Japan. He was 56 years old. The news devastated the gaming community, and tributes to his work continue to this day. But his philosophy endures. "Lateral thinking with withered technology" is still invoked every time someone asks why Nintendo doesn't chase raw power. The Wii outsold the PlayStation 3 and Xbox 360 with hardware that was essentially a slightly upgraded GameCube. The Nintendo Switch uses a mobile processor that was outdated at launch. And both were massive hits, because Nintendo understood what Yokoi taught them decades ago: it's not about the specs. It's about the experience. The Game Boy proved that in 1989. A monochrome screen. An 8-bit processor. Four AA batteries. And a little Russian puzzle game that grabbed you by the brain and wouldn't let go. That was enough. That was more than enough. If you had one, you already know. And if you didn't, well. You missed something special. ## Frequently Asked Questions When did the original Game Boy come out? The Game Boy launched in Japan on April 21, 1989, and in North America on July 31, 1989. The European release followed on September 28, 1990. It launched at $89.95 in the United States, bundled with Tetris. How many Game Boys were sold? The original Game Boy and Game Boy Color combined sold over 118.69 million units worldwide. When you include the Game Boy Advance family, the total Game Boy lineage sold over 200 million units. Why did the Game Boy only have a green screen? Designer Gunpei Yokoi intentionally chose a monochrome screen to maximize battery life (30+ hours on four AA batteries) and keep the price low. Competitors with color screens like the Sega Game Gear and Atari Lynx lasted only 3 to 5 hours on six batteries. Who invented the Game Boy? Gunpei Yokoi led the development of the Game Boy at Nintendo's R&D1 division. Yokoi also created the Game & Watch series and produced franchises like Metroid and Kid Icarus. His design philosophy of "lateral thinking with withered technology" defined Nintendo's approach to hardware for decades. When was the Game Boy discontinued? Nintendo never made a formal announcement. The Game Boy Color was last produced around 2003. The Game Boy Advance line continued until around 2008-2010, when remaining stock was sold off. The Nintendo DS effectively replaced the Game Boy brand starting in 2004. Did Pokemon really save the Game Boy? Yes. By the mid-1990s, Game Boy sales were declining significantly. Pokemon Red and Green launched in Japan in February 1996 and revived hardware sales almost overnight. When the games reached North America in 1998, they sold four million copies by year's end and made the nearly decade-old Game Boy the hottest gaming device on the market again. --- # How the iMac G3 Saved Apple and Killed the Beige Box Forever URL: https://404memoryfound.com/posts/how-imac-g3-saved-apple-killed-beige-box.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-13 Topics: Hardware, Then vs Now In October 1997, Michael Dell was asked at a technology conference what he would do if he ran Apple Computer. His answer: "I'd shut it down and give the money back to the shareholders." At the time, this was not an unreasonable position. Apple had just posted its biggest losses in history. The company had burned through four CEOs in a few years. Its product line was a confusing mess of overlapping models with names like Performa 6400/180 and Power Macintosh 7300/200. Market share had cratered below 4%. Investors were fleeing. The narrative was settled: Apple was done. Ten months later, a translucent blue computer shaped like a gumdrop went on sale for $1,299. Within six weeks, Apple had sold 278,000 of them. Within five months, that number was 800,000. Apple went from losing $878 million in 1997 to posting a $414 million profit in 1998, its first positive year in three. The product was the iMac G3, and it didn't just save Apple from bankruptcy. It fundamentally changed what people expected a computer to look like, how they connected their peripherals, and what an industrial designer could achieve when a company actually listened to them. The story of how this happened is not a simple tale of good design. It's a story about organizational dysfunction, one very specific hiring decision, a bet against the floppy disk, and the moment when a consumer electronics company decided that how something looked was just as important as what it could do. ## The Company Steve Jobs Came Back To To understand why the iMac mattered, you need to understand how bad things were at Apple in 1997 (https://404memoryfound.com/posts/when-apple-almost-went-bankrupt-1997.html). The company had acquired NeXT, Steve Jobs' post-Apple venture, in February 1997 for $429 million. The stated reason was that Apple needed NeXT's operating system technology. The unstated reason was that Apple needed Steve Jobs. Jobs returned as an "advisor" in February 1997. By July, he had effectively taken control of the company as interim CEO (he wouldn't drop the "interim" label until January 2000). What he found was a company that had lost its way so completely that it was selling over a dozen different desktop models simultaneously, many of which differed only in minor configuration details. There were Performas, Power Macintoshes, Quadras, and various other product lines that even Apple employees struggled to keep straight. Jobs famously drew a simple four-quadrant grid on a whiteboard: Consumer and Professional across the top, Desktop and Portable down the side. Four products. That's it. Everything else would be killed. The iMac was designed to fill the Consumer Desktop quadrant, and it needed to do more than just work. It needed to make people care about Apple again. ## Jony Ive Had Been There the Whole Time Here's a detail that gets overlooked in most retellings. Jonathan Ive had been at Apple since 1992. He became head of the Industrial Design Group in 1996, a full year before Jobs returned. He had been designing products at Apple for five years, producing competent but unremarkable work, because the company's leadership had never given design a seat at the strategic table. Jobs changed that immediately. He recognized in Ive what Apple's previous CEOs had missed: a designer who understood materials, form, and emotion at an almost obsessive level. Jobs gave Ive something he'd never had at Apple before: direct access to the CEO and the authority to lead product aesthetics without being overruled by engineering or marketing. The iMac design process started in late 1997. Ive and his team began with a clear mandate from Jobs: this computer needed to be approachable. It needed to look like nothing else on the market. It needed to make people want to touch it. ## Bondi Blue and the Death of Beige Every personal computer in 1998 was beige. Or light gray. Or off-white. This is not an exaggeration. Walk into any CompUSA (https://404memoryfound.com/posts/is-compusa-still-around.html), any Circuit City (https://404memoryfound.com/posts/what-happened-to-circuit-city-electronics.html), any Best Buy in the spring of 1998 and you were looking at rows of identical putty-colored boxes. Dell, Compaq (https://404memoryfound.com/posts/compaq-lost-pc-business-hp-merger.html), HP, Gateway: they all looked like they came out of the same factory, because functionally, they did. Industrial design in the PC industry at that point meant deciding where to put the power button. Ive's team went in the opposite direction entirely. The iMac G3 was egg-shaped, or more precisely, a teardrop. The case was made of translucent polycarbonate plastic, colored in a shade Ive called "Bondi Blue" after the Australian beach. The inspiration came from a piece of beach glass someone on the team brought to the office during early design meetings. The team experimented with solid plastic first, but it looked cheap. Translucent plastic, with the internal components visible as shadowy shapes behind the colored shell, created something that felt alive. Jobs initially wasn't sold on the egg shape. He warmed to it over time, reportedly won over by its playfulness. This is an important detail because it illustrates something about the Jobs-Ive dynamic that would define Apple for the next fifteen years: Ive pushed, and Jobs, despite his reputation as a dictator, was willing to be pushed when the argument was good enough. The iMac was also an all-in-one design, with the monitor, logic board, CD-ROM drive, modem, and speakers all housed in a single enclosure. This was not new for Apple (the original Macintosh from 1984 was also an all-in-one), but it was a deliberate counter to the PC industry's trend of separate towers, monitors, and tangled cable nests. You plugged in the power cord, connected the keyboard and mouse, and you were done. Apple's marketing leaned into this with the famous "three steps" ad: plug in, connect to the internet, there is no step three. ## The Floppy Drive That Wasn't There The iMac shipped without a floppy drive. In 1998, this was seen as borderline insane. The 3.5-inch floppy disk had been the primary method of moving files between computers since the mid-1980s. Removing it from a consumer product, one explicitly targeted at home users and first-time computer buyers, struck many critics as arrogant. One reviewer called it "an astonishing lapse from Jobs." The reaction in tech media was sharp. How were people supposed to transfer files? What about all their existing floppies? Jobs' counterargument was that the floppy was already obsolete, even if people hadn't realized it yet. The iMac included a 56k modem and an Ethernet port. Email attachments and network file sharing were the future. Writable CDs were coming. And the floppy's 1.44MB capacity was increasingly inadequate in a world of multi-megabyte files. He was right, but the timing was aggressive. In 1998, most American households still had dial-up internet (https://404memoryfound.com/posts/dialup-speed-then-vs-now.html), if they had internet at all. Sending a file via email meant waiting minutes for an upload. The practical reality was that for many buyers, the lack of a floppy drive was genuinely inconvenient. Third-party USB floppy drives became a popular accessory for iMac owners, which is both an indictment and a validation of Jobs' bet: people bought the computer anyway, and the market for external floppy drives dried up within two years. ## USB: The Port Nobody Cared About Until Apple Made Them The iMac's other controversial hardware decision turned out to be even more significant in the long run. It had two USB 1.1 ports and nothing else for peripherals. No ADB ports (Apple's proprietary standard). No serial ports. No SCSI. No parallel. Just USB. USB had been introduced in 1996 as a joint standard by Compaq, DEC, IBM, Intel, Microsoft, NEC, and Nortel. By 1998, it had been available for nearly three years and almost nobody was using it. PC manufacturers included USB ports on their machines but also kept all the legacy ports, so there was no incentive for peripheral makers to switch. It was a chicken-and-egg problem that the industry had been unable to solve for three years. Apple solved it in a single product launch. The iMac had USB and nothing else. If you wanted to sell a printer, a scanner, a mouse, a keyboard, or any other peripheral to iMac owners, it had to be USB. What the entire PC industry couldn't accomplish in three years of gradual adoption, Apple accomplished overnight by simply removing the alternative. Within months, the peripheral market transformed. USB devices flooded store shelves. And because USB was a universal standard, those devices worked with PCs too. The iMac didn't just popularize USB for Mac users. It catalyzed USB adoption for the entire computing industry. Every USB device you've ever plugged into a computer owes a small debt to a translucent blue gumdrop from 1998. ## The Sales Numbers That Saved a Company The iMac went on sale August 15, 1998. It was the number one selling desktop computer in American retail stores for its first three months. The 278,000 units sold in the first six weeks represented the strongest launch Apple had seen in years. A significant percentage of buyers, approximately 32% according to Apple's own surveys at the time, were first-time computer owners or people switching from Windows. The iMac was bringing entirely new customers into the Apple ecosystem. Apple's fiscal year 1998 results told the story in financial terms. Revenue stabilized. The $878 million loss of 1997 flipped to a $414 million profit. The stock price, which had been in free fall, began its long climb upward. The iMac did not accomplish all of this alone (Jobs also slashed the product line, cut costs, and renegotiated distribution deals), but it was the centerpiece. It was the product that proved Apple could still make something people wanted. ## Five Flavors and the Design Revolution In January 1999, Apple replaced the original Bondi Blue iMac with five new colors: Blueberry, Strawberry, Tangerine, Grape, and Lime. This was not just a refresh. It was a statement about what computers could be. Personal computers had been sold as tools, like office equipment. The "fruit colors" iMac was being sold as a lifestyle product, something you chose because it reflected your personality. The impact on the broader industry was immediate and, in retrospect, kind of funny. Within a year, translucent plastic started appearing everywhere. George Foreman grills, staple guns, desk lamps, alarm clocks, and yes, competing PCs all suddenly came in translucent blue, green, and orange. The trend was so pervasive that it became a visual shorthand for "late 1990s design." Ive and his team had created a design language so compelling that it leaked out of the computer industry entirely. The iMac G3 continued to evolve through 2003, cycling through slot-loading drives, faster processors, new form factors, and increasingly creative color options (Graphite, Ruby, Sage, Indigo, Snow, Flower Power, Blue Dalmatian). The Flower Power and Blue Dalmatian models, released in 2001, are widely regarded as the low point of the line's design. Even great designers miss sometimes. ## What the iMac G3 Actually Meant The iMac G3's significance extends well beyond Apple's balance sheet. It established several principles that would define the next two decades of consumer electronics. First, it proved that design could be a primary differentiator in technology products. Before the iMac, the conventional wisdom in the PC industry was that consumers bought on specs and price. The iMac demonstrated that a significant number of people would pay a premium for something that looked good and felt different. This insight would drive Apple's entire product strategy going forward, from the iPod (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html) to the iPhone to the Apple Watch. Second, it showed that removing features could be as important as adding them. The no-floppy-drive decision was the prototype for Apple's later moves: removing the optical drive from the MacBook Air in 2008, removing the headphone jack from the iPhone 7 in 2016, removing ports from the MacBook in 2015. Not all of these decisions were equally justified, but the playbook was established with the iMac. Third, it demonstrated that a single product could change a company's trajectory. Apple in early 1998 was a company most analysts expected to fail. Apple in early 1999 was a company with momentum, cash, and a design identity that made it the most talked-about brand in tech. One product did that. One translucent, floppy-drive-lacking, USB-only computer shaped like an egg. Look, the iMac G3 wasn't a perfect machine. The original Bondi Blue model had a tray-loading CD-ROM drive that was prone to jams. The 233 MHz G3 processor was adequate but not exceptional. The 15-inch CRT display was standard for the era but nothing special. The built-in speakers were mediocre. As a pure computer, evaluated on specifications alone, there were better options available for $1,299 in 1998. But that misses the point entirely. The iMac wasn't selling specifications. It was selling the idea that a computer could be delightful, that technology could have personality, that the object sitting on your desk could make you feel something beyond utilitarian satisfaction. That was the insight that saved Apple. And it started with a piece of beach glass, a stubborn designer, and a CEO who, for once, was willing to listen. ## Frequently Asked Questions When did the iMac G3 come out? Steve Jobs unveiled the iMac G3 on May 6, 1998, and it shipped to customers on August 15, 1998. The original model came only in Bondi Blue. Five new colors (Blueberry, Strawberry, Tangerine, Grape, and Lime) were introduced in January 1999. How much did the iMac G3 cost? The original iMac G3 launched at $1,299. This was significantly cheaper than most comparably equipped Macintosh models at the time, which typically cost around $2,000 or more. The lower price point was central to its strategy of attracting first-time computer buyers. Why didn't the iMac G3 have a floppy drive? Steve Jobs argued that the floppy disk's 1.44MB capacity was already obsolete. The iMac included a modem and Ethernet for file transfer. The decision was controversial at the time but accelerated the industry's shift away from floppy disks. Third-party USB floppy drives were available for users who still needed one. Did the iMac really save Apple? It played a critical role. Apple went from losing $878 million in 1997 to earning a $414 million profit in 1998. The iMac sold 800,000 units in its first five months and was the top-selling desktop in US retail for three consecutive months. Combined with Jobs' restructuring of Apple's product line and operations, the iMac gave the company the financial breathing room and brand momentum it needed to survive. Who designed the iMac G3? Jonathan (Jony) Ive led the design as head of Apple's Industrial Design Group. Ive had been at Apple since 1992 but only gained significant creative authority after Steve Jobs returned in 1997. The iMac was the first major product of the Jobs-Ive collaboration that would define Apple's design language for the next two decades. Why was the iMac G3 called "Bondi Blue"? The color was named after Bondi Beach in Sydney, Australia. Jony Ive's design team was inspired by a piece of blue-green beach glass brought to the office during early design sessions. The translucent polycarbonate case was chosen because solid plastic prototypes looked cheap, while the see-through material created a sense of depth and personality. --- # What Happened to Counter-Strike, the Mod That Became Bigger Than the Game URL: https://404memoryfound.com/posts/what-happened-to-counter-strike-half-life-mod.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-13 Topics: Gaming, Software & Apps What Happened to Counter-Strike, the Mod That Became Bigger Than the Game Picture this. It's 1999, and you're a college student with a borrowed copy of Half-Life and an internet connection that screams like a wounded animal at 56 kilobits per second. Two young modders, Minh Le and Jess Cliffe, are tinkering away on something in their spare time. They're not making the next blockbuster title. They're not getting venture capital or signing publishing deals. They're just scratching an itch, building a total conversion mod that reimagines Half-Life as a terrorist vs. counter-terrorist tactical shooter. Neither of them could have predicted that their weekend project would eventually sell over 50 million copies worldwide and fundamentally reshape competitive gaming. This is the story of Counter-Strike, and how a mod became bigger than the game it was built on. ## The Origin Story: Two Kids and a Dream Let's rewind to January 1999. The gaming industry was in a very different place. Half-Life had just launched in November 1998, and while it was critically acclaimed, the modding community was still finding its footing. Minh "Gooseman" Le and Jess Cliffe weren't industry veterans. They were passionate gamers who saw potential in Half-Life's engine and decided to build something that didn't exist yet. They wanted to create a competitive, objective-based tactical shooter that emphasized teamwork, positioning, and strategy over raw reflexes. The mod was officially named on March 15, 1999, during an ICQ (https://404memoryfound.com/posts/what-happened-to-icq-instant-messenger.html) chat session. That simple message in a chat window marked the beginning of something that would change gaming forever. But at that moment, it was just an idea. The real work came when Valve released the Half-Life SDK on April 7, 1999. Now Le and Cliffe had access to the actual tools they needed to turn their vision into reality. They started building in earnest, and by June 19, 1999, they released the first beta version to a small group of testers. Some sources point to July 15, 1999, for the public beta 1 release, but either way, the mod was finally in people's hands. That first beta was rough around the edges, but it contained the seeds of what would become a phenomenon. It had a hostage rescue scenario that became iconic. It had nine weapons that felt distinct and required different tactical approaches. It had four maps that players would memorize, dissect, and dominate over the coming years. And it had one player model for each side, which seems quaint now but was actually a pretty significant detail for team identification. The mod was playable, and more importantly, it was fun in a way that caught people's attention. Word started spreading through forums and gaming websites. People were downloading this little mod and finding something special hidden inside Half-Life. ## The Betas and the Birth of a Community What happened between that first public beta and the 1.0 release was something that rarely happens in gaming. Le and Cliffe listened to their community. They iterated. They patched. They refined. They went through version after version, each numbered beta bringing improvements that felt substantial. The community grew with each release, not as some passive audience waiting for the next update, but as active participants. Players had forums. They had clan matches. They had strategies that they shared and debated. The mod became less like software and more like a living, breathing thing that evolved based on what its users wanted. This was a golden age of modding, before corporate interests had fully colonized the space. Anyone with enough dedication could download the tools and create something. Counter-Strike represented the best of that era: a mod made with passion by people who loved games and wanted to make something better. The development went through numerous iterations as the playerbase grew exponentially. Each beta refined the weapon balance, tweaked the maps, and introduced new features that kept the game fresh and compelling. By the time we hit the mid-betas around version 5.0 in early 2000, something remarkable had happened. Counter-Strike had attracted the attention of Valve Corporation itself. Gabe Newell and the team at Valve weren't just watching this mod succeed. They recognized something rare: they were seeing the future of competitive gaming being built by their community. Rather than shut it down or try to extract value from it, Valve made a pivotal decision. They partnered with Le and Cliffe. They brought them into the fold. This was the moment when Counter-Strike stopped being a hobby project and became an official Valve property. ## The Release That Changed Everything Counter-Strike 1.0 launched on November 9, 2000, and it was the culmination of everything that had come before. This wasn't a mod anymore. This was a retail product that Valve was backing with full marketing support. It was included with Half-Life in some distribution channels. It was available for purchase on its own. It was stable, balanced, and content-complete in a way that the betas never quite were. The moment it hit retail, something shifted in the gaming landscape. Here was this game that nobody was expecting to be huge, that came from the community rather than a major studio, and it was immediately capturing people's imaginations. The numbers tell the story. By 2008, Counter-Strike had sold over 4.2 million retail copies. That doesn't even count the millions of people who played it through Steam or other digital channels over the years. It was one of the best-selling games in history, and it had started as a hobby project by two guys with a good idea. The competitive scene exploded. People formed clans. Tournaments were organized. LAN parties (https://404memoryfound.com/posts/golden-age-of-lan-parties.html) became a thing where Counter-Strike was the main event. You'd see packed arenas with dozens of computers running the game, with spectators watching matches on projector screens, with real money on the line. What made Counter-Strike different from everything else was how it played. This wasn't a run-and-gun shooter like Doom (https://404memoryfound.com/posts/how-doom-changed-pc-gaming-internet.html) or Quake (https://404memoryfound.com/posts/what-happened-to-quake-online-fps.html). It required positioning, communication, and strategy. A single player with a good bomb plant could disrupt an entire round. The economy system, where teams earned money for kills and objectives, meant that tactical decision-making extended beyond the moment-to-moment action. Should you buy full weapons or eco-round it? Should you play for map control or rush the enemy position? These questions made Counter-Strike intellectually engaging in a way that pure reflex shooters weren't. ## Competitive Gaming Goes Mainstream: The CPL Era If Counter-Strike 1.0 was the spark, the Cyberathlete Professional League was the accelerant that turned it into a wildfire. The CPL recognized early on that Counter-Strike had something special. They built their competitive infrastructure around the game. Tournaments were organized with serious prize pools. The 2001 CPL World Championship had a prize pool of $150,000, which was an absolutely staggering amount for esports at the time. Teams competed for glory and for cash. Players became celebrities within gaming circles. You had your favorite teams and your favorite players, and you'd follow their performances across different tournaments. The thing about Counter-Strike's competitive scene was that it felt accessible. You didn't need special hardware. You didn't need to buy exclusive content. You downloaded the game, joined a clan, practiced with your team, and potentially worked your way up to professional competition. The skill ceiling was incredibly high, but the entry barrier was low. Kids in bedrooms across the world were practicing spray patterns and learning smokes and callouts, dreaming of making it big. Some of them actually did. Counter-Strike became proof that esports was viable, that people would watch competitive gaming, that you could build a scene around a single title that would sustain for years. The pro scene developed regional flavors. Certain countries became known for producing exceptional Counter-Strike talent. The Scandinavian countries produced some of the best teams. Teams like SK Gaming and Fnatic dominated international competition. Players like SpawN, Potti, and earlier legends of the game became household names in gaming communities. Every major tournament was an event. Every new team lineup was analyzed and debated. The game had created something that went beyond the game itself: it had created culture, community, and legitimate competition. ## The Steam Transition and Counter-Strike 1.6 In 2003, something happened that cemented Counter-Strike's position in gaming history. Counter-Strike 1.6 launched with Steam on September 12, 2003. Steam was Valve's new digital distribution platform, and it fundamentally changed how PC games were sold and updated. For Counter-Strike, this meant something crucial: the game was now living on Steam, receiving automatic updates, and being part of a larger ecosystem that would eventually dominate PC gaming. The transition to Steam wasn't seamless. There was genuine friction between players who preferred the old WON (World Opponent Network) servers and those who were forced onto Steam. But Valve was resolute. Steam represented the future, and Counter-Strike would be part of that future. The move turned out to be prescient. As Steam grew to become the dominant digital storefront for PC games, Counter-Strike grew with it. New players were constantly discovering the game because it was right there on Steam, often at a bargain price, with an active community and regular updates. Counter-Strike 1.6 remained the competitive standard for years. Even as new versions were released, 1.6 remained the game of choice for professional competitions and serious players. The map design, the weapon balance, the feel of the game: it all became iconic. Maps like Dust2, Inferno, and Nuke became legendary battlegrounds. Players spent thousands of hours learning every pixel of these maps, every sightline, every choke point. The competitive meta evolved constantly, with new strategies emerging and old tactics becoming obsolete. But the core game remained fundamentally sound, and that's why it endured for so long. ## Expansion and Evolution: Source, Global Offensive, and Beyond By 2004, Counter-Strike: Source arrived as a modern reimagining of the classic. Source was built on a newer engine, had updated graphics, and introduced new content. It was supposed to be the future. But something interesting happened. The competitive community largely rejected it in favor of 1.6. Source looked prettier, sure, but it didn't feel the same to play. The weapon mechanics were different. The movement was different. The spray patterns were different. For professional players who had spent years mastering 1.6, switching to Source felt like learning a new game. The community split, with some embracing Source and others sticking religiously to 1.6. This split showed something important about gaming communities. Nostalgia and familiarity matter. Gameplay feel matters more than fancy graphics. The competitive players didn't care that Source looked better. They cared that it played differently, and different was a deal breaker. So Source developed its own following among more casual players and teams that didn't have the professional aspirations of the hardcore crowd. Both versions coexisted, but 1.6 remained the competitive standard for a surprisingly long time. Then came 2012 and Counter-Strike: Global Offensive. This was Valve's attempt to modernize Counter-Strike for a new generation while still maintaining what made the series great. Global Offensive was built on the Source engine but updated extensively. It had modern graphics, refined gameplay mechanics, and was designed from the ground up to support both casual and competitive play. More importantly, it had Valve's full support and marketing muscle behind it. This was going to be the new Counter-Strike. Global Offensive succeeded in a way that Source never quite did. It found acceptance among both casual players and the competitive community. Tournaments gradually migrated from 1.6 to CS:GO. The game introduced cosmetics and the infamous loot box system, which generated massive revenue for Valve. Professional teams adopted it. The esports scene rebuilt itself around CS:GO, and it became bigger than ever. Major tournament prize pools grew. Teams became multinational organizations with sponsors and salaries. Counter-Strike had become not just a game but a legitimate esports franchise that rivaled traditional sports in terms of competitive structure and audience engagement. ## The Modern Era: Counter-Strike 2 In 2023, Valve released Counter-Strike 2, built on a completely overhauled engine with modern technology and features. But here's where the story gets complicated, and where I get a little heated about corporate decision-making. Counter-Strike 2 arrived as a free-to-play replacement for Global Offensive, and the transition was rocky. Valve made the controversial decision to retire CS:GO entirely, forcing the competitive scene and millions of players to migrate whether they were ready or not. The new game had different mechanics, different feel, different everything. The professional community had to adapt quickly. There was genuine anger from players who loved CS:GO exactly as it was and didn't want to learn a new game. The move felt like Valve was more concerned with pushing players toward their newer product than with respecting the preferences of their most dedicated fanbase. That's the reality of modern gaming that gets lost in nostalgic conversations about Counter-Strike's golden age. Back when Le and Cliffe were modding Half-Life, there was no corporate pressure. There were no quarterly earnings calls. There was no need to extract maximum value from every player. They were just making something they loved. But as Counter-Strike grew and became valuable property, it got caught up in the machinery of the game industry. Decisions are made based on revenue potential rather than pure gameplay quality. Players get migrated to new versions whether they like it or not. That said, Counter-Strike 2 isn't a bad game. It's a competent update that maintains the core identity of the series. But it represents something different from what Counter-Strike was at its best. It's a calculated business decision dressed up as an evolution. The fact that the competitive community has largely embraced it suggests that despite my grumbles, the game still has what makes Counter-Strike special. But there's something bittersweet about seeing a mod that started with pure creative passion become another franchise product managed for maximum quarterly revenue. ## Legacy: The Mod That Changed Gaming Let's step back and look at what Counter-Strike accomplished. Starting as a mod in 1999, it grew to become one of the most successful game franchises in history. Over 50 million copies sold across all versions and platforms. At its peak, Counter-Strike 1.6 had over 319,000 concurrent players, an absolutely staggering number for a game from that era. Professional esports tournaments with prize pools that grew larger every year. A competitive scene that attracted global talent and sponsorship deals. But beyond the numbers, Counter-Strike proved something crucial about gaming. It proved that gameplay could trump production values. A mod made by two people in their spare time could out-compete games made by hundred-person studios with massive budgets. It proved that communities matter. The mod would never have succeeded without constant feedback and engagement from players. It proved that competition drives gaming. Esports wasn't invented by Counter-Strike, but Counter-Strike helped establish it as a legitimate pursuit for serious gamers. Counter-Strike also changed how the industry thought about modding. Before Counter-Strike, mods were often seen as cute add-ons that wouldn't impact the bottom line. After Counter-Strike, studios started viewing modding communities as potential goldmines. Valve especially learned to cultivate its modding community, partly because of Counter-Strike's success. Games like Team Fortress 2 and Dota 2 owe their existence to this approach. The game also created a blueprint for how tactical shooters should play. Every tactical shooter that came after Counter-Strike, from Rainbow Six Siege to Valorant, owes something to the design decisions that Le and Cliffe made. The economy system, the objective-based gameplay, the emphasis on positioning and communication: Counter-Strike popularized them and showed that players loved them. Here's what keeps me up at night when I think about Counter-Strike. The mod was special because it came from passion. Le and Cliffe weren't trying to create the next big esports phenomenon. They were just trying to make a game they wanted to play. Could that happen again today? Could two people in their bedroom create a mod that would eventually sell 50 million copies? The answer is probably not. The barriers have gotten higher. The market is more saturated. Counter-Strike belongs to a specific moment in gaming history. It was born when modding was accessible and encouraged. It succeeded because of genuine community engagement rather than marketing budgets. Looking back at Counter-Strike isn't just about missing a great game. It's about missing a moment when games could emerge from passion and luck rather than corporate strategy. ## Frequently Asked Questions When did Counter-Strike first come out as a mod? Counter-Strike was officially named on March 15, 1999, and development began in January 1999. The first public beta was released on June 19, 1999, after Valve released the Half-Life SDK on April 7, 1999. Why is Counter-Strike 1.6 still considered the best version by many pros? Counter-Strike 1.6 became the competitive standard and remained unmatched for years because of its weapon balance, movement mechanics, and map design. The spray patterns, the feel of the weapons, and the overall gameplay became deeply familiar to generations of players. When Counter-Strike: Source came out, the competitive community largely rejected it because it felt different, even though it looked better. What was the CPL and why was it important to Counter-Strike? The Cyberathlete Professional League (CPL) was the first major esports league to build around Counter-Strike's potential. They organized professional tournaments with substantial prize pools, with the 2001 World Championship offering $150,000. The CPL legitimized competitive gaming and proved that esports could be a viable business and career path. How many copies of Counter-Strike have been sold across all versions? The Counter-Strike franchise has sold over 50 million copies across all versions including the original retail releases, Counter-Strike: Source, Counter-Strike: Global Offensive, and Counter-Strike 2. By 2008 alone, the original had sold over 4.2 million retail copies. What happened to Minh Le and Jess Cliffe after Counter-Strike? Minh "Gooseman" Le and Jess Cliffe were brought into Valve when the company partnered with them around Beta 5.0 in early 2000. Counter-Strike 1.0 was released in November 2000 as an official Valve product. Both creators benefited from the game's massive success, though their roles evolved over time as Valve took greater control of the franchise. --- # What Happened to Tumblr, the Platform Yahoo Bought for a Billion and Sold for Pennies URL: https://404memoryfound.com/posts/what-happened-to-tumblr-yahoo-billion-dollar-mistake.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-13 Topics: Internet Culture, Business Blunders What Happened to Tumblr, the Platform Yahoo Bought for a Billion and Sold for Pennies In 2013, Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) paid $1.1 billion for Tumblr. Six years later, Verizon sold it to Automattic for approximately $3 million. That's a 99.7% loss in value, which is not a typo. It's the kind of financial collapse that would be comedy if it didn't represent the death of an entire cultural ecosystem. But here's what makes this story interesting: Tumblr didn't fail because it was fundamentally broken or because no one used it. Tumblr failed because a high school dropout built something genuinely novel, corporate giants systematically misunderstood it, and then one terrible decision destroyed it in under a month. This is the story of how a platform shaped internet culture in ways we're still discovering, and how it got killed anyway. ## The Dropout Who Built It David Karp was 20 years old when he launched Tumblr on February 19, 2007. He had never finished high school. He had been working in tech since he was 14, building websites and learning the web through pure repetition and curiosity. Within two weeks, Tumblr had 75,000 users. Within a year, it had hundreds of thousands. By 2012, just five years after launch, Tumblr hosted 100 million blogs. The unusual part wasn't just the speed. It was that Karp understood something most people didn't: the web was becoming visual, and it was becoming personal. Blogging had been text-first, homepage-first, author-as-authority. Karp saw that people wanted something simpler. They wanted to share images without writing essays. They wanted to remix culture, reblog others' content, build weird niche communities. He built for that instinct before the instinct became obvious. Karp also understood his users in a way that mattered. He didn't treat Tumblr as a product to monetize aggressively. He treated it as a place where people lived. His approach was minimalist by design: give people tools, trust them to build something, and stay out of the way. ## What Made Tumblr Actually Special To understand Tumblr's collapse, you have to understand what made it different. It wasn't just another blog platform. It wasn't Twitter. It wasn't Facebook. It occupied this weird, perfect space that nothing else has fully replicated. Tumblr was built on the reblog. This seems obvious now, but it was revolutionary in 2007. Reblogging meant you could instantly share something with your followers. You could add your own commentary on top. You could participate in conversations without being the original author. This created a culture of remixing and building on each other's ideas that felt fundamentally generous. This is essentially what Twitter's quote-tweet did later, but Tumblr had it first and it worked differently, because it preserved the chain of attribution. Tumblr became the social media platform for subcultures that didn't fit elsewhere. Fandoms found a home. The LGBTQ community, especially trans people and people exploring sexuality, built entire worlds on Tumblr because other platforms were hostile or policed them. The platform said "your weirdness is welcome here" before it became fashionable to say that. This wasn't an accident. Karp built Tumblr with a specific design philosophy: minimal moderation, maximum freedom, and trust in the community. The site's dashboard wasn't an algorithmic feed like Twitter or Facebook. You saw your followed blogs' posts in chronological order. You built your own timeline by following people whose work you cared about. The algorithm didn't decide what mattered. You did. Tumblr also had a particular aesthetic that influenced how the internet looked. The platform made GIFs central. It made visual curation central. You could build a blog that was visually cohesive in a way that other platforms didn't encourage. This led to entire aesthetic movements: dark academia, cottagecore, vaporwave. Tumblr users were curating moods and vibes before anyone had a name for it. This is essentially what Pinterest and Instagram mood boards do now, but Tumblr invented the grammar. And then there were the fandoms. Tumblr became the second home of fanfiction writers, fan artists, and fans in general. Archive of Our Own had the fiction. Tumblr had the community. You could find thousands of people who cared about the same obscure ship, the same anime, the same actor. Some of the most creative people on the internet built their skills and their communities on Tumblr. By early 2014, Tumblr users were posting over 100 million posts per quarter. The platform was active, engaged, and generating something that felt genuinely new in internet culture. ## The Billion Dollar Acquisition On May 20, 2013, Yahoo announced it would acquire Tumblr for $1.1 billion in cash. The deal closed on June 20, 2013. CEO Marissa Mayer personally pushed the acquisition. In a public statement, she promised the company would not "screw it up." Yahoo's logic was sound, in a corporate sort of way. Yahoo was dying. The company had failed at mobile. It had failed to compete with Google in search. It had a large user base but was bleeding cultural relevance. Tumblr was young, hip, mobile-friendly, and culturally dominant in internet subcultures. Buying Tumblr could give Yahoo access to a younger demographic. It could give Yahoo ad inventory. It could make Yahoo relevant again. What Yahoo didn't understand was that Tumblr worked precisely because it didn't operate like a traditional corporation. Karp had kept the company independent and small. There was minimal advertising. The culture was deliberately weird and somewhat anti-corporate. You couldn't bolt Tumblr onto Yahoo without changing what made it valuable in the first place. Still, the early period under Yahoo wasn't catastrophic. The platform continued to grow. User engagement remained high. Karp stayed on as CEO initially. Things didn't immediately fall apart. But Yahoo had a problem that went deeper than Tumblr. Mayer's tenure was defined by acquisitions that never integrated properly. Tumblr was the biggest of these bets, and it needed to start paying off. That meant monetization. That meant ads. That meant turning Tumblr into a revenue machine. ## How Yahoo Systematically Misunderstood Tumblr Yahoo's first mistake was conceptual: they thought they had purchased a blogging platform. What they actually had was a culture platform. These are not the same thing. A blogging platform is a tool for publishing. A culture platform is a place where communities form and express themselves. One is a product. The other is infrastructure for identity. Yahoo tried to monetize Tumblr the way you monetize a content platform. They added ads. They tried to integrate Tumblr into Yahoo's broader ecosystem. They wanted to make Tumblr generate revenue through advertising and data collection, the same playbook they used for everything else. The problem was that Tumblr's users had specifically chosen Tumblr because it wasn't aggressively monetized. The platform had some ads, but they were relatively unobtrusive. The culture was about sharing and community, not about being sold to. When Yahoo started inserting ads more aggressively, users noticed. When Yahoo started integrating Tumblr into Yahoo accounts, users noticed. The vibe shifted. Yahoo's second mistake was neglecting infrastructure. They kept Tumblr's interface and culture roughly the same, which was good. But they didn't invest in moderation tools proportional to the platform's size and diversity. They didn't hire enough people to handle abuse reports. They didn't build systems to help communities self-moderate. This became critically important later. Yahoo's third mistake was passivity. They assumed Tumblr would just keep being Tumblr, without requiring anything beyond letting it exist. They didn't realize that communities need active stewardship. Once users started feeling like Tumblr was extracting money rather than providing value, the social contract started eroding. In 2016, Yahoo wrote down $712 million of Tumblr's value. That reduced the implied valuation to roughly $230 million. The writedown happened before the NSFW ban, before the mass exodus, before the fire sale. Even by Yahoo's own accounting, the acquisition was already a disaster. David Karp left Tumblr in November 2017. By that point, he was increasingly uncomfortable with the direction of the platform under its new corporate owners. Karp, who had built Tumblr on the principle of trusting users and communities, watched his creation become just another corporate asset to optimize. ## The NSFW Ban That Broke Everything On December 3, 2018, Tumblr announced a new content policy: the platform would ban all "adult content." The ban took effect on December 17, 2018. This was not a minor policy tweak. This was an existential decision. A substantial portion of Tumblr's user base was there to share adult content, explore sexuality, build sex-positive communities, or create erotic fan works. The platform had always allowed this. It was one of the few mainstream social media platforms that treated sexuality as a normal part of human expression rather than something to suppress. This attracted sex workers, LGBTQ people, sexual minorities, and anyone interested in exploring identity in a community that felt safe. The ban didn't just target pornography. Tumblr's moderation algorithm, which was clearly built in a rush, flagged anything vaguely sexual. It flagged erotic fanfiction. It flagged discussions of sexuality. It flagged artistic nude photography. It flagged queer relationship content. It flagged people discussing their own bodies. The filter was indiscriminate and clumsy. The consequences were immediate. Similarweb reported that Tumblr lost 30% of its user traffic following the ban. In December 2018, Tumblr had approximately 521 million monthly visits. By April 2019, that number had dropped to roughly 376 million. That's 145 million monthly visits gone in four months. Users, especially LGBTQ users and sex workers, felt specifically targeted. They had been told Tumblr was a safe place. Now that promise was being withdrawn, replaced with an algorithm that treated their expression as something shameful. Long-time users started migrating to Twitter, Reddit, Discord, and smaller platforms. There was no single successor because Tumblr had been unique. The communities dispersed rather than reconsolidated. The NSFW ban was the catalyst, not the root cause. The root cause was years of corporate mismanagement. But the ban crystallized everything. It was the moment everyone realized there was a fundamental problem, and they couldn't trust the people running the platform to make good decisions. ## The Three-Million-Dollar Fire Sale By 2017, Verizon had inherited Tumblr when it acquired Yahoo. A telecommunications corporation managing a culture platform is, to put it gently, a mismatch. Verizon had even less understanding of what Tumblr was than Yahoo did. On August 12, 2019, Verizon sold Tumblr to Automattic, the company behind WordPress.com. The price was approximately $3 million. Yahoo bought Tumblr for $1.1 billion. Verizon sold it for $3 million. A 99.7% decrease in value over six years. From a pure accounting perspective, Verizon's decision to cut losses was rational. Tumblr generated far less advertising revenue than comparable platforms. It had a culture that resisted aggressive monetization. It required significant infrastructure and moderation costs. The user base was shrinking. Keeping it made no financial sense. Automattic's acquisition was more interesting. WordPress is open-source and community-focused. Automattic has demonstrated willingness to respect user freedom in ways that corporate social media platforms typically don't. There was at least a possibility that Automattic would understand what Tumblr actually needed: less extraction, more stewardship. ## Where Tumblr Is Now Since 2019, Tumblr has stabilized without recovering. As of mid-2025, the platform receives approximately 141.9 million monthly visits. That's a 73% decline from its peak, but it's not death. Millions of people still use Tumblr every month. Automattic has made cautious improvements. They loosened the NSFW restrictions, though not completely. They invested in community features. They treated Tumblr more like a living ecosystem and less like a revenue extraction machine. But the damage was structural. Many of the communities that made Tumblr culturally significant migrated and never came back. The fandoms moved to Twitter and Discord. The sex-positive communities moved to Twitter and OnlyFans. The aesthetic bloggers moved to Instagram and Pinterest. The critical mass that made Tumblr feel alive, that created the sense that something was always happening, is gone. Tumblr is not dead. But it is no longer the place where internet culture gets made. It's a community of people who remember when it was. ## The Legacy: What Tumblr Actually Meant If you want to understand internet culture between 2010 and 2018, you have to understand Tumblr. Not as a product or a platform, but as a cultural moment. Tumblr gave subcultures infrastructure at exactly the moment they needed it. When the web was becoming visual and mobile, Tumblr provided tools for communities to be visual and mobile. The fandoms found each other. LGBTQ people found each other. The weird people found each other. And they built something new. Tumblr popularized the reblog as a form of collective creation. You could remix and build on other people's work while preserving a chain of attribution. This influenced how people think about content, credit, and community across the entire internet. Tumblr made aesthetics a form of identity. The visual curation of Tumblr blogs created entire aesthetic movements: cottagecore, dark academia, vaporwave, soft grunge. Instagram later adopted this language and scaled it, but Tumblr originated the grammar. Tumblr was central to fanfiction and fan art culture. The fanfiction ecosystem predated Tumblr, but Tumblr made it visual and social. It created a context where fan creators could build communities, share work, and receive feedback at a scale that hadn't existed before. And Tumblr was enormously important for LGBTQ people, especially trans people and sexual minorities. The platform allowed exploration of identity and sexuality in a way that most mainstream social media didn't. For many users, Tumblr was where they figured out who they were. That matters, and losing it mattered. Perhaps most importantly, Tumblr demonstrated that you could build something culturally significant that wasn't designed primarily to extract value from users. Karp's original vision was that Tumblr should be a place first and a product second. That meant respecting users, trusting communities, and not trying to monetize every interaction. The collapse of Tumblr was a lesson in what happens when you break that contract. Yahoo and Verizon tried to turn Tumblr into just another ad platform. It didn't work because the value came from communities and trust, and you can't extract value from trust without destroying the trust itself. As we watch X struggle under new ownership, as Instagram becomes increasingly algorithmic, as Reddit prioritizes API revenue over user experience, the Tumblr story remains instructive. Platforms are valuable because communities use them. Communities use them because they trust that the platform respects them. Once that trust breaks, migration happens fast and recovery is slow, if it happens at all. ## Frequently Asked Questions - Is Tumblr still active in 2025? Yes. Tumblr receives approximately 141.9 million monthly visits as of mid-2025. That's a significant decline from its peak of over 500 million monthly visits, but it remains a functioning platform with active communities. Automattic, which acquired Tumblr from Verizon in 2019, continues to operate and develop it. - Why did Yahoo buy Tumblr for $1.1 billion? Yahoo CEO Marissa Mayer saw Tumblr as a way to reach younger demographics and become relevant in social media. Yahoo was losing ground to Google and Facebook. Tumblr had 100 million blogs, strong engagement, and cultural influence among younger internet users. The logic was that Tumblr's audience and cultural position could revitalize Yahoo's brand. The problem was that Yahoo never figured out how to monetize Tumblr without destroying what made it valuable. - Did the NSFW ban really destroy Tumblr? The NSFW ban, which took effect December 17, 2018, was the most visible cause of Tumblr's decline. The platform lost approximately 30% of its traffic in four months. But the ban was a symptom of deeper problems: years of corporate mismanagement, failure to invest in moderation, and a fundamental misunderstanding of what made Tumblr special. The ban was the breaking point, not the only problem. - What happened to David Karp after leaving Tumblr? David Karp left Tumblr in November 2017, before the NSFW ban but after years of growing frustration with Yahoo and Verizon's management. He has maintained a relatively low public profile since departing. Karp was 20 when he launched Tumblr and was 31 when he left. He made significant money from the Yahoo acquisition, but by most accounts his departure was driven by disagreement over the platform's direction rather than financial motivation. - Could Tumblr make a comeback? A full recovery to its former cultural position is unlikely. The communities that made Tumblr special have dispersed across multiple platforms, and rebuilding that critical mass would require something extraordinary. However, Tumblr still has millions of monthly users and a dedicated community. Under Automattic's more respectful management, the platform has stabilized. It may not become what it was, but it could remain a meaningful, smaller community for people who value what Tumblr offers. --- # What Happened to Arcades: The Rise, Fall, and Unlikely Return of Quarter-Powered Gaming URL: https://404memoryfound.com/posts/what-happened-to-arcades-golden-age-gaming.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-13 Topics: Gaming, Then vs Now ## What Happened to Arcades: The Rise, Fall, and Unlikely Return of Quarter-Powered Gaming Picture this: It's 1981, and you've got three dollars in your pocket, a best friend who smells like soda and arcade dust, and zero responsibilities except maybe getting home before the streetlights come on. You push through glass doors that stick slightly because they're perpetually grimy, and you're immediately assaulted by the most beautiful noise you've ever heard. Bleeps. Bloops. The mechanical whirring of change machines. A high score table somewhere declaring that TODD is still the king of Galaga, which has absolutely destroyed your self-esteem. The air is thick with cigarette smoke and possibility. You can practically taste the carpet. This was the arcade, and for about five glorious minutes of human history, it was the most important place on Earth. I grew up thinking arcades would last forever. They felt like a permanent fixture of civilization, like libraries or parking lots. You passed them on the way to anywhere worth going. Malls had them. Bowling alleys had them. Beach boardwalks had them. That one weird little shop downtown had them. And then, somehow, they just sort of disappeared. One day you turned around and realized that the last arcade you visited wasn't actually the last arcade you'd ever visit, but you didn't know it was the last one at the time. That's how it worked. That's how they all closed. But here's the thing: the story of what happened to arcades isn't just a story about video games or technology or entertainment trends. It's a story about money and nostalgia and the specific weirdness of American culture in the late twentieth century. It's about the fact that at one point in our history, more people were spending money in arcades than in movie theaters, and barely anyone remembers that now. So let's rewind. ## The Golden Age Exploded Out of Nowhere When I say arcades came out of nowhere, I mean that literally nobody expected video games to become the most culturally dominant entertainment force of the early 1980s. Before 1978, arcade games basically didn't exist in any meaningful way. There were some scattered machines, sure, but nothing that made anyone think, "Yes, this is the future." And then Taito released Space Invaders in 1978, and absolutely everything changed. Here's what happened: Space Invaders was simple. You controlled a little cannon at the bottom of the screen. Aliens moved down from the top. You shot the aliens. The aliens moved faster. You shot faster. You died. But the genius part wasn't the concept. The genius part was that it was addictive in a way that nothing had ever been before. People couldn't stop playing it. They lined up for it. They brought their friends to watch them play it. They spent money they probably didn't have playing it. And arcade owners, who suddenly realized they had literal money printing machines on their floors, started buying more machines and putting them everywhere. The golden age didn't really start until after 1978, but once it started, it moved fast. Pac-Man hit in 1980 from Namco (published by Midway in North America), and if Space Invaders was the reason people went to arcades, Pac-Man was the reason they stayed and brought their entire family. Pac-Man made over a billion dollars by 1981. Let me say that again: one billion dollars. In 1981. From a game where you moved a yellow circle around eating dots. By 1982, there were seven billion coins collected in Pac-Man machines worldwide. Seven billion. The machine had become so ubiquitous that there were approximately 400,000 Pac-Man cabinets installed globally. And here's where it gets interesting. There's this whole story that gets told about how Space Invaders caused a shortage of 100-yen coins in Japan because people were using so much currency to play the game that there literally weren't enough coins in circulation. It's a great story. It's a fun story. It's also completely false. That's an urban legend that people keep repeating because it sounds plausible and amazing. The Japanese Mint's PR office confirmed that increased yen production in the late 1970s had nothing to do with Space Invaders, and the game's own designer, Toshiro Nishikado, described it as a wild rumor. The actual story is somehow less dramatic but still impressive: arcades were just extraordinarily profitable, and nobody had anticipated that level of profitability. By 1982, the arcade industry in the United States alone was pulling in eight billion dollars annually. Eight billion. To put that in perspective, the entire American film industry was making three billion dollars a year. The music industry (https://404memoryfound.com/posts/napster-destroyed-music.html) was making four billion dollars a year. Arcades were outearning Hollywood. Arcades were outearning pop music. A bunch of machines that did nothing but display blocky graphics and beep noises were the most profitable entertainment sector in the entire country. There were 10,000 arcades scattered across North America. The number of arcades had doubled between 1980 and 1982. Kids were cutting school to spend quarters. Parents were confused about why their children suddenly cared about a pizza-eating cartoon character. ## And Then It All Crashed The brutal part about this story is that the decline started almost immediately after the peak. We're not talking about a slow fade over decades. We're talking about a sharp drop that started in 1982 and accelerated from there. In 1981, arcades were pulling in eight billion dollars. By 1983, that number had dropped to five billion. By 1984, it was down to four billion. By the end of the 1980s, arcades were a fraction of what they'd been at their peak. Something had gone catastrophically wrong. The easy answer is that home consoles killed the arcades. The Atari (https://404memoryfound.com/posts/who-owns-atari-now.html) 2600 came out in 1977, before the arcade boom even really started, but it was terrible at first. The games were worse. The graphics were worse. The sound was worse. If you wanted to play a real video game, you went to an arcade. But then technology improved, and licensing happened, and suddenly you could play versions of Space Invaders and Pac-Man on your television at home. Not the exact same versions, sure. The graphics weren't as good. The gameplay was slightly different. But they were close enough, and more importantly, they were free after the initial hardware investment. The Nintendo Entertainment System, which launched in North America in 1985, delivered the real blow. The NES proved that home gaming wasn't just a cheap imitation of the arcade experience. It was its own thing, with its own identity and its own library of games that couldn't be found in any arcade. Super Mario Bros. didn't need an arcade cabinet. The Legend of Zelda didn't need quarters. Home consoles stopped being arcade substitutes and started being something entirely different and better in their own way. But it goes deeper than that. The arcade industry had gotten bloated and weird. The market was oversaturated with machines. Every shopping mall had three different arcades competing with each other. Arcade operators had been making so much money so quickly that some of them had gotten sloppy. Games broke down. Cabinets weren't maintained properly. Quarters got jammed. And more importantly, the novelty started wearing off. The miracle of watching a machine play a game on a screen became normal. The games themselves got more complex and expensive. The quarter-eating increased. If you played Pac-Man a lot, you knew exactly how much money you were dumping into it. Quarters add up fast. And here's the part that I get genuinely heated about: the arcade industry also became kind of exploitative. Operators knew that kids were addicted to these games, and they started making them harder and less fair specifically so people would have to spend more money to get the same amount of playtime. Games were designed not to be beaten, but to be constantly fed with quarters. The business model worked great when there was no alternative, but the moment home consoles became viable, suddenly people realized they could play at their own pace, fail without losing money, and pause whenever they wanted. The arcade model of extracting maximum quarters from minimum entertainment suddenly looked pretty hostile. ## The Fighting Game Renaissance Briefly Changed Everything And here's where it gets interesting. Right when you thought arcades were completely finished, Capcom released Street Fighter II in 1991, and suddenly people cared again. Not the same way they cared in 1981, but they cared. Street Fighter II brought arcade fighting to a level of sophistication and competition that had never existed before. This wasn't a game you played alone to get a high score. This was a game you played against other human beings, in public, where everyone watching could judge your skill level. The machine became a gathering place again, but for a different reason. These weren't kids spending their allowance. These were competitive players who would travel to different arcades to find better opponents. There were actual tournaments. There were sponsorships. There were players who got legitimately famous. Mortal Kombat added to the phenomenon in 1992, bringing violence and shock value to fighting games in a way that Street Fighter hadn't. The combination of Street Fighter and Mortal Kombat created a brief window where arcades were culturally relevant again. But I want to be honest: this renaissance was regional and demographic-specific. It wasn't a return to the golden age. It wasn't children and adults and families all gathering together. It was mostly young men in specific cities with strong arcade communities. In Japan, arcades never really died the same way they did in North America, so the fighting game scene there was able to grow organically. Japanese game centers, sometimes called game sentos, evolved into social destinations that served a broader cultural purpose. In America, the fighting game revival extended the life of arcades, but it couldn't save them. By 1999, arcade revenue had dropped to 1.3 billion dollars. The fighting game scene kept some arcades alive and created a small but devoted community of competitive players, but the broader arcade culture was gone. ## The Mall Arcade Died and Nobody Noticed I want to talk about mall arcades specifically because I think they deserve their own funeral. The mall arcade was where most people experienced arcades in their childhood. You'd go to the mall, get some money from your parents, find the arcade tucked between the food court and the movie theater, and spend two hours of pure contentment while your parents shopped for clothes they'd never wear. Mall arcades were the gateway drug to arcade culture. The mall arcade decline happened in stages. First, the big dedicated arcade chains started closing. Then the mall operators realized that dedicated arcade spaces were expensive and didn't drive traffic the way they used to. Then they started replacing arcades with other stores. Then they started making arcades smaller. Then they started putting them in food courts where nobody really wanted to hang out. Then they were just gone, and when I walk through a mall now, I get confused because something essential is missing from the mall experience, and I can't quite figure out what it is until I remember: there are no arcades anymore. The standalone neighborhood arcades died for similar reasons. The economics just didn't work. Real estate costs went up. Game replacement costs went up. Labor costs went up. But the amount people were willing to spend on quarters stayed the same or went down. An arcade operator in 2005 was basically running a museum of games from the 1980s and 1990s, which nobody cared about because they could play better games on better hardware at home. The weirdest part about the arcade collapse is that it happened in real time but nobody really acknowledged it. There was no moment where someone said, "We're closing all the arcades." It was just a slow dissolution. You'd go back to a mall from your childhood and the arcade would be a cell phone store. You'd check on an arcade you used to visit and there would be a sign saying "Available for Lease." Within about fifteen years of the early 1990s fighting game peak, arcades had basically vanished from the cultural landscape. ## The Unlikely Revival and Where We Are Now And then something unexpected happened. People got nostalgic. And I don't mean the normal kind of nostalgia where you think about good times you had. I mean the specific kind of nostalgia that happens when you realize something you loved is completely gone, and you start investing emotional energy in bringing it back. Dave & Buster's, which started as a sports bar and arcade hybrid in 1982, became one of the primary vehicles for arcade revival. It wasn't a pure arcade experience. It was arcades mixed with alcohol and appetizers and a sports bar atmosphere. But it meant that adults who loved arcades could go to an arcade without feeling weird about it. You were an adult hanging out at a bar that happened to have games. That was fine. That was normal. And Dave & Buster's locations proliferated. Then the barcade phenomenon started. These were spaces that were explicitly designed as bars first and arcades second, but they featured real vintage arcade machines alongside craft cocktails. The appeal was partly nostalgic and partly about the specific aesthetic of vintage arcade cabinets. A cabinet from 1982 looks cooler in some ways than a modern machine. It has weight. It has history. It smells like old electronics and possibility. More recently, Round1 Entertainment and similar operations have brought some of the Japanese arcade philosophy to North America. Round1 locations feature a mix of arcade games, crane machines, and a general arcade-focused atmosphere. They're designed for families and groups, not just hardcore gamers. They've successfully created arcade spaces that feel relevant to contemporary culture rather than trapped in the past. The modern arcade revival isn't a return to the golden age, and I don't think anyone expects it to be. The cultural moment that made arcades the dominant entertainment format of 1982 is gone forever. Home gaming is too good. Mobile gaming exists. Streaming exists. The reasons people went to arcades in 1982 aren't compelling anymore for most people. But arcades have survived, adapted, and found a new role as experience spaces and nostalgia destinations rather than the cutting edge of gaming technology. And honestly, that's kind of beautiful. There's something specific about playing Galaga on an original cabinet that you can't replicate on a modern system. The weight of the joystick. The specific feel of the buttons. The way the screen glows. The fact that you can see yourself reflected in the screen between levels. These are things that matter to people who experienced arcades the first time around, and they're starting to matter again to people who didn't. ## What I Miss Most I've been thinking about what I actually miss about the arcade experience, and I don't think it's the games themselves. The games were great, sure, but they're still available in various forms. What I miss is the specificity of the experience. I miss the particular smell of a 1980s arcade. I miss the specific way conversations happened in those spaces. I miss the way you could watch someone play a game and understand their skill level immediately. I miss the ritualistic feeling of putting a quarter in a machine and resetting to the start screen. I miss the way high scores meant something in a community where maybe fifty people were playing the same machine. I miss the fact that gaming was a social activity that happened in public spaces. Now gaming is something that happens alone or online, and while that has its own value, it's a different thing entirely. The arcade was about showing off and being shown up and learning from watching other people and the constant feedback of a community. That's gone in a way that can't really be replicated, even if arcades themselves are coming back. But I'm also weirdly optimistic about where we are now. The fact that arcades are returning, even in a different form, suggests that people genuinely value that experience. It's not just nostalgic old people trying to relive the past. It's families and young people discovering what made arcades special. It's a recognition that there's something about communal gaming and shared physical spaces that matters, even in an age of hyper-connected digital entertainment. ## FAQ: Understanding Arcade History and Culture ### Why did arcades become so popular in the early 1980s? Arcades exploded in popularity because Space Invaders and Pac-Man offered something completely new: addictive, simple games that required no explanation and could hook a player immediately. The arcade business model of charging quarters per play was also perfect for this moment in consumer culture. People had never experienced anything like these games before, and the community aspect of playing in public spaces made them even more appealing. The combination of novelty, accessibility, and profitability created the perfect storm for expansion. ### Is the story about Space Invaders causing a yen shortage in Japan actually true? No, this is an urban legend. While Space Invaders was extraordinarily popular and consumed enormous numbers of coins, it did not create a national shortage of 100-yen coins in Japan. The Japanese Mint confirmed that increased yen production in the late 1970s was unrelated to the game, and Space Invaders designer Toshiro Nishikado himself described it as a wild rumor. The myth persists because it captures something true about how dominant the game became, but the economic data doesn't support it. ### What's the difference between the original arcade golden age and the modern arcade revival? The original golden age was driven by cutting-edge technology and widespread accessibility. Arcades were the best way to play games, period. The modern revival is driven by nostalgia, craft culture (think barcades), and the specific experience of communal gaming. Modern arcades are intentional destinations rather than ubiquitous gathering spaces, and they typically serve adults and families rather than being the universal experience they once were. ### Did home consoles really kill arcades? Home consoles were a major factor, but they weren't the only reason. Market oversaturation, unsustainable business models that prioritized quarter-extraction over player experience, rising operating costs, and changing consumer preferences all played roles. The NES proved that home gaming could be its own thing, not just a lesser version of arcades. Home consoles provided a viable alternative that made the arcade's pay-per-play model less compelling, but the industry had structural problems before consoles delivered the final blow. ### Are arcades coming back, or is this just a temporary revival? Arcades are unlikely to return to their 1982 dominance as a primary gaming platform, but they've found a stable niche as experiential entertainment spaces. Places like Dave & Buster's, barcades, and Round1 locations suggest that the arcade concept has lasting appeal when adapted to contemporary contexts. The question isn't whether arcades will become mainstream again, but whether they'll continue to serve as valued community spaces and nostalgia destinations. ### Why did Japan keep its arcade culture while North America lost it? Japanese arcades, known as game centers, were understood primarily as social spaces and community gathering places, not just profit-maximizing operations. When the games themselves became less cutting-edge compared to home consoles, the spaces remained culturally valuable. North American arcades were primarily defined by their newest games and highest profit margins, so when those advantages disappeared, the spaces themselves lost their reason to exist. --- # What Happened to LiveJournal, the Platform That Invented Modern Blogging URL: https://404memoryfound.com/posts/what-happened-to-livejournal-blogging-platform.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-13 Topics: Internet Culture, Software & Apps In 2007, LiveJournal had 18 million monthly unique visitors. By 2026, it's essentially a Russian-language blogging platform that most English-speaking internet users have completely forgotten about. This is the story of how a platform that literally invented modern social media features became collateral damage in geopolitics and corporate apathy. ## The Accidental Revolution Brad Fitzpatrick created LiveJournal on April 15, 1999 because he wanted a better way to keep his high school friends updated. He was a college student and programmer, which meant he had the skills to build it and the arrogance to think he could do it better than everything else available. He was right. What made LiveJournal different wasn't the journal feature itself. Countless websites already let you post diary entries online. What made it revolutionary was everything wrapped around the writing. LiveJournal pioneered friends lists, custom access controls that let you set different visibility rules for different groups of people, communities where users could gather around shared interests, mood indicators next to posts, and customizable user icons. These weren't minor features. These were the foundational concepts that would define social media for the next two decades. Facebook would later repackage the friends list and access controls. Twitter would strip everything down to public broadcasting. But LiveJournal got there first, and it got it right. The platform also invented memcached, an open source caching system that became so useful it would later power Facebook, Wikipedia, and basically every large-scale website that needed to serve data quickly. This is worth noting because it shows Fitzpatrick understood infrastructure at a level most social media founders never bothered with. By 2005, LiveJournal had grown to millions of users. The blogging boom was real. Teenagers were writing angst-filled entries. Writers were experimenting with serialized fiction. Fanfiction communities were organizing themselves across the platform. Scientists, activists, and programmers were using it to document their work and thoughts. It was messy and human and genuinely interesting. ## The First Mistake In January 2005, Six Apart purchased Danga Interactive, the company that owned LiveJournal, from Fitzpatrick. This wasn't inherently bad. Companies get acquired. Founders cash out. That's how venture capital works. But Six Apart's approach to running LiveJournal would prove to be the beginning of a very long decline. Six Apart was founded by Ben Trott and Mena Trott, who had previously created Movable Type, a different blogging platform. They understood blogging infrastructure. They understood the market. What they apparently didn't understand was community management or the value of user trust. The warning sign came in 2007 with the Strikethrough controversy. Six Apart suspended over 500 LiveJournal accounts without warning. The stated reason was suspected content violations. Users had no idea why they'd been suspended and no way to appeal. The accounts included fanfiction archives, LGBTQ+ communities, and spaces discussing topics that Six Apart's reviewers had flagged as problematic. Look, content moderation is hard. Removing genuinely harmful material is necessary. But removing hundreds of accounts simultaneously without explanation and without allowing users to defend themselves is how you destroy a community's trust in a single action. Six Apart did exactly that. The users who could rebuild elsewhere started looking for alternatives. The users who couldn't left anyway. LiveJournal was damaged in a way that wouldn't immediately show up in user metrics, but the damage was real. ## The Russian Acquisition In December 2007, roughly nine months after the Strikethrough controversy, Six Apart sold LiveJournal to SUP Media, a Russian company. The terms were never publicly disclosed. Fitzpatrick had already left Six Apart in August, moving on to Google and then eventually to co-founding Tailscale in 2020. He was no longer involved with the platform he'd created. SUP Media's reasoning was simple and sensible from a business perspective. Russia accounted for 28 percent of LiveJournal's traffic. Russian-language users dominated the platform's activity. Why not sell it to a company that could serve that market more effectively? Except that Russian ownership of a major social media platform meant Russian law would eventually apply to the content hosted there. And Russian law in the 2010s and 2020s became increasingly restrictive about what people could say online. Here's the thing: this wouldn't have been catastrophic if Six Apart hadn't already damaged user trust and if LiveJournal still retained the loyalty of English-speaking users. But it didn't. The English-speaking user base had already started migrating to other platforms. Tumblr (https://404memoryfound.com/posts/what-happened-to-tumblr-yahoo-billion-dollar-mistake.html) was becoming the new gathering place for fanfiction communities. WordPress was absorbing serious bloggers. Twitter and Facebook were pulling in everyone else. ## The Move That Changed Everything In December 2016, without announcement, LiveJournal moved its servers from California to Russia. Users discovered this when their content suddenly started loading differently from outside Russia. The company later confirmed it had moved the infrastructure physically across the world, overnight, with no warning. This wasn't technically illegal. SUP Media owned the servers. But it was a statement about where LiveJournal's future lay. It wasn't with Western users. It wasn't with the English-speaking community that had built the platform's culture. It was with Russian users, under Russian legal jurisdiction. Four months later, in April 2017, SUP Media updated LiveJournal's terms of service. Users now had to comply with Russian law. Blogs that received more than 3,000 daily views would be classified as media outlets and face additional restrictions. The informal, personal space that LiveJournal had been since 1999 was transforming into something else entirely. Blogger Aleksei Kungurov was sentenced to two years in prison for opinions he'd written on LiveJournal about Russia's actions in Syria. This was not a hypothetical scenario. This was the direct consequence of a platform's legal jurisdiction determining what its users could say. ## The Alternatives and the Exodus Former LiveJournal employees, seeing what was happening to the platform, created Dreamwidth. Dreamwidth used LiveJournal's open source code, which Fitzpatrick had made freely available, to build a new platform with the same features but without the geopolitical complications. Dreamwidth still exists and is still actively maintained. It has never been purchased by a larger company. Its code remains open source. Its legal jurisdiction remains in the United States. Dreamwidth didn't explode in popularity. This is essentially what Mastodon did with Twitter fifteen years later: creating a decentralized alternative when the original platform became unacceptable to its most engaged users. Like Mastodon, Dreamwidth attracted a committed user base that valued the principles underlying the platform more than its size. The fanfiction communities that had made LiveJournal culturally relevant migrated en masse to Tumblr, where they'd eventually migrate again to Archive of Our Own (AO3), an explicitly non-profit platform run by volunteers. This pattern is worth understanding: when communities lose trust in a platform's ownership and governance, they don't just find a replacement. They build one themselves. What had once been 18 million monthly unique visitors became a fraction of that, almost entirely concentrated in Russia and Russian-speaking countries. LiveJournal still exists. You can still access it. But it's not the same thing that existed from 1999 to 2007. That platform is gone. ## What Made LiveJournal Matter To understand why this story matters beyond nostalgia, consider what LiveJournal proved was possible. Before Facebook launched in 2004, LiveJournal had friends lists with granular access control. Before Twitter, LiveJournal had communities organized around interests and identity. Before Instagram, LiveJournal had user profiles and custom imagery. Fitzpatrick built these features because he understood that social media wasn't about broadcasting to the widest possible audience. It was about maintaining different social contexts simultaneously. You talk differently with your family than you talk with your friends. You talk differently with your professional colleagues than you talk with strangers on the internet. LiveJournal's access controls acknowledged this reality. The platform also proved something else: that a technical founder who understood infrastructure could build a social network that actually scaled. Memcached wasn't an afterthought. It was central to the design from the beginning. Fitzpatrick understood that if you want millions of users, you need to be able to serve them efficiently. This is less common now than you'd expect. Modern social platforms are built on the assumption of venture capital funding and exponential growth. They're designed to be acquired or to go public. They're not designed to be preserved as communities. ## The Real Question The question isn't really what happened to LiveJournal. We know what happened. A venture-backed company bought it and mismanaged it. A Russian company bought it and subjected it to Russian law. Users migrated to other platforms. Communities rebuilt themselves on infrastructure they controlled. The real question is why the same pattern keeps repeating. Communities are built on platforms they don't own. Users trust companies with their data, their writing, their relationships. Then those companies prioritize profit over community, or they get acquired by entities with different values, and the cycle begins again. LiveJournal's story is simultaneously completely unique and completely generic. It's a platform-specific history and a parable about digital ownership. Fitzpatrick built something genuinely innovative. He proved it was possible to create social infrastructure that respected user privacy and community autonomy. Then venture capital bought it, corporate incompetence damaged it, geopolitics transformed it, and the users who cared about those principles scattered to rebuild elsewhere. The platform is still running. It's mostly in Russian. If you wanted to, you could post on LiveJournal right now. But the LiveJournal that mattered, the one that changed how social media worked, the one that had millions of active users writing and connecting and building communities, that one doesn't exist anymore. Which, for context, is roughly what happens to every social platform eventually. This is just the one that invented the features everyone else still uses today. ## Frequently Asked Questions ### Is LiveJournal completely dead? No. LiveJournal still exists and still operates. The platform continues to have active Russian-language users and communities. You can access it, create an account, and post content. What's gone is its presence in English-speaking internet culture and its role as a major social platform. It persists as a functioning service, but not as the thing it once was. ### Can I still access my old LiveJournal account? If you created an account before the platform went primarily Russian, you can generally still access it. Your content should still be there. Whether you'd want to use the platform given that it's now subject to Russian law is a different question. Many users have exported their content for archival purposes or migrated to platforms like Dreamwidth. ### What is Brad Fitzpatrick doing now? Fitzpatrick left Six Apart in 2007, worked at Google as a Staff Software Engineer on the Go programming language team, and co-founded Tailscale in 2020. Tailscale is a VPN tool built on top of WireGuard that simplifies creating secure networks across devices. It's a very different kind of company than LiveJournal, but it's one that Fitzpatrick leads and controls. ### Why did Six Apart's sale to SUP Media matter so much? SUP Media was a Russian company, which meant LiveJournal would eventually become subject to Russian law. As Russia's regulations around online speech became more restrictive through the 2010s, the platform became increasingly inhospitable to users seeking free expression. The server relocation to Russia in 2016 and the terms of service changes in 2017 were direct consequences of Russian ownership. The sale wasn't necessarily a mistake from a business perspective, since Russian users were already the majority, but it had profound implications for everyone else on the platform. ### What is Dreamwidth? Dreamwidth is an alternative blogging platform created by former LiveJournal employees who wanted to preserve the features and community values that made LiveJournal special without the corporate and geopolitical complications. It runs on LiveJournal's open source code, operates as a non-profit, and maintains many of the access control and community features that LiveJournal pioneered. It's much smaller than LiveJournal at its peak, but it remains actively maintained. ### Where did the fanfiction communities go? Fanfiction communities initially migrated from LiveJournal to Tumblr starting around 2007. As Tumblr changed ownership and policies over the years, many communities migrated again to Archive of Our Own (AO3), a non-profit, volunteer-run platform explicitly designed for transformative fiction. AO3 was itself partly inspired by the failures of corporate platforms to serve creative communities. The pattern of migration from corporate platforms to community-controlled ones is one of LiveJournal's most lasting legacies. ### Did LiveJournal invent blogging? Not exactly. Online diaries and web logs existed before LiveJournal's 1999 launch. What LiveJournal invented, or at least pioneered in a mainstream way, were the social features that transformed blogging from a solitary activity into a networked one: friends lists, communities, access controls, and mood indicators. These features would become the foundation of what we now call social media. LiveJournal didn't invent blogging, but it invented the social layer that made blogging matter. --- # Is Quake Still Playable in 2026? Who Owns It Today URL: https://404memoryfound.com/posts/what-happened-to-quake-online-fps.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-10 Updated: 2026-09-18 Topics: Gaming, Software & Apps **Summary:** Quake is still playable in 2026, thirty years after id Software released the shareware episode on June 22, 1996. Microsoft owns it today, through ZeniMax Media, which it acquired on March 9, 2021, and the current enhanced release got a free 19-map episode called Dawn of the Machine on August 6, 2026. Community QuakeWorld servers, kept alive by the GPL engine source id released in December 1999, are still running. **Key facts:** - Launched: Shareware episode, June 22, 1996, id Software - Shareware price: $9.95 for the US disc, $49.95 by phone to unlock the full game - Engine source: Released under the GPL in December 1999 - Owner today: id Software, part of ZeniMax Media, a Microsoft subsidiary since March 9, 2021 - Status today: Still sold and still played. A free 19-map episode, Dawn of the Machine, shipped August 6, 2026 for the 30th anniversary ## How did a $9.95 shareware disc put Quake on PCs in 1996? id Software released the Quake shareware episode on June 22, 1996, pushing it to FTP servers and bulletin boards rather than to stores. Anyone could download the first episode, play it for free and pass the file to a friend, which is exactly how Doom had spread three years earlier. In US retail id tried a twist on the same idea. The boxed disc cost $9.95 and held the whole game in locked form, and a phone call plus $49.95 bought the code that unlocked the remaining three episodes. According to heise online in 2026, the unlock scheme was cracked quickly and id abandoned it. The money was real but smaller than expected. heise reports that Quake's sales "lagged behind 'Doom II' and thus behind expectations", which it puts down to thin home internet access in 1996 and steep hardware requirements. Quake wanted a fast Pentium, and a lot of people did not have one. The same shareware playbook that made Doom a phenomenon in 1993 (https://404memoryfound.com/posts/how-doom-changed-pc-gaming-internet.html) was already running into its limits. ## What did QuakeWorld actually fix about playing online? Quake shipped with internet play built in, and for most people in 1996 it was close to unplayable. Every keypress had to travel to the server and back before the character moved, so a 200 ms modem connection meant a fifth of a second of lag on every step. id's programmers had fast connections at the office and at home, and the problem went unnoticed until players had the game. The fix arrived later in 1996 as QuakeWorld, a separate build meant only for internet play. John Carmack rewrote the network code so the client guessed the result of your own movement immediately and corrected itself when the server's answer arrived. Writing in his .plan file in 1996, Carmack described the goal plainly, saying "the player movement feels exactly like you are playing a single player game". That idea, client-side prediction, is now standard in almost every online shooter. It also made a Quake match over a modem feel close enough to a match on a LAN that the two scenes stopped being separate hobbies, and the LAN party scene of the late 1990s (https://404memoryfound.com/posts/golden-age-of-lan-parties.html) grew alongside it rather than instead of it. ## Which games grew out of the Quake mod scene? id shipped Quake with QuakeC, a scripting language that let anyone rewrite the game's rules without touching the engine. Three Australians took it up within days. Robin Walker, John Cook and Ian Caughley built Team Fortress, a class-based team mod, and uploaded it to ftp.cdrom.com in August 1996 with five classes. Walker told Shacknews in 2018 that there was no plan behind it, only speed, saying "literally the day they released Quake C, we started on TF". Cook told the same outlet they put it online with a short readme and then forgot about it until players wrote in asking for updates. Valve hired the team, and Team Fortress Classic shipped in 1999 on the Half-Life engine, which was itself built on licensed Quake technology. Half-Life's own mod community then produced Counter-Strike, which started life as a free mod in 1999 (https://404memoryfound.com/posts/what-happened-to-counter-strike-half-life-mod.html). The line from Quake's scripting language to two of the biggest multiplayer franchises in gaming is short and direct. ## Why did id Software give the Quake code away? In December 1999, id released the Quake engine source code under the GNU General Public License. The repository id still maintains on GitHub describes the drop in one line, saying "this is the complete source code for winquake, glquake, quakeworld, and glquakeworld". The licence covers the code and nothing else. id's own README is explicit that the game's data files "remain copyrighted and licensed under the original terms", so a source port is legal to distribute but the maps, sounds and textures still have to come from a copy you bought or from the shareware episode. The effect was that Quake never needed id's permission to survive. Volunteers rebuilt the engine for operating systems that did not exist in 1996, added modern renderers and widescreen support, and kept the QuakeWorld branch alive through community clients such as ezQuake, which the nQuake package installs in one step. Thirty years on, the version most competitive players use is maintained by the community, not by a publisher. ## What does the Quake scene look like in 2026? Quake is not a mass-market multiplayer game in 2026 and has not been one for two decades, but it has never gone dark. Community QuakeWorld servers are still listed and still populated at any hour, mostly in Europe and North America, and the deathmatch and capture the flag rulesets people play are the ones the community settled on years ago. Ownership sits a long way from where it started. id Software was bought by ZeniMax Media in 2009, and Microsoft closed its acquisition of ZeniMax on March 9, 2021. Writing on Xbox Wire that day, Xbox head Phil Spencer said it was "an honor to welcome the eight incredibly talented development studios" joining Microsoft, id Software among them. The anniversary is being marked officially. heise reports that QuakeCon in August 2026 celebrated 30 years of both Quake and the convention itself, with all four id Software founders expected as guests, at the Gaylord Texan in Grapevine, Texas. ## How much does Quake cost now, and what comes with it? The version on sale is the enhanced re-release that Nightdive Studios, MachineGames and id Software built for Bethesda, first shown at QuakeCon on August 19, 2021. One purchase covers both the enhanced build and the untouched 1996 original, plus the two 1997 expansions, the newer MachineGames episodes, online and local multiplayer, and crossplay between PC, Xbox, PlayStation and Switch. On August 6, 2026, id Software and MachineGames added a new episode called Dawn of the Machine as a free update for everyone who already owned the game. Bethesda's announcement lists 19 new single-player maps, a deathmatch map and a new soundtrack, across PC, Game Pass, PlayStation 4 and 5, Xbox One and Series X and S, and both Nintendo Switch models. It sells on Steam, the Epic Games Store, the Microsoft Store and GOG, and it goes on sale often. The DRM-free build on GOG (https://www.gog.com/en/games?query=quake) is the one worth having if you want a copy that keeps working without a launcher, and it installs alongside the community source ports without any extra work. ## Frequently Asked Questions ### Is Quake still playable online in 2026? Yes. Quake, released by id Software in 1996, still has populated community servers in 2026, reached through the free nQuake package and the ezQuake client for the QuakeWorld branch. The 2021 enhanced re-release also runs its own matchmaking with crossplay between PC, Xbox, PlayStation and Nintendo Switch. ### Who owns Quake and id Software now? Microsoft owns Quake. id Software, which created Quake in 1996, was acquired by ZeniMax Media in 2009, and Microsoft completed its purchase of ZeniMax on March 9, 2021, taking id and seven other studios with it. Bethesda Softworks, also part of ZeniMax, remains the publisher on the store pages. ### Do you need the original 1996 CD to play Quake today? No. Buying the current Quake release on Steam, GOG, the Epic Games Store or the Microsoft Store gives you the 1996 original alongside the enhanced version, so no disc is needed. The engine source code has been free under the GPL since December 1999, but id Software still requires the game data files to come from a copy you own. **Sources:** - heise online, 30 Facts About 30 Years of Quake (2026): https://www.heise.de/en/background/30-Facts-About-30-Years-of-Quake-11339666.html - John Carmack, .plan file on QuakeWorld (1996), archived by Fabien Sanglard: https://fabiensanglard.net/quakeSource/johnc-log.aug.htm - id Software, Quake GPL Source Release (GitHub): https://github.com/id-Software/Quake - Shacknews, Threading the Needle: The Making of Quake Team Fortress (2018): https://www.shacknews.com/article/103827/threading-the-needle-the-making-of-quake-team-fortress - Bethesda, Quake Dawn of the Machine Available Now (2026): https://slayersclub.bethesda.net/en-US/news/quake-dawn-of-the-machine-available-now --- # What Happened to WebTV, the Box That Promised Internet for Everyone URL: https://404memoryfound.com/posts/what-happened-to-webtv-internet-television.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-10 Topics: Hardware, Internet Culture ## In 1996, WebTV Sold a Promise. By 2013, Nobody Was Buying. Here's a question that sounds absurd now but was deadly serious in 1996: what if most people never buy a computer? In the mid-1990s, this was a legitimate strategic concern. Personal computers were expensive, complicated, and intimidating. A decent desktop setup cost $1,500 to $2,000. You needed to install software from CD-ROMs. You needed to configure your modem. You needed to understand file systems, drivers, and the difference between RAM and hard drive space. For tens of millions of American households, the internet might as well have been on the moon. Three former Apple engineers looked at this problem and saw an opportunity. Steve Perlman, Bruce Leak, and Phil Goldman founded WebTV Networks in June 1995 with a radical premise: put the internet on the device that every household already owned. The television. No keyboard required. No operating system to learn. No $2,000 investment. Just a box, a remote, and a monthly fee. The internet, delivered like cable. WebTV launched on September 18, 1996, with set-top boxes manufactured by Sony and Philips, priced at $325. The monthly service fee was $19.95 for unlimited web surfing and email. Within twenty months of founding, Microsoft acquired the company for $425 million. At its peak, WebTV had roughly 1.1 million subscribers. And then, slowly and quietly, it faded into irrelevance, rebranded as MSN TV in 2001, and finally shut down on September 30, 2013, after eighteen years of service. This is the story of a product that correctly identified a massive problem, built an elegant solution, and still lost. Understanding why requires looking at what WebTV got right, what it got wrong, and what changed underneath it while nobody was watching. ## The Founders Knew Something Everyone Else Missed Steve Perlman wasn't guessing when he designed WebTV. He had spent years at Apple, working on QuickTime and multimedia technology. Bruce Leak had been the lead engineer on QuickTime. Phil Goldman had deep experience in hardware and software integration. These were not hobbyists or first-time entrepreneurs. They were senior engineers who understood both the technical constraints and the user experience challenges of putting internet content on a television screen. The problem they were solving was real. In 1996, only about 23 percent of American households had a personal computer with internet access. The rest, roughly 77 percent, were offline. Not because they didn't want the internet. Because the barrier to entry was too high. A WebTV box at $325 was less than a quarter of what a basic computer setup cost. The monthly fee of $19.95 was competitive with ISP subscriptions that required you to already own a computer. The value proposition was clear: internet access for the price of a nice VCR. The technical execution was clever. WebTV was a thin client in the form of a set-top box that connected to your television via standard composite video cables and to the internet through a built-in 33.6 kbit/s dial-up modem. The device didn't store web pages locally in any meaningful way. It connected to WebTV's proprietary servers, which would fetch, reformat, and compress web content before sending it to the box. This meant WebTV could optimize pages for television resolution and aspect ratio, something that mattered enormously when you were trying to read text designed for a computer monitor on a 27-inch CRT TV from across the living room. The system used 128-bit encryption for communication with its service, which was actually stronger than what most websites used at the time. An optional wireless keyboard was available for $50, though the system was designed to be navigable with just the remote control. The entire setup process was designed to take minutes, not hours. ## The Sony and Philips Strategy Was Brilliant Perlman and his co-founders made a decision early on that showed real business sophistication: they didn't manufacture the hardware themselves. Instead, they created a reference design and licensed it to Sony and Philips, two of the most trusted consumer electronics brands in the world. This was smart for several reasons. First, it gave WebTV instant credibility. A box with the Sony logo on it signaled quality in a way that a box from an unknown startup never could. Second, it distributed the manufacturing risk. WebTV didn't need to build factories or manage supply chains. Third, it created competitive pressure between the licensees. When Philips signed on, Sony, which had initially been hesitant, reversed course and joined as well. The threat of a competitor getting exclusive access to the WebTV platform pushed both companies to commit. From a user psychology perspective, this was also important. In 1996, consumers buying their first internet device were anxious about making the wrong choice. Seeing the Sony or Philips name on the box reduced that anxiety. WebTV was essentially using established brand trust as a substitute for the product awareness it hadn't had time to build. ## Microsoft Saw the Future (and Overpaid for It) On April 6, 1997, Craig Mundie announced at the National Association of Broadcasters conference in Las Vegas that Microsoft had acquired WebTV Networks. The negotiations had taken only six weeks. The announced price was $425 million, though the full deal was valued at $503 million when accounting for unvested employee stock options. To put that in context, WebTV was less than two years old. It had roughly 150,000 subscribers at the time of acquisition. Microsoft paid over $2,800 per existing subscriber, which, even by late-1990s standards, was aggressive. But Microsoft wasn't buying subscribers. It was buying a position. Bill Gates had been talking about internet-television convergence since at least 1995. His book "The Road Ahead" predicted a future where interactive content would flow through every screen in the house. WebTV was the closest thing to a working implementation of that vision. Microsoft saw it as a strategic hedge: if computers remained too expensive or complicated for mainstream adoption, WebTV offered an alternative path to getting Microsoft services in front of consumers. The acquisition also fit Microsoft's broader strategy of controlling the platforms through which people accessed the internet. In 1997, Microsoft was in the middle of the browser wars with Netscape (https://404memoryfound.com/posts/is-netscape-still-around.html). It was bundling Internet Explorer with Windows. It was investing in MSN. Adding a television-based internet platform to that portfolio made strategic sense, even if the economics didn't quite work yet. ## The Peak That Wasn't High Enough Under Microsoft's ownership, WebTV grew. The subscriber base expanded from approximately 150,000 at acquisition to around 800,000 by mid-1999. It peaked near 1.1 million subscribers in the early 2000s. A second-generation product, WebTV Plus, launched in 1998 with enhanced features and a monthly subscription of $24.95 (the original Classic service remained at $19.95). Those numbers sound reasonable until you compare them to what was happening in the broader market. By 2000, more than half of American households had a computer, and internet penetration was climbing rapidly. The price of a basic desktop computer had dropped below $1,000. Dell was selling machines directly to consumers online. Gateway was shipping computers in cow-spotted boxes to suburban doorsteps. America Online was carpet-bombing the country with free trial CDs, making internet setup as simple as inserting a disc. The fundamental assumption behind WebTV, that most people would never buy a computer, was being disproven in real time. Computers weren't getting simpler, exactly. But they were getting cheaper, faster, and more necessary. Email was becoming essential for work. Kids needed computers for school. The web was evolving from a novelty into a utility. And a utility that ran on a dial-up connection through a television screen, with no ability to run software, download files, or do anything beyond basic browsing and email, started to feel limiting. ## The Television Screen Problem Nobody Solved There was a deeper issue with WebTV that no amount of engineering could fix: television screens were terrible for displaying web content. In 1996, the average American television was a 27-inch CRT running at 480i resolution. Web pages were designed for computer monitors running at 640x480 or 800x600, viewed from two feet away. Trying to read those same pages on a TV screen from eight feet across a living room was, to put it diplomatically, suboptimal. WebTV's servers reformatted pages to make them more readable on television, but there were limits to what reformatting could do. Text-heavy pages were hard to read. Complex layouts broke. Websites that relied on Java applets or certain plug-ins simply didn't work. The web was not designed for television, and retrofitting it onto a TV experience created constant friction. This is a pattern that shows up repeatedly in technology: the "just put X on Y" approach almost never works as well as purpose-built solutions. Putting the web on a TV was like putting a newspaper on a billboard. The content doesn't change, but the context changes everything. Reading a 500-word article is fine on a computer monitor at arm's length. On a TV screen across the room, through a dial-up connection, navigating with a remote control, it becomes an exercise in frustration. ## The Rebrand That Signaled the End In July 2001, Microsoft rebranded WebTV as MSN TV and absorbed it into its MSN division. The rebrand was more than cosmetic. It signaled that Microsoft no longer saw television-based internet access as a distinct product category worth investing in heavily. WebTV became a feature of MSN rather than a platform in its own right. MSN TV 2 launched in 2004 with broadband support and improved hardware, but by then the market had moved decisively. Broadband internet was becoming standard in American homes, accessed through computers that could do everything MSN TV could do and vastly more. The price gap between a set-top box and a basic computer had narrowed to the point where the value proposition no longer held. Smartphones were on the horizon. Netflix was starting to stream video. The living room internet experience that WebTV had imagined was arriving, but through entirely different devices and business models. The subscriber count declined steadily through the 2000s. On July 1, 2013, Microsoft sent an email to remaining subscribers informing them that MSN TV would shut down on September 30, 2013. Subscribers were advised to migrate their favorites and data to Microsoft's SkyDrive service. After eighteen years, the service that had promised to bring the internet to every television quietly turned off its servers. ## What WebTV Got Right, and Why It Still Lost The frustrating thing about WebTV is that its founders weren't wrong about the problem. They were right that computers were too expensive and complicated for most people. They were right that the television was an underutilized screen. They were right that there was a massive market of people who wanted internet access but couldn't or wouldn't buy a PC. The demographic they targeted, older adults, lower-income households, people intimidated by technology, was real and underserved. What they couldn't have predicted was how quickly the landscape would shift. PC prices dropped faster than anyone expected. AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html)'s aggressive marketing made getting online feel easy. Broadband replaced dial-up, transforming the internet from a text-and-images medium into a multimedia platform that demanded more than a set-top box could deliver. And eventually, smartphones did what WebTV had tried to do: put the internet in everyone's hands without requiring a computer. They just did it on a screen you carried in your pocket instead of one mounted on your wall. There's a lesson here about timing and technology bets. WebTV was built for a world where personal computers stayed expensive and complicated indefinitely. That world existed in 1996. It didn't exist by 2002. The product solved the right problem at the right time, but the problem itself was temporary. The founders built a bridge, and then the river moved. ## The Real Legacy WebTV's direct legacy is small. The product is mostly forgotten. Its technology was absorbed into Microsoft's broader platform and eventually discontinued. But its indirect legacy is more interesting. WebTV proved that there was a market for internet access on non-computer devices. It demonstrated that consumers would pay a monthly subscription for a simplified internet experience. It showed that hardware could be subsidized by service revenue. These ideas, dismissed by many in the late 1990s, became the foundation of the modern streaming economy. Netflix, Roku, Apple TV, Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html) Fire Stick: they are all, in a sense, descendants of WebTV's core insight that the television is a platform, and people will pay for convenient access to content on it. Steve Perlman went on to found Rearden Studios and develop OnLive, one of the first cloud gaming services. Bruce Leak continued working in technology. Phil Goldman passed away on December 26, 2003, at the age of 39, from heart failure. The company they built together lasted eighteen years, far longer than most startups, even if its peak influence was concentrated in a narrow window of the late 1990s. WebTV is worth remembering not because it succeeded, but because it was right about almost everything except the timeline. The vision of internet content on every screen, accessible to everyone regardless of technical sophistication, is the world we live in now. WebTV just arrived fifteen years too early, on the wrong screen, through the wrong pipe. ## Frequently Asked Questions ### When did WebTV launch? WebTV launched on September 18, 1996. Set-top boxes were manufactured by Sony and Philips, priced at approximately $325, with a monthly service fee of $19.95. ### Who founded WebTV? WebTV Networks was founded in June 1995 by three former Apple engineers: Steve Perlman (who conceived the product idea), Bruce Leak (chief operating officer and executive vice president of engineering), and Phil Goldman (senior vice president of engineering). ### Why did Microsoft buy WebTV? Microsoft acquired WebTV in 1997 for $425 million (with the full deal valued at $503 million) as part of its strategy to control internet access platforms. Bill Gates had been advocating for internet-television convergence, and WebTV represented the most viable implementation of that vision at the time. ### How many subscribers did WebTV have at its peak? WebTV peaked at approximately 1.1 million subscribers in the early 2000s, up from about 150,000 at the time of Microsoft's acquisition in 1997. ### When was WebTV discontinued? The service, rebranded as MSN TV in 2001, was officially discontinued on September 30, 2013. Microsoft notified remaining subscribers on July 1, 2013, giving them three months to migrate their data. ### Why did WebTV fail? WebTV failed primarily because the problem it solved was temporary. PC prices dropped dramatically in the late 1990s and early 2000s, broadband internet replaced dial-up (which WebTV relied on), and eventually smartphones provided the accessible internet experience WebTV had envisioned, on a more convenient form factor. --- # What Happened to 3dfx? The Graphics Card That Ruled the '90s URL: https://404memoryfound.com/posts/what-happened-to-3dfx-voodoo-graphics-card.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-10 Topics: Hardware, Gaming Picture this: 1997. You're walking into your local computer store, the kind with the smell of new plastic and electrical equipment hanging in the air like a specific perfume of the era. Your eyes scan the shelf, and there it is, this sleek black and neon box with the name "Voodoo" printed on the side in letters that seemed to glow with some kind of supernatural promise. The marketing copy on the box promises you "true 3D gaming at last," and you know, somehow, that everything is about to change. You reach for your wallet. This is the moment. This is the card that separates the cool kids from everyone else. This is 3dfx's Voodoo, and for a brief, brilliant moment in the late 1990s, it was the most important piece of hardware in PC gaming. Today, 3dfx is a cautionary tale, a name whispered in forums by old gamers who remember when the company owned the world. You can't buy a Voodoo card anymore. You can't even run the latest games on one if you still have it in a box in your attic. But the story of how we got here, how a startup founded by people who used to work at Silicon Graphics convinced an entire generation of PC gamers that their lives were incomplete without a dedicated 3D card, is genuinely fascinating. It's a story about vision, about perfect timing, about one catastrophic decision that unraveled an empire, and about the thin line between dominance and oblivion in the tech industry. The fall of 3dfx is not just about a company that made bad decisions. It's about the complexity of competing in the 3D graphics space, about a market that was moving faster than anyone could predict, and about how a company can be brilliant at one thing and then suddenly irrelevant. But before we get to the collapse, we need to understand how 3dfx got here in the first place. ## Where 3dfx Came From On August 24, 1994, in San Jose, California, three guys from Silicon Graphics decided to start their own company. These weren't random engineers. They knew what they were doing. Ross Tarolli, Gordon Campbell, and Gary Smith had worked at SGI, where they'd spent years thinking about 3D graphics and how to make them fast. They understood, at a level that most people didn't, that PC gaming was about to explode. The home PC market was finally getting powerful enough that serious 3D graphics were possible, but nobody had really figured out how to do it efficiently yet. The three founders looked at the market and saw an opportunity that nobody else seemed to notice. The company was called 3dfx Interactive, and the name itself was meant to evoke something futuristic and powerful. But the really important thing about those early days wasn't the name or the founding story, though that stuff matters. What mattered was that 3dfx attracted the right kind of investment. Gordon Campbell at TechFarm believed in these guys enough to put real money behind them. This wasn't some garage startup that got lucky. This was smart people with real experience backed by investors who understood the industry. From day one, 3dfx had credibility and resources. In those early years, 1994 and 1995, the PC graphics landscape was a complete mess. You could buy a Matrox card, or a Tseng card, or a Trident card, and none of them could really do 3D the way you needed for modern games. Games were still mostly running on the CPU, doing the 3D calculations in software, which meant everything was slow and limited. What 3dfx understood was that the future belonged to companies that could offload the 3D work onto specialized hardware. They weren't the only company thinking this way, but they were the ones who executed best. The company spent its first couple years in stealth mode, basically, developing the technology that would change everything. They weren't making announcements or trying to drum up press coverage. They were heads-down on engineering, trying to solve the problem of how to make 3D graphics fast enough for real-time gaming. And by 1996, they had something. ## The Voodoo Changed Everything When the Voodoo1 finally shipped in October 1996, it wasn't available as a standalone product you could just walk into a store and buy. It came integrated into a card made by Orchid, a company that was one of the original PC graphics card makers. The card was called the Orchid Righteous 3D, and it retailed for around $299. That was a lot of money in 1996, roughly equivalent to $600 in today's money. This was a serious purchase. You didn't buy a Voodoo card on a whim. But the thing about the Voodoo was that it didn't replace your existing graphics card. This is important to understand, because it was strange and unintuitive. Your existing card still handled your 2D graphics, your Windows desktop, your screen resolution and refresh rate. The Voodoo card was connected via a passthrough, which meant it sat between your main card and your monitor, intercepting the 3D data but letting the 2D stuff pass through. This was weird engineering, and it meant you needed to have the right compatible card, and you needed to deal with some cables and configuration, but it worked. And more importantly, it meant that game developers finally had hardware that could actually deliver 3D graphics at speed. The real magic, though, wasn't just the hardware. It was the software interface that 3dfx created to go with it. They called it Glide, and Glide was revolutionary. See, at the time, there was this graphics standard called OpenGL that everyone was supposed to use, and it was pretty comprehensive but also pretty complex. 3dfx looked at OpenGL and decided that what the gaming industry really needed was something simpler, something that was specifically designed for fast, streamlined 3D graphics without all the bloat. So they built Glide. And they gave it away to developers for free. The first game that really showed what the Voodoo could do was a modified version of id Software's Quake (https://404memoryfound.com/posts/what-happened-to-quake-online-fps.html). John Carmack, the legendary programmer at id, worked with 3dfx to create a version of Quake that used Glide instead of the existing 3D renderers. This wasn't just a port. This was Quake looking better, running faster, and basically showing the world what PC 3D gaming could be if you had the right hardware and the right software working together. People saw GLQuake running on a Voodoo and they basically lost their minds. This is what they'd been waiting for. This is what they'd been dreaming about. The response was extraordinary. Within months, other games started getting Glide support. Descent, Descent 2, Jedi Knight, Quake 2, all the major games of the era started shipping with Voodoo support. It became this thing where if you wanted to play the latest games with the best visuals and the best performance, you needed a Voodoo card. The gaming community coalesced around 3dfx not because they had some regulatory advantage or because they had some evil monopoly practices, but because they had the best product and they'd made it easy for developers to support it. This is the part of the story where 3dfx did everything right. ## Total Market Domination By the end of 1997, 3dfx controlled somewhere between 80 and 85 percent of the dedicated 3D graphics card market. This is not a small number. This is domination. This is the kind of market share that makes everybody else in the industry nervous and makes your company feel invincible. And in 1997, 3dfx went public. June 25, 1997, to be exact. The IPO was priced at $11 per share, and it raised about $33 million for the company. The stock market was telling 3dfx that this was the future, that they were sitting on something special, that this company was going to be huge. And here's where it gets interesting: the revenue numbers back that up. In 1996, 3dfx had revenues of about $44.1 million. By 1997, that number had jumped to $202.6 million. We're talking about more than quadrupling revenue in a year. This is the kind of growth that venture capitalists dream about. This is the kind of growth that makes investors believe they've found the next thing. The company was riding a wave, and nobody seemed to think the wave would ever crash. Every month, more people bought Voodoo cards. Every month, more games got Glide support. Every month, 3dfx looked more and more like the company that was going to define the future of gaming. The stock price reflected this euphoria. People who bought the IPO and held on saw their investment multiply many times over. There were stories in tech magazines about how 3dfx was going to be the next Intel, the next Microsoft, the next huge tech company that would be around for decades. The narrative was almost written. The only thing that could stop 3dfx was if they made some kind of serious strategic mistake, if they got distracted, if they took their eye off the ball. And spoiler alert: that's exactly what happened, but we're getting ahead of ourselves. In 1997, 3dfx was the undisputed king of PC 3D graphics. They had enemies, certainly. Matrox was still making cards. NVIDIA was working on something. ATI was in the market. But none of these companies could touch 3dfx's market share. None of them had the same developer support. None of them had convinced gamers the way 3dfx had that their card was the essential purchase. When you walked into a computer store in 1997 and looked at the 3D graphics cards, the Voodoo was what everyone wanted. The passthrough arrangement was a little weird, sure, but the performance was so much better than everything else that the weird arrangement didn't matter. You put up with it because the payoff was worth it. ## The Voodoo2 and Peak Power In March 1998, about nine months after the IPO, 3dfx released the Voodoo2. If the original Voodoo had been a revelation, the Voodoo2 was absolutely dominant. The card was faster, more capable, and it introduced something called SLI technology, which stood for Scan Line Interleaving. This meant you could put two Voodoo2 cards in your system at the same time, and they would work together to render a single image. This was wild. This was sci-fi territory for consumer hardware. You could basically double your 3D performance by buying two cards instead of one. It was insane, but it worked, and it was thrilling to gamers who had the money and the ambition to do it. The Voodoo2 also brought 1024x768 resolution support, which meant you could finally play your 3D games at a resolution that didn't look like it had been rendered for a monitor from 1985. The original Voodoo had been limited to 640x480 or 800x600, which looked okay if you squinted but wasn't exactly gorgeous. The Voodoo2 changed that calculus. You could play Quake 2 or Unreal at 1024x768 with Voodoo2 SLI and it would look incredible. It would look like the future was here. And for a brief moment, it actually felt like that's where the technology was heading. This is also when the LAN party culture really exploded. If you were a gamer in 1998, 1999, you probably went to at least one LAN party, where gamers would haul their computers to some convention center or university and all plug into a local network to play games together. Quake 2 was the game, and if you showed up to a LAN party with a Voodoo2 SLI system, you were basically a god. You had the hardware that everyone wanted. The Voodoo2 became synonymous with competitive gaming, with serious gamers, with people who took PC gaming seriously. It wasn't just a graphics card. It was a status symbol. It was proof that you cared enough about gaming to invest serious money into your rig. The Voodoo2 sold phenomenally well. 3dfx's market share remained in that 80-85 percent range. Revenue continued to climb. The stock price continued to climb. Everything looked good. Everything looked like 3dfx was going to keep dominating forever. But here's the thing: they were already making the decision that would unravel everything. ## The STB Mistake That Started the Collapse On December 14, 1998, 3dfx announced that they were acquiring STB Systems for $141 million. STB was a company that made graphics cards, but they also had manufacturing capability, they had relationships with PC manufacturers, they had a distribution network. On the surface, the acquisition made sense. 3dfx wanted to move beyond being a chipmaker to actually being a graphics card manufacturer. They wanted to own the entire supply chain. They wanted to be more integrated, more in control of their own destiny. This was the moment where everything started to go wrong. Here's why: the OEM partners that 3dfx had been working with, companies like Diamond Multimedia that made the Diamond Monster line of Voodoo cards, they were already uncomfortable. They made their living selling 3dfx chips as finished graphics cards. The more that 3dfx integrated vertically, the more they became a competitor to these OEM partners rather than just a supplier. Diamond Multimedia had been one of the biggest Voodoo enthusiasts, making some of the most popular versions of the card. But after the STB acquisition, Diamond saw the writing on the wall. 3dfx was going to compete with them directly. So Diamond started looking for alternatives. They started working with NVIDIA. And here's where it gets worse: the STB acquisition was expensive, and it pulled 3dfx's focus away from what they should have been doing, which was staying ahead in the 3D graphics arms race. The company suddenly had to manage a manufacturing operation, deal with supply chain issues, handle inventory management. These are not things that a graphics chip company should be spending its time on. These are distractions. And while 3dfx was dealing with the STB integration, the competitive landscape was shifting. NVIDIA, a company that had been working in relative obscurity up until this point, was preparing to release something called the GeForce 256. NVIDIA had been founded in 1993, so they were only a few years younger than 3dfx, but they'd been keeping their head down, developing technology, building relationships with hardware manufacturers. They didn't have 3dfx's market dominance, but they had something that was about to become more important than market dominance: they had support for Direct3D. ## NVIDIA and the End Direct3D was Microsoft's graphics API, and it was the counterweight to Glide. While 3dfx had built Glide as this streamlined, simple interface for 3D graphics, Direct3D was more comprehensive, more heavyweight, but more importantly, it was backed by Microsoft. And Microsoft had this thing called Windows, which was running on pretty much every PC in the world. As the late 1990s progressed, the idea of using a proprietary API like Glide seemed increasingly risky. What if Microsoft decided to do something that hurt Glide? What if Direct3D got better tools and better support? What if game developers started seeing Glide as a limiting factor rather than a feature? The GeForce 256, released in 1999, supported Direct3D. It wasn't dramatically faster than a Voodoo2, but it was competitive, it was available from multiple manufacturers who weren't owned by NVIDIA, and it had the backing of Microsoft. And here's the thing that really mattered: it didn't need a passthrough cable. You plugged it into your PCI slot and it handled both 2D and 3D graphics. It was simpler. It was cleaner. It was more integrated into the Windows experience. Meanwhile, 3dfx was working on the Voodoo3 and the Voodoo5, which were supposed to combine 3D and 2D graphics on a single card, finally getting rid of the passthrough arrangement. But these cards were delayed. The engineering was harder than expected. The market was moving faster than anyone had predicted. And when the Voodoo3 finally came out, it was too late. NVIDIA had momentum. NVIDIA had developer support that was growing by the day. NVIDIA had a clear strategy. 3dfx, meanwhile, was still dealing with the STB integration, still trying to figure out how to manufacture cards efficiently, still dealing with all the chaos that comes from a merger. And here's where it gets really interesting: 3dfx's core problem wasn't that they made bad products. The Voodoo3 and Voodoo5 were solid cards. The problem was that the market had shifted underneath them. The idea that you needed a specialized 3D graphics card was becoming less necessary as integrated graphics got better and CPUs got faster. The idea that you needed a proprietary API like Glide was becoming quaint as Direct3D got better and OpenGL continued to improve. 3dfx had built their dominance on these two things: specialized hardware and a proprietary API. But the industry was moving toward open standards and integrated solutions. And 3dfx couldn't pivot fast enough to keep up. By 2000, it was clear that 3dfx was in serious trouble. The stock price had crashed. The revenue had stopped growing. The market share was eroding. Diamond Multimedia had completely abandoned them. Other OEM partners were following suit. And NVIDIA, this company that nobody had really heard of a few years before, was becoming the dominant player in 3D graphics. It happened in what felt like an instant, which is kind of insane when you think about it. You go from owning 80 percent of the market to being in a death spiral in less than three years. In the fall of 2000, NVIDIA approached 3dfx about an acquisition. The deal was for around $112 million, which sounds like a lot until you remember that 3dfx had raised $33 million in their IPO and had been making hundreds of millions of dollars in revenue. The company that once seemed like it was going to be the next Microsoft was being bought for a fraction of what the market had once valued it at. Shareholders who had bought the IPO watched their investment crater. Employees who had held company stock watched their stock options become worthless. The gold rush was over. And here's the thing that really stung: NVIDIA didn't even want to keep 3dfx operating as a company. They wanted the technology, they wanted the patents, they wanted to integrate some of the engineering talent, but they didn't want to be in the business of manufacturing graphics cards based on 3dfx's architecture. They wanted to kill Glide, they wanted to move everyone over to Direct3D and OpenGL, they wanted to make it clear that the 3dfx era was over. On February 15, 2001, NVIDIA officially ceased support for the Voodoo line. Any Voodoo card sitting in a system on that date suddenly became a historical artifact rather than a usable piece of hardware. The Glide API had been supplanted. The era had ended. And exacerbating all of this was the formal bankruptcy filing. On October 15, 2002, 3dfx formally filed for bankruptcy. This was years after the NVIDIA acquisition, years after the company had effectively ceased to exist as an independent entity. But the bankruptcy filing was the official acknowledgment that the company was gone. All the promises of the 1997 IPO, all the hype about being the future of gaming, all the dominance of the late 1990s, it all came down to this: a bankruptcy filing, a company that no longer existed, a footnote in technology history. ## What 3dfx Left Behind The collapse of 3dfx is taught in business schools as a case study in how to squander market dominance. How do you go from owning 85 percent of a market to bankruptcy in four years? How do you make that many strategic mistakes? And the answer, of course, is complicated. The STB acquisition was the obvious mistake, the moment where the company took its eye off the ball. But there were other factors too. The company was growing too fast. The executive team might not have had the right expertise. The board might not have been asking the right questions. And the market was moving faster than anyone had predicted. Direct3D and OpenGL were improving faster than anyone expected. NVIDIA was executing better than anyone expected. Sometimes, the thing that makes you the dominant player in one era makes you obsolete in the next era. The very specialization that had been 3dfx's strength became their weakness. But the interesting thing about 3dfx's legacy is that they didn't go to waste. NVIDIA bought their patents, bought their technology, bought some of their people. And if you trace the lineage of NVIDIA's technology forward, you can see the fingerprints of 3dfx all over it. The things that 3dfx learned about efficiently rendering 3D graphics, about building specialized hardware, about serving the gaming market, those insights didn't disappear. They got absorbed into NVIDIA. And NVIDIA built on those insights, improved on them, and became the dominant player that 3dfx had been. And there's something else too: the nostalgia market. Today, there are retro computer enthusiasts who collect Voodoo cards, who get them working in old systems, who try to play the games from the 1990s the way they were meant to be played. These cards are valuable now in a way that has nothing to do with their technical capabilities and everything to do with what they represent. They represent a moment in time, a pivotal moment when PC gaming changed forever. They represent the late 1990s, the LAN parties (https://404memoryfound.com/posts/golden-age-of-lan-parties.html), the excitement of seeing Quake 2 running at 1024x768 with real-time 3D for the first time. If you want to experience what gaming was like in 1998, you need a Voodoo2 and a Voodoo2. There's no digital equivalent. You can't just run these games on modern hardware and get the same experience. The technology was so specific to that era that it's become historical. The fall of 3dfx also serves as a reminder of how quickly things change in the technology industry. There are no permanent winners. There's no guaranteed success, no matter how dominant you are. 3dfx had the best product, the most developer support, the highest market share, and a huge IPO. And in less than five years, they were bankrupt. It's a humbling thing to think about if you're running a tech company. It's a reminder to stay focused, to keep innovating, to not get distracted by things that aren't core to your mission, to pay attention to what your competitors are doing. 3dfx failed at most of these things. And the industry moved on. ## FAQ Q: Can you still buy a Voodoo graphics card? A: Not from any manufacturer. 3dfx is gone, and NVIDIA stopped production of Voodoo cards years ago. However, you can find used Voodoo cards on the secondhand market, eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html), and retro computer forums. Prices vary wildly depending on the model and condition. A Voodoo2 in good condition can fetch hundreds of dollars from collectors. Q: What's the difference between the Voodoo1, Voodoo2, and Voodoo3? A: The Voodoo1 was the original, released in 1996 via the Orchid Righteous 3D. It required a passthrough cable to work with your existing graphics card. The Voodoo2, released in 1998, was faster and introduced SLI technology and 1024x768 resolution support. The Voodoo3, released in 1999, was supposed to combine 3D and 2D graphics on a single card and eliminate the passthrough arrangement, but it came too late to save the company. Q: Did NVIDIA use 3dfx's technology? A: NVIDIA acquired 3dfx and its patents in 2000. They integrated some of the technology and engineering talent into NVIDIA's own graphics architecture. However, Direct3D and OpenGL became the dominant APIs rather than Glide, so the proprietary parts of 3dfx's technology weren't as important as the patent portfolio and engineering expertise. Q: Why did Glide fail? A: Glide was a proprietary API created by 3dfx. While it was streamlined for gaming, the industry ultimately embraced open standards like Direct3D and OpenGL. As these standards improved, the incentive to develop specifically for Glide decreased. Microsoft's backing of Direct3D, combined with its dominance in the PC operating system market, meant that supporting Direct3D was more practical for game developers than maintaining Glide support. Q: Are there any games that still require a Voodoo card? A: Not in the modern sense. Games from the late 1990s were optimized for Voodoo hardware, but they can all run on modern systems through emulation or through ports to modern APIs like OpenGL. If you want to play these games on a Voodoo card the way they were originally intended, you would need an old PC with the appropriate hardware and operating system. Q: What was the STB acquisition? A: In December 1998, 3dfx acquired STB Systems, a graphics card manufacturer, for $141 million. This was intended to help 3dfx move from being primarily a chipmaker to being an end-to-end graphics card manufacturer. However, the acquisition distracted the company from its core business, and it alienated OEM partners like Diamond Multimedia, who became competitors instead of allies. Many analysts consider this acquisition to be the critical mistake that doomed 3dfx. Q: When did 3dfx go bankrupt? A: 3dfx filed for bankruptcy on October 15, 2002. However, the company had effectively ceased to exist much earlier. NVIDIA acquired 3dfx in 2000, and support for Voodoo cards was discontinued on February 15, 2001. The bankruptcy filing was more of a formal legal acknowledgment of what had already happened. --- # What Happened to Flickr, the Photo Site Yahoo Left to Die URL: https://404memoryfound.com/posts/what-happened-to-flickr-yahoo-photo-sharing.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-10 Topics: Internet Culture, Business Blunders In 2011, Flickr hosted approximately 6 billion photographs. By 2017, when Verizon finally sold the company to SmugMug, that number had barely budged despite being one of the internet's oldest and most fundamental photo repositories. The company had effectively stopped growing years earlier, yet the archive itself sat there like a fossil, preserved but inert. This is the story of what happens when a bold new platform gets acquired by a company that doesn't understand what made it bold in the first place. ## A Game That Accidentally Became a Photo Site Flickr's origin story is deceptively simple: it was an accident, a feature that escaped from something else entirely. In 2002, a small Vancouver-based game company called Ludicorp was working on an online multiplayer game called "Game Neverending." Stewart Butterfield, Caterina Fake, and Jason Classon needed a way for players to trade items and show them off to each other. They built a photo-sharing tool into the game as a workaround for that problem. Nothing revolutionary, just a practical solution to a game design problem. The weird part: people stopped playing the game and just started using the photo tool. This is the kind of moment that happens in tech more often than you'd think, but rarely with such clarity. The photo feature was so much more useful, so much more interesting, that it completely overshadowed the actual game. By 2004, Butterfield and Fake realized what they had. They shut down Game Neverending, kept the photo tool, refined it obsessively, and officially launched Flickr on February 10, 2004. Here's the thing: Flickr launched as an invitation-only beta. This sounds like a marketing gimmick now, but in 2004 it was genius. The early users were photo enthusiasts, photographers, and digital archivists. They were people who cared deeply about organization, metadata, and community. Flickr wasn't trying to be mainstream immediately. It was building credibility with the people who actually understood photography. The core insight Butterfield and Fake had was this: people wanted to share photos, but they also wanted to own the moment of sharing. They wanted to organize them, control them, explain them, preserve them for posterity. Most web companies in 2004 treated photos as disposable content, temporary artifacts in the stream. Flickr treated them as artifacts worth preserving and contextualizing. That's why the photo tool from a failed game became something people genuinely cared about, something worth building a platform around. In the early days, Flickr's growth was organic and almost quiet. The team didn't launch with massive marketing. They launched with a product that photographers needed, and photographers told other photographers. By 2005, a year after launch, Flickr had become the standard for photo-sharing among enthusiasts. Professional photographers, hobbyists, and people who just cared about their digital memories all gravitated toward the platform. The features were thoughtfully designed. The interface was clean. The philosophy was clear: your photos matter, and we're going to treat them that way. ## What Made Flickr Special If you used Flickr between 2004 and 2008, you experienced something that basically doesn't exist anymore. The platform wasn't trying to monetize you directly. It wasn't pushing ads into your face. It was building tools for photographers, and photographers were building a genuine community around those tools. The incentives were aligned in a way that is increasingly rare on the internet. Start with tags. This seems obvious now because every platform uses tagging, but Flickr invented the concept of social tagging at scale. Before Flickr, metadata was something librarians created, something formal and controlled and removed from the users themselves. Flickr showed that communities could create metadata collaboratively, that users would tag their own photos with detail and care, and that tags could be discovered and traversed across millions of images. You could click "sunset" and travel through hundreds of thousands of photographs. The tag became a navigational tool and a form of folk taxonomy that was legitimately useful. It was democratic metadata, created by the people who understood the photos best. Then there was Creative Commons integration. Flickr didn't just allow people to share photos. It allowed them to designate which creative rights came with those photos. Photographers could say: use this for non-commercial purposes with attribution, or use this freely with no restrictions, or don't use this at all, I'm keeping full copyright. This sounds bureaucratic. It was actually revolutionary. It gave photographers control and it gave the internet a massive repository of images that creators and writers and journalists could legally use. This was 2004, before Creative Commons was mainstream, before Wikipedia existed at scale, before anyone really thought about how the internet would actually produce content. Flickr was ahead of this curve by years. The API was something else entirely. Flickr published an open API that let developers build things on top of the platform. This wasn't done grudgingly. This wasn't a afterthought. This was a core design philosophy. You could write a script that organized your photos, that edited them, that displayed them in custom ways. You could build websites powered by Flickr photos. You could create tools that Flickr itself would never create because they were too niche or specific. The platform became an ecosystem instead of just a silo. Developers got excited about building on Flickr because the API was good and the company wasn't being hostile to third-party tools. The community aspects were genuinely thriving. Groups were organized by geography, by interest, by photographic style. Photographers would enter weekly competitions. Comments were thoughtful and constructive. People learned from each other. The platform attracted serious amateurs and professionals who treated the site as something worth investing time in. You could spend an hour on Flickr just discovering amazing photography from people you'd never heard of. For a moment in web history, Flickr felt like a space where the incentive structures were actually aligned: the company made money by serving photographers well, and photographers thrived because the company understood their needs and wasn't trying to exploit them. What's remarkable about this period is that Flickr was profitable. By 2005, roughly a year after launch, the company was generating real revenue from premium subscriptions. Users were willing to pay for better features, more storage, more control. This wasn't a company losing money and desperately seeking an exit. This was a company that had found a sustainable business model based on actually serving its users. That's a rarity. That's something most web companies never achieve. And then Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) showed up. ## Yahoo Writes a Check In March 2005, roughly one year after launch, Yahoo acquired Flickr for somewhere between 22 and 35 million dollars. Estimates vary because Yahoo never disclosed the exact number, which is itself telling. The acquisition wasn't controversial at the time. It looked smart: a hot startup with genuine growth, acquired by a company that could give it resources and reach. This was during the height of web 2.0 venture activity. Acquisitions like this happened constantly. Yahoo's strategic logic was straightforward: Flickr was building an index of the world's photographs, with rich metadata created by users themselves. Yahoo owned one of the web's major search engines. If you could attach that photo index to Yahoo's search infrastructure, you could create a search product that indexed images better than anyone else. You could search for photos by geography, by tagging, by community recommendation. This was a play for search dominance in a growing sector. It made sense in a board meeting. It made sense to investors. It made sense on a spreadsheet. The problem was that Yahoo thought it understood what Flickr was better than it actually did. Yahoo saw an image repository. Flickr's actual product was community and tools and creative control and a philosophy that respected photographers. Yahoo was buying technology and reach. It was acquiring the wrong thing. Butterfield and Fake stayed on briefly to run the product. The first year post-acquisition wasn't catastrophic. Flickr kept growing. It kept shipping features. Yahoo gave it resources, and the product improved in some ways. New servers. Better infrastructure. The ability to hire more engineers. In the short term, acquisition resources looked like success. But the incentive structures shifted almost immediately. Yahoo needed to monetize this. Yahoo needed to integrate this into the broader Yahoo ecosystem. Yahoo needed to make this profitable on a scale that would move the needle for a company Yahoo's size. And that's when the story turns. ## Death by a Thousand Paper Cuts The first major mistake was the Yahoo ID migration. Yahoo forced Flickr users to convert their Flickr accounts into Yahoo IDs. This was presented as integration, as streamlining, as progress. What it actually did was tie Flickr into Yahoo's dying authentication system and signal to users that Flickr was no longer its own thing. It was now a subsidiary, a property, a feature that was being absorbed into the larger machine. Users hated it. You'd have a Flickr identity, something you'd built, something that represented your photography and your community. Then Yahoo would tell you that identity was being merged into your Yahoo ID, a generic corporate login that also gave you access to Yahoo Mail and Yahoo News and Yahoo's dozen other failing properties. This wasn't integration. This was identity erasure. But more importantly, it signaled to the team that the company's priorities had shifted. Flickr wasn't being run as a product anymore. It was being managed as an asset to be integrated into a larger portfolio. The people making decisions weren't thinking about photographers. They were thinking about corporate structure and platform unification and synergy. The talent drain started immediately after. Butterfield and Fake left in 2008, roughly three years after the acquisition. Both would later go on to other things, and both would succeed dramatically. But at Flickr, their departure created a power vacuum. The people who understood the culture, who had the vision, who could fight for the product's soul within a larger organization, they were gone. What remained was a team executing against Yahoo's business priorities instead of Flickr's product priorities. Resources dried up. Not immediately, not in a dramatic way. But the resources that were allocated to Flickr were increasingly constrained by what Yahoo thought made sense for its broader strategy. Flickr didn't get the investment it needed to compete with the inevitable mobile shift that was coming. The team couldn't innovate quickly because they had to work through Yahoo's bureaucratic processes. When talented engineers started leaving for startups and Google and Facebook, they weren't being replaced at the same rate. The team got smaller. The roadmap got slower. The pace of innovation decelerated to a crawl. The features that made Flickr distinctive started getting ignored. The API that developers loved wasn't getting improved. The tagging infrastructure that made the platform unique wasn't being enhanced. New photographers were looking for tools that Flickr wasn't building because Flickr didn't have the resources or the mandate to build them. Instagram would eventually eat Flickr's lunch, but that's not because Instagram was technically superior. It's because Instagram had a focused team building for the moment, while Flickr had a fractured team building to corporate requirements. Look, this is what happens when a strategic buyer acquires a company. The buyer has its own priorities, its own roadmap, its own financial constraints. A startup that was optimizing for product excellence suddenly has to optimize for integration into a larger system. Sometimes that works out. Usually it doesn't. In Flickr's case, it didn't. By 2010, Flickr was coasting. It was still the largest photo-sharing platform in the world. It had the most photos, the best archive, the most sophisticated features. But it wasn't exciting anymore. It wasn't innovating. It was just sitting there, slowly losing momentum, while the world moved on to the next thing. ## The Mobile Miss Here's a timeline that matters. In 2007, Apple released the iPhone. The phone had a camera. It was basic by today's standards, 2 megapixels, no flash, no optical zoom, but it existed, and people started using it to take photos. In October 2010, Kevin Systrom and Mike Krieger launched Instagram as an iPhone app. Instagram let you take a photo on your phone, apply a filter, and share it instantly to a social network. It was elegant. It was frictionless. It understood what the mobile moment actually meant. Flickr didn't launch an official iPhone app until late 2009, and the original version was mediocre. This matters because Flickr's entire advantage was that it understood photographers and photography. But Instagram understood something more important: they understood the moment when the photographer became casual, when taking photos stopped being something you did with special equipment and started being something you did because you had a phone in your pocket. They understood that the smartphone camera was going to fundamentally change how people thought about photography. Flickr assumed that photographers wanted tools. Instagram understood that everyone with a phone is a photographer now. Instagram made it effortless to take a photo, make it look good, and share it. The user experience was elegant. The filters made ordinary photos look interesting and cohesive. The sharing infrastructure meant that your friends would instantly see what you posted. All of this was frictionless. Flickr required you to upload photos from your computer. Or you used their half-baked mobile app. Or you used a third-party solution like Shoebox. Or you used your phone's camera app and then used Instagram, which was faster and more fun. This is essentially what every company that dominated the desktop web had to learn in the 2000s: mobile wasn't a secondary platform that you could address later. Mobile was going to be the primary platform for an entire generation of users. If you didn't build mobile-first, you were building for the past. Instagram grew explosively. By 2012, it had 100 million users. By 2014, when Facebook bought it for $1 billion, it had become clear that Instagram had won the photo-sharing war. Flickr was no longer the destination. It was the archive. It was where photographers went to backup their photos and manage their portfolios. It was no longer where moments happened. This wasn't because Flickr's features were bad. This wasn't because Instagram had better photographers. Instagram won because they understood that mobile-first wasn't an option, it was the inevitability. And they built for that inevitability before it arrived. Flickr was still treating mobile as a secondary feature in 2010, and by then it was too late. ## The Mayer Hail Mary Marissa Mayer became CEO of Yahoo in July 2012. She was young, she was competent, she understood technology. She came in with a theory that Yahoo could be revived, that there were products inside Yahoo that could be saved if they were just run better. Flickr was one of those products. She saw potential in the archive, in the community, in the technical infrastructure. In March 2013, Yahoo announced that Flickr would offer 1 terabyte of free storage. For context, this is roughly what you'd pay $120 per year for on most cloud storage platforms. The strategy was bold: flood the market with free storage, get people to use Flickr again, rebuild the user base. It was a Hail Mary pass, and Mayer had the resources to make it work. She was willing to burn cash to revive products she believed in. For a moment, it looked like it might actually work. Users started uploading photos again. The platform got attention. The Flickr forums lit up with activity. Maybe, just maybe, Yahoo had realized the mistake and was going to fix it. Maybe the archive was going to become relevant again. Then the limitations became apparent. Yes, you got 1 terabyte of free storage. But the mobile app still wasn't great. The community features that made Flickr special were still neglected. The API that developers loved still wasn't being improved. You could store unlimited photos, but the tools for organizing them, sharing them, discovering them were all still basically frozen in 2010. You had vast storage but limited tools to use it. The real question is: why did Mayer think free storage was the answer when the actual problem was that Flickr had stopped innovating? You could offer unlimited storage, but if the product wasn't changing, if the team wasn't shipping new features, if the vision wasn't evolving, users would still leave. And they did. The bump from 1TB free storage lasted a few months. Then the growth plateaued again. Mayer's theory was right: Flickr could have been saved. But it required more than throwing storage at the problem. It required giving the team the autonomy to innovate, the resources to build for mobile properly, the mandate to compete with Instagram and Facebook on their own terms. It required someone to admit that the product had been neglected and was going to need years of serious investment to recover. It required strategic focus and competitive aggression. Yahoo wasn't willing to make that commitment. So by 2015, Flickr was stagnating again. Yahoo's stock was declining. The company was being dismantled. And Flickr was, once again, left to fend for itself. ## SmugMug and the Afterlife In June 2017, Verizon bought Yahoo's operating assets for $4.5 billion. Verizon had no interest in running a photo platform. They acquired Yahoo for patents, for intellectual property, for the user base and the search business. Flickr was a footnote in that deal. In April 2018, SmugMug, a company that has been quietly running a professional photo platform since 2002, acquired Flickr from Verizon. The purchase price was never disclosed, but Verizon was essentially giving away the company at that point. The data center costs alone were more valuable than whatever they could get in a fire sale. SmugMug's approach has been straightforward: salvage what can be salvaged, shut down what doesn't make sense, return the service to sustainability. Within a few months of taking over, they ended the 1TB free storage plan. Flickr now allows free users to store up to 1000 photos for free. You can still view all your photos, but you can only upload 1000 of them for free. Premium users get unlimited storage and features. This is actually a sensible business model. It's sustainable. It doesn't bankrupt the company. It allows serious photographers to continue using the platform while generating actual revenue. But it's also an admission of defeat. Flickr isn't trying to dominate anymore. It's trying to survive. SmugMug has kept the site running. They've fixed some technical debt. They've kept the archive intact. If you used Flickr in 2008 and you login today, all your photos are still there, still organized, still accessible. The company respects the history and the data in a way that Yahoo fundamentally didn't. But Flickr isn't a growth story anymore. It's a legacy property. It's where photographers store their work, where the history of digital photography lives, where creative licensing models were first tested at scale. The site still hosts billions of photos. It still has features that no other platform offers. The API still exists. The community still exists, though it's much smaller. Flickr isn't dead. But it's not the future anymore. It's the past. It's the archive. It's the proof that even great ideas can get squashed by companies that don't understand them. ## The Stewart Butterfield Footnote Here's something worth noting: after Butterfield left Flickr, he went on to found Slack. Slack started as an internal communication tool for a gaming company called Tiny Spoon that also failed. When the game company shut down, the team realized that the communication tool was more valuable than the actual product. They spun it out, refined it, and by 2014 it had become one of the fastest-growing software products in history. The pattern is identical to Flickr. This is essentially the same pattern as Flickr. A tool built for one purpose, accidentally becoming more valuable than the original product, getting spun out and polished into something genuinely important. The difference is that Slack remained independent, remained focused, remained willing to iterate and improve based on what users actually needed. Slack had the autonomy to be great. Butterfield learned from the Flickr experience. When Slack was acquired by Salesforce for $27.7 billion in 2021, he made sure the integration was handled carefully. Slack remained its own product. It kept its own CEO. It maintained its independence within a larger organization. He wasn't going to let what happened to Flickr happen to Slack. He'd seen what happened when a thriving product gets absorbed into a larger corporate structure. The pattern recognition here is: great products that start as accidents can thrive or die depending entirely on how they're managed post-acquisition. Flickr had every advantage. It had users who cared. It had technology that worked. It had a team that understood the vision. What it didn't have was a parent company that respected what had been built. ## FAQ Is Flickr still around? Yes. SmugMug acquired it in 2018 and continues to operate it. You can still upload photos, organize them, share them, browse other people's work. The site isn't dead, but it's not growing either. Can I still access my old Flickr photos? Yes. Unless you deleted them, they're still there. Flickr kept all the data intact through the transitions. This is one thing Yahoo and Verizon got right: they didn't erase the archive. What's the free tier now? Free users can upload up to 1000 photos. You can view and organize all of them, but you can't upload the 1001st until you delete something or upgrade. Premium users get unlimited uploads and storage. Why didn't Yahoo just sell Flickr earlier? Because they were convinced it was valuable as part of a larger portfolio. They thought the photo index plus search was a winning combination. They were wrong. But admitting you're wrong and selling a major property you acquired is hard, especially if you paid 30 million dollars for it. So Yahoo held onto Flickr, slowly starving it, until it was essentially worthless. Did Instagram copy Flickr? Not really. Instagram was designed from scratch for mobile. The founders weren't trying to improve Flickr. They were building for a different platform, a different use case, a different user. But Flickr had shown that photo-sharing could work as a business, that people cared about photography online, that community mattered. Instagram took those lessons and applied them to a mobile-first world. Is there a lesson here? Several. First: acquisition by a larger company usually kills startups unless the parent company respects what's been built. Second: mobile disruption was real and predictable, and every company that didn't take it seriously got demolished. Third: the best product doesn't always win. The product that understands the moment it's operating in wins. Flickr understood the 2004 moment perfectly. It failed to understand the 2010 moment. By then, it was too late. --- # What Happened to Oregon Trail, the Game That Taught America to Die of Dysentery URL: https://404memoryfound.com/posts/what-happened-to-oregon-trail-game.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-09 Topics: Gaming, Then vs Now ## The Origin Story: Three Teachers and a Teletype On December 3, 1971, a student teacher named Don Rawitsch sat in front of a teletype terminal connected to an HP 2100 minicomputer in Minneapolis and watched his 8th graders play a game about dying on the frontier. Two weeks earlier, that game hadn't existed. Within three decades, it would sell over 65 million copies. The Oregon Trail game history begins not with a software company or a business plan, but with a frustrated teacher trying to make the westward expansion interesting to teenagers. Rawitsch was a history major at Carleton College in Northfield, Minnesota, working as a student teacher at Jordan Junior High School. His supervising teacher had assigned him a unit on the Western Expansion of the Mid-19th Century, and he'd been planning a board game about the Oregon Trail. Sheets of paper laid out on his apartment floor, trail routes drawn by hand, random event cards scattered around. His roommates, Bill Heinemann and Paul Dillenberger, both math students with programming experience, walked in and made an observation that would change educational computing: this would work better as a computer program. The three of them built the game in two weeks. They coded it in HP Time-Shared BASIC on a teletype connected to the Minneapolis school district's shared HP 2100 minicomputer. The video game industry barely existed in 1971. There were no resources for game development, no frameworks, no tutorials. They worked from their own programming knowledge and their understanding of what would engage students. The result was a text-based simulation that asked players to make the same kinds of decisions that actual frontier travelers had faced: how much food to buy, how fast to travel, when to rest, how to cross rivers. The constraints of the hardware shaped everything about the design. Text-only output meant no graphics. Shared computing resources meant the program had to be efficient. Limited memory meant every decision point had to matter. Those limitations forced Rawitsch, Heinemann, and Dillenberger to distill the frontier experience down to its essential calculations. What they ended up with was pure game theory wrapped in historical context, and it worked. ## MECC and the Institutional Pipeline The Oregon Trail computer game might have stayed a local classroom experiment. Instead, it became infrastructure. In 1973, the state of Minnesota established the Minnesota Educational Computing Consortium, known as MECC. The organization's mission was straightforward: take advantage of falling computer costs to put educational software in front of students across the state. MECC wasn't a startup. It wasn't venture-backed. It was a state-funded educational cooperative with a public mission and institutional relationships with every school district in Minnesota. Don Rawitsch had joined MECC in 1974. He brought the Oregon Trail code with him. By 1975, MECC had ported it to their mainframe systems and made it available to schools statewide. The game became the most popular piece of software in their entire catalog, drawing thousands of players monthly. For context, this was years before personal computers existed in any meaningful way. Students accessed the game through terminals connected to shared mainframes. Playing Oregon Trail meant scheduling time on limited hardware, which made it feel like an event rather than a routine. The numbers reveal the scale of what MECC accomplished. By the early 1980s, Oregon Trail accounted for roughly one-third of MECC's annual revenue, which hovered around $30 million. That means Oregon Trail alone was generating approximately $10 million per year for an educational nonprofit. More importantly, MECC had negotiated a contract with Apple in 1978 to supply 500 computers to Minnesota schools. That contract tied the Oregon Trail directly to the Apple II ecosystem at exactly the moment that ecosystem was about to expand nationally. In 1979, MECC released Oregon Trail on the Apple II, distributed on a floppy disk labeled "Elementary Volume 6." The Apple II was becoming the computer of choice for American schools. It was affordable enough for district budgets. Teachers could use it for multiple subjects. By choosing to support this platform, MECC positioned Oregon Trail precisely where it needed to be to reach millions of students. What's notable about MECC's approach is what they chose not to do. They didn't lock the software down with restrictive licensing. They didn't price it beyond school budgets. They didn't treat Oregon Trail as intellectual property to be maximized. They treated it as educational infrastructure to be distributed. The game succeeded because the organization behind it was designed for distribution, not extraction. ## The 1985 Version: When Oregon Trail Became Universal The 1985 release of Oregon Trail for the Apple II is the version most people remember. It had actual graphics: a pixelated covered wagon, green landscapes, river crossing animations, and the iconic death screens that were somehow both morbid and hilarious to an 11-year-old sitting in a school computer lab. "You have died of dysentery" became a phrase that transcended the game itself, entering the broader cultural vocabulary. The 1985 version succeeded because it balanced accessibility with depth. The core mechanics remained unchanged from the 1971 original: manage your resources, set your pace, make decisions about hunting, trading, and river crossings. But the graphics made the experience more immersive without making it more complicated. Students could visualize what was happening. The covered wagon actually moved across the screen. Animals appeared when you hunted. Rivers looked like rivers. Between 1985 and 1995, Oregon Trail achieved something unusual in software: near-universal cultural penetration within its target demographic. Nearly every American school child encountered it at some point. The game was discussed in newspaper articles, referenced on television, mentioned by parents who'd heard about it from their kids. Oregon Trail wasn't just popular educational software. It was the shared experience of an entire generation's relationship with computers. The game worked as pedagogy because it respected its players. It didn't explain the westward expansion through text boxes or quizzes. It made students experience the economic and physical realities of frontier travel through consequential decision-making. If you went too fast, your oxen tired and people died. If you didn't buy enough food, you starved. If you tried to ford a deep river without adequate preparation, you lost supplies and sometimes lives. The game taught systems thinking through play: every decision had cascading consequences, and the only way to learn was to make mistakes and adjust. This is essentially what modern game designers call "experiential learning," except MECC was doing it in 1985 without the vocabulary or the academic framework. They just knew it worked because students were engaged, and teachers reported better outcomes. ## The Acquisition: When Mission Changed In October 1995, SoftKey International acquired MECC for $370 million in stock. The deal made financial sense from SoftKey's perspective. MECC was profitable, had established distribution channels into schools, and owned the most recognized educational software brand in America. For SoftKey, it was a strategic consolidation play in a market they were attempting to dominate. But the acquisition represented a fundamental shift in how Oregon Trail would be managed. MECC had been a nonprofit cooperative focused on educational access. SoftKey was a for-profit company focused on growth through acquisition and revenue optimization. The incentive structure changed completely. The question was no longer "how do we get this software into as many schools as possible?" It became "how do we extract maximum revenue from this software asset?" SoftKey eventually became The Learning Company, which continued acquiring educational software publishers throughout the late 1990s. Oregon Trail was ported to Windows, Macintosh, and eventually CD-ROM formats. New versions were released with updated graphics and expanded features. The brand was extended into sequels and spinoffs. From a purely commercial standpoint, the strategy made sense. The problem was subtler. Fragmenting Oregon Trail across platforms and versions dissolved the singular shared experience that had made it culturally powerful. A kid playing Oregon Trail II on a home Windows PC in 1998 had a different experience than a kid playing the 1985 Apple II version in a school lab. The game was being optimized for commercial distribution rather than for the institutional channels that had made it pervasive. ## The Learning Company Collapse The Learning Company went public during the dot-com era, positioning itself as an educational technology play with significant growth potential. The stock was promoted aggressively. But the company had serious operational problems beneath the surface. Acquisitions hadn't integrated smoothly. Revenue recognition practices were questionable. The company's growth story was built on buying other companies rather than organic expansion. Mattel acquired The Learning Company in 1999 for $3.5 billion. Within a year, the acquisition was recognized as one of the worst in corporate history. The Learning Company lost $206 million in its first quarter under Mattel's ownership. By 2000, Mattel was writing down the entire acquisition. The CEO who approved the deal was forced out. The Learning Company was eventually sold off for a fraction of what Mattel had paid. During this period, Oregon Trail was minimized. It became a legacy product in a portfolio being liquidated. The game that had been MECC's crown jewel became a line item in a bankruptcy proceeding. The irony is pointed: the game that had thrived under a nonprofit's stewardship became collateral damage in a series of corporate disasters driven by exactly the kind of speculative excess that nonprofits are designed to avoid. By the mid-2000s, Oregon Trail's ownership had been scattered through multiple asset sales. Different companies held rights to different versions. The coherent development history that had stretched from 1971 to 1995 under MECC's care essentially ended. The game wasn't abandoned, exactly. It kept appearing in new forms, on new platforms, through new publishers. But the institutional continuity was gone. ## Why Oregon Trail Worked When Most Educational Games Failed The educational software market has always been littered with failures. Most educational games feel like homework wearing a costume. They prioritize curriculum compliance over engagement. They teach through repetition rather than experience. They treat students as passive recipients of information rather than active decision-makers. Oregon Trail succeeded because it inverted this formula. The game didn't explain the westward expansion. It simulated the decision-making environment that real frontier travelers faced. The mechanics weren't educational scaffolding. They were the education itself. Understanding resource management, risk assessment, and trade-off analysis was the point, and the frontier setting provided the context that made those abstract concepts tangible. The game also benefited from what might be called "institutional fit." MECC understood schools. They understood budgets, scheduling, hardware constraints, and teacher needs. They didn't try to sell Oregon Trail through consumer channels or mass marketing. They sold it through relationships with school districts, through integration with school hardware purchases, through a distribution model that matched how schools actually acquired software. The game reached students because the organization delivering it was designed to reach schools. There's also a design lesson in the original's simplicity. Because the 1971 version was text-only, players had to engage their imagination. You weren't watching a story unfold on screen. You were making decisions about scenarios you had to visualize. That active engagement, that requirement for imagination, may have been more pedagogically effective than the graphically richer versions that followed. ## The Digital Afterlife Oregon Trail was inducted into the World Video Game Hall of Fame in 2016. The recognition placed it alongside Pac-Man, Tetris, and Doom (https://404memoryfound.com/posts/how-doom-changed-pc-gaming-internet.html) as one of the most culturally significant games ever created. For a text-based educational simulation coded on a teletype in 1971, that's a remarkable trajectory. The internet, which had marginalized Oregon Trail as an active teaching tool, paradoxically became its preservation mechanism. Emulators, web-based ports, and digital archives made various versions of the game accessible again. New generations could play it, not because a teacher assigned it, but because they wanted to understand what had been so important to their parents' generation. Modern mobile versions have been released, including a 2021 Apple Arcade version commemorating the game's 50th anniversary. These versions update the graphics and interface while preserving the core mechanics. The franchise has generated over 65 million copies sold across all versions and platforms, making it one of the most successful educational products ever created. What the digital afterlife revealed was that Oregon Trail's appeal wasn't primarily nostalgic. The core mechanic of managing resources under uncertainty, making calculated trade-offs between competing priorities, and living with the consequences of your decisions is genuinely timeless. It maps onto fundamental principles of decision theory and systems thinking. The frontier setting provided flavor, but the underlying structure was universal. ## What Oregon Trail Teaches About Software Stewardship The arc of the Oregon Trail game history offers a case study in how institutional values shape software outcomes. Under MECC's nonprofit stewardship, the game was treated as a public good. It was priced for accessibility, distributed through institutional channels, and maintained with educational mission as the primary objective. The result was three decades of cultural relevance and near-universal adoption in American schools. Under commercial ownership, the game was treated as an asset. It was fragmented across platforms, extended into sequels of varying quality, and eventually caught up in corporate collapses that had nothing to do with the game itself. The software didn't fail. The organizations managing it failed, or more precisely, they succeeded at objectives that were misaligned with what had made the game valuable. In the current era of debates about tech companies, platform governance, and the tension between public benefit and private profit, Oregon Trail provides a historical data point. The game worked best when it was managed by an organization whose mission was aligned with its users' needs. It declined when that alignment broke. The 65 million copies figure is impressive, but the more meaningful statistic is harder to quantify: how many students learned to think in systems because a text-based game on a teletype made them manage oxen and ford rivers. That's the real legacy of what happened to Oregon Trail, and it's a legacy that no corporate restructuring can diminish. ## Frequently Asked Questions Who created Oregon Trail? Don Rawitsch, Bill Heinemann, and Paul Dillenberger created Oregon Trail on December 3, 1971. They were student teachers at Carleton College in Northfield, Minnesota. Rawitsch was teaching 8th grade history at Jordan Junior High School in Minneapolis and wanted to make the westward expansion more engaging for his students. What computer was Oregon Trail originally written for? The original version was written for an HP 2100 minicomputer using HP Time-Shared BASIC. Players accessed it through a teletype terminal. The game was entirely text-based with no graphics. What was MECC? MECC, the Minnesota Educational Computing Consortium, was a state-funded educational organization founded in 1973. MECC distributed Oregon Trail and other educational software to schools across Minnesota and eventually nationwide. MECC was acquired by SoftKey in October 1995 for $370 million in stock. How many copies of Oregon Trail were sold? Over 65 million copies of Oregon Trail have been sold across all versions and platforms from 1971 through the present day, making it one of the most successful educational games in history. What was the most popular version of Oregon Trail? The 1985 Apple II version is the most widely remembered. It featured color graphics, a visible covered wagon, hunting animations, and the famous "You have died of dysentery" death screen. This version was ubiquitous in American school computer labs throughout the late 1980s and early 1990s. Why did Oregon Trail decline in popularity? Several factors contributed. The rise of home computers and the internet gave students access to more graphically sophisticated games. After MECC was acquired by SoftKey in 1995, the game's institutional distribution model was replaced by a commercial one. The subsequent collapse of The Learning Company, which inherited Oregon Trail through corporate acquisitions, scattered the game's ownership across multiple entities. Is Oregon Trail still available today? Yes. Various versions are available including mobile apps, web-based emulations of classic versions, and a 2021 Apple Arcade release commemorating the 50th anniversary. The original versions can also be played through browser-based emulators and digital preservation sites. Why was Oregon Trail inducted into the Video Game Hall of Fame? Oregon Trail was inducted into the World Video Game Hall of Fame in 2016 for its cultural impact, its influence on educational computing, and its longevity. The game demonstrated that educational software could be both engaging and pedagogically effective, influencing how an entire generation experienced computers in schools. What made Oregon Trail different from other educational games? Most educational games teach through repetition or direct instruction. Oregon Trail taught through simulation and consequential decision-making. Players experienced the economic and physical realities of frontier travel by managing resources, assessing risks, and living with the outcomes of their choices. The game made abstract concepts like systems thinking and trade-off analysis tangible through play. --- # What Happened to GoldenEye 007, the Game That Proved FPS Could Work on Console URL: https://404memoryfound.com/posts/what-happened-to-goldeneye-007-n64.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-09 Topics: Gaming, Then vs Now Picture this: 1997. You're standing in an Electronics Boutique at the mall, and you've got birthday money burning a hole in your pocket. One N64 game. That's what you can afford. The cartridge you grab is heavier than you expect, substantial in that way that only cartridges ever were, and the label catches the fluorescent light just right. GoldenEye 007. You had absolutely no idea you were holding the game that would prove first-person shooters could work on a console. You just knew it looked cool and your friend's older brother said it was amazing. For anyone who lived through the late 1990s, GoldenEye 007 wasn't just a game. It was the game. It was the reason you bought an N64. It was what happened at every sleepover, every birthday party, every afternoon when four people could scrape together enough controllers. Before Halo made online shooters mainstream, before Call of Duty became a cultural force, before anyone used the word "esports" without getting laughed at, GoldenEye 007 was the biggest multiplayer experience most of us had ever known. And then, for nearly two decades, it basically vanished. The GoldenEye 007 history is one of the most frustrating stories in gaming. It's a story about creative brilliance colliding with corporate licensing nightmares, about a game so good that it transcended everything around it, and about how the business side of entertainment can lock away something an entire generation loved. But before we get to the frustrating part, we need to talk about the miracle that was GoldenEye's creation. Because what happened at a small British studio in the English countryside shouldn't have been possible. ## Ten People, Zero Experience, One Masterpiece Rare was a British video game developer based in Twycross, Leicestershire, known primarily for making licensed games that were, at best, competent. Not revolutionary. Not boundary-pushing. Just solid, professional work that publishers could rely on. When Nintendo approached them about making a game based on the upcoming James Bond film GoldenEye, it was treated as another licensed project. Another IP to turn into something playable. Another job. The original concept, proposed by director Martin Hollis, started as a side-scrolling action game for the Super Nintendo. A 2D shooter. Perfectly respectable. Perfectly forgettable. Then the Nintendo 64 appeared on the horizon, and Hollis had an idea that most people in the industry would have called delusional: what if they made a 3D first-person shooter instead? Here's the thing you have to understand about 1995. FPS games on consoles were considered basically impossible. The mouse and keyboard setup that powered Doom (https://404memoryfound.com/posts/how-doom-changed-pc-gaming-internet.html) and Quake (https://404memoryfound.com/posts/what-happened-to-quake-online-fps.html) felt essential to the genre. How could you possibly control a first-person camera with an analog stick? The entire industry had essentially agreed it couldn't be done well. Martin Hollis and his team decided to find out if everyone was wrong. The team that took on this challenge was tiny. About ten people total. And here's the detail that still blows my mind: for all but two of them, GoldenEye was their first game ever. First-time developers, working on hardware that didn't fully exist yet, building a game engine from scratch using graphics textbooks because they couldn't afford middleware. They were learning 3D programming while simultaneously inventing solutions that nobody in the industry had figured out. The whole thing cost about $2 million to make, which even in 1990s money was almost nothing for what they accomplished. The development took about two and a half years. During that time, the team was working in relative isolation at Rare's rural studio, building something that nobody outside those walls fully understood. They drew inspiration from Doom, the 1993 id Software classic that defined the FPS genre, and from Virtua Cop, the 1994 arcade shooter that proved gunplay could feel visceral and satisfying. But GoldenEye wasn't copying either game. It was taking the best ideas from both and building something entirely new on top of them. ## The Game That Proved Everyone Wrong GoldenEye 007 launched on August 25, 1997, two years after the film it was based on. In the world of licensed games, releasing two years late was usually a death sentence. Movie tie-in games were supposed to ride the marketing wave of the film's release. A game arriving two years after the movie had left theaters should have been irrelevant. Instead, it was revolutionary. The first thing you noticed when you loaded up GoldenEye was that the controls actually worked. The N64's analog stick, that weird little mushroom cap that felt like nothing else in gaming, turned out to be perfect for camera control when someone actually bothered to design around it. Rare had cracked something that the entire industry thought was impossible. Moving through 3D spaces felt natural, responsive, intuitive. The game just felt right in your hands. But the controls were just the foundation. The real revelation was the game design. Unlike most shooters of the era, GoldenEye wasn't just about killing everything that moved. It was about completing objectives. In the Facility level, you could charge in guns blazing, or you could use your watch laser on a security camera, slip through the shadows, and complete your mission without anyone knowing you were there. The game supported stealth in an era when stealth mechanics in shooters were practically unheard of. Each level was designed like a puzzle box. Guards called for reinforcements if you weren't careful. Difficulty settings didn't just make enemies tougher. They actually changed the levels themselves, adding new objectives, new security systems, new challenges. The same stage became a genuinely different experience on Agent versus 00 Agent difficulty. You could spend dozens of hours mastering the campaign, and people did. People absolutely did. The single-player campaign alone would have made GoldenEye legendary. Sixteen missions, each feeling distinct and purposeful. The level design was phenomenal. The weapons had weight and kick. The gunshot sounds cracked through your television speakers with satisfying sharpness. When you fired the PP7 and watched a guard react, stagger, grab his arm where you'd hit him, it sold the impact in a way that no console game had managed before. Playing GoldenEye was a full sensory experience, and it was unlike anything else available on any console. The game won Console Game of the Year at the AIAS awards. Publications that had been skeptical about FPS games on consoles suddenly had to reconsider everything they thought they knew. Rare had done the impossible. And then the impossible got even more impossible. ## The Multiplayer That Almost Didn't Exist And here's where it gets really wild. The multiplayer mode in GoldenEye 007, the thing that turned it from a great game into a cultural phenomenon, wasn't part of the original plan. It wasn't in the design document. Management didn't request it. A programmer named Steve Ellis basically built it as a side project during the final weeks of development. He wanted to see if split-screen deathmatches could work on the N64, so he just started coding it. Without asking permission from Rare's management. Without telling Nintendo. He and a few team members just made it happen in roughly six weeks. Think about that for a second. The defining multiplayer experience of an entire console generation was created essentially in secret, as an unauthorized side project, by a programmer who was curious whether it could be done. Nobody approved it. Nobody budgeted for it. Nobody even knew it was happening until it was basically finished. And it was incredible. Four players, split-screen, on one television. Characters from the campaign. Maps based on single-player levels. A weapon selection that made every match feel chaotic and unpredictable. Proximity Mines in the Facility bathroom. Power Weapons on the Complex. Slappers Only on whatever map someone had the audacity to suggest. Every match was different. Every match was ridiculous and fun and endlessly replayable. Four controllers plugged into one console, four friends crowded around a television, arguing about whether Oddjob was cheating because he was too short to hit. That was the peak of multiplayer gaming for an entire generation, and it happened almost by accident. Here's a detail I love: Shigeru Miyamoto at Nintendo caught wind of the multiplayer and had concerns. He sent a fax to Rare suggesting that there was too much close-up killing, that watching friends shoot each other in the face at point-blank range felt uncomfortable. Rather than toning down the violence, the team came up with the iconic curtain call credits sequence. At the end of the single-player campaign, all the characters, heroes and villains alike, come out and take a bow, like actors at the end of a play. The message was clear: this was all a performance. Nobody really got hurt. Even Jaws bowed. Even Oddjob bowed. It became one of the most beloved touches in the entire game, and it came directly from Miyamoto's discomfort with the violence. ## Eight Million Copies and a Cultural Takeover GoldenEye 007 sold 8 million copies worldwide, making it the third best-selling game on the Nintendo 64. But that number doesn't begin to capture how dominant it was culturally. Every gamer of a certain age played GoldenEye. It was universal in a way that very few games ever achieve. It wasn't just popular. It was the shared language of a generation. The game had literally invented new conventions for how FPS games could work on consoles. The analog stick control scheme became the template that every subsequent console shooter would build on. Objective-based mission design became standard. The idea that difficulty should modify level design, not just enemy health, influenced game designers for years afterward. Rare's spiritual successor, Perfect Dark, launched in 2000 and expanded on everything GoldenEye had accomplished, adding AI-controlled bots called Simulants to the multiplayer and pushing the N64 hardware to its absolute limits. What made the GoldenEye N64 experience special was that it felt complete. The campaign was meaty and substantial. The multiplayer was pure joy. The character models were iconic. The sound design was incredible. Nothing felt like an afterthought. Nothing was there to fill time or pad a feature list. Every element served a purpose. And the multiplayer, oh man. GoldenEye multiplayer didn't require you to adjust configuration files or understand network settings. You didn't need a broadband connection or a separate screen for each player. You plugged in four controllers and started playing. That simplicity was revolutionary. It made competitive FPS gaming accessible to anyone who could hold a controller, and it turned living rooms across the country into arenas. ## Licensing Hell: How Corporations Killed Access Here's where the story turns cruel. GoldenEye 007 was based on a James Bond film. That detail, the one that made the game appealing in the first place, became the thing that made it nearly impossible to play for almost two decades. The rights situation was a nightmare. Nintendo published the original game. Rare developed it. MGM and Eon Productions owned the James Bond property. Microsoft acquired Rare in 2002, gaining the studio but not the game's publishing rights. Everyone had a piece of the puzzle and nobody had the complete picture. As long as physical copies were circulating on the secondhand market, nobody really cared. But when the N64 faded from mainstream relevance, maintaining the overlapping licensing agreements became more trouble than it was worth. By the late 2000s, GoldenEye was becoming a ghost. New copies were gone from store shelves. Used cartridges commanded premium prices. The game couldn't be sold digitally because the licensing parties couldn't agree on terms. A completed Xbox 360 remaster sat finished and ready to ship, but it never came out. A fully functional game, complete and playable, locked in a vault because four corporations couldn't sort out the paperwork. A finished remaster, ready to release, trapped in licensing purgatory for years. This is what corporate rights management did to one of the greatest games ever created. For years, the only way to play GoldenEye was to own an N64 and a physical cartridge. People hunted through garage sales, swap meets, and eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html) listings. Prices climbed. Hardware degraded. A cultural artifact that had defined an entire console generation was slowly becoming inaccessible. It was like watching someone lock a beloved album in a safe and throw away the combination. ## The Long Absence and the Growing Legend Something strange happened during GoldenEye's years of inaccessibility. The game became more legendary precisely because nobody could play it. People who had experienced it as kids talked about it with a reverence that bordered on religious. "You had to be there," they'd say, and they meant it. The memory of sitting in a room with three friends, all of you crammed around a 27-inch CRT television, screen divided into four tiny quadrants, that memory became sacred. GoldenEye became a rallying point for discussions about game preservation. Why couldn't one of the most important games ever made be legally purchased and played? Why did corporate licensing agreements take precedence over cultural access? The game illustrated, better than almost any other example, how the business structures of entertainment could lock away things that mattered to millions of people. Meanwhile, the FPS genre had evolved dramatically. Halo had redefined console shooters in 2001. Call of Duty had become a global phenomenon. Online multiplayer had replaced local split-screen as the default way to play competitively. The world that GoldenEye had helped create had moved far beyond what GoldenEye itself could offer technically. But none of that erased what the game had accomplished. None of it diminished the memories. Fan communities kept the game alive through emulation, speedrunning, and modding. Speedrunners discovered glitches and techniques that the developers never anticipated, completing levels in times that seemed physically impossible. Modders recreated GoldenEye's levels in other engines. An entire underground ecosystem existed around a game that its rights holders seemed content to let disappear. ## The Return: January 27, 2023 On January 27, 2023, after years of negotiations that probably required more lawyers than the original game had developers, GoldenEye 007 finally came back. Remasters launched on Nintendo Switch through Nintendo Switch Online and on Xbox through Game Pass. The licensing nightmare had somehow been untangled. The game was playable again, legally, officially, on modern hardware. The Switch version offered online multiplayer, which meant you could play with people anywhere in the world. The Xbox version featured widescreen support and up to 4K resolution on supported hardware with local split-screen. All the characters were there. All the maps. All the weapons and modes. The game that had haunted gaming culture for two decades was suddenly accessible again. And here's the bittersweet truth about the resurrection: while everyone celebrated that the game was back, everyone also understood that the world had changed. The experience that made GoldenEye special, four friends in one room, one television, split-screen chaos, was inseparable from the era that produced it. You could play GoldenEye online with strangers in 2023, and it was fun. But it wasn't the same. The magic of GoldenEye wasn't just in the code. It was in the context: the physical togetherness, the limited hardware, the shared screen, the elbowing and the trash talk and the pizza getting cold because nobody wanted to pause. Still, the remaster proved something important. The game design held up. The levels were still brilliant. The mechanics were still satisfying. GoldenEye's achievement wasn't dependent on nostalgia. Play it for the first time in 2023 or 2024, with no childhood memories attached, and you'll still find a remarkably well-designed shooter with levels that reward exploration and mastery. The graphics are charming rather than impressive. The AI is simple by modern standards. But the fundamental design is timeless. ## What GoldenEye 007 Really Meant Understanding what happened to GoldenEye means understanding what it actually was beneath the surface. On one level, it was a movie tie-in game that arrived two years late. On another level, it was the most important FPS ever made for consoles. On yet another level, it was a cultural event: a perfect collision of talent, timing, hardware capability, and creative freedom that produced something nobody could have predicted or planned. A team of ten people, mostly first-timers, working in a rural English studio with no budget for fancy tools, building an engine from graphics textbooks and sheer determination, created something that an entire generation grew up with. They proved that FPS games could work on consoles. They proved that licensed games didn't have to be disposable. They proved that four people in a living room could create memories that would outlast the hardware, the publisher, and even the studio itself. The game sold 8 million copies on a budget of roughly $2 million. That's one of the most remarkable return-on-investment stories in entertainment history. And the cultural impact extends far beyond what sales figures can capture. Every console FPS that came after GoldenEye was building on the foundation Rare laid. The control schemes. The mission design. The understanding that console players wanted accessible, immediate, social experiences. All of it traces back to a tiny team in Twycross who didn't know they were supposed to fail. What happened to GoldenEye 007 is ultimately a story about the tension between creative achievement and corporate control. The people who made it created something extraordinary. The corporations who owned the various rights to it made that creation inaccessible for nearly twenty years. But the game survived anyway, in the memories of millions of people who experienced something genuine and irreplaceable when they played it. No licensing agreement could ever erase that. ## Frequently Asked Questions Who developed GoldenEye 007? GoldenEye 007 was developed by Rare, a British studio based in Twycross, Leicestershire. The game was directed by Martin Hollis. The development team consisted of about ten people, and for all but two of them, it was their first video game. Development took approximately two and a half years. Was the multiplayer mode planned from the start? No. Programmer Steve Ellis created the multiplayer mode during the final weeks of development, without permission from Rare's management or Nintendo. He built it in roughly six weeks as an unauthorized side project. The mode became the game's most celebrated feature and one of the defining multiplayer experiences in gaming history. How many copies did GoldenEye 007 sell? GoldenEye 007 sold approximately 8 million copies worldwide, making it the third best-selling game on the Nintendo 64. The game was developed on a budget of roughly $2 million. Why was GoldenEye unavailable for so long? The game was based on the James Bond franchise, owned by MGM and Eon Productions. Nintendo published the original. Microsoft acquired Rare in 2002. These overlapping rights made it impossible for any single party to re-release the game without agreement from all the others. A completed Xbox 360 remaster was finished but never released because the licensing parties couldn't reach terms. The situation wasn't resolved until 2023. When did the GoldenEye remaster come out? GoldenEye 007 was re-released on January 27, 2023, on Nintendo Switch through Nintendo Switch Online and on Xbox through Game Pass. The Switch version included online multiplayer. The Xbox version featured enhanced resolution and local split-screen support. What did Shigeru Miyamoto think of GoldenEye? Miyamoto sent a fax to the development team expressing concern that there was too much close-up killing in the game. Rather than reducing the violence, the team created the iconic curtain call credits sequence at the end of the single-player campaign, where all characters take a bow as if they were actors in a performance. This touch became one of the most beloved features in the game. How did GoldenEye influence modern FPS games? GoldenEye established the template for console FPS design. Its analog stick control scheme became the industry standard. Its objective-based mission structure influenced countless subsequent shooters. Its demonstration that FPS games could work brilliantly on consoles paved the way for Halo, Call of Duty, and every major console shooter that followed. What was Perfect Dark? Perfect Dark was Rare's spiritual successor to GoldenEye, released in 2000 for the N64. It expanded on GoldenEye's design with more complex missions, additional weapons, and AI-controlled bots called Simulants in the multiplayer mode. While critically acclaimed, it arrived late in the N64's lifecycle and never achieved the same cultural penetration as GoldenEye. --- # What Happened to IRC, the Chat Protocol That Built the Internet URL: https://404memoryfound.com/posts/what-happened-to-irc-chat.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-09 Topics: Internet Culture, Software & Apps Here's something that will date you immediately: in the early 1990s, if you wanted to talk to strangers on the internet in real time, there was basically one option. Not a website. Not an app. A protocol called IRC, which stood for Internet Relay Chat. No graphics. No usernames with verified checkmarks. No algorithm deciding what you should see. Just text, typed into a client program, relayed through servers that anyone could set up. By the year 2000, over a million people were connected to IRC simultaneously. Think about that number for a moment. The web itself wasn't even a decade old yet. Smartphones didn't exist. The idea that you could just open a program and talk to people from Japan, Australia, Brazil, right now, in real time, felt like genuinely powerful magic. And it was built on a protocol so straightforward that you could understand how it worked if someone explained it to you for five minutes. ## The Finnish Accidental Breakthrough IRC was created in August 1988 by a 23-year-old computer science student named Jarkko Oikarinen at the University of Oulu in Finland. He wasn't trying to change the internet. He was trying to improve something much smaller. The university had a system called MUT, short for MultiUser Talk, which let people on the same BBS system chat with each other. But it only worked on one machine. Oikarinen thought, why not make it work across multiple machines on the network. He built the first IRC server on a machine called tolsun.oulu.fi. Released it. People started using it. And then something unexpected happened. By November 1988, just three months later, the protocol had spread across the entire internet. By mid-1989, there were roughly 40 IRC servers running worldwide. The thing had gone from university project to global network without anyone planning it that way. What made IRC spread so fast wasn't hype. It wasn't funding or venture capital. Those things didn't exist in that context. It spread because the protocol was simple enough that other people could implement it, and it solved a real problem that a lot of people had. Which is to say, it was the opposite of how most internet services work now. ## The Gulf War and the First Global Event For most of its early years, IRC was a curiosity. Text-only chat for technical people, mostly developers and academics. You needed to know how to use a Unix command line. You needed to know IRC commands. It wasn't exactly mass market. And then, on January 16, 1991, something happened that changed how the internet reported the news. The Gulf War began. And something extraordinary occurred on IRC. When coalition forces began bombing Iraq, the television networks in the United States cut away from their regular programming to cover the attacks. But in the first week of the war, Saudi Arabia shut down radio and television broadcasts. The country was under a news blackout. No one outside Saudi Arabia was getting information. Except on IRC. A user on IRC named Johan was in Kuwait. Or connected through Kuwait. The exact details are murky because this was 1991 and nobody was writing things down. But Johan was relaying live updates from what was happening on the ground. Real-time information. People who wanted to know what was actually going on weren't watching CNN or the BBC. They were on IRC, reading text being typed in by someone who was there. For the first time, IRC had more than a few hundred concurrent users. It hit over 300 simultaneous users, which at the time felt enormous. The protocol held up. The servers didn't crash. And suddenly, a lot of people understood that this wasn't just a chat system for nerds. This was something that could matter. ## The Network Wars The problem with IRC was that it was decentralized by design. Anyone could set up a server. Anyone could connect their server to the network. Which sounds great in theory, until you realize that this meant no one actually controlled anything, so problems escalated fast. By the early 1990s, the IRC network started fragmenting. Server operators would argue about how things should be run. Some wanted stricter controls. Others wanted to stay more open. In October 1992, someone took down a bunch of servers as part of some dispute nobody can quite explain anymore. When they came back online, the network had split. You had EFnet, which was the "original" network, and Undernet, which was the breakaway. Both claimed to be the real IRC. This happened again. In 1996, EFnet split into EFnet and IRCnet, with the European servers going one way and US servers going another. Different philosophies about server operators and how the network should be managed led to different networks running the same protocol. Then there was DALnet, which formed because people were sick of constant splits and lag and takeovers on the existing networks. The origin story is genuinely funny. A group of people were hanging out in a channel called #StarTrek and got tired of the drama. So they split off and created their own network. It grew to become one of the biggest. Understanding why these splits happened requires understanding what made running IRC servers hard. Servers could be resource-intensive. They could crash. They could get compromised. And because there was no central authority, if a server operator decided to do something controversial, the only recourse was to split off and create a new network. This happened so often that it became almost routine. A server goes down. There's a dispute about how the network should be run. Someone decides to fork the protocol and start their own thing. By 2002, you had probably a dozen major IRC networks. Each with different server configurations. Each with different bot capabilities. Each with different channel limits. Each with different moderation policies. Some networks allowed more automation. Some restricted bots heavily. Some had channels with thousands of users. Some networks were smaller and more curated. Pick your network based on what you wanted. It was fragmented but it worked. The problem was always the same though. Network splits would happen. You'd get logged in one day and suddenly you couldn't see half your friends. They were on a different split of the network. Or you could see them but messages weren't getting through. The technical infrastructure was fragile. Every server was a single point of failure. The network was resilient in theory, decentralized by design, but in practice it kept fragmenting. By the late 1990s, if you wanted to use IRC, you had to pick a network. EFnet. Undernet. IRCnet. DALnet. QuakeNet (which started as a network for Quake (https://404memoryfound.com/posts/what-happened-to-quake-online-fps.html) players but became enormous). Each had its own server list. Each had its own culture. Each had its own peak concurrent user count that they'd brag about. It was like parallel universes, all running the same protocol, all fundamentally unable to talk to each other. The weirdest part is that this was fine. People understood it. You picked a network based on where your friends were or what communities you wanted to join. Nobody complained that you couldn't message someone on EFnet from Undernet. That was just how it worked. This is essentially what TikTok's algorithm did twenty years later, except instead of splitting into separate servers, they just showed you different feeds. ## The Golden Age That Nobody Realized Was Golden Between 2000 and 2005, IRC hit its absolute peak. Peak concurrent users across all networks was over a million people at any given moment. QuakeNet alone had over 240,000 people online simultaneously. There were entire communities living on IRC. People would log in and stay logged in for hours. Channels had bots that would do things. IRC had become a platform. For gaming, IRC was everything. Clans organized matches over IRC. People ran tournaments through IRC channels. The esports infrastructure, before it was called esports, was built on IRC and people typing commands into bots. The mIRC client was genuinely elegant for its time. You'd launch it and you'd get a window with a list of networks. Click on a network. Click on a channel. Boom, you're there. You're talking to people. And because it was Windows shareware, it was accessible. You could run it on any Windows machine. You didn't need to know how to use the command line. You didn't need to be technical. You just needed to know how to click buttons. For people using Unix systems, the experience was different. You'd SSH into a server, usually one hosted by your ISP or your university, and you'd run an IRC client from the command line. BitchX was popular. So was ircII. These were much more technical to use but much more powerful once you understood them. You could configure everything. You could write scripts that would do things. You could automate your entire IRC experience if you knew what you were doing. And then there were the bots. Bots on IRC were legitimate pieces of infrastructure. A bot would manage channel modes, kick troublemakers, maintain channel topics, store information, run games, anything you could imagine. Bots like Eggdrop became legendary. People would spend hours configuring their bots, writing scripts, making them do useful things. It was like having a little robot hanging out in your channel, following whatever rules you told it to follow. For open source development, IRC was the water cooler. Linux kernel developers hung out in IRC channels. Free software communities coordinated on IRC. If you wanted to contribute to a major open source project in the 1990s, you needed to be on IRC. The dominant Windows client was mIRC, which was shareware. You could use it for 30 days free, and then it would nag you to pay something like 20 dollars. A lot of people paid. It became one of the most successful shareware programs ever. The author, Khaled Mardam-Bey, maintained it for decades and still maintains it now, which is genuinely remarkable for a piece of software from that era. On Unix systems, there were clients like ircII and BitchX and ircle on the Mac. The fragmentation was real, but people didn't really care. You got to know your client and you used it. ## The Decline: Two Turning Points IRC didn't die overnight. It had two major turning points that fractured it. The first was the rise of web-based chat. As broadband became standard and web technology got better, people realized you could just log into a website and chat. You didn't need to install a client. You didn't need to remember commands. You could do it on your phone browser, eventually. Slowly, communities started moving to web platforms. Then to proprietary platforms. Then to Discord. Discord launched in 2015. It was IRC with a better UI, better voice integration, and, most importantly, it was one unified network that you didn't have to configure. There were no server splits because Discord Inc. owned all the servers. There were no network fragments because everyone was on the same platform. It was the logical conclusion of what IRC always wanted to be but couldn't quite achieve because of its own architecture. The second turning point was more dramatic. In May 2021, Libera Chat, which had become the largest IRC network, underwent a hostile takeover. The company that owned the domain essentially seized control of the platform. The staff all resigned. The communities, which had been there for decades, had to leave. They migrated to a new network called Libera Chat that was set up by the people who had left. It was basically a refugee crisis in real time, and it exposed something that people had been able to ignore: IRC networks had concentrated power in whoever controlled the domain and the servers. You could own a piece of the network, but you didn't own your community. Between 2003 and 2012, IRC had lost about 60 percent of its users. By the 2020s, it had become something people used if they were in certain niche communities, certain open source projects, certain gaming clans. But the general public had moved on. The Rust programming language community, which had a significant presence on IRC, officially migrated to Discord in 2019. Mozilla deprecated IRC support the same year. ## What IRC Actually Was The thing about IRC is that it was never trying to be a consumer product. It was a protocol. A set of rules for how machines and clients should talk to each other. And because it was just a protocol, anyone could build anything on top of it. You could build clients with whatever UI you wanted. You could build bots that did specific tasks. You could build networks with specific rules and cultures. It was extensible and hackable in a way that closed platforms never could be. The problem with that approach is that it means you're constantly dealing with fragmentation. Different networks. Different clients. Different conventions for the same basic thing. From a platform owner's perspective, this is chaos. From a user's perspective, it's freedom. And the internet chose differently. The internet chose platforms. Discord is essentially modern IRC, but owned by a company. Slack is modern IRC for businesses. Both of them copied IRC's core idea: channels, direct messages, text-based communication, bots that can do things. But they wrapped it in a unified platform that someone controls. And for the general public, that was better. Cleaner. Simpler. Less fragmented. But something was lost. IRC was genuinely weird. There were real communities there. You could create a channel and there was no algorithm deciding who saw it. If you wanted to run a network, you could. If you wanted to fork the protocol and do something different, you could. It was built on the assumption that the internet was something people would tinker with and modify and improve. Not something they would consume. That assumption turned out to be too optimistic. ## The Freenode Disaster and What It Meant When Freenode, the other massive IRC network, went through the hostile takeover in May 2021, it was genuinely shocking. The network had over 200,000 concurrent users. Decades of accumulated communities. The open source infrastructure of the internet essentially running on it. And then it was gone because someone owned the domain and decided to take it back. The staff migrated everyone to Libera Chat. But not everyone moved. Some people were on other networks. Some people had drifted to Discord already. Some people just gave up on IRC entirely. The network never recovered to its previous size. What happened to Freenode and Libera is the exact opposite of how IRC was supposed to work. You weren't supposed to lose everything because one person decided to take over a domain. The whole point of a decentralized network was that it was supposed to be resilient. But in practice, decentralization meant that anyone could disrupt everything by controlling one key piece. ## Where IRC Is Now IRC still exists. You can go find communities on various networks. The old networks like EFnet and Undernet still have active users. Libera Chat rebuilt and has communities. There are smaller networks with specific communities. Open source projects that care about decentralization still use IRC. The Gentoo Linux community basically lives on IRC. But these are niche communities now. The casual user, the person who wants to chat with friends online, is on Discord or Slack or WhatsApp or TikTok or any of a hundred other platforms. They're not thinking about IRC. They might not even know it exists. And that's probably fine. IRC was a tool for a specific era of the internet. An era when the internet was something you built, not something you consumed. When protocols mattered more than platforms. When you expected to tinker with things. That era is over. But if you use Discord, you're using something that is philosophically descended from IRC. If you use Slack, you're using modern IRC for work. The DNA is there. The basic idea, that people should be able to talk to each other in real time in organized channels, was so good that it survived the transition to closed, proprietary platforms. IRC didn't die so much as it evolved into something that didn't need to fragment because someone owned the whole thing. Which, if you think about it, is exactly what the internet was supposed to prevent. But here we are. ## FAQ What does IRC stand for? Internet Relay Chat. It's a protocol, not a specific service, which is why it's confusing to people used to thinking about internet communication through apps. Can I still use IRC? Yes. It's still operational. You can find various IRC networks and connect using a client like HexChat or Irssi. It's just not where the general population is anymore. Is IRC secure? Not by modern standards. Messages were transmitted in plain text and could be read by anyone on the network. Modern chat applications use encryption, which IRC does not inherently support, though some servers and clients have added TLS support. Why did Discord win instead of IRC continuing to evolve? Discord offered a unified platform, better UI, integrated voice and video, and didn't require technical knowledge to use. IRC required you to choose a network, install a client, and understand how everything connected. For most people, that friction was enough to switch. Did Discord copy IRC? Not exactly. Discord was designed as a modern alternative with a very similar feature set (channels, direct messages, communities). The philosophy is different though. Discord is centralized and owned by a company. IRC was decentralized by design. What was the most popular IRC network? QuakeNet was the largest by concurrent users at its peak, with over 240,000 simultaneous users. But Freenode was arguably more important to internet infrastructure because of its role in open source communities. Is IRC dead? Not dead, but definitely a niche product now. It peaked between 2000 and 2005. It still has dedicated users, but it's not where new people are learning to communicate online. --- # Is There a New Tony Hawk Game? The Series in 2026 URL: https://404memoryfound.com/posts/what-happened-to-tony-hawks-pro-skater.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-09 Updated: 2026-09-10 Topics: Gaming, Business Blunders **Summary:** There is no new Tony Hawk game announced as of September 2026. The most recent release is Tony Hawk's Pro Skater 3 + 4, a remake built by Iron Galaxy that shipped on July 11, 2025 to positive reviews. Activision still owns the series, and Microsoft has owned Activision since its $68.7 billion purchase cleared in October 2023. **Key facts:** - Launched: 1999, made by Neversoft for Activision - Franchise revenue: $1.2 billion through September 30, 2007 (Activision filing); more than $1.4 billion in sales by 2025 (CNBC) - Owner today: Activision, owned by Microsoft since October 2023 - Latest game: Tony Hawk's Pro Skater 3 + 4, July 11, 2025, by Iron Galaxy - Status in 2026: No new game announced; Iron Galaxy cut up to 90 jobs in April 2026 ## Who actually made the 1999 original? Neversoft built it. The California studio had never shipped a sports game when Activision handed it a skateboarding license, and the result reached American stores in 1999 on Sony's PlayStation. Activision owned the studio outright, and years later still introduced it in earnings releases as "the creators behind the Tony Hawk franchise." That one line carried a lot of weight for a company that at the time was better known for licensed movie tie-ins. Ports followed to the Nintendo 64, the Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html) Color and Sega's Dreamcast (https://404memoryfound.com/posts/why-sega-stopped-making-consoles-dreamcast.html), which was in the middle of its own short life. The design was simple enough to explain in one sentence: a two minute clock, a list of goals, and a trick system loose enough that a first timer could chain a combo by accident. The soundtrack, heavy on punk and hip hop, did the rest. ## How much did Tony Hawk earn from the Activision deal? Before the first game shipped, Activision offered Hawk a one time buyout of his future royalties worth $500,000. He turned it down and took a cut of every copy instead. Speaking at a technology conference in Stockholm in 2025, in remarks reported by CNBC, he was blunt about it: "That was absolutely the best financial decision of my life." The licensing agreement ran 16 years, from 1999 to 2015, and CNBC reported in 2025 that the series has generated more than $1.4 billion in sales. Activision's own filings back up the scale. In its quarterly report for the period ended September 30, 2007, the company told investors it had released eight titles in the Tony Hawk franchise with cumulative net revenues of $1.2 billion, and that its agreement granted it exclusive rights to publish games using Hawk's name and likeness through 2015. ## When did the series actually start to slide? Not immediately after Underground. Activision kept the franchise near the top of the charts for years. That same September 2007 filing noted that Tony Hawk's Project 8 was one of only three launch titles Activision had ready for the PlayStation 3, that Tony Hawk's Downhill Jam was one of five for the Wii, and that for the eighth straight year the franchise put a game in the US top 10 for December. The problem was the pace. A new Tony Hawk game shipped almost every holiday, and Game Developer counted the franchise at "around 20 games in as many years" by 2020. Underground and Project 8 pushed the formula toward open worlds and story modes. Later entries pushed it toward a plastic skateboard peripheral you stood on in your living room. By the time the PlayStation 2 (https://404memoryfound.com/posts/playstation-2-best-selling-console-ever.html) generation ended, the audience that had bought a Tony Hawk game every year had stopped. Neversoft moved onto Guitar Hero, then onto Call of Duty support work, and Activision folded the studio into Infinity Ward in 2014. ## How bad was Tony Hawk's Pro Skater 5 in 2015? Bad enough to end the run. Tony Hawk's Pro Skater 5 arrived in 2015, the final year of Activision's license, and OpenCritic recorded a score of 32 across 43 critics with zero percent of them recommending it. The reviews were not mixed. Eurogamer's 2015 verdict, quoted on OpenCritic, read: "Glitch-ridden and seemingly unfinished, this is a tragic swansong for Tony Hawk's video game career." IGN's line, also from 2015, was that "rare moments of nostalgic joy are drowned out by its abundance of faults." Giant Bomb's summary ran four words: "Don't play this game." Much of the game's content shipped in a day one patch larger than the disc content itself, which is how a skateboarding game about precision ended up remembered for skaters falling through the floor. After that, the license lapsed and the series went quiet for five years. ## Did the 1 + 2 and 3 + 4 remakes actually work? Yes, both times. Tony Hawk's Pro Skater 1 + 2, rebuilt by Vicarious Visions, arrived in September 2020 and passed one million copies sold through in under two weeks. Activision said at the time that it "is now the fastest selling game in the storied franchise," a record set against roughly two decades of releases. Vicarious Visions did not get to make another one. NME reported that Activision moved the Albany, New York studio under Blizzard in January 2021 and formally merged it in April 2022, with the team "fully dedicated to Blizzard games." The follow up went to a different developer. Game Informer (https://404memoryfound.com/posts/who-owns-game-informer-now.html) reported in May 2025 that Iron Galaxy would ship Tony Hawk's Pro Skater 3 + 4 on July 11, 2025, and it landed to positive reviews. Design manager Mike Rossi told the outlet his brief to the level team was blunt: "The big thing I told the level design team was that this needs to feel like it was left on the cutting-room floor." The main complaint from critics was structural, since Pro Skater 4's open ended career was replaced with the timed run format of the earlier games. ## Who owns the series now, and is anything in development? Activision owns Tony Hawk's Pro Skater, and Microsoft owns Activision. TechCrunch reported that the $68.7 billion purchase cleared its last regulatory hurdle in October 2023, which puts the skateboarding series in the same portfolio as Call of Duty and the Xbox hardware business (https://404memoryfound.com/posts/what-happened-to-original-xbox-microsoft-gaming.html). As of September 2026, no new Tony Hawk game has been announced. The signal pointing the other way is Iron Galaxy. Kotaku reported in April 2026 that the studio had laid off as many as 90 people, by one count it cited, its second round after 66 cuts the year before, with the studio saying it was "impossible for us to sustain the team size that we've carried this past year." Hawk himself has been careful about promising anything, though he is clear about what the games did. "There is a whole generation of people who skate now who have told me specifically that they started skating because they played the game," he told NPR's Here and Now in 2025. Where to find one today: the original PlayStation and PlayStation 2 discs never got a digital re-release, so secondhand copies are the only way to play the versions people remember. Sellers list boxed and loose discs on Etsy's Tony Hawk's Pro Skater listings (https://www.etsy.com/search?q=tony+hawk+pro+skater+playstation). The 2020 and 2025 remakes are the only versions sold new. ## Frequently Asked Questions ### Is there a new Tony Hawk game coming in 2026? No new Tony Hawk game has been announced as of September 2026. The most recent release is Tony Hawk's Pro Skater 3 + 4, which Iron Galaxy shipped in July 2025. Kotaku reported in April 2026 that Iron Galaxy laid off as many as 90 staff, its second round of cuts in two years. ### What happened to Neversoft, the studio behind Tony Hawk's Pro Skater? Neversoft created Tony Hawk's Pro Skater for Activision in 1999 and made the series its own for a decade before moving to Guitar Hero and then Call of Duty support work. Activision folded Neversoft into Infinity Ward in 2014, and the name stopped being used. The 2020 and 2025 Tony Hawk remakes were made by Vicarious Visions and Iron Galaxy instead. ### How much money has the Tony Hawk game series made? Activision told investors that through September 30, 2007 it had released eight Tony Hawk titles with cumulative net revenues of $1.2 billion. CNBC reported in 2025 that the series has generated more than $1.4 billion in sales across its full run since 1999. Tony Hawk turned down a $500,000 buyout of his royalties before the first game shipped and took a per copy cut instead. **Sources:** - Activision, Form 10-Q for the quarter ended September 30, 2007 (SEC EDGAR): https://www.sec.gov/Archives/edgar/data/718877/000110465907080748/a07-26061_110q.htm - NBC 7 San Diego (CNBC), Tony Hawk on turning down $500,000 for Pro Skater, 2025: https://www.nbcsandiego.com/news/business/money-report/tony-hawk-on-turning-down-500000-for-pro-skater-the-best-financial-decision-of-my-life/3760064/ - Game Developer, Tony Hawk's Pro Skater 1+2 sells 1 million copies in less than two weeks, 2020: https://www.gamedeveloper.com/game-platforms/-i-tony-hawk-s-pro-skater-1-2-i-sells-1-million-copies-in-less-than-two-weeks - OpenCritic, Tony Hawk's Pro Skater 5 review scores: https://opencritic.com/game/1545/tony-hawks-pro-skater-5 - Kotaku, Iron Galaxy Slashes More Developers In An Attempt To Survive, 2026: https://kotaku.com/iron-galaxy-layoffs-tony-hawk-last-us-part-2-remaster-2000688573 --- # What Happened to the Sega Game Gear, the Handheld That Deserved Better URL: https://404memoryfound.com/posts/what-happened-to-sega-game-gear-handheld.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-09 Topics: Gaming, Hardware, Business Blunders ## Picture This: 1990, and Sega Was About to Change Handheld Gaming Forever Picture this: 1990. You're walking through a video game store, the smell of arcade cabinets and fresh plastic hanging in the air. Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html) is at the absolute peak of their confidence. The Genesis has absolutely demolished the NES. They own the console space. They're making arcade games that feel like arcade games. They're winning. And now they walk into the handheld market with something that looks genuinely like the future: a full color, backlit handheld that makes the Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html) look like it belongs in 1989, which it kind of does. On October 6, 1990, Sega released the Game Gear in Japan. The specs were wild for the time. A 3.2-inch backlit color screen with 4,096 colors. A Z80 CPU running at 3.58 MHz. The ability to play Master System games with an adapter. This wasn't just a handheld gaming device. This was a statement. This was Sega saying, "We're not just going to compete with Nintendo's Game Boy. We're going to embarrass it." The response in Japan was immediate and overwhelming. Sega sold 40,000 units in the first two days. 90,000 in the first month. By the time the Game Gear hit America in April 1991 at $149.99, there were over 600,000 back orders. The momentum was real. The demand was real. And somehow, despite having objectively superior hardware, better games were coming, and genuine consumer enthusiasm, the Sega Game Gear became one of the most famous failures in handheld gaming history. This is the story of how Sega won the technology war and lost the actual war, spectacularly. ## The Hardware Wasn't the Problem. The Batteries Absolutely Were. Here's where it gets interesting. The Game Gear had one fatal flaw, and it was so obvious that it somehow took everyone by surprise anyway. That beautiful backlit color screen? It was a power hog. A magnificent, electricity-devouring power hog. The Game Gear ran on six AA batteries. Six. And it would eat through them in about three to five hours of continuous play. Think about that for a second. Three to five hours. You go on a family road trip, and your Game Gear is dead before you hit the state line. Meanwhile, Nintendo's Game Boy ran on four AA batteries and got 30 hours or more. Thirty hours. Not three. Thirty. The contrast was so stark that it wasn't even funny. It was the kind of thing that made parents laugh when they saw the battery requirement on the Game Gear box. "Wait, this needs six batteries?" And that laugh, that simple laugh of disbelief, was the sound of a console's future dying. The Game Gear did have some solutions. There was an optional rechargeable battery pack. There was an external power supply. There was a car adapter. But none of these solved the fundamental problem: they were all additional purchases. They were all clunky. And they all screamed the same message: the Game Gear, this marvel of technology, this portable color gaming machine, was kind of a nightmare to actually use. When your handheld gaming device requires more accessories than a camping trip, something has gone wrong. Sega knew this was a problem. They knew it immediately. But by then, momentum is a powerful thing. The Game Gear had launched. The expectation had been set. The Game Boy, by contrast, had already spent years in the market. It had millions of users. It had a library of games that was almost comically deep. And it just worked. You threw four batteries in, and you played for a month. That simplicity, that reliability, that "just works" quality, ended up mattering more than all of Sega's technical superiority. ## The Game Boy's Dominance Wasn't Magic. It Was Boring Excellence. Here's the thing about the Game Boy that people sometimes get wrong. It wasn't a technical marvel. Nintendo engineers had to actively make choices to keep it from being more advanced, because the system was deliberately designed around two principles: battery life and durability. The monochrome green screen that everyone made fun of? That was a choice. That was Nintendo saying, "We could make this fancier, but we're making it more practical instead." The Game Boy had a simple processor. A simple screen. Simple sound. It was boring. It looked outdated the moment the Game Gear came out. But boring works. Boring survives. Boring lets a parent take their kid on a six-hour road trip and not hear complaints about dead batteries. By March 1996, when Sega discontinued the Game Gear after about six years of production, it had sold 10.62 million units globally. Impressive numbers on the surface. Except the Game Boy had sold 118.69 million units. The Game Boy had lapped the Game Gear twice. This wasn't because Game Boy owners were more loyal. It wasn't because Nintendo made better games, though they absolutely did, at least for a significant chunk of the Game Gear's life. This happened because when you're asking someone to carry something in their pocket or backpack, practicality matters more than specs. A device that actually works beats a device that's theoretically superior but needs a truck full of batteries. And here's where the business logic starts to break down a little. Nintendo didn't win because they were smarter about hardware design. Well, they did, but that wasn't the main reason. Nintendo won because they understood a fundamental truth about consumer electronics: the best device is the one people will actually use. And nobody wants to use a device that dies after three hours. ## The Master System Adapter Was a Good Idea That Proved Nothing One of the more interesting things Sega did with the Game Gear was release an adapter that let you play Master System cartridges on the handheld. This is actually kind of brilliant from a business perspective. You already own Master System games? Use them on the Game Gear. You already invested in our library. Now it's portable. Come with us to handheld gaming without losing your existing library. But here's the thing that people don't talk about enough: this strategy works fine if your library is robust. The Master System had a decent game library, but it was nothing compared to the NES and certainly nothing compared to what the Game Boy had access to through third-party publishers. So the adapter was an interesting technology solution to a business problem it didn't actually solve. More people owned Game Boys. More people wanted to play Game Boy games. Sega's strategy made logical sense on a spreadsheet, and it failed in the real world. And which is kind of insane, because Sega had other advantages. The Game Gear's processing power meant games could do more impressive things. The color screen meant certain genres could shine. But Sega was spread so thin across hardware platforms that they couldn't capitalize on any of it effectively. ## But Here's the Real Problem: Sega Was Trying to Win Six Different Wars at Once This is where the narrative gets genuinely tragic. The Game Gear wasn't killed by the Game Boy. It was killed by Sega's own chaos. In the early 1990s, Sega was supporting the Genesis, the Game Gear, the Sega CD (https://404memoryfound.com/posts/what-happened-to-sega-cd-add-on.html) add-on, the 32X (https://404memoryfound.com/posts/what-happened-to-sega-32x-add-on.html) add-on, arcade cabinets, and Saturn development. That's not a product strategy. That's a company eating itself. Every resource committed to Game Gear development was a resource not committed to making sure the Game Gear had the games it needed to compete. Nintendo, by contrast, was completely focused. One console. One handheld. Tons of resources flowing to both. Sega's structure meant they were fighting a multi-front war with divided attention, divided budgets, and divided talent. The Game Boy didn't win because it was better at being a Game Boy. It won because Nintendo could commit fully to making it indispensable. Sega's inability to maintain focus destroyed what could have been a dominant position. The Game Gear launched strong. It had momentum. It had a better screen. It had better hardware. And then Sega spread resources so thin across so many platforms that the Game Gear's potential just evaporated. By 1997, when Sega discontinued it, the Genesis and Saturn were where the real battles were happening. The Game Gear had become a side project for a company that couldn't afford side projects. ## The Budget Relaunch Almost Saved It, and Then Didn't In 2000, something interesting happened. Majesco, the video game publisher, licensed the Game Gear and re-released it at $30. Thirty dollars. The original launch price was $149.99. For a $30 handheld with $15 games, suddenly the battery issue seemed less critical. Parents were willing to accept the tradeoffs. The price made the hassles forgivable. And honestly, for what it was, the budget Game Gear was a decent product. But it was ten years too late. The Game Boy had spent that entire decade establishing market dominance. Game Boy Pocket had come out. Game Boy Color had come out. Tamagotchi (https://404memoryfound.com/posts/tamagotchi-digital-pet-that-made-us-feel.html) was eating the casual market. The Pokemon phenomena had made Game Boy synonymous with gaming itself for an entire generation. A $30 Game Gear was interesting as a curiosity. As a market competitor, it was an afterthought. Majesco tried. They gave it a legitimate effort. But there's no amount of price cutting that overcomes ten years of being the second player. The Game Gear's chance had been 1991. Not 2000. ## What the Game Gear Taught Us About Hardware, Software, and Market Reality The Game Gear is one of the most instructive failures in gaming history because it's not a failure of execution or ideas. The Game Gear failed because of market forces that pure technical superiority couldn't overcome. And that's a lesson that echoes through gaming, consumer electronics, and technology in general. You can have the better hardware. You can have the newer technology. You can have the more impressive specs. But if the fundamental user experience is inconvenient, if the software library doesn't match the competition, if your company is too chaotic to properly support the product, and if your competitor has already locked in the market position, better hardware doesn't matter. At all. The Game Gear was a device ahead of its time in some ways and fundamentally impractical in others. It was a device that demanded infrastructure, accessories, and patience from users who were perfectly happy with a simpler solution. It was a device that represented Sega's technical confidence and strategic confusion all at once. And it was ultimately a device that lost not because it was bad, but because nothing it could offer was worth dealing with dead batteries and limited software. The Atari Lynx (https://404memoryfound.com/posts/what-happened-to-atari-lynx-handheld.html) faced similar problems and did even worse, selling only around 500,000 units total. The Lynx was technically superior to both the Game Boy and the Game Gear, and nobody cared. It had the same battery problems plus even fewer games. So the Game Gear's 10.62 million units start looking almost respectable when you realize Sega was competing not just against the Game Boy but against user expectations shaped by decades of Nintendo handheld dominance. In 2020, Sega released the Game Gear Micro for their 60th anniversary. It's a novelty device with 50 games built in, tiny screen, no cartridge slot, and it costs more than the original Game Gear did at launch. It's a museum piece for collectors, a way of saying, "Remember this thing?" And yeah, people do remember it. They remember it as the handheld that was too ambitious, too power-hungry, and ultimately too late. But they remember it fondly, because hardware enthusiasm is real even when market success isn't. The Game Gear was a remarkable device. It just existed at a moment when remarkable wasn't enough. ## The Games Library Problem: Superior Hardware Needs Superior Games Here's something that gets overlooked in the Game Gear versus Game Boy narrative: hardware is only half the battle. The other half is software. And the Game Gear had a legitimate games library, but it never matched the Game Boy's depth or quality. This is partly a chicken-and-egg problem. Game developers want to develop for platforms where people will buy games. Game Boy had way more users, so developers prioritized Game Boy games. This meant fewer Game Gear exclusives. This meant fewer reasons for Game Boy users to switch. The Game Gear got some genuinely excellent games. Sonic Chaos. Gunstar Heroes. The Shinobi series. Dragon Kingdom. Some of these games were better than their Game Boy counterparts. But there were fewer of them. And in a market where Network effects matter, fewer games means fewer reasons to buy the console, which means fewer users, which means fewer reasons for developers to make games. This is a strategy problem that no amount of hardware superiority can solve. You can have the better system, but if there are fewer games, you lose. Sega's strategy of trying to use Master System compatibility as a workaround was theoretically sound but practically limited. Not everyone owned a Master System, and Master System games often didn't translate well to the smaller screen anyway. By the mid-1990s, the Game Boy had such a deep library that any new user saw value immediately. The Game Gear had to prove its worth with fewer options. Marketing can only take you so far. At some point, the customer looks at the game library and makes a rational choice based on what's available. The Game Boy had more options. Game Boy won this battle decisively. ## The Broader Lesson: Market Position Beats Technical Superiority Looking back at the Game Gear story, the real tragedy is that it was a great device that arrived at exactly the wrong moment. Not too early, not too late, but in a market where first-mover advantage was already insurmountable. Nintendo's Game Boy didn't get there first because it was the most advanced handheld. It got there first because Nintendo had been making handhelds since the 1980s with the Game and Watch series. The Game Boy launched into an existing user base of Nintendo handheld fans. The Game Gear represented a technical revolution. But revolutions in consumer electronics don't usually win. Incremental improvements backed by existing market position win. The Game Gear tried to leap-frog the Game Boy's position with better hardware and lost the chess match to simple market momentum. This is true across technology. The technically superior product often loses to the entrenched leader. Not because consumers are stupid. Because switching costs are high, because network effects matter, because inertia is real, and because "good enough and already here" beats "better and asking me to change" surprisingly often. Sega learned this the hard way, multiple times, in multiple markets. ## Accessories as a Warning Sign: When Your Product Needs Too Much Stuff One of the clearest signs that the Game Gear was fighting an uphill battle was the accessory ecosystem. Gaming devices don't need a ton of accessories to be successful. The Game Boy worked fine with just batteries and a game. The Game Gear came with a recommendation for a rechargeable battery pack, a car charger, a magnifying screen lens, a carrying case, and a light adapter. Some of this was optional. Some of it was almost mandatory if you actually wanted to use the device seriously. This creates a perception problem. When customers see a console that requires that many accessories, they start questioning whether the company itself believes in the product. It sends a signal that the device as shipped isn't quite sufficient. You're not buying a Game Gear. You're buying the foundation of a Game Gear and then spending another $100 in accessories to make it actually usable. The Game Boy, by contrast, required nothing. Just batteries, which you could get anywhere, and a game. That simplicity communicated confidence. The Game Gear's accessory burden communicated desperation. This is a principle that applies broadly in consumer electronics. If your product works fine by itself, people will buy it. If your product requires accessories to be good, people get suspicious. They wonder why you didn't just make the product better to begin with. The accessory problem for the Game Gear was partly a technical issue, but it was also a marketing and positioning problem. Every accessory sold was an admission that the base product had limitations. ## The Collector's Perspective: Why Game Gear Hardware Still Matters Today Interestingly, the Game Gear has maintained a passionate collector and enthusiast community decades after its discontinuation. Modern retro gaming collectors actively seek out Game Gear hardware, especially certain versions. The appeal is straightforward: it's a technically impressive device from an important era of gaming history. The full color screen still looks better than the Game Boy's monochrome display, even though the Game Boy won the market war. This split between market success and technical achievement is important to understand. The Game Gear lost the commercial battle, but it won the technical battle decisively. That technical achievement hasn't gone away. The Game Gear Micro released in 2020 sold out repeatedly in Japan. The 50 built-in games, the recreated hardware design, the acknowledgment that the Game Gear mattered: these created real demand, decades later, from people who appreciate what Sega achieved technically even if they understand why it failed commercially. This is one of the few consolations for Sega in the Game Gear story. You can lose the market and still win in history. The Game Boy won more decisively, but the Game Gear didn't disappear. It's remembered fondly by people who actually used it. It's respected by people who understand handheld gaming history. And it's preserved in the way that truly important technological achievements are preserved: with reverence even if not with market dominance. ## The Parallel Universe Where Sega Wins: What Would Have Changed Imagine a scenario where Sega had solved the battery problem. Not with accessories. Actually solved it in the hardware. Imagine a Game Gear that ran 20+ hours on six AA batteries instead of 3-5 hours. Or imagine a Game Gear that used four AA batteries instead of six. What would have changed? Honestly, it might not have mattered. The Game Boy still had the installed base. The Game Boy still had the game library. But it would have removed the most compelling complaint. It would have made the battery argument disappear, and the Game Gear would have been left with all of its advantages. The better screen. The better processor. Master System compatibility. Sega's arcade libraries ported over. None of these were enough to overcome the battery problem and the library disparity. But maybe they would have been if Sega had just solved the battery problem. This is speculative, but it's instructive. Sometimes in business, there's one thing that's so wrong that it overshadows everything that's right. Fix that one thing, and the whole dynamic might have shifted. The Game Gear's battery life was that one thing. And Sega never quite fixed it convincingly. Even with rechargeable options and car chargers, they never delivered a game that ran as long as the Game Boy on comparable batteries. That single failure to solve a core problem meant that no amount of other advantages could overcome it. ## FAQ: Questions About the Sega Game Gear Q: Was the Game Gear actually better than the Game Boy? A: Technically, yes. The color screen was superior. The processing power was stronger. The sound was better. But consumer electronics aren't evaluated solely on tech specs. They're evaluated on practicality, game library, price, and overall user experience. The Game Gear lost decisively on practicality due to battery life and on game library due to Nintendo's dominance and Sega's distracted support. Q: Could Sega have won the handheld market if they'd focused? A: Possibly. If Sega had committed resources fully to Game Gear development, if they'd shipped it with a better battery solution or made it less power-hungry, and if they'd maintained development support instead of spreading themselves across the Genesis, Sega CD, 32X, and Saturn, the Game Gear might have been genuinely competitive. But this would have required a level of strategic focus Sega didn't exhibit at the time. Q: Why did the monochrome Game Boy beat color handhelds? A: Because battery life matters more than visual fidelity when you're carrying a device in your pocket. A Game Boy that runs for 30 hours is infinitely more useful than a Game Gear that runs for 3-5 hours, no matter how much better the screen looks. Nintendo bet on practicality. Sega bet on specs. Practicality won. Q: What happened to Game Gear developers after 1997? A: Many moved to Saturn development. Some left Sega. The Game Gear's discontinuation meant developers had to migrate to whichever Sega platform was still current, which at that point was the Saturn, already struggling against the PlayStation. It was a cascade of disruption that illustrated just how damaging Sega's hardware diversity strategy was. Q: Is the Game Gear Micro worth buying? A: If you're a gaming collector or Sega enthusiast, probably yes. It's a novelty product that serves as both a piece of gaming history and a compact entertainment device. But it's not a practical handheld and isn't meant to be. It's meant to be a commemorative piece from Sega acknowledging their handheld heritage. --- # What Happened to Eudora, the Email Client That Taught the Internet to Write URL: https://404memoryfound.com/posts/what-happened-to-eudora-email-client.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-04-09 Topics: Software & Apps, Internet Culture, Then vs Now ## In February 1996, Eudora Owned Email. By 2006, It Was Dead. Here's a number that feels impossible in retrospect: in February 1996, Eudora had 10 million users. By December 1996, that number had grown to 18 million. In 1995, Qualcomm, the company that owned Eudora, claimed 64.7 percent of all email software revenues. The year before, 1994, the company had generated $4.2 million in revenue. By 1996, that number had jumped to $33.7 million. Eudora wasn't just popular. Eudora was the category. This was the internet's golden moment for a specific kind of software. Not web-based. Desktop-based. A program you installed on your computer. A tool that sat on your machine and handled one job: email. And Eudora didn't just do email. It taught people how to use email properly. It had features that seemed obvious once Eudora did them, and revolutionary before that. It had a search interface that actually worked. It had organizational tools that made sense. It had filters that could process mail automatically. It had a way of displaying threads that showed you what you were replying to. And then, slowly, it disappeared. The last Qualcomm version shipped on October 11, 2006. Version 7.1.0.9 for Windows. Version 6.2.4 for Mac. That was it. No new versions. No updates. Just discontinuation. Today, Eudora is remembered primarily by people who used it, which is to say it's almost entirely forgotten. This is the story of how the internet's first truly modern email client got replaced by webmail, and why that replacement feels inevitable in retrospect but was basically nobody's plan at the time. ## Before Eudora, Email Clients Were a Disaster You have to understand the baseline here. Before Eudora, email on personal computers was pretty rough. Email wasn't unified. You might check email from work on a different system than you checked email at home. There was no standardization. There were a bunch of competing protocols and terrible user interfaces. The experience of checking email was genuinely unpleasant. It required technical knowledge. It required patience. And then in 1988, a graduate student named Steve Dorner at the University of Illinois Urbana-Champaign built Eudora. He built it for the Mac, because that's what he had access to. He built it with something like 50,000 lines of C code. And he named it after Eudora Welty, the Mississippi writer, specifically after her short story "Why I Live at the P.O." The tagline became "Bringing the P.O. to Where You Live," and if you think about it, that's actually a pretty good description of what email does. Eudora worked. It had a clean interface. It had features that made sense once you saw them. It let you organize mail. It let you search. It let you set up filters so mail could be automatically processed and filed. These seem like obvious features now. In the early 1990s, they were not obvious. They were revelatory. Dorner was working remotely from Illinois when Qualcomm licensed Eudora from the University of Illinois in 1991 and hired him to continue development. This is where the story gets interesting. Qualcomm is a telecom giant. They're in the business of networking, telecommunications, semiconductors. Email software seems like a weird thing for them to own. But they owned it, and they committed resources to it, and they built it into a dominant product through the 1990s. ## The Three-Tier Model: Free, Sponsored, and Pro One of the reasons Eudora became so dominant was their business model. They offered three tiers: Eudora Light, which was free. Eudora Pro, which was paid at $49.95. And Eudora Sponsored, which was free but showed advertising. This was a smart approach. It let everyone use the product. It got students using Eudora, which meant that when those students got jobs, they wanted to use Eudora at work. It let people try before they paid. And it gave Eudora users a choice about how they wanted to engage with the product. By the late 1990s, Eudora had installed itself into the infrastructure of email. Businesses used it. ISPs recommended it. People used it at home and at work. Eudora's dominance wasn't fragile. It was structural. The revenue kept growing. The user base kept expanding. Eudora owned the segment of standalone email clients so completely that the question of what the second-place email client was seemed almost irrelevant. Then, in 1999, the company started offering Eudora Sponsored, an ad-supported free version. This model had worked before, but ad-supported software was becoming more culturally sensitive. People didn't necessarily love the idea of advertisements in their email client. And by 2000, Eudora Sponsored had hit 1 million active users. Which sounds good until you realize it's 1 million people who would have been paying customers and instead were ad-supported users. The math got worse every year. The real question is this: they weren't stupid. They just didn't see what was coming. Nobody did. ## The Bundling Problem: How Microsoft Destroyed Eudora Without Really Trying In 1997, Microsoft released Windows 95 with Outlook Express bundled in. Outlook Express was not as good as Eudora. It had fewer features. It was simpler, which was sometimes an advantage and sometimes a limitation. But it was free, and it came with Windows, and for a significant portion of computer users, having an email client pre-installed was enough. Why buy Eudora Pro when Outlook Express comes with your operating system? This is essentially what every smartphone company did years later when they bundled email clients into iOS and Android. But in 1997, when it happened to Eudora, it was devastating. Eudora's install base started flattening. New users saw no reason to install a third-party email client when Windows came with one. Eudora's growth narrative broke. The user numbers that had been accelerating started stabilizing. The financial dynamics started shifting. Qualcomm tried to adapt. They made Eudora more powerful, more feature-rich, more sophisticated. But sophisticated email clients are actually harder to use, and a lot of users just wanted simple email. And if they wanted simple email, why not use what came with their operating system? ## Webmail Arrived and Changed Everything Then Hotmail (https://404memoryfound.com/posts/why-everyone-had-hotmail-account.html) launched in 1996. This is where the logic starts to break down, at least from Qualcomm's perspective. Hotmail introduced the concept of email that you didn't need to install anything to access. You just went to a website. You logged in. You checked your email. You didn't need a client. You didn't need software. You just needed a web browser. For people who checked email from multiple computers, webmail was enormously appealing. Eudora required you to set up email on each machine separately. Webmail just worked from any computer. For people with laptops who checked email on the road, webmail started looking like the future. You didn't need to download messages. You didn't need to manage storage. You just accessed your email from anywhere. Hotmail grew incredibly fast. Microsoft bought Hotmail in 1997 and integrated it into their services. More competition in webmail launched. Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) Mail. AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html) Mail. Webmail went from novelty to normal incredibly quickly. By the early 2000s, webmail was where email growth was happening. Desktop email clients were becoming artifacts of the previous era. ## Gmail Arrived and Finished the Job And then, in 2004, Google launched Gmail. Gmail was webmail, but it was webmail with conversation threading, powerful search, spam filtering that actually worked, and enough free storage that nobody ever needed to delete emails. Gmail was everything Eudora had been for the standalone client market, but delivered through a web browser instead of an application. Gmail didn't kill Eudora overnight. But Gmail represented the end of an era. The real question is this: why would you use a desktop email client anymore? Why would you install software? Why wouldn't you just use Gmail? Gmail had everything. Search that worked. Organization that made sense. Accessibility from anywhere. No installation. No storage management. Just email. Qualcomm's last version of Eudora shipped in October 2006, two years after Gmail's launch. It took two years for the company to acknowledge that the market had fundamentally changed. By then, it was already over. Email as a category had moved entirely to the web. Eudora was a remnant. An excellent remnant, but a remnant. ## The Open Source Attempt: Too Late, Too Little Qualcomm did try one more thing. They open-sourced Eudora as a project called Penelope, based on the Mozilla Thunderbird framework. The idea was that a community of developers could keep the software alive and relevant. But here's the thing: the market had already moved. Nobody was interested in developing or using a desktop email client anymore. The effort continued until around 2010, then was fully deprecated by 2013. The code was eventually donated to the Computer History Museum in 2018, where it sits as a historical artifact. This is what happened to Eudora. Not a dramatic collapse. Not a technical failure. Just obsolescence. The email market shifted to webmail, and desktop clients became unnecessary. Eudora was the best desktop email client ever built. But desktop email clients stopped being relevant. The market didn't need the best desktop email client. It just needed email. And email, increasingly, existed on the web. ## Why This Matters Now Look at what happened: a company with dominant market position, a great product, significant revenue, and technical excellence got completely displaced because the category itself changed. This wasn't about execution. Qualcomm executed well. This was about the market fundamentally shifting from desktop software to web services. And once that shift happened, no amount of product quality could change the outcome. This is a lesson that echoes through software constantly. Dominance in a shrinking market is just a slow decline. Eudora had 18 million users and owned 64.7 percent of email software revenue. And none of that mattered once email itself moved to the web. The category got disrupted. The leader in the old category got disrupted with it. Eudora's legacy is weird because it's simultaneously huge and invisible. Eudora established conventions that every email client since has followed. Threading. Search. Filters. Good spam handling. Conversation views. Eudora did these things first and did them right. Gmail and every other email service since has built on Eudora's design language, probably without even realizing it. So Eudora won in design but lost in business. It taught the internet how to do email. And then it vanished. ## The Transition Was Inevitable, But Not Obvious at the Time This is the weird thing about Eudora's decline: there's no villain in this story. Qualcomm didn't make bad decisions. They made rational decisions based on available information. They controlled 64.7 percent of email software revenue. They had millions of users. Revenue was growing. And then the category itself shifted, and none of that mattered anymore. This happens in technology regularly, and it's almost always surprising to the people living through it. The shift from desktop email to webmail wasn't obvious in 1996. Desktop email seemed like the natural evolution. You install software on your computer. It manages your email locally. You have control. You have privacy, sort of. You have your data on your machine. This seemed obviously superior to the idea of logging into a website to check email, which felt weird and fragile. It required internet connectivity every time. It meant your email was on someone else's servers. But convenience beats theoretical concerns. Webmail was convenient. You could check email from any computer. You didn't need to set up clients on multiple machines. You didn't need to manage storage and downloads. It just worked. And as broadband internet became standard, the connectivity concern evaporated. As companies started offering more and more storage, the local storage argument disappeared. The advantages of desktop email just kept getting smaller. By 2004, when Gmail launched, the outcome was already inevitable. Google didn't kill desktop email. Gmail just delivered the final blow to something that was already dying. Eudora's fate was sealed the moment webmail became practical. Which means it was sealed in the late 1990s, though nobody acknowledged it at the time. ## What Qualcomm Got Right and Why It Still Didn't Matter Qualcomm's business model decisions were actually quite good. The three-tier system worked. Free users who later paid. Corporate users who needed features. Students who became corporate users. The Sponsored tier created a revenue stream even when people didn't pay. From a pure business perspective, this was sound. And it worked. Eudora's revenue trajectory was upward through the 1990s and into the early 2000s. The problem was that they were optimizing for a market that was disappearing. All the revenue in the world doesn't matter if the category you're dominating is being disrupted. This is the trap that every incumbent faces. You can be the best in a category. You can be profitable. You can be growing. And then the category itself becomes irrelevant, and all of your advantages become liabilities. Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html) faced this with arcades. Kodak (https://404memoryfound.com/posts/what-happened-to-kodak-digital-camera-bankruptcy.html) faced this with film. Blockbuster (https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html) faced this with video rental. Eudora faced this with desktop email. Dominance in an obsolete market is just a slow decline. You're still winning, right until you're not winning anymore. And by then, it's too late to reposition. ## The Legacy: Eudora Defined What Email Should Be Despite being completely forgotten by the general public, Eudora's influence on email design is massive and ongoing. Everything you expect from email, Eudora probably invented or popularized. Threading. Search. Filters. Spam handling. Organization. Conversation views. These seem obvious now because Eudora made them obvious. But they weren't obvious in 1988. Steve Dorner figured these things out and built them into Eudora, and every email system since has borrowed from that template. Gmail's interface looks nothing like Eudora's, but Gmail's functionality is basically Eudora's design philosophy implemented for the web. Which is maybe the highest compliment you can pay a product: the category evolved to reflect your design decisions, even though everyone forgot you were the one who made them. The Computer History Museum has Eudora's source code now. It sits there as a historical artifact, a reminder of when a company could own a category completely and still become irrelevant because the category itself became irrelevant. That's the real lesson of Eudora. Not that they made bad decisions. But that being the best at something that's becoming obsolete is the worst position you can be in. ## The Business Model Mismatch: Charging for Software That Wanted to Be Free One of the fundamental challenges Eudora faced was philosophical. The email client market wanted to move toward free. Outlook Express was free because it came with Windows. Netscape Navigator had an email client and it was free. The open source email clients were free. And then Eudora was charging $49.95 for Eudora Pro, or asking people to look at advertisements if they wanted the free version. This positioned Eudora as a premium product in a market that was increasingly skeptical of premium pricing for commodity software. In 1995, when email was still somewhat exotic, people would pay $49.95 for a good email client. By 2000, when email was becoming ubiquitous, charging for email software seemed almost offensive. Why pay when you could get something free? The Sponsored tier with advertising tried to bridge this gap, but advertising in software was becoming increasingly unpopular. People didn't like ads in their operating system. They didn't like ads in their applications. The ad-supported model worked for some software categories, but for something as personal as email, most users wanted it ad-free or free, not ad-supported. This left Eudora in an impossible position: charge for it and lose users to free alternatives, or offer it ad-supported and watch users complain about the ads. ## Why Microsoft's Bundling Strategy Was So Devastatingly Effective When Microsoft bundled Outlook Express with Windows 95, they weren't trying to destroy Eudora specifically. They were just including an email client with their operating system. But the effect was devastating. Every Windows computer came with an email client. Every computer manufacturer installed Windows. Every user who turned on a new computer saw Outlook Express sitting there, ready to use, with no need to install anything or pay any money. This is the power of platform bundling. You don't need to make the best product. You just need to make a good enough product and bundle it with something that everyone needs. This strategy worked for Internet Explorer against Netscape. It worked for Outlook Express against every competing email client. And it set the stage for why Gmail, built on the platform of the web, would eventually dominate. Qualcomm couldn't compete with this strategy because they weren't a platform company. They made software. They sold it to users. Microsoft was a platform company that used their platform to distribute software. Those are different games, and the platform company has structural advantages that a software company can't overcome. Eudora could be better than Outlook Express. It didn't matter. Outlook Express was already there. Already free. Already the default. ## The Email Wars Were Fought on Three Different Battlefields and Nobody Noticed The decline of Eudora actually happened across three distinct technology shifts, and it's easy to get confused about which one actually killed it. First, there was the shift from bundled email to free email clients. Outlook Express beat Eudora here because Microsoft could bundle for free. Second, there was the shift from desktop to webmail. Hotmail beat desktop clients because webmail didn't require installation. Third, there was the shift from free webmail to useful webmail. Gmail beat Hotmail and other webmail providers because Gmail actually worked well. Eudora lost on the first battlefield. If they'd won there, they might have competed on the second. If they'd won on the second, they couldn't have competed on the third, because Gmail's advantages were in web infrastructure, not client software. But the first loss was decisive. Once Outlook Express had the installed base, Eudora's advantages evaporated. The trajectory was set. It just took a decade to complete. ## The Open Source Attempt Was Noble and Doomed When Qualcomm released Eudora as open source under the Penelope project, it was a genuinely interesting decision. They're acknowledging that the market for commercial email clients is gone. They're trying to keep the product alive through community contribution. They're being generous with intellectual property. And it failed completely, which tells you something about markets: open source only works when there's a market demand for the category. Nobody wanted an open source desktop email client because nobody wanted a desktop email client anymore. The code was good. The product was still useful. But useful for a category that nobody cared about anymore is just irrelevance with good code. By 2010, when the open source effort was basically dead, Gmail had already won definitively. There was no market to capture. No users waiting to switch. No developers interested in maintaining a platform that was already obsolete. Eudora's story ends not with a bang but with a museum donation. The Computer History Museum has the source code, preserved like all artifacts from eras that have fully passed. It's a fitting end. Not destroyed. Not forgotten. Just archived, with the understanding that this was important once, and now it's history. ## FAQ: Questions About Eudora Q: Was Eudora really used by 18 million people? A: Yes, by December 1996. This was measured as installed base, not daily active users. But even accounting for that, Eudora had genuinely achieved dominant market position in desktop email clients. It was the clear market leader by a significant margin. Q: Why didn't Eudora pivot to webmail? A: Qualcomm did invest in webmail offerings, but they were always secondary to the desktop client. By the time they fully committed to webmail, Gmail had launched with better technology, better design, better infrastructure, and better funding. They couldn't compete. Sometimes being the leader in the previous category doesn't help you compete in the new one. Q: Could Eudora still exist today? A: Theoretically, yes. Thunderbird, which is based on the same Mozilla framework that the open source Eudora attempted to use, still exists and still has users. But nobody wants to use desktop email clients anymore. The ecosystem has moved entirely to web and mobile. A desktop-only email client can't compete with that reality. Q: What made Eudora better than Outlook Express? A: More features, better interface design, better search, better filtering, better organization tools. Outlook Express was simpler, which was sometimes an advantage. But Eudora was more powerful. The problem was that "simpler and free" beat "more powerful and paid" when the market had already shifted to webmail, which was even simpler. Q: Did Steve Dorner continue working on email software after Eudora? A: Dorner continued leading Eudora development at Qualcomm for many years. When desktop email clients became obsolete, so did the role. The transition from being the most important email software developer in the world to having an obsolete skill set happened very quickly. --- # What Happened to SimCity and How EA Destroyed It URL: https://404memoryfound.com/posts/what-happened-to-simcity-ea.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-08 Topics: Gaming, Business Blunders ## The Game That Taught a Generation to Think Like a Mayor In 1985, a college dropout named Will Wright was building maps for a helicopter game called Raid on Bungeling Bay. He kept noticing something strange: he was having more fun placing buildings, roads, and power lines on his custom maps than he was actually playing the game. That observation, as mundane as it sounds, would eventually generate one of the most influential franchises in PC gaming history. It would also, decades later, produce one of the most spectacular launch disasters the industry has ever seen. The story of SimCity is not just about a game. It is about how a single creative insight turned into a billion-dollar franchise, how corporate acquisition slowly eroded the thing that made it special, and how one catastrophic decision in 2013 killed a studio that had been making beloved games for nearly three decades. ## Will Wright, Jeff Braun, and the Pizza Party That Started Everything Will Wright was not a typical game designer. He had dropped out of Louisiana State University, then the New School in Manhattan, and eventually landed in the Bay Area with an obsessive interest in systems, urban planning, and the writings of Jay Forrester, an MIT professor who pioneered the field of system dynamics. Forrester’s 1969 book Urban Dynamics modeled cities as interconnected feedback loops: population growth drives housing demand, which drives tax revenue, which funds infrastructure, which attracts more population. Wright wanted to turn that into something you could play. The problem was that nobody wanted to publish it. Wright shopped his city simulator around to multiple publishers through the late 1980s. Every single one passed. The consensus was that a game without a win condition, without enemies, without a clear objective, was not a game at all. It was a toy. And toys, the thinking went, do not sell. Then, in 1986, Wright attended what he later called \"the world’s most important pizza party.\" There he met Jeff Braun, an entrepreneur looking to break into the software business. Braun saw the potential immediately. Together, they founded Maxis in 1987 in Orinda, California. Their first release would be the game no one else wanted to publish. ## SimCity Arrives and Breaks Every Rule SimCity launched in February 1989, first on the Amiga and Macintosh, then on IBM PC and Commodore (https://404memoryfound.com/posts/what-happened-to-commodore-computers.html) 64 later that year. The game dropped players into the role of mayor of a blank plot of land. You zoned residential, commercial, and industrial areas. You built roads and rails. You placed power plants and watched the electrical grid come alive. You raised and lowered taxes. And then you waited, watching your tiny simulated citizens, called \"Sims,\" respond to your decisions in real time. There was no way to \"win\" SimCity. There was no final boss, no score threshold, no ending screen. You could build a thriving metropolis or a crime-ridden wasteland. You could unleash disasters for fun: earthquakes, tornadoes, Godzilla-like monsters stomping through your downtown. The game trusted you to set your own goals, which was practically unheard of in 1989. It worked. SimCity sold one million copies by late 1992, split roughly evenly between home computers and the Super Nintendo port. Macworld named it the Best Simulation Game of 1989. It won the Origins Award for Best Military or Strategy Computer Game. Schools started using it in classrooms to teach urban planning concepts. The game that every publisher had rejected became a cultural phenomenon. ## SimCity 2000: The Masterpiece If the original SimCity proved the concept, SimCity 2000, released in 1993, perfected it. Wright co-designed the sequel with Fred Haslam, and the improvements were staggering. The flat, top-down grid gave way to an isometric perspective that made cities feel three-dimensional. Terrain had elevation. You could build underground water pipes. Highways and on-ramps appeared. Arcologies, those massive self-contained mega-structures from science fiction, became the ultimate late-game reward. SimCity 2000 also introduced the newspaper feature, a small touch that had an outsized impact on immersion. After each year in the simulation, a procedurally generated newspaper would summarize events in your city: budget surpluses, crime waves, industrial growth. It made the simulation feel alive in a way the original never quite achieved. The game was a commercial juggernaut. SimCity 2000 sold an estimated 4.23 million copies worldwide across all platforms. It won Best Simulation at the 1994 Codie Awards, marking the fifth consecutive year Maxis had taken that category. It also won the Origins Award for Best Strategy Game of 1994. For many players, especially those who grew up in the mid-1990s, SimCity 2000 remains the definitive version of the franchise. The one that got everything right. ## Maxis Grows, Stumbles, and Gets Acquired Success brought expansion, and expansion brought risk. Through the mid-1990s, Maxis launched an ambitious slate of \"Sim\" branded games: SimEarth, SimAnt, SimFarm, SimTower, SimCopter, Streets of SimCity. Some were creative experiments. Most were commercial disappointments. SimCopter, released in 1996, was particularly troubled, shipping with bugs and an infamous hidden Easter egg inserted by a disgruntled developer that made international news. By 1997, Maxis was struggling financially. The company had overextended itself, releasing too many niche titles that diluted the SimCity brand without generating enough revenue to sustain the studio. That year, Electronic Arts acquired Maxis for approximately $125 million. Will Wright stayed on, and the deal initially seemed like a lifeline. EA had the resources and distribution network that Maxis needed. The trade-off, of course, was creative independence. ## SimCity 3000 and SimCity 4: The Steady Middle Years SimCity 3000 arrived in 1999, and while it was a solid game, it felt more like an incremental update than a revolution. The isometric perspective returned. Graphics improved. The simulation grew more complex. But the core loop was familiar: zone, build, manage, grow. It sold well enough, contributing to a franchise total of five million units by the end of the decade, but it did not redefine the genre the way its predecessors had. SimCity 4, released in January 2003, was more ambitious. It introduced a regional system where multiple cities could exist on the same map, sharing resources and commuters. The simulation engine was significantly more detailed, modeling individual Sims traveling to work, shopping, and returning home. The game also introduced the \"My Sim\" mode, letting you import characters from The Sims (https://404memoryfound.com/posts/original-sims-best-selling-pc-game.html) to walk around your city. SimCity 4 was critically acclaimed but commercially modest compared to The Sims, which had exploded into a global phenomenon after its 2000 launch. Within EA, the priorities were clear. The Sims was generating hundreds of millions in revenue through expansion packs. SimCity, by comparison, was a prestige title with a loyal but smaller audience. ## The Ten-Year Gap and the Decision That Changed Everything After SimCity 4, the franchise went quiet. Ten years passed without a new mainline entry. During that decade, the gaming landscape shifted dramatically. Always-online games became normalized through World of Warcraft and its imitators. Digital distribution through Steam began replacing physical retail. And within EA, a philosophy took hold that single-player, offline experiences were relics of an earlier era. When EA and Maxis finally announced a new SimCity in 2012, the excitement was enormous. The franchise had been dormant for a decade. A new generation of players had grown up hearing about SimCity but never playing one at launch. Pre-orders surged. Preview coverage was overwhelmingly positive. The new \"GlassBox\" simulation engine promised to model every individual Sim, every car, every unit of electricity flowing through the grid. There was one catch. The new SimCity would require a persistent internet connection to play. Even in single-player mode. EA and Maxis framed this as a feature, not a restriction. The game’s creative director Ocean Quigley and general manager Lucy Bradshaw explained that the simulation relied on server-side computing, that cities in the same region would interact and share resources, that the always-online requirement was fundamental to the game’s design. ## March 5, 2013: The Launch That Broke Everything SimCity launched at 12:01 AM Eastern on March 5, 2013. Within hours, the servers were overwhelmed. Players who had purchased the game, many of them digitally for $60, could not log in. Those who managed to connect faced rubberbanding, lost save files, and disconnections that erased hours of progress. The game was, in the words of multiple reviewers, \"unplayably broken.\" The scale of the failure was remarkable. Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html) pulled the digital version from its store and posted a warning to customers. Review outlets, including Polygon and Kotaku, refused to publish final scores, instead posting ongoing \"review in progress\" updates as the situation deteriorated. Kotaku published a comprehensive guide titled \"Your Complete Guide to the SimCity Disaster\" that cataloged the cascading failures in real time. On the evening of March 7, Lucy Bradshaw issued a public statement acknowledging the problems. She reported that more than 700,000 cities had been built in 24 hours, framing the crisis partly as a victim-of-success story. On March 8, she published a blog post announcing that server capacity had been increased by 120 percent and that error rates had dropped by 80 percent. She also announced that EA would offer every affected player a free game from the EA catalog as compensation. The goodwill gesture came too late. The damage to SimCity’s reputation was already done. And then, things got worse. ## The Offline Discovery and the Trust Collapse Within days of launch, modders began poking at SimCity’s code. What they found was explosive. A modder discovered that by commenting out a single line of code, the game could run offline indefinitely. The simulation did not, in fact, require server-side computing. The cities functioned perfectly well on the player’s local machine. A source inside Maxis confirmed to the gaming site Rock Paper Shotgun that the always-online requirement was not a technical necessity. It was a DRM decision dressed up as a design philosophy. This revelation demolished whatever remaining trust players had in EA and Maxis. The company had spent months insisting that the online requirement was integral to the game’s design, that the simulation literally could not function without server support. That turned out to be false. The backlash was swift and brutal. SimCity’s user reviews cratered. The game became a case study in how not to implement digital rights management. In March 2014, a full year after launch, EA finally released an offline mode for SimCity. By then, the gesture felt more like an admission of defeat than a victory for players. ## The Fall of Maxis Emeryville On March 4, 2015, exactly two years after the SimCity launch disaster, EA announced the closure of Maxis Emeryville. The studio that Will Wright and Jeff Braun had founded in 1987, the studio that had created SimCity, SimCity 2000, The Sims, and Spore, was shutting its doors. Approximately 100 employees lost their jobs. Guillaume Pierre, a Maxis designer, announced the closure on Twitter: \"A fun 12 years, but it’s time to turn off the lights and put the key under the door.\" EA stated that Maxis IP development would be consolidated to studios in Redwood Shores, Salt Lake City, Helsinki, and Melbourne. The Maxis name would survive, technically, but the original studio, the one with the history and the institutional memory, was gone. The SimCity 2013 disaster was not the sole cause of the closure, but it was the most visible symptom of a deeper problem. EA had gradually transformed Maxis from an independent creative studio into a division that served corporate priorities. The always-online decision for SimCity was not a Maxis idea. It was an EA mandate, driven by concerns about piracy and a desire to push players toward microtransactions and online engagement metrics. ## What SimCity’s Story Tells Us About the Industry Look at the arc of SimCity and you see a pattern that has repeated across the gaming industry. A visionary creator builds something original. It succeeds because of its uniqueness. A larger corporation acquires the creator’s studio. The corporation gradually imposes its own priorities: recurring revenue, online engagement, DRM, franchise extension. The original vision gets diluted. The audience notices. And eventually, the thing that made the franchise special in the first place gets lost entirely. Will Wright left EA in 2009, four years before the SimCity disaster. He had seen the trajectory. The Sims franchise had been carved into dozens of expansion packs and stuff packs, each one smaller and more cynical than the last. Spore, his most ambitious project, had shipped in 2008 with aggressive DRM that limited installations to three computers, sparking a consumer revolt that prefigured the SimCity backlash by five years. The irony is that SimCity’s core design philosophy, the idea that complex systems can be made accessible and fun, has never been more relevant. Cities: Skylines, developed by Colossal Order and published by Paradox Interactive, launched in 2015, the same year Maxis Emeryville closed. It sold one million copies in its first month. It delivered exactly what SimCity fans had wanted: a deep, offline, moddable city builder. It succeeded precisely because it did not make the mistakes that killed SimCity. The real question is not why SimCity failed in 2013. The mechanics of the failure are straightforward: bad DRM, insufficient servers, dishonest communication. The real question is why the people making the decisions did not see it coming. And the answer, most likely, is that they did see it coming. They just believed the business model was worth the risk. They were wrong. ## Frequently Asked Questions When was the original SimCity released? \nThe original SimCity was released in February 1989 for the Amiga and Macintosh, with IBM PC and Commodore 64 versions following later that year. It was created by Will Wright and published by Maxis, the company he co-founded with Jeff Braun in 1987. Why did SimCity 2013 require an always-online connection? \nEA and Maxis initially claimed the always-online requirement was necessary because the simulation relied on server-side computing. This was later shown to be false when modders demonstrated the game could run offline by modifying a single line of code. The requirement was primarily a DRM (digital rights management) measure. What happened to Maxis after SimCity 2013? \nMaxis Emeryville, the original studio founded by Will Wright, was closed by EA on March 4, 2015. Approximately 100 employees were affected. EA consolidated Maxis development to other studios in Redwood Shores, Salt Lake City, Helsinki, and Melbourne. How many copies did SimCity 2000 sell? \nSimCity 2000 sold an estimated 4.23 million copies worldwide across all platforms. It remains one of the best-selling entries in the franchise and is widely considered the definitive SimCity experience. What replaced SimCity after the franchise stalled? \nCities: Skylines, developed by Colossal Order and published by Paradox Interactive, launched in March 2015 and quickly became the spiritual successor to SimCity. It sold one million copies in its first month and offered the deep, offline, moddable city-building experience that SimCity 2013 had failed to deliver. Will there ever be a new SimCity game? \nAs of early 2026, EA has not announced a new mainline SimCity title. The Maxis brand continues to be used primarily for The Sims franchise. Whether EA will revisit SimCity remains an open question, though the success of Cities: Skylines and its sequel suggests the audience for city builders remains strong. \n --- # What Happened to RollerCoaster Tycoon, Built by One Man in Assembly URL: https://404memoryfound.com/posts/what-happened-to-rollercoaster-tycoon.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-08 Topics: Gaming, Software & Apps ## One Guy, Assembly Code, and the Greatest Theme Park Ever Built Picture this: 1997. Somewhere in a small village near Dunblane, Scotland, a guy named Chris Sawyer is sitting alone at his desk. No team. No studio. No corner office with a whiteboard covered in sprint planning sticky notes. Just one man, a compiler, and what might be the most absurdly ambitious solo development project in the history of video games. He is writing a theme park simulation, from scratch, in x86 assembly language. And he is going to do this for the next two years. If you played PC games in the late 1990s, you already know how this story ends. RollerCoaster Tycoon became one of the best-selling PC games of its era, moved over four million copies in the United States alone, and generated roughly $180 million in total revenue. Chris Sawyer personally earned around $30 million in royalties. And he did almost all of it by himself, writing 99% of the code in a programming language that most developers considered obsolete even in 1997. This is the story of how that happened. And honestly, it is kind of insane. ## Before the Tycoon: Transport Tycoon and the Assembly Habit To understand why Chris Sawyer wrote RollerCoaster Tycoon in assembly, you have to understand what came before it. In 1994, Sawyer released Transport Tycoon through MicroProse, a business simulation where you built transportation networks connecting cities with buses, trains, ships, and planes. It was a hit in Europe, particularly in the UK, and it established Sawyer as a developer who could build deeply complex simulations that were somehow still fun to play. Transport Tycoon was also written largely in assembly language. This was not a random choice or an act of stubbornness. Sawyer had been programming in assembly since the early 1980s, starting on 8-bit machines where assembly was not optional, it was the only way to get acceptable performance. By the time he moved to PC development, he had built up years of hand-tuned libraries, tools, and routines. His assembly codebase was, in a very real sense, his competitive advantage. He could write machine code faster and more reliably than most developers could write C. An enhanced version, Transport Tycoon Deluxe, followed in 1995. And then Sawyer started thinking about roller coasters. ## The Roller Coaster Obsession Sawyer has said in interviews that he had been interested in the engineering side of roller coasters for years before starting RollerCoaster Tycoon. The physics of it fascinated him: the way potential energy converts to kinetic energy on a drop, the forces that pin you to your seat on a loop, the precise calculations that determine whether a coaster is thrilling or terrifying. He saw an opportunity to build a game that captured that engineering challenge while wrapping it in the addictive management loop of a tycoon game. The basic pitch was simple. You manage a theme park. You build rides, set admission prices, hire staff, maintain the grounds, and try to attract enough guests to meet financial targets. Each scenario gave you a park with specific goals: reach a certain guest count, achieve a target park rating, or accumulate a set amount of cash. Some scenarios started you with an empty plot. Others gave you a struggling park that needed to be turned around. Simple pitch, extraordinary execution. ## 99% Assembly: The Technical Marvel Here is the fact that makes every programmer do a double take. Chris Sawyer wrote 99% of RollerCoaster Tycoon’s code in x86 assembly language, using the Microsoft Macro Assembler. The remaining 1% was written in C, and that tiny slice existed solely to interface with Microsoft Windows and DirectX. Everything else, the simulation engine, the pathfinding, the ride physics, the guest AI, the rendering, the UI, all of it was hand-written assembly. For non-programmers, here is what that means. Most game developers in the late 1990s wrote their code in C or C++, high-level languages that let you express complex ideas in relatively readable syntax. The compiler then translates that code into machine instructions the processor can execute. Assembly language skips the middleman. You are writing instructions that the processor executes directly. It is faster, because there is no compiler making guesses about optimization. It is also dramatically harder, because you are managing memory, registers, and processor states manually. A task that takes 10 lines of C might take 50 or 100 lines of assembly. Sawyer chose assembly for a practical reason: the game needed to simulate hundreds of individual guests, each with their own preferences, happiness levels, hunger, thirst, and pathfinding logic, while simultaneously running physics calculations on dozens of rides, all rendered in an isometric view that scrolled smoothly on late-1990s hardware. The machines of that era, Pentium and Pentium II processors running at 200 to 400 MHz, did not have power to waste. Assembly let Sawyer squeeze every cycle out of the CPU. The result was a game that ran beautifully on modest hardware. While competitors struggled with frame rate issues and memory problems, RollerCoaster Tycoon was silky smooth. Guests wandered the paths. Roller coasters climbed, dropped, and looped with fluid animation. Rain fell. Fireworks exploded. And the whole thing ran on a machine that would be considered a calculator by today’s standards. ## MicroProse, Hasbro, and Getting the Game to Market Sawyer developed RollerCoaster Tycoon over approximately two years in his home near Dunblane, Scotland. The publisher was MicroProse, the same company that had released Transport Tycoon. But MicroProse in the late 1990s was not the MicroProse of the Sid Meier glory days. The company had been acquired by Spectrum HoloByte in 1993, which was then acquired by Hasbro Interactive in 1998. By the time RollerCoaster Tycoon was ready to ship, MicroProse was essentially a brand name within Hasbro’s game division. The game launched in North America on March 22, 1999. And it took off. ## The Sales Explosion RollerCoaster Tycoon sold over 700,000 copies in 1999 alone, making it the third highest-grossing PC game of the year with $19.6 million in revenue. That is impressive on its own. What makes it extraordinary is the trajectory. The game did not have a massive marketing campaign. It did not have a cinematic trailer or a celebrity endorsement. It sold on word of mouth, positive reviews, and the simple fact that it was incredibly fun. And the sales kept accelerating. As the game spread into international markets, the numbers climbed. Within three years, RollerCoaster Tycoon had sold more than four million copies in the United States alone. Total worldwide revenue reached approximately $180 million. For Hasbro, the game single-handedly justified the money they had spent acquiring MicroProse. For Chris Sawyer, it meant roughly $30 million in royalties, an almost unheard-of payout for a solo developer. Two expansion packs followed: Added Attractions (released as Corkscrew Follies in North America) in 1999 and Loopy Landscapes in 2000. Both added new scenarios, rides, and scenery objects. Both sold well. The RollerCoaster Tycoon franchise was, by any measure, a phenomenon. ## Why It Worked: The Design Genius RollerCoaster Tycoon was not the first theme park management game. Bullfrog Productions had released Theme Park in 1994, and it was a great game. But RollerCoaster Tycoon succeeded because of a handful of design decisions that elevated it from good to unforgettable. First, the coaster builder itself was genuinely deep. You were not selecting pre-built rides from a menu. You were laying track piece by piece, adjusting elevation, banking turns, adding inversions, and then testing the result with a physics simulation that calculated g-forces, excitement ratings, intensity ratings, and nausea ratings. A well-designed coaster that balanced thrills with comfort would attract long lines. A poorly designed one would make guests vomit on the pathways. There was real skill involved, and players spent hours perfecting their designs. Second, the guest AI was remarkably detailed for its era. Each guest was an individual with specific traits. Some guests preferred gentle rides. Others craved intensity. Guests got hungry, thirsty, and tired. They complained about litter, long queues, and high prices. They got lost if your paths were confusing. Watching a guest’s thought bubbles became a diagnostic tool: \"I’m not paying that much to go on Mega Coaster 1,\" a guest might think, and you would know it was time to lower the price. Third, the scenarios were brilliantly paced. The game started with simple parks on flat terrain and gradually introduced more challenging conditions: hilly landscapes, pre-built parks with problems, strict time limits, limited budgets. Each scenario felt like a puzzle with multiple solutions. The difficulty curve was steep enough to be satisfying but gentle enough to avoid frustration. You always felt like you were learning something. And fourth, the game had personality. The isometric art style was colorful and detailed. The sound design was full of charming touches: the clatter of a wooden coaster, the screams of riders, the cheerful music that played when things were going well. It was the kind of game where you would look up from your screen and realize three hours had passed. Which, honestly, is the highest compliment you can pay a simulation game. ## RollerCoaster Tycoon 2: Sawyer Does It Again In October 2002, Chris Sawyer released RollerCoaster Tycoon 2, again coded almost entirely in assembly by himself. The sequel was published by Infogrames Interactive, which had acquired Hasbro Interactive in 2001 and would later rebrand as Atari (https://404memoryfound.com/posts/who-owns-atari-now.html). The game used an enhanced version of the original engine with improved graphics, more ride types, and a scenario editor that let players create and share their own parks. RollerCoaster Tycoon 2 was well received, though some critics noted that it felt more like an expansion pack than a true sequel. The core gameplay was essentially identical to the original. For most fans, that was fine. The original formula was so strong that \"more of the same\" was exactly what they wanted. The game sold well, though exact figures are harder to pin down than the original’s. ## RollerCoaster Tycoon 3: Sawyer Steps Back For the third game in the franchise, released in October 2004, Chris Sawyer stepped back from development entirely. He served only as a consultant while Frontier Developments, a UK studio founded by David Braben of Elite fame, handled the actual development. The publisher was Atari. RollerCoaster Tycoon 3 was a significant technical leap. It moved the series into full 3D, replacing the isometric perspective with a free-roaming camera. A \"Coaster Cam\" let you ride your creations from a first-person viewpoint. Day-night cycles and fireworks shows added visual spectacle. It was, by most accounts, a solid game. But something was lost in the transition. The charm of the isometric art style, the buttery smoothness that Sawyer’s assembly code had delivered, the feeling that every pixel had been placed with intention, those qualities were harder to replicate in a 3D engine built by a team rather than a single obsessive craftsman. RollerCoaster Tycoon 3 was good. It was not the same. ## The Franchise After Sawyer What happened next is a familiar story in gaming. Atari, perpetually struggling financially, continued to exploit the RollerCoaster Tycoon brand without Sawyer’s involvement. RollerCoaster Tycoon 3D for the Nintendo 3DS arrived in 2012 to mediocre reviews. RollerCoaster Tycoon 4 Mobile launched in 2014 as a free-to-play game stuffed with microtransactions, earning scathing reviews from fans who saw it as a betrayal of everything the franchise stood for. RollerCoaster Tycoon World, a PC title developed by multiple studios after a chaotic development process, launched in early access in 2016 to overwhelmingly negative reception. Meanwhile, Frontier Developments went on to create Planet Coaster in 2016, a spiritual successor that captured much of the original’s magic in a modern 3D engine. It was, in many ways, the true follow-up to RollerCoaster Tycoon that fans had been waiting for. In November 2005, Sawyer sued Atari, claiming the company had failed to pay him certain royalties owed from the franchise. Atari counter-sued in 2007. The two parties settled out of court in February 2008 for an undisclosed sum paid to Sawyer. ## The Legacy: Why Assembly Matters RollerCoaster Tycoon’s development story has become something of a legend in the programming community, and for good reason. It is proof that a single talented individual can compete with entire studios. That choosing the right tool, even an unconventional one, can produce results that brute-force team size cannot. That optimization is not just about making things faster. It is about making things possible. Chris Sawyer did not write RollerCoaster Tycoon in assembly to show off. He did it because he had decades of experience in the language, because he had a library of proven routines from previous projects, and because the game’s technical requirements demanded it. The hundreds of simultaneous pathfinding calculations, the real-time physics on dozens of rides, the smooth scrolling across a detailed isometric world, none of that would have been feasible on late-1990s hardware without the kind of fine-grained control that assembly provides. Today, writing a commercial game in assembly would be considered somewhere between eccentric and reckless. Modern hardware is powerful enough that the performance gains rarely justify the development cost. But in 1999, on a Pentium running at 233 MHz, it was the difference between a game that ran and a game that soared. And here is the thing that still gets me. When you loaded up RollerCoaster Tycoon for the first time, you did not think about assembly code. You did not think about optimization or register allocation or memory management. You thought about where to put the roller coaster. You thought about whether to build a food court near the entrance or the back of the park. You thought about that one guest who kept getting lost near the hedge maze. The best technology disappears. You do not notice it. You just notice that the game runs perfectly, that it responds instantly, that everything feels right. Chris Sawyer understood that better than almost anyone. He spent two years writing tens of thousands of lines of assembly code so that millions of players could have the experience of building the perfect theme park without ever thinking about the code underneath. That is what great engineering looks like. ## Frequently Asked Questions Was RollerCoaster Tycoon really written in assembly language? \nYes. Chris Sawyer wrote 99% of RollerCoaster Tycoon’s code in x86 assembly language using the Microsoft Macro Assembler (MASM). The remaining 1% was written in C to interface with Microsoft Windows and DirectX. How many copies did RollerCoaster Tycoon sell? \nRollerCoaster Tycoon sold over four million copies in the United States alone within three years of release. Total worldwide revenue reached approximately $180 million. Chris Sawyer earned roughly $30 million in royalties. Did Chris Sawyer make RollerCoaster Tycoon completely alone? \nSawyer did virtually all of the programming and game design himself. The graphics were created by artist Simon Foster, and the music was composed by Allister Brimble. But the code, simulation engine, ride physics, AI, and overall design were Sawyer’s solo work. What happened to the RollerCoaster Tycoon franchise after Chris Sawyer? \nAfter RollerCoaster Tycoon 2, Sawyer stepped back to a consultant role for RCT3, which was developed by Frontier Developments. Subsequent entries made without Sawyer’s involvement, including RCT 4 Mobile and RCT World, were poorly received. Frontier went on to create Planet Coaster in 2016 as a spiritual successor. Why did Chris Sawyer choose assembly language instead of C or C++? \nSawyer had been programming in assembly since the 1980s and had built extensive libraries of optimized routines from previous projects like Transport Tycoon. He could write assembly faster and more efficiently than high-level languages, and the game’s complex simulation demanded the performance that hand-tuned machine code provided on late-1990s hardware. Is there a connection between RollerCoaster Tycoon and Transport Tycoon? \nBoth games were created by Chris Sawyer and share a similar development philosophy: complex simulations made accessible through intuitive management interfaces. Sawyer began developing RollerCoaster Tycoon while working on the engine for a Transport Tycoon sequel, reusing and adapting code from his earlier projects. \n --- # Newgrounds Is Still Online in 2026, 31 Years Later URL: https://404memoryfound.com/posts/what-happened-to-newgrounds.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-08 Updated: 2026-09-18 Topics: Internet Culture, Music & Entertainment **Summary:** Newgrounds is still online in 2026 and still independent, more than 31 years after Tom Fulp founded it on July 6, 1995. Founder Tom Fulp and three staff run it from Glenside, Pennsylvania, with no investors and no parent company. Adobe killed Flash on December 31, 2020, but the archive survived through the Newgrounds Player and the Ruffle emulator, and the site now pays its bills with supporter subscriptions starting at $5 a month. **Key facts:** - Status today: Online and independent in 2026, 31 years after launch - Launched: July 6, 1995, by Tom Fulp, from a 1991 paper fanzine - Owner today: Newgrounds, Inc., never sold, no outside investors - Who runs it: Tom Fulp and three staff in Glenside, Pennsylvania - Cost today: Free to use; Supporter from $5 a month or $36 a year (2026) ## How did a 13-year-old's Neo Geo zine turn into a website? Newgrounds started on paper. In 1991 Tom Fulp, then 13, published a Neo Geo fanzine called New Ground from his parents' basement in Perkasie, Pennsylvania, and mailed sporadic issues to around 100 members of a club on the Prodigy (https://404memoryfound.com/posts/what-happened-to-prodigy-online-service.html) online service. In 1995 Fulp got his own web space and carried the name online as New Ground Remix. Newgrounds counts July 6, 1995 as its founding date, which made July 6, 2025 its thirtieth birthday as a website. The turn came in 1998, when Fulp started experimenting with Macromedia Flash. After a call from the TV show Inside Edition about his game Assassin, he registered a domain that would be easy to remember on screen and paid $33 a month out of pocket to host it. "In 1998, I picked up Flash and my life changed forever," Fulp writes on the Newgrounds about page. Traffic grew faster than the money. By 1999 his host wanted more than $1,000 a month and his only good ad company had dropped him over the site's content, so Fulp partnered with the film studio Troma, which hosted Newgrounds in exchange for a cut of ad revenue. That arrangement lasted until February 2003. ## What actually came out of the Portal? The Portal started in 1999 as a page where Fulp dumped small projects and then began hosting other people's Flash files, each one hand coded into its own HTML page. On April 6, 2000, Newgrounds automated it. By the site's own account it was the first website anywhere to allow instant self-serve publishing for movies and games, four and a half years before YouTube launched. That single change turned a personal showcase into a career pipeline. Fulp's Pico's School (1999) became the reference point for what Flash 3 could do. Alien Hominid, built with artist Dan Paladin in 2002, started as a Portal game; the two formed The Behemoth on April 1, 2003, and the console version reached stores in November 2004. The studio's follow-up, Castle Crashers, arrived on Xbox Live Arcade in August 2008. "In the 11 years since its launch, the game has sold over 20 million copies," The Behemoth wrote in 2019. Animation did the same thing. Salad Fingers, Eddsworld and the Madness series all grew inside the Portal, and the Numa Numa Dance premiered on Newgrounds on December 6, 2004, a full year before YouTube opened. The Audio Portal followed in 2003 so musicians could be credited in other people's work, and an Art Portal arrived in June 2009. Much like Homestar Runner (https://404memoryfound.com/posts/homestar-runner-still-updating.html), the site proved that a small crew could hold an audience without a network behind it. ## When did Newgrounds peak, and what knocked it down? The clearest peak markers are bandwidth bills. Newgrounds went from a 100mbps cap to 300mbps during 2003, was consuming 500mbps by the end of 2004, and blew past 800mbps during peak hours in December 2005. Time ranked it among the 50 best websites of 2010. Two things ended that run. The first was video. YouTube made watching easier than the SWF format ever would, and by 2012 Fulp described a site where his own efforts were not changing the trajectory and everyone was, in his words on the history wiki, "livin' la vida YouTube." The second was ad blocking. Newgrounds ran a 2009 experiment showing banners only to ad block users, which mostly made them angry. "we would go on to lose well over half our revenue to adblock," Fulp wrote on the Newgrounds history wiki, "and it's a big reason why you guys don't get to have nice things." A 2013 peering dispute between Verizon and Cogent, the site's bandwidth provider, throttled it further. Plenty of contemporaries did not survive that stretch at all, as the shutdown of GeoCities (https://404memoryfound.com/posts/history-of-geocities-websites.html) showed. ## What happened when Adobe killed Flash? Adobe set the date years in advance and hit it. "Flash will be completely removed from all browsers by December 31, 2020, via Windows Update," Microsoft told administrators in its lifecycle notice in 2020. Adobe began blocking Flash content from running on January 12, 2021, which Fulp called the true end of Flash day. Newgrounds had been preparing since 2011, when staffer Mike built Swivel to convert SWF files to video. In April 2019 the site shipped Newgrounds Player, a desktop application that keeps Flash content playable, with support for medals, high scores and shared creations. In August 2019 it introduced Ruffle, an open-source Flash emulator written in Rust, and started switching classic content over to it in 2020. The migration is still running. Ruffle gained working medals and high scores for ActionScript 2 games in October 2020, picked up Flash video support in October 2021, and pushed into ActionScript 3 from 2023, after which Newgrounds switched on emulation for thousands more games in the second half of 2024. Newgrounds put $6,000 into Ruffle development in 2021 alone. The full story of the format is in the history of Macromedia Flash (https://404memoryfound.com/posts/definitive-history-macromedia-flash.html). The archive survived, and so did the audience. Friday Night Funkin' debuted on Newgrounds on November 1, 2020, and when its Week 7 update landed on April 18, 2021 the site broke its one-day record with 1.2 million views on a single piece of content. The game's Kickstarter, launched the same day, raised more than $2 million. ## Who runs Newgrounds in 2026? Tom Fulp does, and he never sold. Pages across the site carry a Newgrounds, Inc. copyright line reading 1995 to 2026, and the staff page lists him as founder and guy in charge alongside three others: senior programmers Josh Tuttle, who joined in 2008, and James Holloway, who joined in 2002, plus staff artist Jeff Bandelin, who joined in 2006. The company works out of Glenside, Pennsylvania, in a building Fulp bought in 2005. "Newgrounds has been self-sustained since its inception, free of investor and corporate interests," Fulp writes on the about page. There is no parent company and no acquisition anywhere in the site's history. The recent work is unglamorous. A large denial-of-service attack in September 2025 led to a screening layer called NG Guard, and the site added age verification for United Kingdom users to comply with the Online Safety Act. Fulp's own summary of 2026 on the history wiki is that overhauls and regulatory compliance remain the heavy themes, alongside a mobile refresh in May 2026. He is still making games too, including a long-delayed console project called Nightmare Cops. ## How does Newgrounds pay for itself now? Mostly through readers, and not comfortably. Newgrounds launched its Supporter upgrade in September 2012 at $25 for a year of ad-free browsing. It took until Thanksgiving 2018 to reach 2,000 active supporters, a jump helped by Tumblr (https://404memoryfound.com/posts/what-happened-to-tumblr-yahoo-billion-dollar-mistake.html)'s ban on adult content, which sent more than 1,000 new artists to the Art Portal within days. The numbers have improved without solving the problem. "We finished the year with nearly 5,800 Supporters but NG continued to run at a loss," Fulp wrote in the 2021 entry of the history wiki, in a year when the site also sent more than $40,000 to the community in prizes and drawing tablets. In October 2024 Newgrounds added Takeover Tuesday, which lets each supporter feature one item on the front page each week. In 2026 the pitch is a monthly subscription starting at $5 or a yearly one from $36, which works out to $3 a month paid up front. The site still runs a small amount of advertising on A-rated games and movies. Fulp's stated targets on the history wiki are to "stop losing money, keep increasing the monthly prizes and gradually phase out the last remaining ads." Monthly creator prizes passed $5,000 for the Best of June 2026. ## Frequently Asked Questions ### Is Newgrounds still active in 2026? Yes. Newgrounds is online and publishing new games, animation, art and music in 2026, more than 31 years after Tom Fulp founded it on July 6, 1995. The site ran its eleventh annual Pixel Day on January 23, 2026 and its sixth annual Flash Forward game jam that spring. ### Can you still play old Flash games on Newgrounds? Yes. Adobe Flash Player reached end of life on December 31, 2020, but Newgrounds runs its classic archive through Ruffle, an open-source Flash emulator it began sponsoring in 2019, and offers Newgrounds Player, a desktop application released in April 2019, for content Ruffle cannot handle yet. Thousands more games were switched over to Ruffle in the second half of 2024. ### Who owns Newgrounds? Newgrounds is owned by Newgrounds, Inc., the private company founded by Tom Fulp, who still runs it from Glenside, Pennsylvania. It has never been acquired and has no outside investors, and the staff page listed four people in 2026. **Sources:** - Newgrounds Wiki: About Newgrounds: https://www.newgrounds.com/wiki/about-newgrounds - Newgrounds Wiki: History: https://www.newgrounds.com/wiki/about-newgrounds/history - Support Newgrounds (Supporter pricing and perks): https://www.newgrounds.com/supporter - Microsoft Lifecycle: Adobe Flash end of support on December 31, 2020: https://learn.microsoft.com/en-us/lifecycle/announcements/adobe-flash-end-of-support - The Behemoth: Castle Crashers Remastered Has Arrived (2019): https://blog.thebehemoth.com/2019/09/17/castle-crashers-remastered-has-arrived/ --- # What Happened to BonziBuddy, the Internet's Friendliest Spyware URL: https://404memoryfound.com/posts/what-happened-to-bonzibuddy-spyware.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-08 Topics: Software & Apps, Internet Culture ## The Numbers Behind the Purple Gorilla In 1999, a piece of software called BonziBuddy was downloaded by millions of people who genuinely believed it would make their computers better. It was free. It featured a friendly purple gorilla who told jokes, sang songs, managed your downloads, and read your email aloud using text-to-speech. It was, by every surface-level metric, a delightful little program. It was also spyware. Not in the metaphorical sense. Not in the "well, every app collects some data" sense that people use today. BonziBuddy was classified as actual spyware and adware by every major security researcher who examined it. It tracked your browsing habits. It changed your homepage. It installed browser toolbars you didn't ask for. It served pop-up ads. It collected personal information from users, including children, without consent. And it did all of this while a cartoon gorilla smiled at you from the corner of your screen. The story of BonziBuddy is not a story about evil software engineers who set out to deceive the world. It is a story about a specific moment in internet history when nobody had figured out the rules yet, when "free software" didn't automatically trigger suspicion, and when the line between a helpful tool and an invasive one was so blurry that millions of people couldn't tell the difference. It is also a story about what happens when the Federal Trade Commission gets involved. ## Joe and Jay Bonzi Build a Desktop Companion BonziBuddy was created by two brothers, Joe and Jay Bonzi, and released in 1999 through their company, Bonzi Software Inc. The concept was straightforward. Microsoft had recently introduced Microsoft Agent, the technology that powered animated characters like Clippy (https://404memoryfound.com/posts/what-happened-to-clippy-microsoft.html) in Microsoft Office. Microsoft Agent provided a framework for creating interactive desktop characters with text-to-speech capabilities, predefined animations, and the ability to respond to user input. It was the same underlying technology behind the Office Assistant that everyone complained about, except Microsoft had made it available for third-party developers. The Bonzi brothers saw an opportunity. They built a desktop assistant using Microsoft Agent that could do things like search the web, manage downloads, tell jokes, and read email aloud. The initial version used Peedy, a green parrot character from Microsoft's default Agent library. However, Microsoft did not approve of their character being used in BonziBuddy. In May 2000, the software was updated with an original character: a purple gorilla named Bonzi. The gorilla was cartoonish, expressive, and designed to appeal to a broad audience. It waved. It danced. It spoke in a synthesized voice that sounded vaguely friendly. And it was completely free to download. The "free" part is critical. In 1999 and 2000, the concept of "if you're not paying for the product, you are the product" did not exist in the public vocabulary. Free software was just free software. People downloaded it because it didn't cost anything and it seemed fun. There was no widespread understanding that a free application might be monetizing your data or your attention in ways you hadn't agreed to. BonziBuddy exploited that gap between what users expected and what the software actually did. ## What BonziBuddy Actually Installed on Your Computer The advertised features of BonziBuddy were innocuous enough. It could perform web searches. It could read text aloud using Microsoft Agent's text-to-speech engine. It had a joke database. It could check the weather. For a certain kind of user, particularly children and less technically sophisticated adults, it felt like having a little digital friend living on your desktop. The unadvertised features were considerably less friendly. BonziBuddy installed additional software components without clear disclosure. It modified browser settings, changing homepages and default search engines. It tracked browsing activity and transmitted that data back to Bonzi Software's servers. It served pop-up advertisements. It installed browser helper objects, the Internet Explorer equivalent of browser extensions, that could monitor web activity and inject ads into pages. By 2002, Consumer Reports had examined BonziBuddy and classified it as spyware. The publication went further, stating that BonziBuddy functioned as a "Backdoor Santa," a term they used for software that collected information about users and transmitted it to third parties without adequate disclosure. Multiple antivirus vendors added BonziBuddy to their detection databases. The security community was unambiguous: this was not a benign desktop companion. The fundamental problem was one of consent, or rather, the absence of it. BonziBuddy's end-user license agreement technically disclosed some of its data collection practices. But the EULA was dense, legalistic, and written in a way that almost no one read. The gap between what users understood they were agreeing to ("I get a free talking gorilla") and what they were actually agreeing to ("I grant this software permission to monitor my browsing and serve me targeted advertisements") was enormous. ## The Banner Ads That Looked Like Windows Alerts If BonziBuddy's data collection was the quiet part, its advertising practices were the loud part. Bonzi Software ran banner advertisements across the web that were deliberately designed to look like Microsoft Windows system dialog boxes. These ads displayed messages like "Your IP address is being broadcast" or "Warning: Your computer may be at risk," complete with button styles that mimicked Windows interface elements. The ads were not warning you about anything real. They were designed to trigger alarm in users who didn't know the difference between an actual Windows security alert and a banner advertisement styled to look like one. Clicking the ad would redirect you to download BonziBuddy or one of Bonzi Software's other products. It was social engineering at scale, and it worked precisely because most internet users in 2001 and 2002 had no framework for distinguishing legitimate system messages from deceptive advertisements. On December 4, 2002, a class-action lawsuit was filed against Bonzi Software over these deceptive advertising practices. The suit accused the company of using its banner advertisements to mislead users into believing their computers were compromised, then directing them to download software under false pretenses. The case was settled on May 27, 2003. Bonzi Software agreed to modify its pop-up advertisements to reduce their resemblance to Microsoft Windows dialog boxes. Look: the settlement required them to make their ads look less like fake Windows alerts. That was the remedy. Not to stop advertising entirely. Not to disclose their data practices. Just to make the deception slightly less convincing. Which tells you something about the regulatory environment of the time. ## The FTC Steps In: COPPA and the $75,000 Fine The class-action lawsuit was one thing. What came next was worse for Bonzi Software, and more significant for the internet as a whole. On February 18, 2004, the Federal Trade Commission announced that Bonzi Software Inc. had agreed to pay $75,000 in civil penalties for violating the Children's Online Privacy Protection Act, commonly known as COPPA. The law, enacted in 1998 and effective since April 2000, required websites and online services that collected personal information from children under 13 to obtain verifiable parental consent before doing so. BonziBuddy had collected personal information, including email addresses and age data, from children without obtaining parental consent. The software's installation process and registration forms gathered this data from anyone who downloaded it, regardless of age. Given that BonziBuddy's friendly purple gorilla character was particularly appealing to children, a significant portion of its user base was under 13. The FTC case was historically significant for a specific reason: it was the first COPPA enforcement action that challenged the information collection practices of an online service in connection with a software product, as opposed to a website. Previous COPPA cases had targeted websites that collected children's data through registration forms and cookies. The BonziBuddy case established that COPPA applied equally to downloadable software that collected data during installation and use. That precedent matters. It extended children's privacy protections into a domain that hadn't been explicitly tested before. The $75,000 fine might sound small by today's standards, where tech companies sometimes pay billions in regulatory penalties. But in 2004, for a small software company, it was meaningful. More importantly, it was accompanied by a consent order that imposed ongoing compliance requirements on Bonzi Software. ## The Second FTC Settlement and the End The COPPA case was not the end of Bonzi Software's legal troubles. In September 2004, the FTC settled separate charges against Bonzi Software and its principals, Joe Bonzi and Jay Bonzi, individually. These charges related to unsubstantiated claims about the security and privacy protections of their software products. The company had marketed its software as protecting users' privacy and security, when in fact it was doing the opposite. By the time the second settlement was finalized, BonziBuddy was already effectively dead. The software had been discontinued in 2004. The combination of lawsuits, FTC enforcement actions, and the growing awareness among users and security researchers that BonziBuddy was spyware had made the product unsustainable. The purple gorilla disappeared from desktops, and Bonzi Software faded into the increasingly long list of early internet companies that didn't survive the transition from the Wild West era to the regulated one. ## What BonziBuddy Reveals About Early Internet Trust The BonziBuddy story is often told as a cautionary tale, and it is one. But it is worth examining why it succeeded for as long as it did, because the reasons are not flattering to any of the parties involved. Users trusted BonziBuddy because they had no reason not to. In 1999 and 2000, the concept of malware that disguised itself as legitimate software was not part of mainstream awareness. Viruses were things that corrupted your hard drive. Spyware, as a category, was barely a word. When a piece of software offered itself for free and had a friendly character and entertaining features, most people took it at face value. The idea that software could simultaneously entertain you and exploit you was, for the average user, genuinely novel. The technology industry failed users because it had not yet developed the norms and tools that would later become standard. Browsers did not warn you about potentially unwanted programs. Operating systems did not sandbox applications or limit their access to system resources. Antivirus software existed but was primarily designed to catch viruses, not adware. The entire ecosystem was built on the assumption that software installed by the user was software the user wanted, which was exactly the assumption that BonziBuddy exploited. Regulators were slow because the internet was new. COPPA had only been effective since April 2000. The FTC was still figuring out how to apply existing consumer protection frameworks to software distribution. The BonziBuddy COPPA case, filed in 2004, was enforcement against practices that had been occurring since at least 2000. Four years is a long time in internet years. Millions of children's data had already been collected by the time the government acted. This is essentially what happens every time a new technology platform emerges: there is a window between when exploitative practices become possible and when the legal and social frameworks catch up to prevent them. BonziBuddy existed in that window. So did early social media data harvesting. So did mobile app permission abuse. So did the first generation of smart TV data collection. The technology changes, but the pattern does not. ## The Purple Gorilla's Legacy BonziBuddy has become an internet meme. The purple gorilla appears in YouTube compilations, ironic nostalgia posts, and retro-internet humor alongside other artifacts of the early 2000s like dial-up sounds and "You've Got Mail (https://404memoryfound.com/posts/aol-youve-got-mail-voice-elwood-edwards.html)." There is something almost charming about BonziBuddy in retrospect, the same way people find old scam emails charming: the deception was so crude, so obvious by modern standards, that it has aged into comedy rather than outrage. But the underlying dynamic that BonziBuddy represented has not gone away. It has scaled. The modern internet runs on the same fundamental exchange that BonziBuddy pioneered: users receive a free service, and in return, their data and attention are monetized in ways they don't fully understand or meaningfully consent to. Facebook does not feature a purple gorilla. Google does not install browser toolbars without disclosure. But the core transaction, free services funded by surveillance-adjacent data practices, is the dominant business model of the consumer internet. The difference between BonziBuddy and a modern social media platform is not one of kind but of sophistication. BonziBuddy changed your homepage and served pop-up ads. Modern platforms build psychological profiles from your behavior and serve targeted content designed to maximize engagement. BonziBuddy collected email addresses from children. Modern platforms track children's online behavior across dozens of data points using methods that are technically legal under current frameworks. The gorilla was just more honest about being in your face. Joe and Jay Bonzi were not criminal masterminds. They were two brothers who built a piece of software, monetized it the way they knew how, and pushed the boundaries until regulators pushed back. They were not stupid. They just did not see what was coming. And the internet, as it tends to do, moved on to more sophisticated versions of the same game. There is one more detail worth noting. BonziBuddy was built on Microsoft Agent, which Microsoft itself eventually discontinued. The technology that powered Clippy, Merlin, Rover, and every other animated Office Assistant was quietly removed from Windows after Vista. Microsoft apparently decided that anthropomorphic desktop characters were more trouble than they were worth. BonziBuddy may have been the most extreme example of why. When your technology framework becomes best known for powering spyware, that tends to accelerate the deprecation timeline. ## Then vs Now: The Consent Problem Never Went Away In 2000, BonziBuddy collected browsing data from users who clicked through a license agreement they didn't read. In 2026, every major social media platform collects behavioral data from users who click through privacy policies they don't read. The mechanism is identical. The scale is different by orders of magnitude. BonziBuddy was fined $75,000 for collecting data from children. In 2019, YouTube was fined $170 million by the FTC for tracking children's viewing habits. In 2022, Epic Games paid $275 million for COPPA violations related to Fortnite. The numbers have gotten bigger. The underlying violation has not changed. Software companies collect data from children without proper consent, get caught, pay a fine, and the industry moves on. The most telling comparison is BonziBuddy's deceptive advertising. Those banner ads that mimicked Windows system alerts were considered so egregious that they generated a class-action lawsuit. Today, dark patterns, which are interface designs specifically crafted to trick users into actions they didn't intend, are everywhere. Cookie consent banners that make "Accept All" a bright button and hide "Reject" in tiny gray text. Subscription cancellation flows that require multiple clicks and guilt-tripping confirmation screens. App notifications designed to mimic urgent system messages. The deception has become more polished, but the principle is the same. BonziBuddy was crude. It was a purple gorilla that changed your homepage and served pop-ups. But it asked a question that the internet still has not answered satisfactorily: what does meaningful consent look like when the user and the software maker have fundamentally asymmetric information about what the software is actually doing? Joe and Jay Bonzi did not answer that question well. Neither has anyone else. ## Frequently Asked Questions ### What was BonziBuddy? BonziBuddy was a freeware desktop virtual assistant created by Joe and Jay Bonzi, released in 1999 by Bonzi Software Inc. It featured a purple gorilla character that could tell jokes, search the web, read email aloud, and manage downloads using Microsoft Agent text-to-speech technology. It was later classified as spyware and adware due to its data collection practices, browser modifications, and pop-up advertising. ### Why was BonziBuddy considered spyware? BonziBuddy tracked users' browsing habits, modified browser settings (including changing homepages and default search engines), installed browser helper objects without clear consent, served pop-up advertisements, and transmitted personal data back to Bonzi Software's servers. Consumer Reports classified it as spyware in 2002, and multiple antivirus vendors added it to their detection databases. ### What happened to Bonzi Software after the FTC fine? Bonzi Software Inc. was ordered to pay $75,000 in February 2004 for violating COPPA by collecting personal information from children under 13 without parental consent. In September 2004, the FTC settled additional charges against the company and its founders, Joe and Jay Bonzi, for making unsubstantiated claims about their software's security features. BonziBuddy was discontinued in 2004. ### Was BonziBuddy illegal? BonziBuddy's data collection practices violated COPPA, the Children's Online Privacy Protection Act, which requires parental consent before collecting personal information from children under 13. Its deceptive advertising practices, specifically banner ads designed to mimic Windows system alerts, led to a class-action lawsuit settled in 2003. While the software itself was not banned outright, the company faced multiple legal actions that effectively ended the product. ### How many people downloaded BonziBuddy? Exact download numbers for BonziBuddy have not been publicly verified, but the software was downloaded millions of times during its peak popularity between 2000 and 2002. Its free price point and appeal to children and less technically sophisticated users contributed to its wide distribution. The FTC's COPPA enforcement action confirmed that a significant number of users were children under 13. ### Is BonziBuddy still available? BonziBuddy was officially discontinued in 2004. The software is no longer available for download from legitimate sources. Various unofficial archives and emulators exist online, but running the original software on modern systems would require compatibility workarounds and is generally not recommended due to its documented spyware functionality. --- # The Original Sims: How a Virtual Dollhouse Became the Best-Selling PC Game Ever URL: https://404memoryfound.com/posts/original-sims-best-selling-pc-game.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-07 Topics: Gaming, Software & Apps ## The Game Nobody at Maxis Wanted to Make Picture this: 1993. Will Wright, the guy who created SimCity (https://404memoryfound.com/posts/what-happened-to-simcity-ea.html), one of the most successful PC games of all time, walks into a meeting at Maxis and pitches his next idea. It’s a game where you control a little digital person. You make them eat breakfast. You make them go to the bathroom. You watch them sleep. That’s it. That’s the pitch. The Maxis board looked at him like he’d lost his mind. They weren’t entirely wrong to be skeptical. This was the mid-90s. PC gaming was all about faster, louder, more polygons. Doom (https://404memoryfound.com/posts/how-doom-changed-pc-gaming-internet.html) had just blown the doors off the industry. Warcraft was turning real-time strategy into a phenomenon. And here was Will Wright, arguably the most respected designer in the business, saying he wanted to make a game about doing laundry. The project was internally called "Dollhouse." And that name tells you everything about why the suits at Maxis hated it. In a boardroom full of guys chasing the next shooter or strategy blockbuster, the word "dollhouse" was a death sentence. It sounded small, soft, and fundamentally unmarketable. Who would buy this? Who was the audience? What was the hook? Seven years later, The Sims would become the best-selling PC game of all time, move over 16 million copies of the base game alone, spawn a franchise worth over 200 million units sold, and attract an audience that was nearly 60% women in an industry that barely acknowledged female gamers existed. It changed everything. And it almost never happened. ## A House Fire Started It All The origin story of The Sims begins with a disaster. On October 20, 1991, the Oakland firestorm tore through the hills east of San Francisco, destroying over 3,000 homes and killing 25 people. One of those homes belonged to Will Wright. Wright lost everything. And then he had to rebuild. He had to deal with architects, pick out furniture, figure out floor plans, argue with contractors about where the kitchen should go. For most people, this would just be a miserable bureaucratic nightmare. For Will Wright, it was a game idea. He started thinking about the process of building and furnishing a home as a system. There were constraints. There was a budget. There were competing priorities. You had to balance aesthetics against function, space against cost. And then the people who lived in that house had their own needs, their own routines, their own personalities that clashed with each other and with the space itself. Wright had already proven with SimCity that complex systems could be turned into compelling games. SimCity let you build a city and watch it evolve. What if he zoomed all the way in? What if instead of managing a city of thousands, you managed one household? One family? The economics of a kitchen renovation. The social dynamics of roommates who didn’t get along. The mundane, relatable chaos of everyday domestic life. He started prototyping in 1993. The working title was "Dollhouse," and right from the beginning, the name was a problem. ## Project X and the Women Who Saved It The Maxis board didn’t just dislike the Dollhouse concept. They actively wanted it dead. The company was gearing up for an IPO in the mid-90s, and the last thing they needed was their star designer spending years on a game about making beds and cooking spaghetti. Maxis had already been burned by a string of underperforming titles. The gaming market was getting more expensive to compete in, and the company had no appetite for risk. Wright knew he had a perception problem. So he did something clever: he rebranded the project internally. "Dollhouse" became "Project X," and he pitched it as a "tactical domestic simulator." Which is both ridiculous and kind of brilliant, because it’s technically accurate. You are, in fact, running tactical operations in a domestic environment. You’re managing resources, optimizing schedules, making strategic decisions about furniture placement. Calling it that made it sound like something the board could take seriously. But the rebrand only bought him time. By the end of 1996, Maxis was in financial trouble. The company’s stock was sinking, layoffs were happening, and the board pulled the plug on Project X. Wright’s game was officially dead. Here’s where the story takes a turn that nobody expected. Electronic Arts, the company that would later become famous for buying studios and grinding them into dust, actually saved The Sims. EA had acquired Maxis in 1997 for about $125 million, primarily because they wanted SimCity. But in the process, they inherited Will Wright and his weird little cancelled project. The people at EA who championed the game were, notably, not the usual suspects. It wasn’t the guys in marketing who sold shooters. It was a group that included several women at the company who immediately understood what the Maxis board had missed: the game wasn’t about managing a house. It was about people. And people are inherently interesting to everyone, not just the 18-to-34 male demographic that the entire industry was obsessed with. ## February 4, 2000: Launch Day The Sims shipped on February 4, 2000. EA’s expectations were modest. This was a game that didn’t fit neatly into any genre. It wasn’t a strategy game, though it had strategy elements. It wasn’t a simulation in the traditional sense, though it simulated human behavior. There was no win condition. No final boss. No way to "beat" it. You just played, and kept playing, and things happened. It was the best-selling PC game of 2000. And 2001. And 2002. By March 2002, The Sims had sold over 6.3 million copies worldwide, surpassing Myst as the best-selling PC game in history. This was a record that Myst had held for years, and The Sims demolished it. By 2005, the base game alone had shipped over 16 million copies worldwide. But the sales numbers, as impressive as they were, weren’t the most remarkable thing about The Sims. The demographics were. ## The Game That Changed Who "Gamers" Were In 2000, the PC gaming audience was overwhelmingly male. Not exclusively, but the industry marketed almost entirely to men. Game covers featured soldiers, warriors, race cars, and explosions. The assumption, baked into every marketing budget and focus group, was that women didn’t really play games. At least not "real" games. Maybe Tetris. Maybe some puzzle games. But not the big releases. The Sims blew that assumption apart. Within months of launch, surveys showed that nearly 60% of Sims players were women. This wasn’t a niche finding. This was the best-selling PC game in the world, and the majority of its audience was a demographic that the industry had been ignoring for decades. Will Wright talked about this in interviews, and he was characteristically matter-of-fact about it. The game appealed to women because it was about social relationships, domestic life, and human behavior. These weren’t "women’s interests." They were universal interests that the gaming industry had simply never bothered to turn into a game before. The Sims didn’t discover a new market. It revealed an existing one that had been invisible. And those players weren’t casual about it. They played obsessively. They spent hours building houses, designing rooms, creating elaborate family dramas. The game had a modding community that exploded almost immediately, with players creating custom furniture, wallpaper, clothing, and character skins. This was before Steam Workshop, before modding was mainstream. Players were downloading .zip files from fan sites, extracting them into game folders, and troubleshooting file conflicts manually. And they loved it. ## The Expansion Pack Machine If The Sims base game was a hit, the expansion packs turned it into an empire. EA released seven expansion packs between 2000 and 2003, and every single one of them sold in the millions. The first, Livin’ Large, launched in August 2000, just six months after the base game. It added new career paths, items, and lot sizes, and it sold well enough to remain the sixth-highest-selling PC game in the United States through 2001, generating $22.9 million in domestic revenue that year. House Party followed in March 2001, adding social events and party mechanics. Hot Date came in November 2001 and introduced downtown areas and dating systems. Then came Vacation in 2002, Unleashed later that year (pets, which would become a franchise tradition), Superstar in 2003, and finally Makin’ Magic in October 2003. Each one layered new systems onto the base game, and the cumulative effect was a game that, by 2003, bore almost no resemblance to the relatively simple life sim that had launched three years earlier. The expansion pack model was nothing new. PC games had been releasing expansions for years. But The Sims perfected it in a way that nobody had before. Each pack felt less like a cash grab and more like a genuine evolution of the core experience. The players who had been building elaborate houses and creating complex family trees suddenly had new tools, new spaces, new possibilities. The game kept growing, and the audience kept growing with it. By the time the final expansion pack dropped, The Sims and its add-ons had combined to sell tens of millions of units. More importantly, they had trained an audience to expect ongoing content for a single game. This model, releasing a base game and then selling modular add-ons over several years, would become the template for how EA (and eventually the entire industry) monetized franchises going forward. ## The Pool Ladder and the Dark Side of Suburbia You can’t write about The Sims without talking about the pool ladder. It’s become one of the most enduring memes in gaming history, and if you played the original game, you know exactly what I’m talking about. Here’s what would happen: you’d build your Sim a nice house. Put a swimming pool in the backyard. Your Sim would go for a swim. And then, while they were in the pool, you’d delete the ladder. The Sim couldn’t get out. They’d swim around in circles, getting more and more tired, until eventually they drowned. You just killed someone with architecture. This was never an intended feature. It was a consequence of the game’s AI. Sims couldn’t climb out of pools without ladders because the pathfinding system required a specific exit point. When you removed it, the Sim was trapped. And players discovered this almost immediately and turned it into a ritual. The pool ladder thing says something fascinating about what The Sims actually was. On the surface, it was a cheerful domestic simulator with bright colors and quirky music. Underneath, it was a sandbox for human psychology. Players didn’t just build happy homes. They built miserable ones. They trapped Sims in rooms with no doors. They removed toilets and watched the consequences. They created elaborate torture chambers disguised as suburban houses. Will Wright understood this completely. In interviews, he talked about how the game was really a platform for storytelling, and stories need conflict. The ability to be cruel was just as important as the ability to be kind. The Sims gave players power over digital lives, and players immediately explored every dimension of that power, including the dark ones. This wasn’t something EA marketed, obviously. But it was fundamental to why the game worked. The Sims was never really a game about happiness. It was a game about control. And that turns out to be something that almost everybody finds irresistible. ## The Sims 2 and the Franchise Takes Off The Sims 2 launched on September 14, 2004, and it made the original look like a tech demo. Full 3D graphics replaced the isometric view. Sims now aged, going from babies to elders over the course of a game. Genetics meant that children actually resembled their parents. Aspirations gave Sims individual goals and desires. The emotional depth went from "happy or sad" to a complex web of wants, fears, and memories. It sold over one million copies in its first ten days, making it the fastest-selling PC game at that point in history. The franchise was now a blockbuster on the level of the biggest console titles, except it existed almost entirely on PC, a platform that the mainstream gaming press was starting to write off in favor of the PlayStation 2 (https://404memoryfound.com/posts/playstation-2-best-selling-console-ever.html) and Xbox. The Sims 2 eventually received eight expansion packs and nine "stuff packs" (smaller add-ons focused on specific item categories), establishing a monetization cadence that EA would refine with each subsequent generation. The game also came to consoles for the first time in a meaningful way, with versions on PlayStation 2, Xbox, Nintendo DS, and Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html) Advance. ## What The Sims Got Right That Nobody Else Could Copy The gaming industry has tried, repeatedly, to replicate what The Sims did. And almost nobody has succeeded. There have been life simulation games, virtual dollhouse games, people management games. None of them have come close to the cultural impact or commercial success of The Sims. The reason is that Will Wright understood something about game design that most developers miss. The Sims wasn’t really a game about Sims. It was a game about you. Every house you built was a reflection of your taste. Every family you created was a projection of your relationships, your fantasies, your anxieties. The pool you filled with water and then took the ladder out of was a mirror. The game gave you a set of extremely flexible tools and then got out of the way. There were no scripted storylines. No cutscenes telling you what to feel. No objectives beyond the ones you set for yourself. This was radical in 2000, and honestly, it’s still rare today. Most games are terrified of giving the player that much freedom, because freedom means some players will do "wrong" things, or boring things, or cruel things. The Sims embraced all of it. The other thing Wright got right was the mundane. Before The Sims, nobody would have believed that watching a digital character make a sandwich could be compelling. But it was, because the game turned every mundane action into a potential story beat. Your Sim burns the sandwich. The kitchen catches fire. The fire department comes. Your Sim’s roommate falls in love with the firefighter. Suddenly you have a narrative, and it emerged entirely from the game’s systems interacting with each other. This emergent storytelling is incredibly hard to design. It requires systems that are simple enough to be readable but complex enough to surprise. The Sims nailed it on the first try, and two decades later, it’s still the gold standard. ## 200 Million Units and Counting The Sims franchise has now sold over 200 million copies across all titles and platforms. The Sims 4, released in 2014, has been the longest-running entry in the series and went free-to-play in 2022. The franchise generates hundreds of millions in annual revenue through expansion packs, game packs, stuff packs, and kits. But the legacy of The Sims goes beyond sales numbers. It proved that games didn’t have to be about combat to be massive. It proved that female gamers were a real, enormous, lucrative market. It proved that mundane subject matter could be endlessly compelling if the systems were good enough. And it proved that a weird little project about making digital people eat breakfast, the one that everybody at Maxis wanted to kill, was the most commercially important PC game ever made. Will Wright left EA in 2009. Maxis, the studio he co-founded, was effectively shut down in 2015 after the disastrous launch of SimCity (2013). The Sims franchise continues under EA’s internal teams, and the community remains passionate, vocal, and creative. Not bad for a game they called "Dollhouse." ## Frequently Asked Questions When was the original Sims released? The Sims was released on February 4, 2000, for Microsoft Windows. It was developed by Maxis and published by Electronic Arts. How many copies did The Sims sell? The original Sims base game sold over 11.3 million copies. The entire franchise has sold over 200 million copies across all platforms and sequels. Why did Maxis reject The Sims initially? The Maxis board didn’t believe a domestic life simulator would sell, especially during a period when the company was preparing for an IPO and avoiding risky projects. The game was internally called "Dollhouse," which didn’t help its perception. Why was The Sims so popular with women? Nearly 60% of Sims players were women. The game focused on social relationships, home design, and human behavior, topics that appealed to a broad audience the gaming industry had largely ignored. What is the pool ladder meme about? In the original Sims, players discovered they could trap Sims in swimming pools by removing the exit ladder, causing the Sim to drown. This became one of gaming’s most iconic dark humor memes. How many expansion packs did the original Sims have? The original Sims had seven expansion packs: Livin’ Large, House Party, Hot Date, Vacation, Unleashed, Superstar, and Makin’ Magic, released between 2000 and 2003. Is Will Wright still involved with The Sims? No. Will Wright left Electronic Arts in 2009. The Sims franchise is now managed by EA’s internal development teams. --- # What Happened to Polaroid, the Camera That Died and Came Back URL: https://404memoryfound.com/posts/what-happened-to-polaroid-camera.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-07 Topics: Hardware, Then vs Now ## A Three-Year-Old Asked a Question. It Created a $3 Billion Company. In 1943, Edwin Land was on vacation in Santa Fe, New Mexico, with his three-year-old daughter Jennifer. He took her photograph with a conventional camera. She asked why she couldn't see the picture right away. Land, who was already one of the most prolific inventors in America with patents in light polarization technology, spent the next hour walking around Santa Fe working out the chemistry, optics, and mechanics of instant photography in his head. Four years later, in 1947, he demonstrated the first working instant camera to the Optical Society of America. The following year, in 1948, the Polaroid Model 95 went on sale at a Jordan Marsh department store in Boston. It sold out on the first day. The story of Polaroid is one of the cleanest arcs in corporate history. A genuine technological breakthrough. Decades of dominance. A peak so high it seemed permanent. Then a refusal to adapt, two bankruptcies, and near-total erasure. And then, against every reasonable expectation, a resurrection led by people who had no business pulling it off. ## The Invention That Changed Photography To understand what made Polaroid special, you have to understand what photography was like before instant film. In the 1940s, taking a photograph was an act of faith. You pressed the shutter button, wound the film, and then waited. The roll sat in the camera until it was full, which could take days or weeks. Then you brought it to a drugstore or a photo lab, waited another few days, and finally saw your pictures. If you'd held the camera wrong, or the exposure was off, or someone blinked, you found out a week later. There was no feedback loop. Land's system eliminated all of that. You took the picture, pulled a tab, waited 60 seconds, and peeled apart the print. There was your image. Instantly. The chemistry was astonishingly complex, involving multiple layers of reagents, dyes, and timing mechanisms that had to work in precise sequence at room temperature. Land held over 500 patents, second only to Thomas Edison at the time. He wasn't just an inventor. He was a one-man R&D department. The Polaroid camera became a cultural fixture almost immediately. It showed up at birthday parties, family reunions, and holidays. It was the camera you used when you wanted to see the picture now, when the moment mattered more than the quality. Professional photographers like Ansel Adams, Andy Warhol, and Walker Evans used SX-70 cameras for their work. Warhol, in particular, became famous for his Polaroid portraits, using the medium's imperfections as an aesthetic tool. ## The SX-70 and the Peak Years Polaroid's most important product was the SX-70, introduced in 1972. It was the first camera to produce a self-developing print that didn't require peeling, timing, or coating. You pressed the button, the camera ejected a white square, and over the next few minutes the image materialized in front of you. It was, as Land described it to shareholders, "absolute one-step photography." The SX-70 was also a design masterpiece. It folded flat, small enough to fit in a coat pocket. The leather and chrome construction made it look more like a luxury accessory than a gadget. Land demonstrated it on stage at the 1972 annual meeting by pulling a folded camera from his suit jacket, unfolding it, and snapping five photos in ten seconds. Both actions had been impossible with previous Land cameras. The technology was expensive to develop. Estimates put the total R&D cost for the SX-70 system at around $600 million, an enormous sum for the early 1970s. But the bet paid off. Instant camera sales across the industry peaked at about 13 million units in 1978, with Polaroid controlling roughly two-thirds of the U.S. market. By the early 1990s, Polaroid's annual revenue hit $3 billion. During this period, Polaroid was more than a camera company. It was a technology institution. Land ran it like a research lab that happened to sell consumer products. Engineers and chemists were given extraordinary freedom. The company attracted talent from MIT and Harvard. The corporate culture was closer to Bell Labs than to Kodak (https://404memoryfound.com/posts/what-happened-to-kodak-digital-camera-bankruptcy.html). ## Polavision and the First Crack The first real sign of trouble came in 1977, when Land introduced Polavision, an instant movie system. The logic seemed sound: if people loved instant photographs, they'd love instant movies. But the timing was catastrophic. Sony's Betamax (https://404memoryfound.com/posts/betamax-vs-vhs-format-war.html) had launched in 1975, and VHS arrived in 1976. Home video was already a thing, and it was recordable, reusable, and played on your television. Polavision produced silent, 2.5-minute clips on proprietary film cartridges that could only be viewed on a special player. The picture quality was poor. The cost was high. And the world had already moved on. Sony founder Akio Morita reportedly warned Land directly that Polavision was arriving a decade too late. Land pushed forward anyway. The result was a $68.5 million writeoff when the unsold inventory was liquidated in 1979. It was Polaroid's first major commercial failure, and it shook the company. More importantly, it shook confidence in Land. After four decades as chairman, Edwin Land was pressured into resigning as CEO in 1980. He left the company entirely in 1982 and died in 1991. With Land gone, Polaroid lost more than its founder. It lost the person whose instinct for breakthrough technology had defined the company's identity. What remained was a $3 billion revenue machine that was very good at selling instant film and very uncertain about what to do next. ## The Digital Question Nobody Wanted to Answer Here's the part of the story that frustrates engineers and business school professors in equal measure. Polaroid saw digital photography coming. Not vaguely, not theoretically. They saw it in detail, years before it arrived. Polaroid's own engineers built a working digital camera prototype in 1990. The company held significant patents in digital imaging technology. Internal reports from the early 1990s explicitly outlined the trajectory: digital sensors were getting cheaper, storage was getting denser, and within a decade or two, chemical film would face serious competition from electronic alternatives. The company's response was to do essentially nothing. According to former Polaroid vice president Sheldon Buckler, the problem wasn't ignorance. It was identity. "From his point of view, instant photography was going to be his legacy," Buckler said of Land's influence on the company culture. The idea that some fancy new technology from the physics side was going to displace the creation rooted in chemistry was not something the organization was prepared to accept. Even after Land left, his philosophy permeated the company. Polaroid was a chemistry company. Digital was physics. Those were different worlds. There was also a financial logic to the inaction, at least in the short term. Polaroid's business model was built on the razor-and-blade principle. Cameras were sold near cost, and the real money came from film cartridges. Every time someone pressed the shutter button, Polaroid made money. Digital photography would eliminate the recurring revenue entirely. Going digital meant cannibalizing the company's most profitable product line to compete in a market where the margins were lower and the competition was fiercer. This is a pattern that shows up repeatedly in corporate history. The company that dominates a technology often can't bring itself to kill it, even when the replacement is obvious. Kodak had the same problem with digital photography. Blockbuster (https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html) had it with streaming. BlackBerry (https://404memoryfound.com/posts/what-happened-to-blackberry-smartphone.html) had it with touchscreens. Polaroid was arguably the earliest and cleanest example of what Clayton Christensen would later formalize as "the innovator's dilemma." ## Bankruptcy No. 1: October 2001 By the late 1990s, the numbers were ugly. Instant film sales had been declining for years. Digital cameras were getting good enough and cheap enough to compete. Polaroid tried to diversify, launching digital cameras and even a line of sunglasses (a callback to the company's original polarization technology), but none of it gained traction. The company was carrying significant debt and burning through cash. On October 11, 2001, Polaroid Corporation filed for Chapter 11 bankruptcy protection. The timing was brutal: exactly one month after September 11th, when financial markets were already in freefall and nobody was in the mood to rescue a struggling camera company. Within ten months, the business was sold off. Bank One's One Equity Partners acquired the Polaroid name and most of its assets. The company that Edwin Land had built from a polarization filter patent in 1937 was now a brand name owned by a private equity firm. ## The Brand Becomes a Zombie What followed was a decade of corporate indignity. The Polaroid brand changed hands multiple times, each new owner licensing the name onto products that had nothing to do with instant photography. Polaroid-branded flat-screen TVs. Polaroid-branded DVD players. Polaroid-branded tablet computers. The name was slapped onto cheap consumer electronics manufactured in China, a ghost of the company that used to employ thousands of chemists and optical engineers. The most cynical episode came in December 2008, when the post-reorganization Polaroid Corporation filed for bankruptcy a second time. Its CEO, Tom Petters, was later convicted of running a $3.65 billion Ponzi scheme that had used Polaroid as one of its vehicles. The brand that once represented American ingenuity was now associated with financial fraud. Meanwhile, something strange was happening on the ground floor. Instant film enthusiasts, the people who had been shooting with Polaroid cameras for decades and now faced a future without film, were getting organized. The cameras still worked. The film was the problem. If someone could make new film, the cameras could live again. ## The Impossible Project In June 2008, two men met at the closing event for Polaroid's last remaining film factory in Enschede, the Netherlands. Florian Kaps was an Austrian instant photography enthusiast who ran a website called Polanoid.net. Andre Bosman was the factory's production manager who had spent his career making Polaroid film. Together, they hatched a plan that most people in the photography industry considered absurd. They were going to buy the factory's equipment and start making instant film themselves. In October 2008, they founded The Impossible Project and purchased Polaroid's remaining production machinery for $3.1 million. They leased the north building of the old factory. And then they ran into a wall. The machines were there, but the chemical formulas were not. Polaroid's proprietary film chemistry had been closely guarded, and key suppliers had gone out of business. The original film contained over 20 chemical layers, each engineered to react in a specific sequence. Recreating this from scratch was, to put it mildly, a challenge. It took two years. In 2010, The Impossible Project released its first films: black-and-white and color cartridges for SX-70 and 600-series cameras. The early batches were imperfect. Colors were muted, development times were long, and the results were unpredictable. But they worked. And for the community of instant photography devotees who had been hoarding expired Polaroid film in their refrigerators, imperfect new film was infinitely better than no film at all. ## From Impossible to Polaroid Again The Impossible Project spent the next several years refining its film chemistry and expanding its product line. Each generation of film got closer to the quality of the original Polaroid stock. The company also started making its own cameras, bridging the gap for people who didn't want to hunt for vintage hardware on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html). In September 2017, The Impossible Project's largest shareholder, Polish billionaire Wiaczeslaw Smolokowski, acquired the Polaroid brand itself. The Impossible Project rebranded as Polaroid Originals and released the OneStep 2, a new instant camera using a modified "i-Type" film that was more affordable to produce (it dropped the battery from each cartridge, a cost-saving measure the original SX-70 system never attempted). In March 2020, the name simplified further, from Polaroid Originals to just Polaroid. The circle was complete. A group of enthusiasts had bought a dead factory, reverse-engineered the chemistry, built a company from nothing, and eventually purchased the brand itself. The Impossible Project didn't just save Polaroid film. It became Polaroid. ## The Gen Z Revival Nobody Predicted The final twist in the Polaroid story is the one that would have surprised Edwin Land the most. Instant photography is growing again, and the people driving that growth are teenagers and twenty-somethings who grew up in a completely digital world. The appeal is counterintuitive. In an era where everyone has a 48-megapixel camera in their pocket and can take unlimited photos for free, why would anyone pay roughly $1 per shot for a low-resolution image on a small piece of cardboard? The answer has nothing to do with image quality and everything to do with what a photograph means. A digital photo lives on a phone. It gets buried in a camera roll of thousands. It might get posted to Instagram, where it joins a feed that scrolls past in seconds. An instant print is a physical object. It exists in exactly one copy. You can hold it, pin it to a wall, hand it to someone. There's a ritual to the process: press the button, hear the mechanical whir, watch the print slide out, shake it (even though you're not supposed to), wait for the image to appear. In a world of infinite, disposable digital images, the scarcity and tangibility of instant film is precisely what makes it feel special. Fujifilm's Instax line has been the biggest commercial beneficiary of this trend, selling millions of cameras annually. But Polaroid, now under its new ownership, has carved out a meaningful position as the premium, heritage option. The brand carries a cultural weight that Fujifilm can't replicate. When someone says "take a Polaroid," they're not talking about a product. They're talking about a type of moment. ## What Polaroid's Story Actually Teaches The standard business school reading of Polaroid is a cautionary tale about failing to adapt. And that reading isn't wrong. The company absolutely failed to transition from analog to digital, and that failure was fatal. But the story is more nuanced than a simple "they didn't innovate" narrative. Polaroid innovated constantly. The SX-70 was one of the most ambitious consumer products ever engineered. The company held hundreds of patents. Its R&D spending was enormous. The problem wasn't a lack of innovation. It was an inability to innovate in the direction that mattered. Polaroid kept making better instant film while the world was moving beyond film entirely. The resurrection is equally instructive. The Impossible Project succeeded not because it had better technology or more money (it had neither). It succeeded because it understood something that the private equity firms and licensors who owned the Polaroid brand in the mid-2000s did not: the value of Polaroid was never in the brand name on a flat-screen TV. It was in the experience of watching a photograph appear in your hands. That experience was the product. Everything else was packaging. Edwin Land, who spent his entire career chasing the magic of instant chemistry, would probably have appreciated that conclusion. The three-year-old who asked "why can't I see it now?" got her answer. And 80 years later, people are still asking the same question, still paying for the privilege of watching a photograph develop in real time, still choosing the imperfect and physical over the perfect and digital. Some products don't die because the need they serve isn't really about the technology. It's about the moment. ## Frequently Asked Questions When did Polaroid go bankrupt? Polaroid filed for Chapter 11 bankruptcy twice: first on October 11, 2001, and again in December 2008. The second bankruptcy was entangled with the Tom Petters Ponzi scheme. What was the Impossible Project? The Impossible Project was founded in October 2008 by Florian Kaps and Andre Bosman. They purchased Polaroid's last film factory equipment in Enschede, Netherlands for $3.1 million and spent two years reverse-engineering instant film production. They eventually acquired the Polaroid brand in 2017. Why did Polaroid fail? Polaroid failed primarily because it couldn't transition from chemical instant film to digital photography. Despite having digital camera prototypes as early as 1990, the company chose to protect its profitable film business rather than cannibalize it with digital products. Is Polaroid still making cameras today? Yes. After The Impossible Project acquired the Polaroid brand in 2017 (rebranding to Polaroid Originals, then simply Polaroid in 2020), the company produces both instant cameras and film. The Enschede factory in the Netherlands remains the only facility on Earth producing Polaroid film. What was Polaroid's peak revenue? Polaroid's annual revenue peaked at approximately $3 billion in 1991. Who invented the Polaroid camera? Edwin Land invented the Polaroid instant camera, demonstrating the technology in 1947 and releasing the first commercial model (the Model 95) in 1948. Land held over 500 patents and co-founded the Polaroid Corporation in 1937. Why are Polaroid cameras popular again? Instant photography has seen a resurgence, particularly among younger generations, because the physical, one-of-a-kind nature of instant prints feels meaningful in a world of infinite digital photos. The ritual of taking and developing an instant photo offers an experience that smartphone cameras cannot replicate. --- # What Happened to Iomega Zip Drives and the Click of Death URL: https://404memoryfound.com/posts/what-happened-to-iomega-zip-drives.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-07 Topics: Hardware, Then vs Now ## You Used to Be Able to Hold Your Entire Life on One Disk Picture this: 1995. You're standing in a CompUSA (https://404memoryfound.com/posts/is-compusa-still-around.html), staring at a wall of beige peripherals, and a store employee hands you something that looks like a chunky floppy disk on steroids. "A hundred megabytes," he says, like he's letting you in on a secret. A hundred megabytes. On a single removable disk. For context, a standard floppy held 1.44 megabytes. This thing held almost seventy times that. You could fit entire folders of Word documents, a few small programs, maybe even a handful of MP3s if you were already on that wave. It cost $199 for the drive and about $20 per disk, and it felt like the future had just shown up at your local electronics store. The Iomega Zip drive launched in March 1995, and it didn't just sell well. It sold out. First day. Gone. Every unit. The thing moved so fast that Iomega, a company that had been struggling in the early '90s with a stock price hovering around $2 a share, suddenly couldn't manufacture enough of them. And the press went wild. Every computer magazine on the shelf was running reviews, and every review said basically the same thing: this changes everything. They weren't wrong. Not yet, anyway. ## From Two Bucks to Seven Billion Here's where the Iomega story gets genuinely insane. Before the Zip drive, this was a company on life support. Founded in 1980 in Roy, Utah, Iomega had been making Bernoulli drives, which were removable storage devices that cost thousands of dollars and never really caught on outside of professional settings. The original Bernoulli drive ran about $2,700. That's not a typo. The company had talent and technology, but no product that regular people could afford or wanted to buy. Then the Zip drive happened, and everything changed overnight. Revenue hit $362 million in 1995, the Zip drive's launch year. Then it exploded to $1.2 billion in 1996. Let that sink in. They more than tripled revenue in a single year. The stock price experienced a 2,135% surge by 1996, rocketing from around $2 to well over $100 per share. At its peak, Iomega was valued at nearly $7 billion. Seven. Billion. For a company that made removable storage disks. Wall Street lost its mind. Individual investors piled in. There was a whole online community of Iomega stock boosters who called themselves "Iomegans" and traded tips on early internet message boards. This was the mid-'90s version of a meme stock, except nobody had that vocabulary yet. People were taking out loans to buy Iomega shares. Kim Edwards, the CEO who had arrived in January 1994 and greenlit the Zip drive project, was being talked about like a tech visionary. And honestly, the product deserved a lot of the hype. If you were alive and using a computer in 1996 or 1997, you probably had a Zip drive, or you knew someone who did. Graphic designers needed them. College students needed them. Offices ran on them. The 100MB Zip disk became the default way to move large files from one computer to another, because email attachments were still tiny and burning CDs required expensive equipment. Iomega eventually released 250MB and 750MB versions, but the original 100MB drive was the one that mattered. That was the one that changed the game. ## The Sound Nobody Wanted to Hear And then people started hearing it. A rhythmic clicking noise coming from their Zip drives. Click. Click. Click. Not the normal sounds of a drive reading a disk. Something else. Something mechanical and wrong. They called it the Click of Death, and it became one of the most infamous hardware defects in personal computing history. Here's what was happening: the read/write heads inside the Zip drive were becoming misaligned. When that happened, the heads would bounce off their end stops, producing that distinctive clicking sound. A misaligned drive couldn't read your disk properly. But here's the really nasty part: a bad drive could actually damage a good disk. And then that damaged disk could go on to misalign the heads in the next drive you put it in. It was a chain reaction. A contagious hardware failure. The technical root cause was almost painfully preventable. Inside earlier Zip drives, there was a small doughnut-shaped foam washer at the end of the thin steel bearing that the actuator arm slid on. This washer acted as a cushion, protecting the heads from slamming into the end stops. At some point during manufacturing, someone at Iomega decided to cut costs by removing this part. It was a tiny component. Probably saved pennies per unit. And it eventually helped destroy the company. Dust was also a factor. Zip disks weren't sealed like hard drives. Tiny particles could get inside and cause the heads to misalign. Magnetic fields from nearby speakers or monitors could do it too, since the drives lacked internal shielding. But the removed foam washer was the big one. That was the decision that turned a normal engineering vulnerability into a widespread catastrophe. ## The Internet Turns Against Iomega Now, this was the late '90s, and the internet was just becoming a place where angry consumers could organize. A guy named Steve Gibson, who ran a website called Gibson Research Corporation, became the unofficial chronicler of the Click of Death. He set up a dedicated page documenting the problem, collecting user reports, and publishing technical analysis of what was going wrong inside the drives. His site became a rallying point for frustrated Zip drive owners who felt like Iomega was ignoring the problem. And Iomega was, for a while, ignoring the problem. Or at least downplaying it. The company's official position was that the Click of Death affected only a small percentage of drives. They offered replacement drives under warranty but didn't acknowledge any systemic defect. For users who had lost important data, irreplaceable files, college theses, graphic design work, family photos that existed nowhere else, "a small percentage" wasn't much comfort. The frustration built. Message boards lit up. Gibson's site got millions of hits. And in September 1998, a class action lawsuit was filed against Iomega in federal court, alleging violations of the Delaware Consumer Fraud Act and claiming that the company had known about the defect and failed to warn consumers. The case was settled in March 2001. The settlement wasn't exactly a windfall for affected users. Iomega agreed to provide a free 24/7 troubleshooting hotline and offered rebates of up to $40 on future Iomega products. If your Zip drive ate your thesis, you got a coupon. That was the deal. ## The World Moved On Without Them Even without the Click of Death, Iomega was running into a wall. The same year the Zip drive launched, CD-R drives were starting to drop in price. By 1998, CD-RW drives, the ones that could burn rewritable CDs, were becoming affordable for regular consumers. A blank CD-R held 650 to 700 megabytes, cost well under a dollar, and worked in any CD-ROM drive. The Zip disk held 100MB and cost $10 to $15. The math stopped working pretty quickly. Then USB flash drives showed up around 2000 and 2001, and it was over. Smaller, faster, no moving parts, no special drive required, just plug it into any USB port. The first commercial USB flash drives held 8MB, which wasn't much, but by 2003 you could get a 256MB drive for about $50. By 2005, 1GB drives were common and cheap. The Zip drive's entire reason for existing had been replaced by something that fit on your keychain. Iomega tried to adapt. They released the Zip 250 in 1999 and the Zip 750 in 2002. They branched into network-attached storage and external hard drives under the "Rev" brand. None of it recaptured the magic. The stock price that had peaked above $100 fell to around $1 per share by the early 2000s. The company that had been worth $7 billion was now worth almost nothing. In April 2008, EMC Corporation, one of the biggest names in enterprise storage, announced it would acquire Iomega for $213 million. For a company that had once been valued at $7 billion, $213 million felt like a clearance sale. EMC eventually folded the Iomega brand into its LenovoEMC joint venture, and by 2013 the name was essentially gone. ## Why the Zip Drive Still Matters I still have a Zip drive in a box in my closet. An external USB model, translucent blue, from around 1999. I don't have anything to put in it. I don't have any Zip disks anymore. But I can't bring myself to throw it away, because holding it takes me back to a very specific moment in computing history when portable storage felt exciting. When you could hand someone a disk and say, "everything's on there," and it actually meant something. The Zip drive's story is really a story about timing. Iomega found a gap in the market, a brief window between the floppy disk's decline and the arrival of cheap CD burning and USB storage, and they drove a truck through it. For about three years, they owned that gap completely. They made a product that people genuinely needed and genuinely loved. And then they let a cost-cutting manufacturing decision undermine consumer trust at the exact moment when the technology landscape was shifting under their feet. The Click of Death didn't kill Iomega by itself. CD-RW drives and USB flash drives would have eventually made the Zip drive obsolete no matter what. But the defect accelerated the decline, poisoned the brand at its most vulnerable moment, and turned what could have been a graceful transition into a collapse. Iomega had the chance to become a major storage company, to evolve alongside the technology. Instead, they became a cautionary tale about what happens when you cut the wrong corner. If you ever owned a Zip drive, you remember two things: how cool it felt to carry 100MB in your pocket, and the fear of hearing that click. ## Frequently Asked Questions When did the Iomega Zip drive come out? The Zip drive launched in March 1995 with a retail price of $199 for the drive and approximately $20 per 100MB disk. It sold out on its first day. What was the Click of Death? The Click of Death was a well-documented hardware defect in Zip drives where the read/write heads became misaligned, producing a distinctive clicking noise. A defective drive could damage disks, and those damaged disks could then damage other drives, creating a chain reaction of failures. Why did Zip drives fail? Zip drives declined due to a combination of the Click of Death defect, which eroded consumer trust, and competition from cheaper, higher-capacity alternatives like CD-RW drives and USB flash drives that emerged in the late 1990s and early 2000s. What happened to Iomega as a company? After its stock price fell from over $100 to around $1 per share, Iomega was acquired by EMC Corporation in 2008 for $213 million. The brand was eventually folded into EMC's product line and discontinued. How much could a Zip disk hold? The original Zip disk held 100MB. Later versions included 250MB (released 1999) and 750MB (released 2002) capacities, though neither achieved the same market success as the original. --- # What Happened to Kazaa and the P2P Wars After Napster URL: https://404memoryfound.com/posts/what-happened-to-kazaa-p2p-wars.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-07 Topics: Software & Apps, Internet Culture ## The Most Downloaded Software You Probably Broke the Law With In May 2003, Kazaa Media Desktop broke a record. It became the single most downloaded piece of software in the world, as tracked by CNET's Download.com, surpassing the previous record holder, ICQ (https://404memoryfound.com/posts/what-happened-to-icq-instant-messenger.html), with 229,513,316 cumulative downloads. At its peak, the network had roughly four million simultaneous users sharing files at any given moment. To put that in perspective, Netflix wouldn't reach four million subscribers until 2005. Kazaa wasn't a niche tool for power users. It was mainstream, global, and running on millions of desktops in dorm rooms, bedrooms, and home offices across the world. And yet, almost nobody talks about Kazaa anymore. It exists in a strange historical blind spot, overshadowed by Napster (https://404memoryfound.com/posts/who-owns-napster-now.html)'s dramatic origin story on one side and BitTorrent's technical elegance on the other. Kazaa was the bridge between those two eras, the application that proved peer-to-peer file sharing could survive a legal shutdown and come back stronger. It was also the application that proved the music industry (https://404memoryfound.com/posts/napster-destroyed-music.html) would spend any amount of money to make sure that didn't happen again. The story of Kazaa is really the story of the P2P wars: a five-year stretch from roughly 2001 to 2006 when the recording industry, the courts, and millions of ordinary internet users fought over who owned digital music. The outcome shaped everything that came after, from iTunes to Spotify to the way copyright law works online today. ## Built in Estonia, Launched in the Netherlands, Sued in Australia Kazaa's origin story reads like the setup for a techno-thriller that nobody would believe. The core technology, the FastTrack peer-to-peer protocol, was built by three Estonian programmers: Jaan Tallinn, Ahti Heinla, and Priit Kasesalu, working at a company called BlueMoon Interactive that they had co-founded in 1993. These were serious engineers. Tallinn, in particular, would go on to become one of the founding engineers of Skype (https://404memoryfound.com/posts/skype-shut-down-what-replaced-it.html) and later a prominent figure in artificial intelligence safety research. But in 2000, their immediate project was building a file-sharing protocol that could do what Napster did without Napster's fatal weakness. Napster's problem was architectural. Every search query went through Napster's central servers. When a court ordered those servers shut down in July 2001, the whole network went dark. FastTrack solved this with a concept called "supernodes." Instead of routing all traffic through a central server, the protocol designated certain high-bandwidth computers on the network as indexing points. These supernodes maintained searchable databases of nearby files, and queries were distributed across them. There was no single point of failure. You couldn't kill the network by pulling one plug. The Estonian programmers sold the technology to Niklas Zennstrom, a Swedish entrepreneur, and Janus Friis, a Danish programmer. These two would become recurring characters in early 2000s tech history. After Kazaa, they went on to create Skype in 2003, which used a similar P2P architecture for voice calls, and which Microsoft would eventually acquire for $8.5 billion in 2011. Zennstrom and Friis launched Kazaa through a Dutch company called Consumer Empowerment in March 2001. The timing was almost comically perfect. Napster was in the process of being shut down by court order. Millions of users who had gotten used to free music were suddenly looking for somewhere else to go. Kazaa was right there, ready, and technically superior. ## How Kazaa Actually Worked For anyone who never used it, here's what the experience was like. You downloaded Kazaa Media Desktop from the website. The installer was bundled with a significant amount of adware and, depending on the version, varying degrees of spyware. This wasn't a secret. It was the business model. Kazaa was free to use because it made money by installing advertising software on your computer. For most users in 2002, this seemed like a reasonable trade. Free music in exchange for some pop-up ads. The calculus would shift later. Once installed, you typed what you wanted into a search bar, and results appeared from across the FastTrack network: MP3s, movie files, software, documents, whatever other users were sharing from their hard drives. You clicked download, and the file transferred directly from their computer to yours, often from multiple sources simultaneously to speed up the process. Download speeds varied wildly depending on the era. In 2001, most people were still on dial-up, so a single MP3 might take twenty minutes. By 2003, broadband adoption was accelerating, and you could grab an album in under an hour. The quality of what you downloaded was unpredictable. A file labeled "Eminem - Lose Yourself.mp3" might actually be that song, or it might be a low-bitrate recording, or it might be a completely different track, or it might be a virus. The Recording Industry Association of America, the RIAA, eventually exploited this by flooding the network with corrupted or fake files as a disruption tactic. Searching for popular songs would return dozens of results, and sorting the real from the fake became its own skill. ## Sharman Networks and the Corporate Shell Game One of the most fascinating aspects of Kazaa's story is the corporate structure behind it. After the Dutch company Consumer Empowerment ran into legal trouble, the rights to Kazaa were transferred to Sharman Networks, a company incorporated in Vanuatu, the Pacific island nation. Sharman was run by Nikki Hemming, an Australian businesswoman, and its day-to-day operations were based in Sydney. The FastTrack protocol itself was licensed through Joltid Ltd., a company registered in the British Virgin Islands. This wasn't accidental. The entire structure was designed to make it as difficult as possible for any single country's legal system to shut down the operation. If you sued Sharman in Australia, they could argue the technology was licensed from a company in the British Virgin Islands. If you went after the protocol itself, the people who built it were in Estonia. The corporate entity was in Vanuatu. It was jurisdiction arbitrage, and for a while, it worked. Look, they weren't stupid. They watched what happened to Napster, a company that was based in San Mateo, California, with a clear corporate address and a founder who showed up to court. Napster made it easy. Kazaa's operators made it as hard as possible. ## The RIAA Goes to War The recording industry tried everything. In 2003, the RIAA began filing lawsuits against individual Kazaa users, not just the company but the people downloading music. They sued a twelve-year-old girl named Brianna LaHara, who lived in a New York City housing project. They sued a seventy-one-year-old grandfather in Texas. They sued college students, single mothers, and people who didn't own computers but whose names were on internet service accounts. Between 2003 and 2008, the RIAA filed lawsuits against approximately 35,000 individuals. The strategy was deliberate. The RIAA knew it couldn't sue everyone who used Kazaa. But it could create enough fear to change behavior. Settlement demands typically ranged from $3,000 to $11,000, and most people paid rather than fight. For the recording industry, the math worked: even if each lawsuit cost more to file than it recovered, the chilling effect on the broader public was the real goal. Simultaneously, the industry went after Sharman Networks directly. In 2004, Australian record labels, led by Universal Music Australia, filed suit against Sharman in the Federal Court of Australia. The case, Universal Music Australia Pty Ltd v Sharman License Holdings Ltd, was heard by Justice Murray Wilcox, and the 2005 ruling was a landmark in copyright law. Justice Wilcox found that Sharman had "authorized" its users to infringe copyrights. The key finding was that Sharman knew the predominant use of Kazaa was for sharing copyrighted material and had taken no meaningful steps to prevent it. The court noted that technical measures existed that could have curtailed the sharing of copyrighted files, but Sharman chose not to implement them because doing so would have been against its financial interests. The ruling ordered Kazaa to modify its software so that searches would filter out files matching a list of copyrighted works supplied by the industry. ## The $115 Million Surrender On July 27, 2006, Sharman Networks settled with the global recording industry. The settlement required Sharman to pay $115 million in damages to the four major music companies: Universal Music, Sony BMG, EMI, and Warner Music. Sharman also agreed to convert Kazaa into a legal, licensed music service and to use "all reasonable means" to discourage piracy through its software. The conversion never really worked. By the time Kazaa relaunched as a legal download service, the market had moved on. Apple's iTunes Store had launched in 2003 and was already the dominant legal music download platform. Users who wanted free music had migrated to BitTorrent, which was even more decentralized than FastTrack and didn't require a specific application. Kazaa's brand was permanently associated with viruses, spyware, and RIAA lawsuits. Nobody was coming back. The website lingered for years, technically operational but functionally dead, before finally going offline. The domain changed hands. The software stopped being updated. By 2012, it was over in every meaningful sense. ## What Kazaa Actually Changed Here's the thing about the P2P wars: the recording industry won every legal battle and lost the actual war. Kazaa was shut down, Grokster was shut down, LimeWire (https://404memoryfound.com/posts/what-happened-to-limewire-file-sharing.html) was shut down, and file sharing kept growing. The RIAA's lawsuit campaign against individuals was widely seen as a public relations disaster, generating more sympathy for downloaders than for record labels. And the fundamental insight that Kazaa proved, that people would choose digital convenience over physical media if given the option, turned out to be correct. The industry just needed a legal way to deliver it. That's where the story connects to what came next. Steve Jobs launched the iTunes Store on April 28, 2003, positioning it as "music downloads done right." He cited Kazaa by name during the announcement, contrasting its unreliable, virus-laden experience with iTunes: individual songs for 99 cents, perfectly encoded, instant download. Within its first week, iTunes sold one million songs. The model worked because Jobs understood something the record labels had been fighting for years: the demand was real. People didn't want to steal music. They wanted to access it easily, and they were willing to pay a reasonable price for the convenience. Spotify took this further in 2008, eliminating the purchase entirely and replacing it with a subscription model. The P2P era had demonstrated that people wanted instant access to everything, and streaming delivered that legally. This is essentially what Kazaa's users were doing, just without the licensing agreements. ## The Skype Connection Perhaps the most remarkable legacy of Kazaa is what its creators built next. Niklas Zennstrom, Janus Friis, and the Estonian engineering team, the same people who built the FastTrack protocol, used the same underlying peer-to-peer architecture to create Skype in 2003. The technology that had been used to share pirated MP3s was repurposed to route voice calls over the internet, bypassing traditional telephone networks. Skype's P2P architecture used supernodes in exactly the same way FastTrack did, distributing the load across the network instead of routing everything through central servers. eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html) acquired Skype in 2005 for $2.6 billion, and Microsoft acquired it in 2011 for $8.5 billion. Jaan Tallinn, the Estonian programmer who had helped build FastTrack in the late 1990s, became wealthy enough to devote his career to funding research into existential risks from artificial intelligence. Which brings us to an interesting question about how we assign value to technology. The same protocol that was treated as a criminal instrument when it moved music files was celebrated as a breakthrough when it moved voice data. The architecture didn't change. The application did. ## Frequently Asked Questions When was Kazaa released? Kazaa Media Desktop was publicly released on March 1, 2001, by the Dutch company Consumer Empowerment. It was created using the FastTrack peer-to-peer protocol developed by Estonian programmers at BlueMoon Interactive. How many people used Kazaa? At its peak, Kazaa had approximately four million simultaneous users online at any given time. The software was downloaded over 239 million times, and in May 2003 it became the most downloaded application tracked by CNET's Download.com. Why was Kazaa sued? Kazaa faced lawsuits from the recording industry because its network was predominantly used to share copyrighted music and movies without authorization. The most significant case, heard in Australia in 2005, found that Sharman Networks had "authorized" copyright infringement by knowingly facilitating it. How much did Kazaa pay in the settlement? In July 2006, Sharman Networks agreed to pay $115 million in damages to Universal Music, Sony BMG, EMI, and Warner Music. The company also agreed to convert Kazaa into a legal music service. What is the connection between Kazaa and Skype? Kazaa and Skype were created by the same team. Niklas Zennstrom, Janus Friis, and the Estonian programmers who built the FastTrack protocol used similar P2P technology to create Skype in 2003. Microsoft acquired Skype in 2011 for $8.5 billion. What replaced Kazaa? Users migrated to other P2P applications like LimeWire and BitTorrent clients, while legal alternatives like the iTunes Store (launched 2003) and later Spotify (launched 2008) eventually captured most of the demand for digital music. --- # What Happened to Club Penguin, Disney's Accidental MMO URL: https://404memoryfound.com/posts/what-happened-to-club-penguin.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-07 Topics: Gaming, Internet Culture ## The Computer Lab Where Everything Started Picture this: 2006. You're in a school computer lab somewhere in suburban America. The monitor is a chunky eMac or maybe a beige Dell tower. The room smells like a combination of hand sanitizer and that mysterious scent that only exists in school buildings. You and three other kids are hunched over your keyboards, and your penguin avatars are ice fishing, decorating igloos, and throwing snowballs at each other in a place called Club Penguin. The bell hasn't rung yet. Your teacher isn't even looking. For a brief, perfect moment before lunch detention exists, you're genuinely engaged with something happening on the internet that doesn't involve clicking through banner ads or navigating the hostile wilderness of early 2000s web forums. This scene played out millions of times in the mid-2000s, becoming a genuine cultural touchstone for an entire generation of kids. Club Penguin wasn't supposed to be huge. It wasn't some elaborate experiment by a major tech company or a calculated venture-backed startup with a five-year plan. It was, by many measures, an accident. A beautiful, profitable accident that would eventually sell for $350 million to Disney, reach over 200 million registered accounts, define what online childhood looked like for Gen Z, and then disappear almost entirely by 2017. What makes the Club Penguin story fascinating isn't just how it became a phenomenon, but why it fell apart so thoroughly, and what that tells us about corporate acquisitions, the expectations placed on beloved properties, and how quickly childhood memories can be erased when a balance sheet demands it. ## How Three Guys in Canada Created a Global Phenomenon The story begins not with Club Penguin, but with something called Penguin Chat. In the early 2000s, Lance Priebe, a developer from Kelowna, British Columbia, built a simple Flash-based chatroom with a twist: instead of just text, users got penguin avatars they could move around a virtual space. It was rough, it was basic, but it proved something important. Kids wanted to hang out online in a visual environment, not just type at each other. Priebe teamed up with Lane Merrifield and Dave Krysko to form a company called New Horizon Interactive. Merrifield handled sales and marketing. Krysko covered expenses and business planning. Priebe built the product. Together, they decided to rebuild the Penguin Chat concept from the ground up. They wanted something cleaner, something designed specifically for children, something that parents wouldn't freak out about because it actually had safety mechanisms built in. They called it Club Penguin, and they spent months obsessing over the details. The penguin avatars needed to have personality. The world needed to be colorful and inviting. There needed to be things to do beyond just chatting, because chatting alone was boring and potentially dangerous for kids. What they built was the online equivalent of a perfectly designed public park for kids. There were spaces to hang out in, things to interact with, social structures that rewarded being present and participating. Your penguin had an igloo that you could decorate with furniture you bought or earned. You could catch fish and play simple games. There was a currency system that made sense, a progression that felt real without being overwhelming. And crucially, there were other people, which meant the place felt alive. Club Penguin launched on October 24, 2005, and within weeks, kids started telling their friends about this weird place where you could be a penguin and live in the Arctic with thousands of other kids. ## The Secret Sauce: Why Club Penguin Actually Worked Here's the thing about Club Penguin that people who weren't there often misunderstand: it wasn't just a chatroom where you could see other people. It was a genuine virtual space with intentional design decisions that made it feel like a real place. The geography mattered. You walked between rooms. You could see other penguins on your screen at the same time you were there. The social hierarchy was visible but not brutal. You could be whoever you wanted to be, which for a 10-year-old kid is genuinely liberating. Nobody cared what school you went to or what your real name was or what brand of shoes you were wearing. The membership model was also genius. Club Penguin offered a free-to-play experience that was genuinely enjoyable, but membership unlocked cosmetic items that made your penguin look cooler and gave you access to exclusive rooms and games. This wasn't exploitative, at least not by modern standards. The free experience was complete. Your parents could let you play without spending money, but if you begged hard enough, they might pay for a month of membership. This is the monetization model that actually works because it's based on creating genuine desire, not on creating artificial scarcity or pay-to-win mechanics. The social events were also crucial. Club Penguin held parties. Real, organized parties with specific themes that happened on specific days. You could only experience certain things at certain times, which created urgency and community. The Valentine's Day party. The Halloween party. The Christmas party. These weren't just in-game cosmetics. They were gatherings. They were the digital equivalent of a town square where something was actually happening right now. If you missed it, you missed it, and you'd have to wait until next year. This FOMO, to use a term that wouldn't be coined until much later, was actually a feature, not a bug. It made people come back. And then there were the puffles. These were small, round creatures that you could adopt as pets. They had personalities. They could grow and evolve. You had to feed them and play with them. This was the hook that made Club Penguin less of a chatroom and more of a life simulation game. Your puffle was depending on you. Kids would log in specifically to take care of their puffles, and while they were there, they'd hang out with their friends. The game also introduced Card-Jitsu, a simple card-based battle system that had surprising depth and became genuinely competitive among the hardcore players. Suddenly, Club Penguin wasn't just a place you went to chat. It was a place you went to live. ## When Disney Showed Up With $350 Million Disney acquired Club Penguin in August 2007, and the price tag shocked the gaming industry: $350 million in cash, with an additional $350 million potentially coming through performance bonuses if specific targets were met by 2009. For a game that had only been running for two years and was created by three guys in Kelowna, this was enormous. At the time of the acquisition, Club Penguin had roughly 12 million accounts, with about 700,000 paid subscribers generating around $40 million in annual revenue. Disney had just paid attention to something that the rest of the mainstream media had mostly ignored: kids were spending real time online, and they were willing to pay money for digital experiences. For a brief moment, this acquisition felt like a victory. A Canadian indie success story had caught the eye of the biggest entertainment company on the planet, and the founders would make bank. Disney, meanwhile, seemed to understand what they had. They didn't come in and immediately gut the game or change its culture. They poured money into expansion and marketing. Club Penguin's user base exploded. By 2013, the game had surpassed 200 million registered accounts. Imagine that: two hundred million kids had created penguin avatars and walked through the digital ice at some point. Server capacity became a genuine issue. Disney had to continuously upgrade infrastructure just to handle the load. The peak years of Club Penguin, roughly 2008 to 2012, were genuinely golden. The game had a real economy, a real community, and a real sense of place. Fan sites exploded. Kids wrote stories about their penguins. There was fan art. There were kids who spent hours every day playing, and it was actually socially acceptable to talk about Club Penguin at school because, unlike World of Warcraft, your parents had probably heard of it and didn't think it was weird. For a generation of kids, Club Penguin was the internet. It wasn't a game they played. It was where they lived. ## The Long Decline Nobody Really Wants to Talk About But here's the thing about internet culture, especially kid culture: nothing stays hot forever. Mobile gaming started exploding around 2010. Kids who had been playing Club Penguin on desktop computers started playing Angry Birds on iPhones instead. Attention spans shifted. Social media was emerging as the new place where kids wanted to hang out. Club Penguin's user numbers started sliding. Not dramatically at first, but noticeably. The growth had stalled by 2012, and by 2014, the game had started feeling less like the center of kid culture and more like something your parents had let you play three years ago. Disney noticed this decline, and the response was to try modernizing the experience. But they didn't seem to understand that the aesthetic of Club Penguin wasn't a bug that needed to be fixed. The simplicity, the flat graphics, the slightly retro feeling of the whole thing: this was the point. It was charming. It was accessible. It made kids feel like they were part of something genuine rather than part of a corporate product. More importantly, Disney also seemed to fundamentally misunderstand why Club Penguin had worked in the first place. They thought the magic was in the IP and the mechanics. What they were missing was the community and the continuity. You couldn't just shut down the old Club Penguin and move everyone to a new version. That's not how attachment works. That's not how community works. The old Club Penguin, with all its rough edges and technical limitations, was the place where the memories lived. A new version, no matter how pretty, would always feel like an imitation. ## The Shutdown and the Funeral Nobody Expected On January 30, 2017, Disney announced that Club Penguin would be shutting down on March 29, 2017. Twelve years of memories were being deleted, and players had less than two months to say goodbye. No long, gradual wind-down. Just: here's the date, come say goodbye if you want, then we're turning off the servers. What happened next was beautiful and heartbreaking in equal measure. Kids from around the world logged in for the final party. Thousands of penguins, all crowded into the town center, mostly just standing there. Some were chatting. Some were taking screenshots. Some were probably just sitting in their bedrooms crying, though you couldn't see that on screen. It was like watching the final episode of a show that had defined your childhood, except you didn't get multiple seasons to say goodbye. You got about eight weeks. The final moments before the servers shut down on March 30, 2017 were absolute chaos: so many penguins trying to be in the same space that the game started having trouble rendering everyone. And then it was gone. All 200 million accounts. All the igloos you had decorated. All the puffles you had raised. All the memories of the parties you had attended. All the friendships you had made with people you'd never met in real life. Deleted. Not archived. Not preserved. Just gone. ## Club Penguin Island: When Disney Really Lost the Plot And here's where it gets interesting, and also where it gets frustrating. Disney didn't retire the Club Penguin IP. Instead, on the very same day the original game closed, they launched Club Penguin Island, a mobile-first remake that was supposed to be the future of the franchise. It was completely separate from the original game. It had new graphics. It was optimized for touchscreens. It was, by all accounts, a perfectly functional mobile game. It was also a complete failure. Kids didn't want Club Penguin on their phones. They wanted Club Penguin, the actual place they had spent years of their childhoods. The new version felt soulless and corporate in comparison. It was designed by people who understood game design but didn't understand the actual magic of why Club Penguin had mattered. Worst of all, Club Penguin Island was exclusively available on mobile, which meant that parents couldn't just let their kids play it on a family computer anymore. It required an app download and a login system that was more complicated than the original. Club Penguin Island limped along for less than two years. On September 27, 2018, Disney announced that it too would be shutting down, with servers going offline on December 20, 2018. Twice in less than two years, Club Penguin players were told that their game was being deleted. The second time felt almost like a punchline. You couldn't even preserve the new version as a backup or a replacement for the original because it didn't offer anything of value that the original hadn't. It was just gone, again. ## The Fan-Run Clones and Disney's Legal Awakening But here's what Disney didn't count on: the internet doesn't forget, and passionate communities don't give up. Almost immediately after Club Penguin shut down, fans started creating their own versions. The most famous is probably Club Penguin Rewritten, which launched in 2017 and was built by fans who had actually studied how the original game functioned. It wasn't a perfect recreation, but it was close enough that thousands of people returned to it. They wanted their memories back. They wanted to feel like they were part of Club Penguin again, even if it was an unofficial imitation. Disney was not happy about this. The company sent cease and desist letters. They pursued legal action against the fan-run servers. In 2022, Club Penguin Rewritten was shut down after a criminal investigation. But here's the thing about trying to kill something that the internet has decided to resurrect: you can't stamp it out entirely. For every official takedown, more versions popped up. New private servers emerged. The community kept finding ways to keep the spirit of Club Penguin alive, even as Disney's lawyers worked overtime to shut them down. The existence of these fan servers raises a genuine question about digital ownership and preservation. Club Penguin was a cultural artifact for an entire generation. It was something that shaped how kids understood online community and virtual spaces. Yet Disney, which owned the IP, decided to delete it entirely rather than preserve it in any form. The fans had to step in and do the preservation work themselves. This is part of a larger problem with digital media: when companies own everything, and those companies decide that something isn't profitable anymore, that thing simply ceases to exist. There's no archive. There's no museum. There's just a void. ## The Legacy: What Club Penguin Actually Meant So what is Club Penguin's actual legacy? For one thing, it proved that you didn't need cutting-edge graphics or massive budgets to create something that kids cared about. The simplicity of Club Penguin was a feature, not a limitation. It made the game accessible and charming in a way that a more technically impressive game might not have been. Club Penguin also created a template for what a safe, moderated online space for kids could look like. The game had real moderation. Real consequences for behavior that violated community standards. Adults weren't trolling kids in Club Penguin because Disney had invested in actual human moderation. Compare this to modern platforms where algorithms do the moderation and the result is chaos. Club Penguin proved that you could have a genuine community of kids online without turning it into a hellscape. But perhaps the most important legacy is the one that Disney seems most determined to forget: that digital experiences matter to people, and that erasing them has real emotional consequences. There are adults now in their twenties and thirties who grew up on Club Penguin and still think about it. They remember specific moments, specific friends they made online, specific achievements in the game. These memories are real, even if the place they happened no longer exists. By completely deleting Club Penguin instead of preserving it in some form, Disney erased a piece of cultural history. Club Penguin was, ultimately, an accident. It wasn't supposed to become this huge thing. It wasn't supposed to define a generation's online experience. But it did, and for about five or six years, it was genuinely magical. Kids from around the world met each other in a virtual Arctic, decorated igloos, played card games, raised digital pets, and felt like they were part of something real and meaningful. That mattered. That was important. And the fact that it's gone now doesn't change the fact that it happened or that it meant something. ## Frequently Asked Questions ### When did Club Penguin shut down? Club Penguin's servers went offline on March 30, 2017, after Disney announced the closure on January 30, 2017. There was a final farewell party before the servers were permanently taken down. Disney did not preserve or archive the game in any official capacity. ### Why did Disney shut down Club Penguin? Disney never gave a single detailed explanation, but the primary reason was declining user numbers and revenue. The game's popularity had peaked around 2010 to 2012, and by 2017, it was generating less revenue than Disney considered worthwhile to maintain. Disney also wanted to redirect resources toward Club Penguin Island, its mobile replacement, which itself shut down in December 2018. ### How many people played Club Penguin? By 2013, Club Penguin had surpassed 200 million registered accounts. At the time of Disney's acquisition in 2007, the game had about 12 million accounts with roughly 700,000 paid subscribers. Not all registered accounts were active simultaneously, and many players created multiple accounts. ### How much did Disney pay for Club Penguin? Disney paid $350 million in cash to acquire Club Penguin in August 2007, with up to an additional $350 million in performance-based bonuses tied to meeting specific growth targets by 2009. The total potential deal value was $700 million. ### Was Club Penguin Island any good? Club Penguin Island was a competent mobile game, but it was not the experience that longtime Club Penguin players wanted. It felt disconnected from what made the original game special. It launched on March 29, 2017, the same day the original Club Penguin closed, and shut down on December 20, 2018, lasting less than two years. ### Can I still play Club Penguin anywhere? There are fan-created private servers that attempt to recreate the Club Penguin experience. However, these are unofficial and operate in a legal gray area. Disney has actively pursued legal action against some of the larger fan projects, including Club Penguin Rewritten, which was shut down in 2022. Any remaining fan servers could theoretically be taken down at any time. ### Did Club Penguin have any cultural impact? Yes, significant cultural impact. For an entire generation of kids growing up in the 2000s, Club Penguin was their introduction to online community, virtual worlds, and persistent multiplayer experiences. It shaped expectations for what online spaces could be and influenced how moderated, safe digital environments for children were designed. ### Who created Club Penguin? Club Penguin was created by Lance Priebe, Lane Merrifield, and Dave Krysko through their company New Horizon Interactive, based in Kelowna, British Columbia, Canada. Priebe was the lead developer who had previously created Penguin Chat, while Merrifield handled sales and marketing, and Krysko managed business operations. --- # What Happened to MiniDisc, Sony's Forgotten Music Format URL: https://404memoryfound.com/posts/what-happened-to-minidisc-sony.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-07 Topics: Hardware, Music & Entertainment ## The Format That Won Japan and Lost Everything Else At the height of MiniDisc's popularity in Japan, during the late 1990s, the format was outselling recordable CDs by a wide margin. Sony had built something so compelling that Japanese consumers were choosing it over the format that had already conquered the rest of the world. And yet, two decades later, MiniDisc is remembered primarily as a failure, a footnote in the graveyard of obsolete media formats, right up there with Betamax (https://404memoryfound.com/posts/betamax-vs-vhs-format-war.html) and the Laserdisc. The real story is more complicated, and more interesting. MiniDisc wasn't killed by the iPod (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html). It wasn't a technological dead-end. It was killed by strategy: by DRM restrictions, by Sony's inability to negotiate pre-recorded content deals outside Japan, by a retail price that felt like a punch to the wallet, and by the fundamental wrong-place-wrong-time problem of arriving just as the industry was about to transform completely. Look, MiniDisc matters because it represents something specific about how companies fail. Sony didn't lack vision. They didn't lack engineering talent. What they lacked was the ability to execute globally, and the willingness to make hard choices about control. ## The Birth of an Idea Sony announced MiniDisc in September 1992. The company had already conquered portable audio with the Walkman cassette player, a product that essentially created the category of personal mobile music. They were looking for the next evolution. CDs were getting smaller in players but not small enough. They were also read-only for consumers and fragile enough to skip if you looked at them wrong. Sony wanted something different: a format that was tiny, recordable, and durable enough to withstand the punishment of daily commuting. The technology was magneto-optical storage. Unlike CDs, which used physical pits pressed into polycarbonate, MiniDisc used a magnetic layer that could be written and rewritten thousands of times. The disc itself was about two and a half inches across, sealed in a protective plastic cartridge that made it genuinely pocket-friendly. Sony imagined it as the natural successor to the cassette Walkman. They weren't entirely wrong about that. The problem was execution and timing. MiniDisc used a proprietary compression algorithm called ATRAC, short for Adaptive Transform Acoustic Coding. This is important because it allowed roughly 74 or 80 minutes of near-CD-quality audio to fit on a disc with only about 140 megabytes of storage capacity, compared to the 650 megabytes a standard CD held. The compression was lossy, meaning it permanently discarded some audio data. But it was lossy in a smart way. Sony's researchers had studied psychoacoustics, the science of how human ears perceive sound, and optimized the algorithm to remove data that most listeners wouldn't notice was missing. The first MiniDisc devices hit shelves in November 1992 in Japan and December 1992 in North America and Europe. By 1997, the technology was mature and the devices worked well. The question was whether the world outside Japan would adopt them. ## Japan's Love Affair with MiniDisc In Japan, MiniDisc took off. The format captured a substantial share of the Japanese portable audio market at its peak. In some years during the late 1990s, more MiniDisc units sold in Japan than in the rest of the world combined. This wasn't random. Japanese commuters spent hours on trains and subways. They were willing to adopt new technology if it solved a real problem. MiniDisc solved several: it was small, it held a lot of music, it was shockproof, and crucially, it let you record music yourself. Recording was a bigger deal in Japan because of how the culture consumed media. Students would record songs off the radio or from borrowed CDs. Music fans would create custom compilations. MiniDisc made this faster and more reliable than cassettes, with better audio quality. Sony's market position in Japan was also just stronger. They had distribution networks, relationships with retailers, and cultural cachet that didn't translate as directly to the American market. In the Japanese music industry (https://404memoryfound.com/posts/napster-destroyed-music.html), MiniDisc was positioned as the legitimate, high-quality alternative to dubbing cassettes. Record labels in Japan actually supported MiniDisc pre-recorded releases. This is the key difference. By the late 1990s, major Japanese labels were pressing pre-recorded MiniDisc albums. You could walk into a record store in Tokyo and buy a new album on MiniDisc the same day it came out on CD. This created a virtuous cycle: more music available in the format meant more incentive to buy a player, which meant more pressure on labels to release new content. For a moment, it looked like Japan had found its next standard. ## Why America Said No Thanks The United States was a different story entirely. When Sony launched MiniDisc in North America in December 1992, the pricing was brutal. A playback-only portable unit cost $549. The recording model cost $750. A single blank MiniDisc cartridge cost around $17, compared to a couple of dollars for a CD-R blank by the mid-1990s. Americans had already standardized on CDs and were rapidly adopting CD-R for recording. The technology was cheap, it worked fine, and blank media was abundant. Consumers looked at MiniDisc and asked: why would I spend several times as much for something smaller? The only compelling answer would have been a selection of pre-recorded titles in the stores. And that answer didn't exist. Here's the thing: major American record labels weren't interested in pressing MiniDisc albums. They saw MiniDisc as a Sony proprietary format and viewed it with suspicion. They were wary of the ATRAC compression. They were wary of the whole proposition. And they were just uninterested in fragmenting the market further when CDs were selling perfectly well. This was the strategic bottleneck that killed MiniDisc outside Japan. If the American labels had released albums on MiniDisc, the entire trajectory could have been different. Not that CD-R wouldn't have still existed. Not that the iPod wouldn't have eventually changed everything. But MiniDisc could have become a legitimate mainstream format. Instead, it remained a niche product that early adopters and audio enthusiasts bought while most people never even saw one in person. ## The Technical Brilliance That Wasn't Enough Let's pause on the commercial challenges and acknowledge what Sony actually accomplished technically. ATRAC compression was genuinely sophisticated for its era. The team that built it understood psychoacoustics at a level that most audio engineers couldn't match. The 5:1 compression ratio was aggressive by the standards of the time, reducing CD-quality audio from about 1.4 megabits per second down to around 292 kilobits per second. Modern listeners with trained ears and good headphones would probably hear the difference between ATRAC-compressed music and uncompressed audio. The algorithm did throw away data permanently. But in the late 1990s, on the portable devices of the era, with the headphones people actually used, most casual listeners couldn't reliably tell. It was good engineering applied to a real problem. The other technical advantages were real too. MiniDisc players were genuinely shock-resistant because the format used a buffer memory system that read ahead and stored several seconds of audio in RAM. CD players would skip if jostled. MiniDisc players kept playing. For someone jogging or riding a crowded subway, this mattered enormously. And the re-recordability was genuinely useful for the music fans Sony was targeting. You could record your favorite songs, erase them, record something else. You could create custom compilations and edit track names on the tiny screen. This is essentially what playlists on streaming services do today, except instead of licensing your access, MiniDisc required you to own the hardware and do the recording yourself. But technical elegance is not the same as market success. Sony built a better mousetrap. The world had already committed to a different mousetrap and was quietly building an entirely new kind of mousetrap altogether. ## The DRM Problem That Poisoned Everything Here's where Sony made a decision that haunted the format for its entire lifespan. MiniDisc included SCMS, the Serial Copy Management System. This meant you could make a digital recording from a CD onto a MiniDisc, but you couldn't then make a digital copy of that MiniDisc recording onto another MiniDisc. One generation of digital copying, and that was it. On paper, this made sense. Sony was trying to address label concerns about unlimited digital piracy. In practice, it made MiniDisc less appealing to exactly the users who would have been most enthusiastic about it. The restrictions were the opposite of the value proposition. People wanted to record their music and share it and move it around freely. MiniDisc said: you can do this, but only once, and only according to specific rules. This is the recurring mistake of hardware manufacturers who try to protect their media partners at the expense of their customers. The restrictions make the product worse for legitimate users without actually stopping determined pirates. People who want to pirate will find a way around any protection scheme. People who don't will just buy the product that imposes fewer restrictions on their behavior. By the late 1990s, file-sharing software like Napster (https://404memoryfound.com/posts/who-owns-napster-now.html) was already emerging. The restrictive nature of MiniDisc made it look backward to early adopters who wanted full control over their own music libraries. If Sony had made MiniDisc recordings fully copyable, would it have changed the format's fate? Probably not entirely, since the pricing and content problems were more fundamental. But it would have removed one more obstacle to adoption and might have built more goodwill among the enthusiast community. ## NetMD and Hi-MD: Two Solutions to the Wrong Problem By the early 2000s, Sony was aware that MiniDisc was stagnating outside Japan. They responded with NetMD in 2001, which added USB connectivity to MiniDisc players. You could now transfer music from your computer to a MiniDisc via USB cable, using Sony's SonicStage software. Here's the problem: NetMD actually highlighted why MiniDisc was becoming unnecessary. If you were already at your computer managing music files, why would you transfer them to a MiniDisc when you could transfer them to an MP3 player that didn't require proprietary compression or copy-protection restrictions? MP3 players were getting smaller, cheaper, and more capable every year. Sony had created a solution that inadvertently proved the superiority of the competing approach. In 2004, Sony released Hi-MD, which increased storage capacity to 1 gigabyte per disc and allowed uncompressed PCM recording. This was supposed to be the format's definitive comeback. Except by 2004, the iPod had been on the market for three years. It was already clear that portable music was going to be digital files on hard drives and flash memory, not physical discs of any kind. The real question is whether Sony ever understood that the core problem wasn't MiniDisc the technology but MiniDisc the market strategy. They kept trying to solve the problem by adding features. What they needed to do was abandon the proprietary ecosystem entirely and make devices that played whatever formats consumers wanted. But that would have meant admitting that their control-oriented approach had failed, and Sony in the early 2000s wasn't ready for that admission. ## The iPod Changes the Conversation Apple released the first iPod on October 23, 2001. It held 5 gigabytes of music, roughly 1,000 songs. It used a FireWire connection to sync from iTunes. It cost $399. In terms of raw portability and battery life, it wasn't necessarily better than the best MiniDisc players of the same era. What it was better at was integration. The iPod worked with iTunes, which worked with the music you already had on your computer, in whatever format you had it. You could buy an iPod and it did exactly what you expected without restrictions or proprietary compression mandates or expensive blank media. The experience was seamless in a way that Sony's proprietary stack could never manage. Within a few years, the iPod became the dominant portable music player globally. MiniDisc didn't lose because it was technically inferior. It lost because the entire market realized that the future of portable audio wasn't about specific formats or specific physical media types. It was about devices that could handle any reasonable format and connect easily to the computer that was increasingly becoming the hub of everyone's digital life. Sony could have built the iPod before Apple. They had better brand recognition in consumer electronics, better distribution networks, better relationships with record labels, and decades more experience in portable audio. They didn't build it because they were too invested in the proprietary path. They believed in format control and licensing revenue. Apple believed in simplicity and interoperability. The market made its choice. ## The Quiet End MiniDisc continued on for over a decade after the iPod's launch, though each year the market shrank further. Japan kept buying, but even there the numbers declined steadily. Sony manufactured its last MiniDisc player in March 2013, more than twenty years after the format's introduction. The company continued producing blank MiniDisc media, specifically the MDW80T model, for the Japanese market for years afterward, only announcing the end of blank disc production in early 2025. It's a strange epilogue. Billions of dollars spent on development, marketing, manufacturing, and licensing deals across more than three decades. All of it wound down not because the format failed catastrophically in a single moment but because the world simply, gradually, moved on to something different. ## The Collector Revival In the last few years, MiniDisc has experienced something between a genuine revival and a nostalgia-driven trend. This is largely a Gen Z phenomenon, playing out mostly on TikTok and Instagram, where young people who were born long after MiniDisc's commercial peak have started seeking out the devices and the media. They find the hardware aesthetically appealing. They like the idea of a format that's physical and tactile but smaller and sleeker than vinyl or even CD. They appreciate that recording to MiniDisc requires deliberate choice and intention, which contrasts sharply with the infinite, effortless nature of streaming. This revival has driven up prices in the used market. A decent portable MiniDisc recorder that might have sold for $30 five years ago now goes for $100 to $200 or more. Some rare models, particularly Sony's high-end units like the MZ-RH1, have become genuinely expensive collector's items. People are investing in MiniDisc hardware as a deliberate choice for how to engage with music. The economics of this market are tiny. We're talking about thousands of units globally, not millions. But the community is real and growing. There are sellers specializing in refurbished MiniDisc players. There are detailed online guides to the best recording devices and the best blank media. There are forums where people share their recording setups and custom disc labels. This is essentially what happened to vinyl records a decade earlier, except without the corporate marketing push that fueled the vinyl revival. Young people discovered MiniDisc through parents' closets, thrift stores, and internet nostalgia content, and decided it was interesting. The format itself matters less than the object and the ritual of using it. ## What MiniDisc Teaches Us MiniDisc was a format designed for a world that was already becoming obsolete. Sony built something technically sophisticated at exactly the moment when the industry was beginning to realize that proprietary formats and physical media were both heading toward irrelevance. Sony had defensible reasons for their choices. Those reasons just weren't good enough. The format teaches several lessons. First, technical superiority is not sufficient for market success. MiniDisc was a better-engineered product than CD-R in several measurable ways. It didn't matter. CD-R was good enough for most people and dramatically cheaper. Second, lock-in strategies can backfire catastrophically. SCMS copy protection was supposed to reassure the music industry. It actually made MiniDisc less attractive to the consumers Sony needed most. Restrictions that protect partners at the expense of customers rarely work out well in the long run. Third, you cannot win a format war alone. Sony could have succeeded globally with MiniDisc if they'd convinced the American record labels to release pre-recorded content. They couldn't close those deals. That was the turning point, not the technology, not the price, but the absence of content in the format's most important potential market. And finally, sometimes timing is the most important variable, and it's the one you control least. MiniDisc arrived at the precise moment when computers were about to become the center of how people managed and consumed music. A format that required its own dedicated hardware ecosystem was always going to struggle against a future where everything lived on a hard drive. ## Frequently Asked Questions ### Can you still buy MiniDisc players? Only used. You can find them through online marketplaces, specialty retailers, and auction sites. Prices range from about $30 for basic playback-only models to several hundred dollars for sought-after recording units like the Sony MZ-RH1. No new MiniDisc players have been manufactured since March 2013. ### Can you still buy blank MiniDisc media? Sony continued producing its MDW80T blank MiniDisc for the Japanese market long after discontinuing hardware, but announced the end of blank disc production in early 2025. Remaining new-old-stock blanks can still be found through specialty sellers and online marketplaces, though prices have risen as supply dwindles. Expect to pay $10 to $20 or more per blank disc. ### Is MiniDisc sound quality better than CD? No. Standard MiniDisc uses ATRAC lossy compression, which permanently removes audio data during encoding. A lossless CD recording will always contain more information. However, ATRAC was specifically designed to be perceptually transparent for casual listening, and most people cannot reliably distinguish between a MiniDisc recording and the CD original on typical portable equipment. Hi-MD, introduced in 2004, did support uncompressed PCM recording at full CD quality. ### Why didn't Sony just make devices that played MP3 files? Sony was deeply committed to the ATRAC ecosystem and the licensing revenue it generated from other manufacturers. Creating a MiniDisc player that handled MP3 or other open formats would have undercut their own proprietary advantage. By the time Sony began supporting MP3 playback in later devices, the iPod had already established the market standard and the window of opportunity had closed. ### Did MiniDisc ever have a real chance in the United States? Possibly, but success would have required fundamentally different decisions. If American record labels had released pre-recorded MiniDisc albums and the initial hardware pricing had been lower, adoption could have been stronger. More realistically, MiniDisc was always likely to be niche in the US because CD-R solved the same basic problem at a fraction of the cost. The real market opportunity was Japan, where Sony did succeed. The mistake was assuming that domestic success would translate globally without the content deals to support it. ### Why is MiniDisc trending on TikTok? Gen Z users have discovered MiniDisc through secondhand markets and have become aesthetically and practically interested in the format. The appeal combines several factors: the hardware is compact and visually distinctive, the recording process is intentional and tactile, and the format is rare enough to feel exclusive. This mirrors the broader trend of younger consumers seeking physical, analog-adjacent alternatives to purely digital music consumption. ### Did other companies make MiniDisc devices? Yes. Sony licensed the MiniDisc format to other manufacturers including Aiwa, Sharp, Panasonic, Kenwood, and others. Some of these companies produced excellent devices that occasionally innovated beyond Sony's own designs, particularly Sharp's portable recorders. However, Sony always dominated market share. Third-party manufacturers gradually exited the market as MiniDisc sales declined, leaving Sony as the sole remaining manufacturer by the late 2000s. ### What is the most sought-after MiniDisc player for collectors? The Sony MZ-RH1, released in 2006, is generally considered the most desirable MiniDisc device among collectors and enthusiasts. It was the last high-end portable MiniDisc recorder Sony produced, and it was the only device that could upload recordings from MiniDisc to a computer in real time via USB. Working units in good condition regularly sell for $300 to $500 or more. --- # What Happened to Screensavers: Flying Toasters, 3D Pipes, and Doing Nothing URL: https://404memoryfound.com/posts/what-happened-to-screensavers-flying-toasters.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-06 Topics: Software & Apps, Then vs Now ## You Used to Pay Fifty Bucks for Animated Toasters Picture this: 1994. You're in a CompUSA (https://404memoryfound.com/posts/is-compusa-still-around.html), wandering past towers of shrink-wrapped software boxes, and you stop at a display that's running a screensaver. Not just any screensaver. Chrome toasters with little wings are flying across a pitch-black monitor, gliding alongside slices of toast at varying levels of doneness. There's a crowd of maybe four or five people just standing there, watching. Mesmerized. By toasters. And then someone picks up the box, reads the back, sees the $49.95 price tag, and buys it. For a screensaver. This actually happened. Millions of times. If you grew up in the 1990s, screensavers weren't just background noise. They were a thing. A hobby. A personality statement. The screensaver you chose said something about you. Were you a 3D Pipes person? A Starfield guy? Did you spring for After Dark and its legendary Flying Toasters? Or were you the kid whose dad left the default Windows marquee scrolling "Hello World" across a beige CRT monitor in the den? Today, screensavers are basically extinct. Most people under 25 have never intentionally set one. Windows still technically has them buried in the settings, but they're an afterthought, a relic sitting in the same drawer as fax machine drivers and Internet Explorer shortcuts. So what happened? How did an entire category of software go from cultural phenomenon to digital ghost? To answer that, you have to go back to when screens actually needed saving. ## Burn-In Was Real, and It Was Terrifying Here's something younger people don't realize: screensavers weren't invented as entertainment. They were invented out of genuine necessity. CRT monitors, those big, heavy, glass-fronted boxes that took up half your desk and weighed about thirty pounds, had a very real problem called phosphor burn-in. The way a CRT worked was pretty brutal. An electron gun at the back of the tube fired a beam at a layer of phosphor coating on the inside of the glass. When the beam hit the phosphor, it glowed. That glow was your image. But here's the problem: if the same image sat on the screen for hours, the phosphor in those spots would degrade permanently. The image would literally burn itself into the glass. You could turn the monitor off and still see a ghost of whatever had been displayed. This was especially bad in offices. If a secretary left a WordPerfect (https://404memoryfound.com/posts/what-happened-to-wordperfect-word-processor.html) menu on screen overnight, by morning there'd be a permanent shadow of that menu bar etched into the monitor. Banks, airports, libraries, anywhere with a computer running the same display for hours on end was at risk. Replacing a CRT monitor in the late 1980s could cost $300 to $500, which is roughly $700 to $1,100 in today's dollars. Burn-in wasn't just annoying. It was expensive. The first screensaver was written by a programmer named John Socha and published in the December 1983 issue of Softalk magazine. It was called SCRNSAVE, and all it did was blank the screen after three minutes of inactivity. That's it. No flying toasters. No morphing shapes. Just darkness. And for a while, that was enough. ## Berkeley Systems Turned Screen Protection Into Entertainment The real transformation started in 1989, when a small company in Berkeley, California called Berkeley Systems released a product called After Dark for the Apple Macintosh. The founders, Wes Boyd and Joan Blades (who would later go on to found MoveOn.org, which is a wild career pivot), had a simple insight: if people had to have something running on their screens to prevent burn-in, why not make it fun? After Dark shipped with a collection of animated screensaver modules, and users could pick their favorite. There was a bouncing ball. There was a module called Lunatic Fringe, which was basically a full space combat game disguised as a screensaver. There was one that simulated rain on the screen. But the star of the show, the one that turned After Dark from a utility into a phenomenon, was Flying Toasters. The story behind Flying Toasters is almost too simple. Engineer Jack Eastman was working late one night, wandered into the kitchen, saw a toaster sitting on the counter, and imagined what it would look like with wings. That was it. That's the whole origin story. He coded it up, gave the toasters chrome bodies and stubby little bird wings, added slices of toast at various levels of doneness floating alongside them, and created one of the most recognizable images in 1990s computing. After Dark retailed for around $30 to $50 depending on the version, and people paid it gladly. By the mid-1990s, Berkeley Systems had grown to about 120 employees and was pulling in roughly $30 million in annual revenue. Thirty million dollars. For screensavers. There was even a slider in the Flying Toasters module that let you adjust the darkness of the toast, which is maybe the most 1990s feature ever designed. The Deluxe version added a Flying Toasters Pro module with a choice of background music: either Richard Wagner's Ride of the Valkyries or an original flying toaster anthem with optional karaoke lyrics. I am not making this up. There were karaoke lyrics for animated flying toasters. And people loved it. ## Windows Made Screensavers Universal While Berkeley Systems was selling screensavers as a premium product, Microsoft was about to make them free. Windows 3.1, released in 1992, included a handful of built-in screensavers, and suddenly every PC on the planet had access to them without paying a dime. The selection was modest: Marquee (scrolling text), Mystify (bouncing geometric shapes), Starfield (flying through space), and a few others. They were simple, but they were there. Then Windows 95 arrived in August 1995, and the screensaver game changed completely. The story of how Windows got its most famous screensavers involves the OpenGL team at Microsoft, and it starts with a technical demo that wasn't supposed to ship. When Windows NT 3.5 was being developed, the team needed a way to showcase OpenGL, the 3D graphics technology that was being built into the operating system. Writing a full application was too risky because it might have bugs that would delay the release. But a screensaver? A screensaver was low stakes. If it crashed, nothing important happened. Your computer just went back to showing the desktop. So the Windows OpenGL team built a series of 3D screensavers as tech demos: 3D Text, 3D Flying Objects, 3D Maze, and the one that became an absolute legend, 3D Pipes. A member of the marketing team saw them, thought they were cool, and decided to include them in the shipping product. And just like that, some of the most iconic screensavers in computing history were born from what was essentially a side project. 3D Pipes was hypnotic. It drew an endless, slowly growing network of connected pipes in three dimensions, changing colors at each joint, occasionally throwing in a teapot-shaped connector (an inside joke referencing the Utah Teapot, one of the most famous test objects in computer graphics history). People would sit and watch it for minutes at a time, which is exactly the opposite of what screensavers were supposed to do. The whole point was that nobody should be looking at the screen. 3D Maze was equally mesmerizing. It generated a random first-person maze and then "walked" through it, turning corners, hitting dead ends, and occasionally finding its way to the exit before starting over. If you were a kid in the late 1990s, you probably spent at least one afternoon staring at 3D Maze hoping you could somehow take control and navigate it yourself. ## The Golden Age: When Screensavers Were a Culture By the mid to late 1990s, screensavers had evolved far beyond burn-in prevention. They were entertainment. They were art. They were, in some offices, a form of passive-aggressive self-expression. The IT guy who put the Matrix digital rain screensaver on his monitor was making a statement. The receptionist with the tropical fish aquarium was making a different one. Third-party screensavers became a massive market. Companies like Sierra On-Line got into the game. In 1993, Sierra's subsidiary Dynamix released Johnny Castaway, which was arguably the most ambitious screensaver ever made. Instead of abstract shapes or animated objects, Johnny Castaway told a story. A little cartoon man was stranded on a tiny desert island, and over time, different things would happen to him: he'd try to fish, build a raft, encounter a mermaid, deal with storms, even celebrate holidays in real time. It was like a tiny sitcom running in the corner of your screen. People in offices would check on Johnny throughout the day to see what he was up to. A screensaver with a fan following. That happened. The late 1990s also brought the era of themed screensavers tied to movies, TV shows, and brands. There were screensavers for The X-Files, Star Wars, Star Trek, Jurassic Park, and basically anything with a marketing budget. You could download them from the internet (slowly, over dial-up), and installing a screensaver from a random website was one of the great acts of faith in early internet computing. Half the time, the screensaver worked fine. The other half, you'd accidentally installed a toolbar, three pop-up ad generators, and something called BonziBuddy (https://404memoryfound.com/posts/what-happened-to-bonzibuddy-spyware.html). There were also the aquarium screensavers, which deserve their own paragraph. A former Air Force pilot named Jim Sachs was so disappointed with the aquarium screensaver that shipped with Windows 95 that he started building his own. His creation, SereneScreen Marine Aquarium, became one of the most popular paid screensavers of all time, running for over 25 years. He hand-rendered every fish, every coral, every bubble. People paid real money for a digital fish tank on their monitor. And it was beautiful. It genuinely was. ## The Death Sentence: LCD Screens Didn't Need Saving The beginning of the end for screensavers came from, of all places, the display technology industry. Throughout the late 1990s and into the 2000s, LCD monitors began replacing CRTs. And LCD screens don't use phosphor. They use liquid crystals illuminated by a backlight. There's no electron beam burning anything. There's no phosphor degradation. Which means there's no burn-in. Well, technically, LCDs can develop something called image persistence if a static image is displayed for extremely long periods, but it's temporary and usually fixes itself. It's nothing like the permanent scarring that CRTs suffered from. The fundamental technological reason for screensavers to exist just vanished. On top of that, power management got smarter. Instead of keeping the monitor on and running a screensaver, operating systems started just turning the monitor off after a period of inactivity. Which makes way more sense, when you think about it. Why run an animated program that uses CPU cycles and keeps the screen lit when you could just turn the screen off and save electricity? Energy Star compliance standards pushed manufacturers and OS developers in this direction, and by the mid-2000s, "turn off monitor" had replaced "run screensaver" as the default power setting on most new PCs. Windows Vista in 2007 was really where the shift became obvious. The built-in screensavers were afterthoughts. Some of the classics were gone. 3D Pipes was removed. 3D Maze was removed. The OpenGL screensavers that a generation had grown up watching were simply deleted from the operating system and never came back. Microsoft replaced them with some gentle, forgettable options like Bubbles and Ribbons. It was like replacing a vintage jukebox with a Bluetooth speaker playing elevator music. Windows 10 and 11 still technically support screensavers, but the feature is buried deep in the settings. Most users never touch it. The default is "None." The screen just goes to sleep. ## The Screensaver Economy Collapsed Almost Overnight For Berkeley Systems, the end came before the LCD transition even finished. In 1997, the company was acquired by Sierra On-Line (which was itself part of CUC International, later Cendant, which turned out to be one of the largest accounting fraud cases in American corporate history at the time, but that's another story). Sierra was eventually absorbed into what became Sierra Entertainment under Vivendi Universal Games, and After Dark just sort of disappeared into the corporate shuffle. The entire paid screensaver market collapsed. Why would anyone pay $50 for software that their operating system now included for free, to protect a screen that no longer needed protecting? Third-party screensaver companies that had thrived in the early and mid-1990s either pivoted to other products or simply closed. The category didn't slowly decline. It fell off a cliff. Screensavers had been a multi-hundred-million-dollar global market in the 1990s when you add up retail sales, corporate licensing, and bundled deals. By 2005, the market was essentially zero. That's one of the fastest category extinctions in software history. ## The Weird Afterlife of Screensavers But here's the thing about screensavers: they never fully died. They just transformed. Apple's tvOS includes screensavers as a feature, those beautiful slow-motion aerial videos of cities, landscapes, and oceans that play on your Apple TV when it's idle. They're essentially screensavers, just rebranded. Nobody calls them that, but that's what they are. Windows 11's lock screen with its rotating Bing wallpapers and Spotlight images is, functionally, a screensaver with extra steps. Samsung and LG smart TVs have ambient modes that display art or photographs when the TV is idle. Roku has aquarium and fireplace channels. The concept survived. The name didn't. And retro computing enthusiasts have kept the originals alive. You can still download and run the classic After Dark modules through Internet Archive. Web developers have recreated 3D Pipes in JavaScript that runs in your browser. There's even an open-source project that faithfully recreates the Windows 95 3D Maze screensaver. People build these things not because they need screensavers, but because they miss them. Because watching 3D Pipes grow across a screen for five minutes is oddly meditative in a way that doomscrolling through social media will never be. ## What Screensavers Were Really About Here's what I think people actually miss about screensavers, and it's not the software itself. It's the moment. Screensavers activated when you walked away from your computer. When you went to the kitchen. When you took a phone call. When you just stopped and did something else for a while. They were visual evidence that you had stepped away from the machine. Today, screens never go idle because we never stop looking at them. We go from laptop to phone to tablet to TV and back again. There's no moment where the computer sits there, unattended, doing its own little thing. The screensaver was, in a weird way, a reminder that it was okay to not be looking at a screen. It was the computer's way of saying, "I'm fine. Go live your life. I'll be here when you get back." Flying Toasters weren't just flying toasters. They were the visual equivalent of a deep breath. And maybe that's why, twenty-five years after they stopped mattering, people still remember them so fondly. Because sometimes, the best thing your computer ever did for you was the thing it did when you weren't watching. ## Frequently Asked Questions Why were screensavers invented? Screensavers were invented to prevent phosphor burn-in on CRT monitors. When a static image was displayed for too long, it would permanently etch itself into the screen. The first screensaver, SCRNSAVE, was written by John Socha in 1983 and simply blanked the display after three minutes of inactivity. What was the Flying Toasters screensaver? Flying Toasters was the most famous module in Berkeley Systems' After Dark screensaver software, first released in 1989 for Macintosh and 1991 for Windows. It featured chrome toasters with bird-like wings flying across the screen alongside toast. Engineer Jack Eastman created it after seeing a toaster in a kitchen during a late-night coding session. Why did Windows remove 3D Pipes and 3D Maze? Microsoft removed the classic OpenGL screensavers (3D Pipes, 3D Maze, 3D Flying Objects, and 3D Text) starting with Windows Vista in 2007. As LCD monitors replaced CRTs, the practical need for screensavers disappeared. Microsoft shifted toward power management features that simply turned monitors off instead. Do modern monitors need screensavers? No. LCD and LED monitors don't suffer from phosphor burn-in the way CRTs did. LCDs can develop temporary image persistence from very long static displays, but it typically resolves on its own. OLED screens can experience burn-in, but modern OLED devices include built-in mitigation features like pixel shifting. Can you still get classic screensavers today? Yes. The original After Dark modules are available through the Internet Archive. Web developers have recreated classics like 3D Pipes and 3D Maze in browser-compatible formats. Retro computing communities maintain collections of classic Windows and Mac screensavers that can still run on modern systems with compatibility tools. How much did screensavers cost in the 1990s? Berkeley Systems' After Dark retailed for roughly $30 to $50 depending on the version. The company generated approximately $30 million in annual revenue by the mid-1990s from screensaver sales alone. Other premium screensaver packages from companies like Sierra On-Line were priced similarly. --- # What Happened to Clippy, the Paperclip Everyone Loved to Hate URL: https://404memoryfound.com/posts/what-happened-to-clippy-microsoft.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-06 Topics: Software & Apps, Then vs Now ## 250 Characters Walked Into a Focus Group. The Paperclip Won. In 1996, a team of designers, engineers, and Stanford social psychologists working inside Microsoft's Office division had a problem. They'd spent months developing a new help system for Office 97, one built around an animated virtual assistant that would watch what you were doing and proactively offer guidance. The concept was grounded in real academic research. The execution involved testing roughly 250 different character designs through six months of focus groups. And after all of that research, all of that money, all of those carefully controlled experiments, the winner was a paperclip with googly eyes. His official name was Clippit. The world called him Clippy. And within three years, he would become the most hated feature in the history of personal computing. The story of Clippy is often told as a simple joke. Microsoft made an annoying paperclip, everyone hated it, they got rid of it. But the real story is more interesting than that. It involves a failed predecessor called Microsoft Bob (https://404memoryfound.com/posts/what-happened-to-microsoft-bob-friendly-interface.html), a genuine misunderstanding of academic research, one of the most expensive user testing programs in software history, and a lesson about the gap between what people say they want and what they actually tolerate. It's also, whether Microsoft intended it or not, the most important cautionary tale in the history of AI assistants. ## The Research That Started Everything The intellectual foundation for Clippy came from Stanford University, specifically from the work of two professors: Clifford Nass and Byron Reeves. In the early 1990s, Nass and Reeves developed a theory that would reshape how the tech industry thought about human-computer interaction. Their finding was deceptively simple: people unconsciously apply human social behaviors to computers. The experiments were clever. When a computer program gave users positive feedback about their work, users rated the program more favorably, even when they knew the feedback was automated and meaningless. When a program was "rude," delivering terse, unhelpful responses, users described feeling genuinely offended. People would describe software as "friendly" or "aggressive" without any awareness that they were anthropomorphizing a tool. Nass and Reeves called this the "media equation," and they eventually published their findings in a 1996 book titled "The Media Equation: How People Treat Computers, Television, and New Media Like Real People and Places." Microsoft paid attention. If people were already treating computers as social actors, the reasoning went, why not lean into it? Why not give the computer an actual social presence, a character that could interact with the user, read the room, and offer help at the right moment? The idea wasn't crazy. The execution, however, would prove to be. ## Microsoft Bob: The Predecessor Nobody Remembers Before Clippy, there was Bob. Microsoft Bob launched on March 10, 1995, personally introduced by Bill Gates. The concept was ambitious: replace the standard Windows desktop with a virtual house. Instead of clicking on program icons, you'd walk through rooms. The living room had a calendar. The study had a word processor. A cartoon dog named Rover guided you around, offering tips and encouragement. Bob was designed for people who found Windows intimidating, specifically the first-time computer buyers that the mid-1990s PC boom was attracting. Microsoft hired Nass and Reeves as consultants and invested heavily in making the interface feel warm and approachable. The problem was that Bob treated every user like someone who had never touched a computer before. For anyone with even basic competency, it was condescending. Bob launched at $99 and was discontinued within a year. It became one of the most notorious flops in Microsoft's history. CNET later named it one of the 25 worst tech products of all time. But inside Microsoft, the core idea persisted. The thinking wasn't that social computing was wrong. The thinking was that Bob had been too ambitious, too all-encompassing. What if, instead of replacing the entire desktop, they embedded a social character inside an existing application? Something smaller. Something optional. Something helpful. That thinking led directly to the Office Assistant. ## How Clippy Was Actually Built The Office Assistant project was part of Office 97's development, and it was not a small effort. Microsoft assembled a team that included software engineers, user interface designers, and external consultants including the Stanford social psychologists who had inspired the project in the first place. Illustrator Kevan Atteberry was brought in to design the characters. He and the broader team produced roughly 250 different character concepts. The range was enormous: animals, robots, abstract shapes, household objects, cartoon humans. Each character needed to convey approachability, intelligence, and helpfulness without being annoying. Which, in retrospect, turned out to be the hard part. Microsoft ran these designs through six months of focus groups. They tested which characters people found most trustworthy, most engaging, most endearing. The focus groups involved members of the public evaluating different character designs based on specific social criteria. Atteberry himself contributed about 15 to 20 of the designs personally. The result? Clippy. The paperclip ranked number one in trustworthiness, engagement, and endearment across the focus group data. Out of 250 options, the googly-eyed paperclip beat them all. Here's where it gets interesting. Alan Cooper, the programmer widely known as the "Father of Visual Basic," later described the entire Office Assistant concept as being built on what he called a "tragic misunderstanding" of the Nass and Reeves research. The Stanford work showed that people respond to computers as social actors. But there's a difference between recognizing that tendency and actively exploiting it. Nass and Reeves had demonstrated a subtle cognitive phenomenon. Microsoft built a cartoon character that leaned on your shoulder while you were trying to write a letter. ## Office 97 Ships, and the Complaints Begin Immediately Microsoft Office 97 was released to manufacturing in November 1996 and hit retail shelves in January 1997. Clippy was enabled by default. This is a critical detail. Every single person who installed Office 97 got Clippy whether they wanted him or not. He would appear in the corner of the screen, watching you work. If you started typing what looked like a letter, up he'd pop: "It looks like you're writing a letter. Would you like help?" If you were formatting a document, he'd offer tips. If you were doing absolutely nothing unusual, he'd sometimes show up anyway, just to check in. The problem wasn't that Clippy offered help. It was that he offered help constantly, usually at the wrong time, and the help he offered was rarely what you actually needed. His appearance was interruptive. He would animate, bobbing and twisting and tapping on the inside of your screen, demanding attention when you were trying to concentrate. Dismissing him didn't make him go away permanently. He'd come back. Always. There was also a mismatch between what the focus groups had measured and what real-world usage revealed. In a controlled focus group setting, people rated the paperclip character as trustworthy and endearing. But a focus group interaction lasts minutes. Working with Clippy lasted hours, days, weeks. The character that seemed charming for five minutes became intolerable after five hours. What the research failed to capture was that the thing people find "endearing" in a brief encounter is often the thing that drives them insane with prolonged exposure. This is the gap that Cooper identified. The Nass and Reeves research described how people respond to social cues from computers. It didn't prescribe building a character that interrupts your workflow with unsolicited advice. The research showed that users would be polite to a "friendly" program. It didn't show that users would enjoy being interrupted by one. ## The Backlash Builds Into Something Cultural By 1999, hating Clippy had become something close to a universal experience among computer users. It transcended demographics. Tech journalists wrote columns about how much they hated him. Office workers shared tips on how to disable him. IT departments began including "turn off the Office Assistant" in their standard setup procedures for new machines. Smithsonian Magazine would later call Clippy "one of the worst software design blunders in the annals of computing." Time Magazine included him in a 2010 article listing the fifty worst inventions, alongside asbestos and the Ford Pinto. For context, that list also included chlorofluorocarbons, which destroyed the ozone layer. Clippy was ranked alongside actual environmental disasters. The backlash wasn't just cultural noise. It represented a genuine user experience problem at scale. Microsoft Office was the dominant productivity suite on the planet. Hundreds of millions of people used it daily. And every single one of them had to actively figure out how to disable a feature that the company had spent millions of dollars developing and testing. ## The Removal That Microsoft Turned Into Marketing On April 11, 2001, Microsoft announced that Clippy would be removed from future releases of Office. But the company didn't just quietly kill the feature. They turned the removal into a marketing event, which, depending on your perspective, was either brilliant or deeply strange. On May 31, 2001, during the Office XP launch event in New York City, a man dressed in a Clippit mascot costume interrupted the keynote presentation. He stumbled onto stage, gave a speech begging for his job back, and was then dragged off stage by a comically oversized magnet. The crowd loved it. Microsoft launched a promotional website called officeclippy.com that featured Flash cartoons depicting an unemployed Clippy trying to find work. There was a parody song. There was a mini-game called "Office XP: Xtract Paperclip" where players fought off hordes of Clippys with office supplies. Microsoft had figured out something important: people enjoyed Clippy more as a joke than they ever had as a feature. In Office XP itself, the Office Assistant was still technically present in the code but was no longer installed by default. You had to actively choose to enable it. Almost nobody did. The final removal came with Office 2007, which shipped on January 30, 2007, with the Office Assistant feature fully stripped out. It has not returned in any subsequent version of Microsoft Office. ## The Lesson Microsoft Learned (and the Tech Industry Didn't) Look. The Clippy story is often framed as a simple failure: Microsoft made something annoying, users complained, Microsoft removed it. But the actual lesson is more nuanced than that, and it's one that matters more today than it did in 1997. What went wrong with Clippy wasn't the research. The Nass and Reeves findings about people treating computers as social actors were real and have been validated repeatedly. What went wrong was the application. Microsoft took a descriptive finding, people respond to computers socially, and turned it into a prescriptive design choice: therefore, let's create a social character that proactively inserts itself into the user's workflow. The gap between those two things is enormous. The fact that people unconsciously respond to social cues from software doesn't mean they want software that acts social. People don't want their word processor to have a personality. They want it to work. This is essentially the same tension that exists today with AI assistants. Every major tech company is now building products that proactively offer suggestions, anticipate needs, and insert themselves into workflows. The language has changed. Nobody calls them "Office Assistants" anymore. They're "copilots" and "AI companions" and "intelligent assistants." But the fundamental design question is identical to the one Microsoft faced in 1996: how do you build a helpful assistant that doesn't become an annoying interruption? The answer, which Clippy demonstrated at great expense, is that the assistant must be invisible until summoned. It must never assume it knows what the user wants. It must never demand attention. And it must never, ever pop up to ask if you're writing a letter when you are obviously writing a letter. ## Clippy's Afterlife: From Shame to Icon Here's the strangest part of the Clippy story. After being universally hated for a decade, Clippy became beloved. It started as ironic nostalgia. Around the early 2010s, Clippy memes began appearing on social media. People who had grown up with Clippy and cursed his existence as children now shared "It looks like you're..." jokes with genuine affection. The character that had symbolized everything wrong with late-1990s software design became a mascot for retro internet culture. In July 2021, Microsoft tweeted that if a photo of Clippy received 20,000 likes, they would replace the standard paperclip emoji in Microsoft 365 with Clippy. The tweet received over 170,000 likes, and Microsoft followed through, updating the paperclip emoji to a Clippy-inspired design as part of a broader refresh of 1,800 emojis across its products. The character that Microsoft had publicly fired in 2001 was being invited back, not as a feature, but as a mascot. Kevan Atteberry, Clippy's original designer, has had his own complicated relationship with his creation. He told interviewers that he used to be "so embarrassed" by Clippy that he left it out of his design portfolio for years. But the character's cultural longevity changed his perspective. As he put it, "It's important to me that people remember Clippy because as long as they do, I have cachet." He also revealed, in what might be the most perfect ironic detail in the entire story, that he designed Clippy on an Apple Macintosh. ## What Clippy Got Right (Sort Of) The final irony is this. Clippy was trying to solve a real problem. In the mid-1990s, most computer users were genuinely intimidated by software. Office 97 was complex. Word alone had hundreds of features that most users never discovered. The idea that software should proactively help users find the features they need wasn't wrong. It was just twenty years too early, and the execution was twenty years too crude. Modern AI assistants, the ones built into Word, Google Docs, and every other productivity tool in 2026, do essentially what Clippy was trying to do. They watch what you're working on. They suggest improvements. They offer help. The difference is that they're better at reading context, better at staying quiet when they should, and better at providing help that's actually useful. Clippy was the prototype. A $100 million prototype that everybody hated, but a prototype nonetheless. The research behind it was sound. The character design, according to the focus groups, was optimal. The technology just wasn't ready to deliver on the promise. The gap between "it looks like you're writing a letter" and a genuinely intelligent writing assistant turned out to be about three decades of machine learning research. In that light, Clippy wasn't a failure. He was premature. Which, if you're a paperclip with googly eyes and a legacy that won't die, is probably the kindest interpretation available. ## Frequently Asked Questions What was Clippy and when was it introduced? Clippy, officially named Clippit, was an animated paperclip character that served as the default Office Assistant in Microsoft Office 97, released to retail in January 1997. It would appear on screen to offer help and suggestions based on what the user appeared to be doing, such as writing a letter or formatting a document. Who designed Clippy? Clippy was designed by illustrator Kevan Atteberry as part of a larger project that produced roughly 250 character concepts. The designs were tested through six months of focus groups conducted with the help of Stanford social psychologists. Clippy was selected because focus group participants rated the paperclip character highest in trustworthiness and engagement. Why was Clippy so annoying? Clippy was enabled by default and would proactively interrupt users with unsolicited suggestions. The character animated to grab attention at inopportune moments, and dismissing it didn't prevent it from returning. The focus groups that selected Clippy measured first impressions over brief interactions, failing to predict how the same character traits would feel after hours of prolonged exposure during real work. When was Clippy removed from Microsoft Office? Clippy was disabled by default starting with Office XP in 2001 and fully removed from the software in Office 2007. Microsoft turned the removal into a marketing event, staging a mock firing at the Office XP launch in New York City and creating a promotional website with Flash cartoons depicting an unemployed Clippy. Was Microsoft Bob related to Clippy? Yes. Microsoft Bob, launched in March 1995 and discontinued within a year, was Clippy's direct predecessor. Both products were built on the same academic research from Stanford professors Clifford Nass and Byron Reeves about people treating computers as social actors. When Bob failed as a full desktop replacement, Microsoft scaled the concept down to an in-application assistant, which became Clippy. Is Clippy coming back? Not as a functional assistant, but as a cultural icon. In 2021, Microsoft replaced the standard paperclip emoji in Microsoft 365 with a Clippy-inspired design after a public Twitter vote. Clippy now appears in Microsoft merchandise, stickers, and marketing materials as a nostalgic mascot rather than a product feature. --- # The Night Windows 95 Launched and Changed Everything URL: https://404memoryfound.com/posts/the-night-windows-95-launched.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-04-05 Topics: Software & Apps, Then vs Now ## The Biggest Software Launch in History Started With a Rock Song Picture this: August 24, 1995. It's a Thursday night, and there are people camping outside a CompUSA (https://404memoryfound.com/posts/is-compusa-still-around.html). Not for concert tickets. Not for sneakers. For a piece of software that came in a cardboard box. Windows 95 was about to drop, and somehow, some way, Microsoft had convinced the entire planet that an operating system was the event of the year. And you know what? They were right. I was seven years old when Windows 95 came out. I didn't fully understand what was happening, but I remember my dad talking about it like it was a moon landing. He worked the floor at Circuit City (https://404memoryfound.com/posts/what-happened-to-circuit-city-electronics.html), so he was on the front lines. He told me later that the store opened at midnight and there was a line wrapped around the building. For software. In a box. That cost $209. But here's the thing. Windows 95 wasn't just software. It was a cultural moment. And to understand why people lost their minds over it, you have to understand what came before. ## Before the Start Button, There Was Just a Blinking Cursor If you used a PC before Windows 95, you know the pain. Windows 3.1 was fine, I guess, in the way that a bicycle with one flat tire is fine. It technically worked. You had Program Manager, which was basically a bunch of boxes inside boxes. You had to know what you were doing. There was no Start menu. There was no taskbar. You couldn't right-click on anything useful. Multitasking was a joke, more of a concept than a reality. If one program crashed, the whole system went down with it. And underneath all of it was DOS. You literally had to type commands into a black screen to do basic things. Want to play a game? Better know what "cd c:\games\doom" means. Want to install something? Hope you brought your stack of floppy disks and a lot of patience. I remember watching my dad navigate DOS like it was some kind of secret language. He'd type these commands and the screen would scroll and eventually, if everything went right, a program would start. If it didn't, well, you'd start over and try to figure out what you typed wrong. The Mac was easier to use, sure, but it was also expensive. A Power Macintosh in 1995 could run you $2,500 or more. Most families weren't dropping that kind of money. So the PC world was stuck in this awkward place where the hardware was getting better every year, but the software still felt like it was designed for engineers. Windows 95 was supposed to fix all of that. And honestly, it kind of did. ## The Start Menu Was Revolutionary (No, Seriously) It sounds almost silly to say now, but the Start menu was a genuine breakthrough. Before Windows 95, there was no single, obvious place to go when you wanted to do something on your computer. The Start button gave you exactly that. Click it, and everything was right there: your programs, your settings, your files, the ability to shut down without typing a command. The taskbar at the bottom of the screen showed you what was running. You could switch between programs by clicking their names. You could minimize a window and it wouldn't just vanish into the void. These things sound obvious now, but in 1995, this was a revelation for the average person sitting in front of a beige Compaq (https://404memoryfound.com/posts/compaq-lost-pc-business-hp-merger.html) in their spare bedroom. Windows 95 also introduced long file names. Before this, you were limited to eight characters plus a three-character extension. That's it. Your school paper couldn't be called "English_Essay_Final_Draft.doc." It had to be something like "ENGESS~1.DOC." The fact that you could now name a file something a human being could actually read felt like freedom. And then there was Plug and Play. In theory, you could connect a new printer or modem and Windows would just figure it out. In practice, it was more like Plug and Pray, which became one of the great running jokes of the era. But the intention was right, and it worked often enough to matter. Before Plug and Play, installing a new piece of hardware meant manually configuring IRQ settings and DMA channels, which sounds like technical jargon because it absolutely was. Regular people had no business dealing with that stuff, and Windows 95 was the first real attempt to take it off their plate. The operating system also brought 32-bit computing to the mainstream. Windows 3.1 was essentially a 16-bit system running on top of DOS. Windows 95 was a hybrid, still carrying some 16-bit code for compatibility, but the core was 32-bit. This meant programs could address more memory, run more efficiently, and do things that simply weren't possible before. For software developers, it was a new world. For gamers, it meant DirectX was coming, and that changed everything. ## Microsoft Spent $300 Million to Make You Care About Software Here's where it gets wild. Microsoft didn't just release Windows 95. They launched it. Like a blockbuster movie. Like a stadium concert. The marketing budget was $300 million, which in 1995 dollars is absolutely staggering. For context, the entire production budget of "Jurassic Park" two years earlier was $63 million. Microsoft spent nearly five times that amount just telling people about a piece of software. The centerpiece was the launch event on the Microsoft campus in Redmond, Washington. Over 12,500 people were invited. It took a crew of more than 200 people over 20 days to build the setup. A massive tent was erected on the campus grounds. Jay Leno hosted the whole thing, cracking jokes on stage with Bill Gates while the crowd cheered like it was a rock concert. The whole event was broadcast live via satellite to 42 cities around the world. And speaking of rock concerts: the soundtrack. Microsoft licensed "Start Me Up" by the Rolling Stones as the official theme song. The actual cost was $3 million, paid directly to the band, as later confirmed by former Microsoft COO Bob Herbold. For weeks afterward, rumors swirled that it was $10 to $14 million, numbers that may have been strategically leaked by the Stones' camp to boost their perceived value. Classic move. Either way, it worked. You couldn't hear that riff without thinking of Windows 95. "You make a grown man cry." The Rolling Stones probably didn't write that lyric about installing printer drivers on Windows 95, but it fit. Microsoft also commissioned a 30-minute promotional video starring Jennifer Aniston and Matthew Perry, fresh off the first season of "Friends." It was billed as the first "cyber sitcom," and it was exactly as awkward as that sounds. The two of them wandered around a fake house, pretending to discover the wonders of Windows 95 while delivering scripted banter that aged like milk left on a summer porch. It's on the Internet Archive now, and honestly, it's worth watching just for the time capsule factor. Perry pretends to be confused by email. Aniston acts amazed by clip art. It's beautiful in the worst possible way. The global spectacle didn't stop at Redmond. In New York, the Empire State Building was lit up in red, yellow, and green to match the Windows logo colors. In Toronto, the CN Tower displayed a 300-foot banner featuring the Start button. The Times of London printed an entire edition sponsored by Microsoft, wrapping the paper in Windows 95 branding. This was before tech companies routinely dominated culture. Apple wouldn't do anything close to this level of cultural saturation until the iPod (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html) silhouette ads nearly a decade later. Microsoft was basically inventing the playbook. ## Midnight Lines and the Birth of Tech Hype Culture The night before launch, people lined up outside electronics stores across the country. This was years before the iPhone would make midnight tech launches a regular thing. In 1995, nobody had ever seen anything like it for a piece of software. CompUSA stores opened at midnight. Best Buy had special events. Even small-town computer shops got in on it. My dad said the Circuit City in Panorama City had maybe 40 or 50 people in line, which doesn't sound like a lot until you remember these were people waiting to buy a $209 operating system at midnight on a weekday. Some of them brought lawn chairs. A few had radios. One guy reportedly brought his kid, who fell asleep in the back of a station wagon in the parking lot while his dad waited. Microsoft moved one million copies in the first four days. Seven million copies sold in the first five weeks. By the end of the first year, 40 million copies were out in the wild. Those are numbers that most software companies today would kill for, and this was 1995, when a significant chunk of American households didn't even own a computer yet. The retail packaging itself was a big deal. The box was heavy. It had a sky-and-clouds design that felt oddly aspirational for an operating system. Inside was a CD-ROM (or, if you were unlucky, a set of 13 floppy disks, which is a sentence that should make any younger reader pause and appreciate how far we've come). There was a thick manual, a quick start guide, and that distinctive certificate of authenticity with a holographic sticker. Unboxing it felt like an event, which was exactly what Microsoft wanted. ## What Windows 95 Actually Changed Beyond the hype, Windows 95 genuinely shifted how people interacted with computers. It was the first version of Windows that felt like a complete operating system rather than a graphical shell sitting on top of DOS. Technically, DOS was still under there, but for most users, you never had to see it. It also introduced the concept of the desktop as we still know it today. Icons on a background. A recycle bin in the corner. Double-click to open things. Right-click for options. The basic grammar of how you use a computer in 2026 was largely established in August 1995. The file management experience changed completely. Windows Explorer replaced the old File Manager with a dual-pane view that let you navigate your files like folders in a filing cabinet. The concept of "My Computer" as an icon on the desktop gave people a single place to see their hard drives, floppy drives, and CD-ROM drives. It sounds trivial, but it was the first time most PC users had a visual, intuitive way to understand what was actually inside their machine. And then there was the internet. Windows 95 didn't ship with a web browser in the original release, but the Plus! add-on pack included Internet Explorer 1.0, and later versions bundled IE directly into the OS. This decision would eventually lead to the massive antitrust lawsuit against Microsoft, but in 1995, it meant that millions of people suddenly had a way to get online without figuring it out themselves. Microsoft also included a built-in TCP/IP stack, which made connecting to the internet dramatically easier than it had been on Windows 3.1. The irony is that while Windows 95 helped bring the internet to the masses, Microsoft was actually late to the internet game. Bill Gates famously pivoted the entire company toward the internet in late 1995 after realizing Netscape (https://404memoryfound.com/posts/is-netscape-still-around.html) was eating their lunch. His "Internet Tidal Wave" memo, written in May 1995 just months before the Windows 95 launch, is one of the most important documents in tech history. But for regular people sitting at home, Windows 95 plus a 28.8k modem plus an AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html) CD was the gateway drug to the World Wide Web. ## The Problems Nobody Talks About Let's be honest, though. Windows 95 was not perfect. Not even close. The blue screen of death became a cultural icon for a reason. Windows 95 crashed. A lot. The 16-bit and 32-bit compatibility layer was held together with duct tape and prayers. Memory management was rough. If you ran too many programs, the whole thing would lock up and you'd lose whatever you were working on. There was no auto-save in most applications back then, so a crash could mean hours of lost work. The phrase "did you save?" became a household reflex. Driver support was a nightmare in the early months. Plug and Play was more of an aspiration than a reality for a lot of hardware. Sound cards, printers, modems: getting them all to work together without conflicts required patience and sometimes a physical trip to the store to buy a different card. I remember my dad spending an entire Saturday trying to get our SoundBlaster to work after upgrading to 95. He was on the phone with tech support for two hours. The fix, if I remember right, involved manually editing a configuration file. And the hardware requirements were steep for the time. Microsoft said you needed 4 MB of RAM, but realistically, you wanted at least 8 MB to have a decent experience, and 16 MB if you wanted to run anything serious. A lot of people bought Windows 95 only to discover their 486 PC couldn't really handle it well. The upgrade path often meant buying a new computer entirely, which was a several-hundred-dollar proposition that Microsoft's marketing conveniently glossed over. The initial lack of software was also a real issue. Not every Windows 3.1 program ran perfectly on 95. Some didn't run at all. DOS games, in particular, could be tricky. If you had a kid who wanted to play their DOS games and a parent who wanted the new Windows experience, you sometimes ended up with a dual-boot situation that nobody in the household fully understood. But none of that mattered to the cultural narrative. Windows 95 had won the hearts and minds war. Even its flaws became part of the shared experience. Everyone had a blue screen story. Everyone had a "Plug and Pray" anecdote. It was communal suffering, and somehow, that made people love it more. ## The Aftermath: 98, XP, and the Long Shadow Windows 95 got three major updates: the OEM Service Release versions 1, 2, and 2.5, which added USB support, FAT32, and Internet Explorer 3.0 and later 4.0. These updates fixed a lot of the early issues and kept Windows 95 relevant through 1997 and into 1998, when Windows 98 finally arrived. Windows 98 was essentially Windows 95 with better hardware support, better internet integration, and fewer crashes. It was a refinement, not a revolution. The real next leap came with Windows XP in 2001, which finally ditched the Windows 9x kernel entirely and moved consumer Windows onto the NT architecture. That was the technical revolution. But the interface? The way you actually used the thing? That was still recognizably Windows 95. Windows 95 also birthed DirectX, Microsoft's gaming API that would eventually make Windows the dominant PC gaming platform. The first version of DirectX shipped with Windows 95, and while it was rough around the edges, it laid the groundwork for the Windows gaming ecosystem that still dominates today. Without DirectX, there's no Xbox. That's not an exaggeration. Microsoft has said as much. ## The Legacy That Still Shows Up Every Time You Hit Start Here's what gets me about Windows 95. It's been over 30 years, and the basic design language is still the same. The taskbar. The Start menu. The system tray in the bottom right corner with the clock. The desktop with icons. The recycle bin. Microsoft has redesigned all of these things multiple times across XP, Vista, 7, 8, 10, and 11, but the fundamental layout that an average person uses to navigate a PC was established on August 24, 1995. Windows 8 tried to kill the Start menu in 2012, replacing it with a full-screen tile interface. Users revolted so hard that Microsoft brought the Start menu back in Windows 10 three years later. That's how deeply Windows 95's design is embedded in how people expect a computer to work. The launch also set the template for tech hype culture. The midnight lines, the celebrity endorsements, the massive marketing budgets, the idea that a product release could be an event. Apple would perfect this formula with the iPod, iPhone, and iPad launches in the 2000s, but Microsoft did it first with a beige box and a CD-ROM. Windows 95 didn't just change how computers worked. It changed how we felt about them. For the first time, a computer felt like it was on your side. My dad kept his original Windows 95 CD for years. It sat in a drawer in the kitchen next to batteries and takeout menus, in its jewel case with the cloud-sky artwork on the front. Eventually it got scratched up beyond use, but by then it didn't matter. The world had moved on to 98, then XP, then everything after. But every single one of those sat on the foundation that Windows 95 built. If you were there that night in August 1995, standing in line at a store or watching the news coverage or just hearing your parents talk about it, you know the feeling. Something shifted. Computers stopped being intimidating and started being exciting. And it all started with a Start button, a rock song, and a $300 million bet that regular people were ready for the future. ## Frequently Asked Questions How much did Windows 95 cost at launch? The full retail version of Windows 95 cost $209.95. The upgrade version, for people already running Windows 3.1, was $109.95. Adjusted for inflation, the full version would be roughly $420 today. How many copies did Windows 95 sell? Microsoft sold one million copies in the first four days and seven million in the first five weeks. First-year sales reached approximately 40 million copies. Did Microsoft really pay the Rolling Stones $3 million for "Start Me Up"? Yes. Former Microsoft COO Bob Herbold confirmed the actual licensing fee was $3 million. Early rumors of $10 to $14 million were likely inflated, possibly by the Stones' camp to increase their perceived market value. Could Windows 95 connect to the internet? Yes, but the original release didn't include a web browser. It did include a built-in TCP/IP stack for internet connectivity. Internet Explorer 1.0 was available through the Plus! add-on pack, and later versions of Windows 95 bundled Internet Explorer directly. What were the minimum system requirements for Windows 95? Microsoft listed 4 MB of RAM and a 386DX processor as minimums, but most users needed at least 8 MB of RAM and a 486 for a usable experience. A Pentium processor with 16 MB of RAM was recommended for comfortable performance. Why was the Windows 95 launch such a big deal? It was the first time a software release was marketed like a blockbuster entertainment event. The $300 million marketing campaign, celebrity appearances, and global stunts like lighting up the Empire State Building created a cultural moment that transcended the tech world. It also represented a genuine leap forward in making PCs accessible to everyday people. What happened to Windows 95? Windows 95 was succeeded by Windows 98 in June 1998, which built on its foundation with better hardware support and internet integration. Microsoft officially ended support for Windows 95 on December 31, 2001. The Windows 95 kernel architecture was eventually replaced entirely by the NT-based Windows XP in October 2001. --- # What Happened to Commodore, the Company That Won and Lost URL: https://404memoryfound.com/posts/what-happened-to-commodore-computers.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-05 Topics: Hardware, Business Blunders ## The Best-Selling Home Computer Company You Barely Remember In 1983, Commodore International had roughly 50% of the U.S. home computer market. Not 50% of some niche subcategory. Half of all home computers sold in America had the Commodore logo on them. The Commodore 64, their flagship machine, would go on to become the best-selling single computer model of all time according to the Guinness Book of World Records, with estimated sales somewhere in the range of 12.5 to 17 million units, depending on which source you trust. Some claims push that number to 30 million, though those figures are disputed by historians who have dug into the actual production records. By 1994, Commodore was bankrupt. That's a ten-year slide from total dominance to complete collapse. And the reasons behind it aren't the usual "technology moved on" story. Commodore didn't lose because a better product showed up. They lost because the people running the company after its visionary founder left seemed determined to destroy it from the inside. ## Jack Tramiel Built Commodore by Waging Price Wars You can't understand Commodore without understanding Jack Tramiel. Born Idek Tramielski in Lodz, Poland in 1928, Tramiel survived the Auschwitz concentration camp as a teenager. He immigrated to the United States after the war, joined the Army, learned to repair typewriters, and eventually founded a typewriter repair company in the Bronx in 1954. That company became Commodore Business Machines. Tramiel's entire business philosophy fit into a single phrase he repeated constantly: "We need to build computers for the masses, not the classes." He meant it. While Apple was positioning the Apple II as a premium product and IBM was selling PCs to businesses at premium prices, Tramiel was slashing costs and margins to put a computer in every living room. The Commodore PET arrived in 1977, the same year as the Apple II and the TRS-80. It was one of the first commercially available personal computers, an all-in-one unit with a built-in monitor, keyboard, and cassette drive. The PET found its niche primarily in education, particularly in Canada, where it became a classroom staple. The VIC-20 followed in 1981 and became the first computer to sell one million units, a milestone that seemed remarkable at the time but was about to be dwarfed. Then came the Commodore 64 in August 1982, priced at $595. Within a year, aggressive pricing pushed it below $200, and the home computer price war of 1983 began in earnest. Tramiel's strategy was brutally simple. Commodore owned their own chip fabrication facility, MOS Technology, which they had acquired in 1976. This meant they could produce the C64's custom chips, the SID sound chip and VIC-II graphics chip, at a fraction of what competitors paid for off-the-shelf components. Tramiel used this vertical integration as a weapon, dropping prices until competitors simply couldn't keep up. The SID chip deserves special mention. Designed by Bob Yannes, who would later found Ensoniq, it was a genuine synthesizer on a chip. Three voices, multiple waveforms, a programmable filter. The C64's sound capabilities were so far ahead of anything else in its price range that musicians and demo scene programmers were still finding new tricks for the SID chip well into the 2000s. It wasn't just a beeper. It was an instrument. The casualties of Tramiel's price war were significant. Texas Instruments exited the home computer market in October 1983, booking a $330 million loss on the TI-99/4A. Mattel shut down its Intellivision computer expansion. Coleco's Adam computer was a disaster. Atari (https://404memoryfound.com/posts/who-owns-atari-now.html)'s home computer division was bleeding money. The home computer shakeout of 1983 and 1984 killed or wounded nearly every competitor, and Commodore was the one doing most of the killing. ## The Departure That Started the Decline In January 1984, Jack Tramiel resigned from Commodore. The official story was a disagreement with Irving Gould, the chairman and majority shareholder who had bankrolled the company since the 1960s. The specifics vary depending on who's telling it, but the core issue was control. Tramiel wanted to bring his sons into the business. Gould didn't want that. Tramiel wanted to move into business computers. Gould had different priorities. After years of tension, Tramiel walked. Within months, Tramiel bought the consumer division of Atari from Warner Communications for roughly $50 in cash and $240 million in assumed liabilities. He then began competing against his former company. That detail matters because Commodore's greatest asset, the man who understood pricing strategy, manufacturing economics, and the consumer market better than anyone in the industry, was now working for the other side. What Tramiel left behind was a company with massive market share, a profitable product line, and no one at the top who understood how to sustain it. Gould's approach to management was primarily financial. He understood balance sheets and investment returns. What he didn't understand, or didn't care about, was product vision. And in the personal computer business of the mid-1980s, product vision was everything. ## The Amiga Was a Masterpiece That Nobody Marketed Here's where the story gets painful for anyone who cares about technology being recognized for what it actually is. In 1985, Commodore released the Amiga 1000. The machine had a complicated origin. It was originally designed by a small company called Hi-Torro (later renamed Amiga Corporation) founded by Jay Miner, the engineer who had designed the Atari 2600's graphics chip. Commodore acquired Amiga Corporation in 1984, beating out a competing offer from Atari (now run by Tramiel), in a transaction that generated lasting bitterness between the two companies. From a technical standpoint, the Amiga 1000 was years ahead of everything else on the market. It had a custom chipset, with three co-processors named Agnus, Denise, and Paula, that handled graphics, sound, and I/O independently from the main CPU. This meant it could do things that IBM PCs and Macs simply couldn't. Preemptive multitasking. Stereo sound as standard. A color palette of 4,096 colors when most PCs were stuck with 16. Hardware scrolling and sprite capabilities that made game developers weep with joy. Video capabilities that made it the default choice for TV production and desktop video throughout the late 1980s and early 1990s. The Video Toaster, a third-party hardware and software package developed by NewTek, was used to produce the special effects on shows like "Babylon 5" and "SeaQuest DSV." NewTek chose the Amiga specifically because no other consumer hardware could handle real-time video switching and 3D rendering at that price point. A broadcast-quality video production setup that would have cost $50,000 or more with dedicated hardware could be built around an Amiga for a few thousand dollars. This was not an incremental improvement. It was an order-of-magnitude shift in the economics of video production. In 1994, BYTE magazine wrote: "The Amiga was so far ahead of its time that almost nobody, including Commodore's marketing department, could fully articulate what it was all about. Today, it's obvious the Amiga was the first multimedia computer, but in those days it was derided as a game machine." That BYTE quote captures the central tragedy. Commodore had the most technically advanced personal computer on the market and couldn't figure out how to sell it. Their marketing was inconsistent, underfunded, and confused. They positioned the Amiga 500 as a game machine in one campaign and a business tool in another. They never established a clear identity the way Apple did with the Macintosh or IBM did with the PC. The comparison to Apple is instructive. In 1984, Apple aired the famous "1984" Super Bowl commercial and established the Macintosh as a creative, rebellious alternative to IBM. Apple spent $1.5 million on that single ad. Commodore had a technically superior product in the Amiga and marketed it with the creative energy of a regional car dealership. The ads they did produce were generic, forgettable, and badly targeted. Commodore seemed unable to decide whether they were selling to gamers, video professionals, or businesses, so they failed to convince any of those audiences. ## Mehdi Ali and the Slow-Motion Destruction If Jack Tramiel was the builder, Mehdi Ali was the demolition crew. Ali joined Commodore in 1986 as a special adviser and became president in 1989. His background was in management consulting, not technology, and it showed. Under Ali, Commodore's already weak marketing got worse. Research and development funding was cut repeatedly. And Ali's compensation became a point of genuine outrage within the company and among industry observers. In 1989, Ali earned $1.38 million in salary. By 1990, that figure had risen to approximately $2 million in base salary alone, not counting bonuses. For context, John Akers, the CEO of IBM, one of the largest technology companies on Earth, earned $713,000 that same year. Ali was running a company with a fraction of IBM's revenue and paying himself nearly three times what the IBM chief was making. The engineers who were actually building Commodore's products noticed. The resentment was deep and widespread. The strategic decisions under Ali's leadership ranged from questionable to self-destructive. The Amiga 600, released in 1992, was a cost-reduced replacement for the popular Amiga 500. The problem was that nobody wanted it. It was only marginally cheaper than the 500 but less capable in several ways, with fewer expansion slots and an incompatible form factor that broke compatibility with many peripherals. Meanwhile, the Amiga 1200, which was genuinely excellent and featured the new AGA chipset, couldn't be manufactured in sufficient quantities to meet demand. Commodore was simultaneously overproducing a product nobody wanted and underproducing one that everyone did. This is not a complex problem to diagnose. It is, however, a very difficult problem to solve when the people making decisions aren't listening to the people who understand the market. The CD32, released in 1993, was actually an interesting product: the world's first 32-bit CD-ROM based game console. It sold reasonably well in Europe, where Commodore still had significant brand recognition. But it couldn't be sold in the United States due to a patent dispute with Cadtrak Corporation, which obtained a $10 million judgment against Commodore. Rather than settle the patent issue and get the CD32 into the massive American market, Commodore's management let the injunction stand. The CD32's European success couldn't save the company alone. Ali also reportedly sabotaged a licensing deal with Sun Microsystems, not once but twice, that would have allowed Sun to license Commodore's hardware designs. The details of these negotiations aren't fully documented, but multiple former Commodore employees have described them in interviews and written accounts as potentially transformative deals that Ali killed for reasons that were never adequately explained. By the end, Commodore's engineers actively despised the executive team. Dave Haynie, one of Commodore's most respected hardware engineers, filmed a documentary called "The Deathbed Vigil" on the day Commodore's offices closed in West Chester, Pennsylvania. He walked through the empty halls with a camera, interviewing colleagues as they packed boxes and said goodbye. It's a raw, bitter document of talented people watching their life's work get destroyed by people who didn't understand it. ## The Numbers Tell the Final Story In fiscal year 1993, Commodore reported a loss of $357 million. That's not a typo. A company that had dominated the home computer market a decade earlier lost over a third of a billion dollars in a single year. Revenue had been declining for years. The C64, which had been printing money throughout the mid-1980s, was finally winding down. The Amiga product line, while technologically excellent, was generating nowhere near enough volume to sustain the company, partly because of the marketing and distribution failures described above. On April 26, 1994, Commodore shut down its Amiga division. Three days later, on April 29, the company filed for voluntary liquidation in the Bahamas, where it was incorporated. It was not acquired. It was not restructured. It simply ceased to exist. The brand and technology were eventually purchased at auction by a German company called Escom in April 1995 for approximately $14 million. To put that in perspective, Commodore's annual revenue had been over $1 billion at its peak in the late 1980s. Escom itself went bankrupt the following year, and the Commodore assets passed through several more owners over the next decade, including Gateway 2000 (https://404memoryfound.com/posts/what-happened-to-gateway-2000-cow-box-computer.html) (which bought the Amiga technology in 1997) and various smaller companies that licensed the name without ever producing anything significant. ## Why Commodore Matters More Than You Think The standard narrative about Commodore is that it was a casualty of the IBM PC standard. The PC won, everything else lost, end of story. That narrative is wrong, or at least deeply incomplete. Commodore's actual technical trajectory, from the C64 through the Amiga line, represented a genuinely different approach to personal computing. The custom chipset architecture of the Amiga anticipated how modern computers work, with dedicated hardware for specific tasks rather than forcing the CPU to do everything. The concept of a GPU handling graphics independently? The Amiga had a version of that in 1985. The idea that a personal computer should handle audio and video as first-class capabilities? The Amiga was built around that premise years before "multimedia PC" became a marketing buzzword in the early 1990s. This is essentially what happened with dedicated graphics cards in the late 1990s and early 2000s. Nvidia and ATI didn't invent the concept of offloading graphics processing to specialized hardware. They just brought it to the IBM PC platform a decade after the Amiga had already proven it worked. The Amiga's operating system, AmigaOS, was also remarkably advanced. It offered preemptive multitasking and a microkernel architecture in 1985, at a time when Mac OS was cooperative-multitasking only (and wouldn't get preemptive multitasking until OS X in 2001) and Windows was still a graphical shell on top of DOS. The AmigaOS was lean, fast, and efficient in ways that inspired admiration from developers who worked with it and frustration from those same developers when they realized the company behind it was incapable of capitalizing on its advantages. The difference between Commodore and, say, Apple in the 1990s is that Apple had Steve Jobs come back. Commodore had Mehdi Ali. Apple nearly went bankrupt in 1997 and was rescued by a visionary leader who understood both technology and marketing. Commodore went bankrupt in 1994 under a president who understood neither. They weren't stupid, the engineers who built the Amiga. They just didn't get to choose who ran the company. ## The Collector's Market and the Community That Refused to Die One of the more interesting postscripts to the Commodore story is that the community never really went away. Retro computing enthusiasts have kept the C64 and Amiga alive through emulation, hardware preservation, and new software development. The C64 homebrew scene is still active, with new games and demos being released regularly. The demo scene, in particular, continues to push the C64's SID chip and VIC-II graphics to produce audiovisual experiences that seem physically impossible on 1982 hardware. New Amiga-compatible hardware, like the Vampire accelerator cards and the MiSTer FPGA project, continues to be developed by a small but dedicated community. These aren't museum pieces. People are actively using and developing for these platforms decades after the parent company ceased to exist. Working Commodore 64 units regularly sell for $50 to $200 on the collector's market, depending on condition and included accessories. Rare variants and boxed sets with original packaging can fetch significantly more. Amiga systems, particularly the Amiga 1200 and 4000 models, have become genuinely expensive collector's items, with prices regularly exceeding $500 to $1,000. The THEC64, a licensed full-size recreation of the Commodore 64 with built-in games, a working keyboard, and HDMI output, was released in 2019 and sold well enough to prove that the nostalgia market for Commodore remains real, even among people who weren't alive when the originals were sold. A mini version preceded it in 2018. Both were well-received, which says something about the enduring appeal of a product line that officially died 25 years before the recreations hit shelves. Which brings us to the real question this story raises. Not "what killed Commodore?" but "what could Commodore have become?" If Tramiel had stayed, or if the post-Tramiel leadership had been even marginally competent, would Commodore be a name that sits alongside Apple and Microsoft today? The technology was there. The market position was there. The engineering talent was there. What wasn't there was the leadership to hold it all together. And in the end, that's the most common way great companies die. Not from external competition, but from internal failure. The technology doesn't betray you. The people do. ## Frequently Asked Questions How many Commodore 64s were sold? The Guinness Book of World Records cites approximately 30 million units, making it the best-selling single computer model in history. However, independent researchers and historians place the more realistic figure somewhere between 12.5 and 17 million units. The discrepancy comes from Commodore's inconsistent record-keeping and the difficulty of tracking sales across dozens of international markets over a decade of production. When did Commodore go bankrupt? Commodore shut down its Amiga division on April 26, 1994, and filed for voluntary liquidation on April 29, 1994. The company's assets were eventually sold at auction to the German company Escom in April 1995 for approximately $14 million. Why did Jack Tramiel leave Commodore? Tramiel resigned in January 1984 following a dispute with Irving Gould, the chairman and majority shareholder. The conflict centered on corporate control, including Tramiel's desire to bring his sons into company management. Within months, Tramiel purchased the consumer division of Atari and began competing against Commodore directly. What made the Amiga so special? The Amiga used a custom chipset architecture that handled graphics, sound, and I/O independently from the CPU. This gave it preemptive multitasking, a 4,096-color palette, and stereo sound at a time when most PCs had 16 colors and a single-channel beeper. It became the standard platform for TV and video production in the late 1980s and early 1990s thanks to products like the NewTek Video Toaster. Does Commodore still exist? The Commodore brand name has changed hands multiple times since the original company's bankruptcy. Various companies have licensed the name for products like smartphones and retro gaming devices, but none have any meaningful connection to the original Commodore International or its engineering team. The Amiga technology was purchased by Gateway 2000 in 1997 and has passed through several owners since then. What was the Commodore 64 competing against? The C64's main competitors were the Apple II, the Atari 800, the TRS-80, the Sinclair ZX Spectrum (primarily in Europe), and eventually the IBM PC and its clones. Commodore's aggressive pricing strategy, enabled by their ownership of chip manufacturer MOS Technology, allowed them to undercut nearly all competitors on price while offering comparable or superior capabilities. What was the Commodore demo scene? The demo scene is a computer art subculture that produces audiovisual presentations called "demos" designed to push hardware to its absolute limits. The Commodore 64 demo scene has been active continuously since the 1980s, with programmers discovering new tricks for the SID sound chip and VIC-II graphics processor that the original designers never intended. It remains one of the most active retro computing communities in the world. --- # What Happened to Neopets, the Virtual World That Raised a Generation URL: https://404memoryfound.com/posts/what-happened-to-neopets-virtual-world.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-05 Topics: Internet Culture, Business Blunders In 1999, 150 million registered accounts sounded like a number you would associate with a country, not a website about virtual pets. But that is exactly where Neopets ended up. At its peak, it was one of the stickiest sites on the entire internet, pulling in 4 billion page views a month and converting a generation of kids into amateur capitalists, amateur web designers, and amateur pet enthusiasts, all at the same time. The trajectory of Neopets is not a simple story of a fad that faded. It is a case study in what happens when a product built by passionate creators gets passed through a chain of corporate owners, each one less equipped than the last to understand what made it special. And in 2023, a management buyout tried to undo two decades of damage. Whether it succeeds or not, the story of how we got here is worth understanding. ## Two Students, a Server, and a Lot of Virtual Creatures Adam Powell and Donna Williams (later Donna Powell) were students at the University of Nottingham in England when they started building Neopets in September 1999. The concept was deceptively simple: create a virtual pet, feed it, play games to earn Neopoints, and spend those Neopoints in an economy that the players themselves would drive. The site launched on November 15, 1999. The timing was almost absurdly perfect. The web was still young enough that a site built by two people in their spare time could compete for attention with products backed by millions in venture capital. Neopets did not have a marketing budget. What it had was a product that was genuinely fun, surprisingly deep, and perfectly calibrated for the demographics that the rest of the internet was ignoring: kids and teenagers. Within months, the site was growing faster than Powell and Williams could manage. By 2000, they had relocated from the UK to a small office in Glendale, California, hired a handful of employees, and started building out the features that would define the platform: the Neopian stock market, the auction house, user-created shops, HTML-customizable pet pages, and an ever-expanding library of browser-based mini-games. ## The Immersive Advertising Model Here is the thing about Neopets that most retrospectives gloss over: it was a business, and a surprisingly innovative one. The site pioneered what it called "immersive advertising," a model where brands did not just buy banner ads but became part of the game world itself. General Mills, Disney, and McDonald's all paid to have their products integrated into Neopets. There was a Cheerios-branded virtual cereal. There were McDonald's-themed games. The ads were not beside the content. They were inside it. This was years before anyone used the phrase "native advertising." Neopets figured out that if you made the ad feel like part of the experience, kids would engage with it voluntarily. The model worked. The company was profitable. And that profitability attracted attention from exactly the kind of company that tends to ruin things. ## Viacom Writes a $160 Million Check On June 20, 2005, Viacom acquired Neopets, Inc. for $160 million through its MTV (https://404memoryfound.com/posts/when-mtv-actually-played-music-rise-fall.html) Networks division. At the time of the acquisition, the site had approximately 25 million registered users, with 35 million unique visitors per month generating roughly 4 billion page views. Users had created over 140 million virtual creatures. By any metric, it was a juggernaut. The logic of the acquisition made sense on paper. Viacom owned Nickelodeon, MTV, and a portfolio of brands targeting young audiences. Neopets was the stickiest youth-oriented property on the internet. The average session length was over 40 minutes, a number that would make any social media company jealous even today. But the acquisition came with an immediate philosophical shift. Viacom announced plans to move away from immersive advertising and toward traditional banner ads. The reasoning was straightforward from a corporate perspective: immersive ads were complex to produce, difficult to scale, and raised questions about marketing to children. Banner ads were simple, measurable, and familiar to Viacom's existing advertisers. The problem was that immersive advertising was not just a revenue model. It was part of what made the experience feel cohesive. When you replaced it with banner ads, the site started feeling like what it technically always was: a free-to-play game sustained by advertising. The magic leaked out slowly. Adam Powell and Donna Williams left the company shortly after the acquisition, citing creative differences. This is the polite version of what happens when founders realize they no longer control their own creation. ## The Slow Decline Under Corporate Ownership Viacom did not destroy Neopets overnight. The site continued to operate, continued to attract millions of users, and continued to generate revenue. But the investment in new features slowed. The development team shrank. Bug fixes took longer. New content became less frequent. The site's visual design, which had always been charmingly homemade, started to feel simply outdated. The broader internet was changing, too. Facebook launched in 2004 and went public in 2006. YouTube was founded in 2005. The iPhone arrived in 2007. The attention of young users, the demographic Neopets depended on, was being pulled in a dozen new directions simultaneously. Neopets was a product of the late 1990s web, built on the assumption that users would sit at a desktop computer for long sessions and engage deeply with a single site. That assumption was evaporating. ## JumpStart and the Years of Neglect In March 2014, Viacom sold Neopets to JumpStart Games, a company primarily known for educational software aimed at young children. The sale price was not publicly disclosed, but given that Viacom had paid $160 million nine years earlier, the markdown was almost certainly dramatic. JumpStart's ownership is remembered by the Neopets community as the darkest era. The site became plagued by technical problems: persistent lag, broken features, lost inventory items, and security breaches. A data breach in 2022 exposed the personal information of approximately 69 million user accounts. Development of new content effectively stopped. The Neopets Team, as the developers had always been known, was reduced to a skeleton crew tasked with keeping the servers running and little else. In July 2017, the Chinese company NetDragon acquired JumpStart Games, and Neopets came along as part of the deal. NetDragon was primarily interested in JumpStart's educational technology platform, not a virtual pet site built on aging Flash infrastructure. Neopets was, at best, an afterthought in the acquisition. ## The Flash Apocalypse For most of its existence, Neopets relied heavily on Adobe Flash (https://404memoryfound.com/posts/definitive-history-macromedia-flash.html). Hundreds of mini-games, site features, and interactive elements were built in Flash. When Adobe announced in 2017 that it would end support for Flash by December 31, 2020, it was a death sentence for large portions of the Neopets experience. The Neopets Team began converting Flash content to HTML5, but the scale of the task was enormous. Hundreds of games and features needed to be rebuilt from scratch. With limited resources under JumpStart and NetDragon, the conversion was slow and incomplete. When Flash officially died at the end of 2020, dozens of beloved games simply stopped working. The Neopian economy, which depended on those games as a source of Neopoints, was thrown into further disarray. For longtime players, the Flash apocalypse was not just a technical problem. It felt like watching parts of their childhood become literally inaccessible. Games they had played thousands of times were simply gone. ## The NFT Detour In 2021, amid the crypto boom, NetDragon greenlit a Neopets Metaverse project built on blockchain technology. The announcement was met with immediate and intense backlash from the existing community. Neopets players, many of whom had been on the site since childhood, were overwhelmingly opposed to NFTs being grafted onto their game. The backlash was not just about cryptocurrency skepticism. It reflected a deeper frustration. The community had spent years watching the site deteriorate under corporate owners who did not understand or care about the product. An NFT project felt like the ultimate expression of that disconnect: instead of fixing what was broken, ownership was chasing the latest speculative trend. ## The Management Buyout On July 17, 2023, Neopets announced what it called "a new era." A management buyout, led by Dominic Law (previously the Chief Metaverse Officer, a title that has aged poorly), acquired Neopets from NetDragon. The resulting company, World of Neopia Inc., became fully independent for the first time since the Viacom acquisition eighteen years earlier. The new leadership announced $4 million in investment funding and a roadmap focused on what the community had been asking for: fixing bugs, converting remaining Flash content, modernizing the site's infrastructure, and developing new features that respected the spirit of the original game. The NFT project was quietly shelved. Law acknowledged the obvious challenge in interviews. The site had been operating at a loss for over a decade. The active user base, once in the tens of millions, had contracted to a fraction of that number. The brand still had recognition and nostalgia value, but converting that into a sustainable business would require more than goodwill. ## What Neopets Actually Taught a Generation The lasting significance of Neopets is not really about the pets. It is about what the site accidentally taught millions of young users. The Neopian economy was a functioning market simulation. Kids learned about supply and demand by watching item prices fluctuate. They learned about investment by playing the Neopian stock market. They learned HTML by customizing their pet pages and user profiles. They learned about scams by getting scammed, which is, unfortunately, the most effective way to learn about scams. The site was also, for many users, their first experience with online community. The Neoboards (the site's forums) were where a generation of kids learned how to argue with strangers on the internet, how to form online friendships, and how to navigate the social dynamics of a digital space. These skills, for better or worse, are foundational to how the internet works today. There is a direct line from Neopets to the creator economy, to the gig economy, to the gamification of everything. The site demonstrated, before anyone had a word for it, that if you wrapped economic activity in a layer of play, people would engage with it voluntarily and enthusiastically. Every mobile game with an in-app economy owes something to what Neopets built in 1999. ## The Competitive Landscape That Ate Its Lunch Neopets did not lose its audience to a single competitor. It lost them to the entire internet growing up around it. Club Penguin (https://404memoryfound.com/posts/what-happened-to-club-penguin.html) (launched 2005, acquired by Disney for $350 million in 2007) targeted a similar demographic with a more modern, visually polished experience. Webkinz (launched 2005) tied virtual pets to physical plush toys, creating a hybrid model that drove massive retail sales. And then there were the platforms that were not virtual pet sites at all but competed for the same attention: Facebook games, smartphone apps, YouTube, and eventually TikTok. The common thread among Neopets' competitors was that they adapted to the way the internet was changing. Neopets, trapped under a series of owners who viewed it as a legacy asset rather than a living product, did not. ## Where Neopets Stands Now As of 2024, Neopets is still online. World of Neopia Inc. has made progress on the roadmap announced during the management buyout. The site has been migrated to more modern infrastructure. Some Flash content has been converted. The team has been communicating more transparently with the community than at any point since the Viacom acquisition. Whether this is enough remains an open question. The site's 150 million registered accounts represent two decades of accumulated signups, not active users. The active community is small, dedicated, and predominantly composed of adults who played as children. Attracting new, younger users to a browser-based virtual pet site in 2024 is an entirely different challenge than doing so in 2001. But the fact that Neopets exists at all is, in its own way, remarkable. It has survived three corporate owners, a data breach, the death of Flash, an unwanted NFT project, and over a decade of neglect. The community that kept logging in through all of that is one of the most resilient user bases on the internet. Whether World of Neopia can build something worthy of that loyalty is the last unanswered question in a story that has been going on for over twenty-five years. ## The Economics of Neopia: A Crash Course Nobody Asked For One of the most underappreciated aspects of Neopets was its economy. The Neopian economy was not a simplified play system. It was a remarkably complex market simulation that ran on real supply-and-demand dynamics, and for millions of kids, it was the first economic system they ever participated in. The Neopian stock market, called the NEODAQ, let users buy and sell shares in fictional companies. Prices fluctuated based on algorithms that mimicked real market behavior. Users who paid attention could buy low and sell high, accumulating Neopoints in the process. The forums were filled with stock tips, analysis, and debates about whether "BOOM" (a fictional company) was overvalued. These were twelve-year-olds doing market analysis. Nobody told them to. The game just made it interesting enough that they wanted to. The auction house and user-run shops created a secondary market where rare items could be traded for enormous sums. A rare paintbrush, which could change a pet's appearance, might sell for millions of Neopoints. Users learned about price manipulation, arbitrage, and the concept that scarcity creates value. Some users ran what were essentially virtual businesses, buying items in bulk at low prices from NPC shops and reselling them at a markup in their own shops. The concept of "restocking," refreshing an NPC shop page rapidly to snag rare items before anyone else, was a skill that required fast reflexes and deep knowledge of item values. The site also had a system called the Trading Post for high-value items that exceeded the shop price cap. Negotiation happened through a wish list system, and the back-and-forth of offers and counteroffers taught users the basics of dealmaking. None of this was accidental. The Powell team designed an economy with enough depth to sustain long-term engagement, and they succeeded beyond any reasonable expectation. ## The HTML Generation Neopets offered users the ability to customize their pet pages, shop pages, and user profiles using HTML and CSS. For millions of young users, this was their first encounter with code. They did not think of it as programming. They thought of it as making their Kougra's page look cool. But the effect was the same: they learned to read and write markup, to understand how styling worked, to debug broken layouts, and to view the source code of pages they admired in order to figure out how things worked. The Neopets user page ecosystem was, in retrospect, a massive distributed coding education program. Entire communities formed around sharing HTML templates, teaching CSS tricks, and helping other users customize their pages. Sites like Neopets CSS guides became some of the most popular fan resources on the web. A generation of web developers, designers, and technologists trace their earliest exposure to code back to trying to make their Neopets shop page look better than their friend's. This is the part of the Neopets legacy that is hardest to quantify but might be the most significant. The site did not just entertain kids. It gave them tools and a reason to learn skills that turned out to be valuable in the real world. HTML, basic economics, community management, digital commerce. These were not in the lesson plan. They were side effects of a game that was designed to be engaging, and that happened to require its users to think. ## The Community That Refused to Leave Perhaps the most remarkable thing about Neopets is not its rise or its fall but the tenacity of its community. Through every ownership change, every technical failure, every period of neglect, a core group of users kept logging in. They kept feeding their pets. They kept restocking. They kept participating in the site's seasonal events, even when those events were buggy and half-finished. The Neopets subreddit remains active. Fan sites like Jellyneo and The Daily Neopets continue to update. Artists share Neopets fan art on social media. The community has outlasted three corporate owners, and its members are, by and large, adults in their twenties and thirties who first played as children. They are not there out of habit. They are there because the game gave them something that no subsequent platform has fully replicated: a sense of ownership over a digital world that felt like theirs. This is the paradox at the heart of the Neopets story. The product was neglected for over a decade, and the community sustained itself anyway. That level of loyalty is extraordinarily rare in any medium, let alone a browser-based game from 1999. It suggests that whatever Adam Powell and Donna Williams built in that first year, they got something fundamentally right about what makes people care about a digital space. The question for World of Neopia Inc. is whether that foundation is strong enough to build on, or whether the community's patience, remarkable as it has been, has finally reached its limit. ## Frequently Asked Questions Is Neopets still online? Yes. Neopets is still operational at neopets.com. The site is now owned and operated by World of Neopia Inc., an independent company formed through a management buyout in July 2023. Why did Neopets decline? Multiple factors contributed: Viacom's shift away from the original advertising model, reduced investment in new content, the sale to JumpStart Games in 2014, persistent technical problems, the death of Adobe Flash in 2020, and the broader shift in internet usage away from browser-based games. How many people played Neopets? Neopets accumulated approximately 150 million registered accounts over its lifetime. At its peak around 2005, the site had roughly 35 million unique monthly visitors generating 4 billion page views per month. How much did Viacom pay for Neopets? Viacom acquired Neopets through its MTV Networks division on June 20, 2005, for $160 million. What was immersive advertising? Neopets pioneered a model where brand advertisements were integrated into the game itself rather than displayed as banner ads. Brands like General Mills and McDonald's had their products woven into the Neopian world as in-game items, games, and experiences. Did Neopets try to do NFTs? Yes. In 2021, under NetDragon's ownership, a Neopets Metaverse project built on blockchain technology was announced. It was met with significant backlash from the existing community and was quietly shelved after the 2023 management buyout. --- # How Doom Changed Everything About PC Gaming and the Internet URL: https://404memoryfound.com/posts/how-doom-changed-pc-gaming-internet.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-05 Topics: Gaming, Software & Apps Picture this: December 10, 1993. It is a Friday night. Five guys in a dark office in Mesquite, Texas, the one they nicknamed "Suite 666," are trying to upload a 2.39 megabyte file to an FTP server at the University of Wisconsin. They cannot connect. Not because the server is down, but because thousands of people are already logged in, waiting. The file has not even been uploaded yet, and the server is buckling under the weight of anticipation. When the upload finally completes, 30 minutes later, 10,000 people try to download the file simultaneously. The university's network crashes. Within hours, universities across the country start banning the game from their networks. Within weeks, system administrators at corporations are writing memos about lost productivity. Within months, the game is everywhere, installed on more PCs than most commercial software, and the five guys in Texas are making over $100,000 a day. The file was Doom. And nothing about PC gaming, the internet, or the business of making video games was ever the same. ## The Guys Who Built It To understand Doom, you have to understand the people who made it, because this is not a story about a corporation executing a business plan. This is a story about a small group of obsessives who were, in the most literal sense, building the future in a rented office space. John Carmack was the engine. A programmer from the Kansas City area, largely self-taught, who had an almost supernatural ability to write code that made computers do things they were not supposed to be able to do. Before Doom, he had already proven this with Wolfenstein 3D, the game that established the first-person shooter as a genre. But Wolfenstein was a proof of concept. Doom was the thing he actually wanted to build. John Romero was the designer, the showman, the guy who understood not just how games should work but how they should feel. Where Carmack was quiet and methodical, Romero was loud and instinctive. He designed Doom's levels with a sense of pacing and atmosphere that most game designers still struggle to match. The tension of a dark corridor. The relief of finding a shotgun. The panic of hearing a door open behind you. Adrian Carmack (no relation to John) and Kevin Cloud handled the art. Tom Hall was brought on as a designer but left during development after creative disagreements. Hall wanted Doom to have a story, characters, a narrative arc. Carmack and Romero wanted a game where you moved fast, shot demons, and felt like a force of nature. The story, famously, was "you're a marine on Mars and demons are trying to kill you." That was it. That was enough. ## The Engine That Changed Everything What made Doom technically revolutionary was Carmack's engine. Wolfenstein 3D had used a raycasting technique that created the illusion of 3D space, but it was limited. All walls were the same height. All floors were flat. There was no lighting variation. It felt like navigating a maze, which is exactly what it was. Doom's engine, the id Tech 1 engine, could render rooms of different heights, staircases, outdoor areas, variable lighting, and a texture-mapped environment that felt genuinely three-dimensional. It used a technique called binary space partitioning, or BSP, to efficiently render complex environments on hardware that, by modern standards, had less processing power than a smart thermostat. The technical achievement here is hard to overstate. Doom ran on a 386 processor. It looked and felt like nothing that had ever existed on a personal computer. When people saw it for the first time, the reaction was not "oh, that's a nice game." The reaction was closer to disbelief. This should not have been possible on this hardware. But there it was. ## The Shareware Gamble Id Software had used the shareware model before, with Commander Keen and Wolfenstein 3D, both distributed through Apogee Software. The premise was simple: give away the first part of the game for free, and if players liked it, they would pay for the rest. With Doom, id decided to cut out the middleman entirely and self-publish. Jay Wilbur, id's CEO, made the call. The mainstream gaming press was not particularly interested in covering a shareware game, no matter how good it looked in previews. Retail distribution meant giving up a huge percentage of the revenue. But if id sold directly to customers, they could keep up to 85 percent of the $40 price tag. The math was obvious, if the game was good enough. So they structured Doom as three episodes. The first episode, "Knee-Deep in the Dead," was free. You could download it from a BBS, copy it from a friend's floppy disk, or grab it from an FTP server. The second and third episodes cost $40 total, ordered directly from id by mail or phone. The bet paid off in a way that nobody, including id, fully anticipated. Within days of release, the shareware episode had been downloaded and copied millions of times. Id estimated that only about 1 percent of people who played the free episode would pay for the full game. But 1 percent of millions was still an enormous number. The company was processing thousands of mail orders daily, pulling in over $100,000 every single day. To put that in perspective: Wolfenstein 3D, which was considered a massive hit, had generated about $100,000 per month at its peak. Doom was doing that in a day. By May 1994, the game had sold over 65,000 registered copies directly, and the shareware version had been distributed over a million times. By 1996, the shareware episode alone had been downloaded an estimated 20 million times. ## The Night the Internet Broke The December 10 launch was, in retrospect, one of the first viral events in internet history. This was before the World Wide Web was mainstream. Most people were still on dial-up. Distribution happened through FTP servers, bulletin board systems, and floppy disks passed hand to hand. And yet, the demand was so intense that it overwhelmed infrastructure that was designed for academic use, not for distributing entertainment to millions of people. The University of Wisconsin incident was just the beginning. Within hours, network administrators at universities and corporations across the country were dealing with Doom-related traffic. The game's multiplayer mode used network broadcast packets, which meant that even computers not running Doom were affected by the traffic. On large networks, this could cause slowdowns or outright crashes. Carnegie Mellon University reportedly banned Doom multiplayer from its network. Intel issued internal memos about lost productivity. The U.S. military would later develop its own modified version of the game for training purposes, which is one of those facts that sounds made up but is completely real. Carmack, to his credit, released a patch within hours of the first network complaints, addressing the broadcast packet issue. But the larger problem was not technical. The problem was that Doom was so compelling, so addictive, so unlike anything that existed before it, that people could not stop playing. And when they discovered they could play against each other over a local network, the game became something entirely new. ## Deathmatch: The Birth of Competitive Gaming Doom did not invent multiplayer gaming. But it invented the version of multiplayer gaming that matters. The game's deathmatch mode, a term coined by Romero, dropped up to four players into a level and told them to kill each other. No teams. No objectives. Just pure, fast, chaotic combat. This was possible because of Carmack's networking code, which allowed Doom to be played over a local area network. LAN parties (https://404memoryfound.com/posts/golden-age-of-lan-parties.html), which had existed in a limited form before Doom, exploded in popularity because of it. People hauled their desktop computers, their CRT monitors, their keyboards and mice, to offices, basements, and garages, strung Ethernet cables across the floor, and spent entire weekends playing deathmatch. If you were there, you know the feeling. The weight of a 17-inch CRT on a folding table. The cables taped to the carpet so nobody would trip. The sound of someone yelling from across the room because you just hit them with a rocket launcher. It was social in a way that online gaming, for all its convenience, has never quite replicated. You were in the same room as the people you were competing against. You could hear them react in real time. The competitive gaming scene that eventually became esports can be traced, with a pretty straight line, back to Doom deathmatch. The first major Doom tournament, held at the 1994 Dallas gaming convention, was one of the earliest organized competitive gaming events. The culture of LAN parties, of competitive FPS gaming, of trash talk and highlight reels and the pursuit of the perfect frag, all of it started here. ## The Mod Scene and the Birth of User-Generated Content One of Carmack's most consequential decisions was to make Doom's engine moddable. The game's data files, called WADs (an acronym for "Where's All the Data"), were stored separately from the engine code. This meant that anyone with the technical inclination could create new levels, new textures, new enemies, and new game modes without touching the engine itself. The modding community that grew around Doom was enormous and enormously creative. People made everything from new level packs to total conversions that turned Doom into entirely different games. Aliens TC turned it into an Aliens movie simulation. Batman Doom reimagined it as a Batman game. The creativity was limitless, and it was all free, shared through the same BBS networks and FTP servers that had distributed the original game. This was user-generated content before anyone called it that. It was the open-source ethos applied to entertainment. And it created a pipeline of talent that fed the game industry for decades. Many professional game developers got their start making Doom WADs in their bedrooms. The idea that a game's community could extend and improve the product, that modding was not a threat but an asset, started with Doom. Carmack would later release the full Doom source code on December 23, 1997, initially under a restrictive non-commercial license. It was re-released under the GNU General Public License in 1999, cementing this philosophy. The engine has since been ported to virtually every computing platform that exists, from calculators to ATMs to pregnancy tests. That last one is real, by the way. Someone got Doom running on a pregnancy test. The game has become a benchmark: if a device has a processor, someone will try to run Doom on it. ## The Violence Debate Doom arrived at a moment when America was starting to pay attention to violence in video games. The 1993 congressional hearings on video game violence, prompted largely by Mortal Kombat and Night Trap, led to the creation of the Entertainment Software Rating Board (ESRB) in 1994. Doom, with its graphic depictions of demon-killing and its first-person perspective, became a lightning rod in the debate. The game was blamed, unfairly, for inspiring real-world violence. After the Columbine shooting in 1999, media reports emphasized that the perpetrators had been Doom players and had created custom Doom levels. The connection was specious. Millions of people played Doom without committing acts of violence. But the narrative stuck, and it shaped public perception of video games for years. Looking back, the violence debate was really about the shock of a new medium becoming culturally dominant. The same moral panic had accompanied rock and roll, comic books, and television. Doom was just the latest thing that parents did not understand and therefore feared. The game itself, for all its graphic content, was ultimately about the simple, primal satisfaction of overcoming challenges with fast reflexes and spatial awareness. The demons were not the point. The feeling of mastery was. ## The Business Legacy Doom's financial model proved several things simultaneously. First, that shareware could work at massive scale for high-quality products. Second, that self-publishing was viable if the product was good enough to generate its own word of mouth. Third, that a small team with a brilliant technical lead could compete with, and outperform, studios backed by major publishers. Id Software licensed the Doom engine to other developers, creating another revenue stream and establishing the practice of engine licensing that companies like Epic Games (with the Unreal Engine) would later turn into a multi-billion-dollar business. The id Tech 1 engine powered games like Heretic, Hexen, and Strife. The principle that a game engine was a product in itself, separate from any individual game, was one of id's most enduring contributions to the industry. The company's trajectory after Doom was both triumphant and cautionary. Quake (https://404memoryfound.com/posts/what-happened-to-quake-online-fps.html) (1996) pushed the technology further with true 3D rendering and online multiplayer. Quake III Arena (1999) became a competitive gaming staple. But creative tensions between Carmack and Romero led to Romero's departure in 1996, and the two never worked together again. Romero went on to make Daikatana, one of the most infamous disappointments in gaming history. Carmack continued at id through Doom 3 (2004) and Rage (2011) before leaving to work on virtual reality at Oculus in 2013. Id Software was acquired by ZeniMax Media in 2009 and is now part of Microsoft through the 2021 Bethesda acquisition. The Doom franchise itself is alive and well. Doom (2016) and Doom Eternal (2020) were critical and commercial successes that proved the core formula, fast movement, big guns, demons, still works. But those games, for all their quality, exist in the shadow of what five people in a dark office in Texas built in 1993. ## Why Doom Still Matters Thirty-plus years later, Doom's influence is so deeply embedded in the fabric of gaming that it is almost invisible. First-person shooters are the dominant genre in gaming. Online multiplayer is the default mode of play. Engine licensing is a foundational business model. Modding communities sustain games for decades. Shareware's "try before you buy" logic lives on in every free-to-play game and demo. But beyond the industry impact, Doom matters because of what it represented: the moment when a handful of talented, passionate people, working with minimal resources and maximum conviction, created something that changed an entire medium. No focus groups. No corporate strategy. No market research. Just a brilliant programmer, a visionary designer, a small team, and the belief that if you made something extraordinary, the world would find it. The world found it. At 100,000 downloads per day, crashing university networks and getting banned from corporate offices, the world found it in a way that nobody was prepared for. And that is why, decades later, people are still trying to run Doom on everything from refrigerators to pregnancy tests. The game is not just a classic. It is a proof of concept for an entire way of thinking about what games can be, how they can be distributed, and who gets to make them. ## The Sound of Doom One thing that gets overlooked when people talk about Doom's technical achievements is the sound design. Bobby Prince composed the soundtrack, and he did it by taking obvious inspiration from heavy metal bands like Metallica, Pantera, Slayer, and Alice in Chains, filtering those riffs through MIDI synthesis and creating something that sounded aggressive and propulsive even through a Sound Blaster (https://404memoryfound.com/posts/what-happened-to-creative-labs-sound-blaster.html) card and cheap desktop speakers. The music was not just background noise. It was a pacing mechanism. When a track kicked in with a driving riff, you knew something was about to happen. The quiet moments were genuinely quiet, which made the sudden blast of a shotgun or the growl of an approaching Imp hit harder. Sound design in games was often an afterthought in 1993. In Doom, it was integral to the experience. And then there were the sound effects. The grunt of the Imp. The scream of the Cacodemon. The mechanical whir of a door opening in a dark corridor. The wet, crunching sound of the chainsaw. These sounds became iconic not because they were technically impressive but because they were perfectly matched to the gameplay. Every audio cue told you something about the world: where enemies were, what was happening behind you, whether you were safe or about to die. If you played Doom in 1993, you can probably still hear that shotgun pump in your head right now. ## What the Retail Industry Thought (And How Wrong They Were) The traditional game retail establishment did not take Doom seriously at first. Shareware was seen as the domain of hobbyists and budget software. Real games came in big boxes with printed manuals and were sold at stores like Electronics Boutique, CompUSA (https://404memoryfound.com/posts/is-compusa-still-around.html), and Babbage's. The idea that a game distributed for free through BBS networks and university FTP servers could outsell boxed retail products was, to most industry observers, absurd. Id Software eventually did release a retail version of Doom through GT Interactive in 1995, called "The Ultimate Doom," which included a new fourth episode. It sold well. But by that point, the shareware version had already made id wealthy and had proven a fundamental truth about the emerging digital distribution model: if your product was good enough, you did not need a publisher, a retail chain, or a marketing budget. You just needed an internet connection and a few BBS sysops who were willing to host your file. This lesson took the rest of the industry almost two decades to fully internalize. Steam launched in 2003. The App Store launched in 2008. The free-to-play model did not become dominant until the early 2010s. Doom proved the viability of all of these concepts in 1993. The infrastructure was not there yet, so nobody could scale the model immediately. But the proof of concept was sitting right there, generating $100,000 a day, for anyone who was paying attention. ## The Romero Departure and What Came After The partnership between Carmack and Romero was one of the most productive creative collaborations in gaming history, and like a lot of great partnerships, it eventually fell apart. Carmack was focused on technology, on pushing the engine forward, on solving the next rendering problem. Romero was focused on design, on spectacle, on building a gaming empire that extended beyond code. After Doom II shipped in 1994, the tension between these two visions became unmanageable. Quake's development was notoriously contentious, with the team struggling to define what the game should be while Carmack forged ahead on the technical side. Romero left id Software in 1996, and his subsequent venture, Ion Storm, became a cautionary tale. Daikatana, the game that was supposed to prove Romero could build a studio on his own terms, was released in 2000 after years of delays to devastating reviews. The infamous "John Romero's about to make you his..." ad campaign had aged poorly long before the game even shipped. Carmack, meanwhile, stayed at id and continued to push technology forward through Quake II, Quake III Arena, and Doom 3. His work on 3D rendering techniques, including stencil shadow volumes and megatextures, kept id at the cutting edge of engine technology for over a decade. He left id in 2013 to become CTO at Oculus, pursuing virtual reality, which he saw as the next frontier in the kind of immersive technology that Doom had pioneered twenty years earlier. The divergent paths of Carmack and Romero after Doom tell you something important about the game itself. It was the product of two very different kinds of genius working together at exactly the right moment, on exactly the right project. Neither of them produced anything quite as culturally significant on their own. That is not a knock on either of them. It is a testament to how rare and how valuable that kind of collaboration actually is. ## Frequently Asked Questions When was Doom released? Doom was released on December 10, 1993. Id Software uploaded the shareware episode to an FTP server at the University of Wisconsin, where it was immediately downloaded by thousands of users, crashing the server. How much money did Doom make? At its peak, Doom was generating over $100,000 per day in direct sales. By 1998, the shareware edition alone had yielded $8.74 million in U.S. revenue from 1.36 million units sold. Total franchise revenue, including engine licensing and sequels, was substantially higher. What was the shareware model? Shareware was a distribution method where the first portion of a game was given away for free. Players who enjoyed it could purchase the full version directly from the developer. Doom's first episode was free, and the remaining two episodes cost $40. Did Doom really crash university networks? Yes. The game's release overwhelmed the FTP server at the University of Wisconsin. Within hours, multiple universities banned Doom multiplayer from their networks because the game's broadcast packet system was disrupting network traffic for all users. Who made Doom? Doom was created by id Software. The core team included programmers John Carmack and John Romero, artists Adrian Carmack and Kevin Cloud, and CEO Jay Wilbur. Tom Hall contributed to early development before departing. What was Doom's impact on gaming? Doom popularized the first-person shooter genre, pioneered networked multiplayer deathmatch, established the practice of engine licensing, normalized modding communities, and demonstrated that shareware distribution could work at massive scale. It is widely considered one of the most influential video games ever made. Can you still play the original Doom? Yes. The original Doom is available on virtually every modern platform. The source code was released in 1997, and the game has been ported to hundreds of devices. The shareware episode can still be downloaded for free. --- # Why the Atari Jaguar Failed and What It Costs Now URL: https://404memoryfound.com/posts/what-happened-to-atari-jaguar-64-bit-console.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-04 Updated: 2026-09-20 Topics: Gaming, Hardware, Business Blunders **Summary:** The Atari Jaguar failed because Atari Corporation sold roughly 125,000 of them between November 1993 and the end of 1995, against a library of 63 licensed games and a 64-bit claim the press took apart. Atari Corporation itself was gone by July 1996, folded into hard drive maker JTS in a reverse merger, and the name now belongs to Atari SA in Luxembourg. A working Jaguar costs hundreds of dollars in 2026, and sealed units have sold for $1,199.99. **Key facts:** - Launched: November 23, 1993, Atari Corporation, $249.99 - Units sold: About 125,000 through the end of 1995, per Atari's SEC filing - Games released: 63 licensed titles, 50 on cartridge and 13 on CD - Status today: Discontinued in 1996; Atari Corporation no longer exists - Owner today: Atari SA, Euronext Paris, run by CEO Wade Rosen since 2021 ## What was the Atari Jaguar supposed to be? By 1993 Atari Corporation was a small company living on the memory of what Atari had been in 1980. Jack Tramiel, the founder of Commodore (https://404memoryfound.com/posts/what-happened-to-commodore-computers.html), had bought the consumer division in 1984, and his son Sam Tramiel ran it as president. The 7800 console, the Lynx handheld and the ST computers had all come and gone without changing the company's direction. The Jaguar was the last bet. Atari launched it on November 23, 1993 at $249.99, first in test markets in New York City and San Francisco, with a nationwide rollout in 1994. Atari reported shipping about 17,000 units during that 1993 test window. Inside the case were three chips. One was a Motorola 68000, the same 16-bit CPU that ran the Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html) Genesis. The other two were custom parts designed in the UK and nicknamed Tom and Jerry. Tom carried a 64-bit object processor and blitter for graphics. Jerry was a 32-bit digital signal processor that handled sound. The pack-in game was Cybermorph, a polygon flight shooter with genuine 3D environments, a weak color palette and a set of design flaws that reviewers held up against Star Fox on the Super Nintendo. ## Was the Atari Jaguar really a 64-bit console? Atari's entire marketing plan rested on one number. The Do the Math campaign set 64 beside the 16-bit Genesis and Super Nintendo and let shoppers draw the obvious conclusion. Sony and Sega were both preparing 32-bit machines, so 64 looked like a full generation of lead. The arithmetic was doing a lot of work. Tom's internal bus was 64 bits wide, but the processor running most game code was the 16-bit 68000, and Jerry was 32-bit. Electronic Gaming Monthly ran a short editorial pointing out that by Atari's way of counting, "the Sega Saturn (https://404memoryfound.com/posts/what-happened-to-sega-saturn-console.html) would be a 112-bit monster of a machine." The deeper problem was not the label. Developers were supposed to push the work onto Tom and Jerry and leave the 68000 as a traffic cop. Hardware bugs made synchronizing the three chips difficult, so studios working to a deadline wrote for the 68000 instead. Games built that way ran at roughly Super Nintendo speed on a machine sold as four times more powerful. The controller did not help. It was a wide pad with three face buttons and a full 12-key telephone keypad underneath, driven by paper overlays that slid into the handset so players could remember what each key did. ## How many Atari Jaguars did Atari actually sell? Atari's own filings answer this better than any estimate. In its Form 10-K405 covering 1995, Atari Corporation told the Securities and Exchange Commission that it had sold approximately 125,000 Jaguar units from the console's introduction in late 1993 through the end of 1995. At December 31, 1995 it still held roughly 100,000 unsold units in inventory, close to half of everything it had ever built. The same filing is blunt about what that meant. "There can be no assurance that Atari's substantial unsold inventory of Jaguar and related software can be sold at current or reduced prices if at all," the company warned shareholders. Atari had already cut the Jaguar's retail price and written inventory down, and it told the SEC that Jaguar sales came in substantially below expectations. The trade press had seen the shape of it early. Computer Gaming World called the Jaguar "a great machine in search of a developer/customer base" in January 1994, two months after launch, and said Atari still had to overcome the stigma of its own name after years of thin marketing and poor developer relations. The Sega Saturn reached the United States in May 1995 and the PlayStation in September 1995. Whatever room the Jaguar had left closed that year. ## Did any Jaguar game make the hardware worth it? One did. Tempest 2000, designed by Jeff Minter and released in 1994, rebuilt Dave Theurer's 1981 arcade game as a fast, loud, hallucinatory shooter with a techno soundtrack, and it drove the custom chips the way Atari's engineers had imagined. Electronic Gaming Monthly named it the best Jaguar game of 1994. The trouble was that Minter was an outlier. Most studios never got past the 68000, and Atari's reputation with third parties meant few were willing to spend months finding out whether they could. Across the console's entire commercial life it received 63 licensed games: 50 on cartridge and 13 on CD. The CD add-on made the math worse. The Atari Jaguar CD shipped on September 21, 1995 at $149.95, which put a complete Jaguar setup just under $400 in the same holiday season the Saturn and the PlayStation were on shelves. It earned a reputation for failing drives, its library never filled out, and it was discontinued alongside the console in 1996. Atari had run a version of this play before with the Atari Lynx (https://404memoryfound.com/posts/what-happened-to-atari-lynx-handheld.html), a better handheld than the Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html) on paper and a distant second in sales, and it was not the only casualty of 1995: the 3DO (https://404memoryfound.com/posts/what-happened-to-3do-700-dollar-console.html) was dying on the same shelves. ## Is Atari still in business in 2026, and who owns it? Atari Corporation, the company that built the Jaguar, no longer exists. On February 13, 1996 it agreed to a reverse merger with JTS Inc., a hard disk drive maker, and the deal closed on July 30, 1996. The Jaguar was discontinued, staff were dismissed or moved, and what had been Atari became a holding line inside a drive company. In 1998 JTS sold the Atari brand and the intellectual property of Atari Corporation to Hasbro Interactive for $5 million. The package ran to more than 75 game properties, among them Pong, Centipede, Missile Command, Breakout and Tempest. Hasbro renamed the acquiring subsidiary Atari Interactive. The name later passed to Infogrames, which eventually renamed itself Atari SA. That is who owns it in 2026. Atari SA trades on Euronext Paris under the ticker ALATA, is headquartered in Luxembourg, and has been chaired and led by Wade Rosen since April 2021. Bloomberg reported in May 2026 that preliminary revenue for the year to March 31, 2026 was in the $50 million to $60 million range, up at least 40 percent on the prior year, built on retro hardware such as the Atari 2600+ and on licensing the back catalogue. For the full chain of custody, see who owns Atari now (https://404memoryfound.com/posts/who-owns-atari-now.html). ## Where to find one today The Jaguar is a collector machine now, priced by scarcity rather than by anything it can do. SlashGear tracked a boxed new-condition console selling for $999.99 in May 2023 and a sealed unit at $1,199.99. Used working consoles with a controller trade below that, in the hundreds of dollars, and the price climbs with the box, the manuals and a copy of Tempest 2000. If the goal is to play rather than to own, Atari 50: The Anniversary Celebration carries a set of Jaguar titles including Tempest 2000, running on a Jaguar emulator Digital Eclipse wrote from scratch, and it is available on PC, PlayStation, Xbox and Switch for a fraction of the hardware price. For the real thing, Stone Age Gamer (https://stoneagegamer.com/atari/jaguar/) stocks Jaguar consoles, controllers, accessories and flash carts. ## Frequently Asked Questions ### How many Atari Jaguars were sold? Atari Corporation told the Securities and Exchange Commission that it sold approximately 125,000 Atari Jaguar consoles between the machine's introduction in late 1993 and the end of 1995, and that roughly 100,000 more were still sitting unsold in inventory on December 31, 1995. The Jaguar was discontinued in 1996, when Atari Corporation merged into the hard drive maker JTS. ### Was the Atari Jaguar really 64-bit? Only in part. The Atari Jaguar, launched on November 23, 1993, ran most game code on a 16-bit Motorola 68000, paired with a 32-bit sound chip called Jerry and a graphics chip called Tom whose object processor and blitter were 64-bit. Electronic Gaming Monthly mocked the arithmetic at the time, noting that by the same logic the Sega Saturn would count as a 112-bit machine. ### How much is an Atari Jaguar worth in 2026? The Atari Jaguar is a collector console in 2026 rather than a cheap way into retro gaming. SlashGear recorded a sealed unit selling for $1,199.99 and a boxed new-condition console at $999.99 in May 2023, while used working consoles with a controller change hands for hundreds of dollars. Tempest 2000 and the 1995 Jaguar CD add-on carry premiums of their own. **Sources:** - Atari Corporation, Form 10-K405 for fiscal year 1995 (SEC EDGAR): https://www.sec.gov/Archives/edgar/data/0000802019/000089161896000213/0000891618-96-000213.txt - Atari Jaguar, Wikipedia: https://en.wikipedia.org/wiki/Atari_Jaguar - Atari Goes to Hasbro, GameSpot (1998): https://www.gamespot.com/articles/atari-goes-to-hasbro/1100-2462915/ - Here's How Much An Atari Jaguar Is Worth Today, SlashGear: https://www.slashgear.com/1327569/worth-atari-jaguar-today/ - Meet Atari's Tycoon CEO, Who Is Aiming for a Retro Gaming Revival, Bloomberg (May 2026): https://www.bloomberg.com/news/newsletters/2026-05-15/meet-atari-s-tycoon-ceo-who-is-aiming-for-a-retro-gaming-revival --- # Is WordPerfect Still Around in 2026? Corel Sells It URL: https://404memoryfound.com/posts/what-happened-to-wordperfect-word-processor.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-04-04 Updated: 2026-09-20 Topics: Software & Apps, Business Blunders, Then vs Now **Summary:** WordPerfect still exists in 2026. Corel sells WordPerfect Office 2021, which remains the latest version five years after its release, and Corel itself passed from KKR to Vector Capital in May 2026. The word processor that held 85 percent of the MS-DOS market in 1992 now survives mostly in law firms and government offices. **Key facts:** - Status today: Still sold; WordPerfect Office 2021 remains the latest version - Owner today: Corel Corporation, backed by Vector Capital since May 2026 - Launched: 1979, by Alan Ashton and Bruce Bastian in Orem, Utah - Peak market share: 85 percent of the MS-DOS word processor market in 1992 - Sold for: $1.4 billion to Novell in 1994, about $115 million to Corel in 1996 ## Who built WordPerfect, and where did it come from? WordPerfect began in 1979 at Brigham Young University, where computer science professor Alan Ashton and graduate student Bruce Bastian wrote a word processor for the city of Orem's Data General minicomputer. The program worked well enough to sell, so the two founded a company in Orem, Utah, and moved the software to the IBM PC. The market they walked into belonged to WordStar, released in 1978 by MicroPro International. WordStar leaned on control-key sequences that were hard to learn, its updates for IBM PC hardware came slowly, and a rewrite called WordStar 2000 pushed its own users away. WordPerfect filled the gap. Version 4.2, in 1986, was the breakthrough, and version 5.1, in 1989, made the program the default on MS-DOS. Part of that was reach: it ran on DOS, Unix, VMS, Data General, the Amiga and dozens of other systems, which built an installed base no rival could match. Part of it was service. WordPerfect answered support calls for free, and kept answering them long after the cost stopped making sense. ## What made Reveal Codes worth keeping? Two features made WordPerfect users loyal in a way that looks strange now. The first was the keyboard. Every command sat on a function key combination modified by Shift, Alt or Ctrl: F7 to exit, Shift-F7 to print, Alt-F7 for columns. A cardboard template sat above the F-keys until you stopped needing it. There was no mouse support to miss. The second was Reveal Codes. Alt-F3 split the screen: the document on top, every hidden formatting instruction on the bottom. Bold tags, margin changes, tab stops and font switches were all visible, and all could be deleted one at a time. Microsoft Word treated formatting as something the program handled on your behalf, the same instinct that later produced Clippy (https://404memoryfound.com/posts/what-happened-to-clippy-microsoft.html). WordPerfect handed you the wiring. That is still the reason lawyers keep it. Federal and state courts set exact margin, numbering and page-break requirements, and a filing that misses them gets rejected. Reveal Codes shows precisely which code caused the problem. No other word processor has copied it in the same form. ## Why did WordPerfect lose to Microsoft Word? Windows 3.0 shipped in 1990 and took over the PC inside of two years. WordPerfect Corporation had no real Windows product ready. The first WordPerfect for Windows arrived late in 1991 and behaved like a DOS program wearing a Windows costume. WordPerfect 6.0 for Windows, in 1993, was supposed to fix that. It was slow, it was buggy, and it wanted hardware most buyers did not have. A stable release did not land until 1994, by which point Word for Windows was polished and shipping inside Microsoft Office next to Excel and PowerPoint. The bundle was the real weapon. WordPerfect sold a word processor. Microsoft sold a suite. WordPerfect Corporation had no spreadsheet and no presentation program of its own, so it leaned on Borland's Quattro Pro and on partnerships that never added up to one product. A buyer comparing an integrated suite against a word processor plus Lotus 1-2-3 (https://404memoryfound.com/posts/what-happened-to-lotus-1-2-3-spreadsheet.html) plus something for slides took the suite. Word did not have to be better. It had to be adequate and included. The numbers moved fast. WordPerfect entered 1992 still claiming 85 percent of the MS-DOS word processor market, but on Windows it held roughly a third against more than half for Microsoft. Annual revenue dipped from $622 million to $579 million in 1992. ## What did Novell and Corel actually pay? Ashton and Bastian sold. Novell, the networking company in nearby Provo, Utah, completed a $1.4 billion acquisition of WordPerfect Corporation on June 27, 1994. The payment was mostly stock, and accounts of its value differ: UPI reported the deal at $1.4 billion, while other tallies put the 59 million Novell shares and options exchanged at about $855 million. Novell's plan was to assemble an application suite that could stand against Microsoft Office. It did not work. Novell had no experience running a desktop application business, releases slipped, and the gap with Word widened. Eighteen months later Novell quit. On January 31, 1996, Corel Corporation, the Canadian maker of CorelDRAW, agreed to buy WordPerfect from Novell for roughly $115 million in stock and cash, in a transaction the Deseret News valued at about $180 million once licensing was counted. Corel chairman Michael Cowpland told the paper the purchase gave his company a "best of breed" suite of applications, and told existing customers holding site licenses and upgrade contracts that "We will be trying for maximum continuity." Less than two years separated the two prices. ## Did Microsoft break the law to beat WordPerfect? Novell argued that it did. In 2004 it sued Microsoft under section 2 of the Sherman Act, claiming Microsoft had shared Windows interoperability information, including the namespace extensions WordPerfect needed to sit properly inside the shell, then withdrew it once Microsoft had its own word processor to sell. The disputed period ran up to the Windows 95 launch (https://404memoryfound.com/posts/the-night-windows-95-launched.html). The jury did not reach a verdict, the district court entered judgment for Microsoft as a matter of law, and in 2013 the Tenth Circuit affirmed. Writing for the court in Novell, Inc. v. Microsoft Corp., Judge Neil Gorsuch held that antitrust law rarely forces a firm, even a dominant one, to deal with a rival, and concluded: "With respect to Novell at least, Microsoft did nothing unlawful." The Supreme Court refused to hear Novell's appeal in 2014, ending the case after a decade. The narrower question of whether Microsoft targeted WordPerfect specifically was therefore never answered on the merits. What the record does show is structural: the company that set the rules of the platform also sold the application that won on it. ## Who owns WordPerfect today? Corel Corporation still owns and sells WordPerfect, and Corel's own ownership has changed twice since 2019. Vector Capital held the company from 2003, KKR bought it in 2019, and on February 26, 2026 Corel announced it would separate into two independent companies. Vector Capital took the creativity and productivity brands, including WordPerfect, CorelDRAW, WinZip, PaintShop Pro, Roxio, VideoStudio, Pinnacle and MindManager, while Parallels stayed with KKR. The transaction closed in May 2026, and Corel relaunched as an independent company backed by Vector Capital. The product has moved more slowly than its owners. WordPerfect Office 2021, released in May 2021, is still the current version, and everything since has been patches and hot fixes rather than a new release. It is a Windows product, sold through Corel and through retailers such as Staples, Best Buy and Walmart. Usage matches that. In the American Bar Association's 2021 Legal Technology Survey Report, 98 percent of responding lawyers named Microsoft Word among the word processors they use and 15 percent named Corel WordPerfect. That is a small share, and it is not nothing: a product that lost its market in the mid-1990s was still on roughly one law office desk in seven almost thirty years later. Where to find one today: the physical leftovers turn up regularly from independent sellers. Shrink-wrapped DOS-era boxes, the thick printed manuals and the cardboard function-key templates that slotted above an IBM keyboard all circulate, and Etsy is the practical place to look through vintage listings on Etsy (https://www.etsy.com/search?q=vintage+wordperfect). The software itself is still sold new by Corel for Windows. ## Frequently Asked Questions ### Is WordPerfect still available in 2026? Yes. Corel still sells WordPerfect Office for Windows in 2026. The current version is WordPerfect Office 2021, released in May 2021, and the releases since then have been updates and hot fixes rather than a new numbered version. ### Who owns WordPerfect now? WordPerfect is owned by Corel Corporation, which is backed by Vector Capital. Corel announced on February 26, 2026 that it would split into two companies, with Vector Capital taking WordPerfect and the rest of the creativity and productivity portfolio from KKR; that deal closed in May 2026. ### Why do law firms still use WordPerfect? Law firms keep WordPerfect for Reveal Codes, the split-screen view that exposes every hidden formatting instruction in a document so it can be deleted individually. Court filings carry exact margin, numbering and page-break rules, and in the American Bar Association's 2021 Legal Technology Survey Report 15 percent of responding lawyers still named Corel WordPerfect as a word processor they use. **Sources:** - WordPerfect Corporation company history, Encyclopedia.com: https://www.encyclopedia.com/books/politics-and-business-magazines/wordperfect-corporation - Novell completes deal for WordPerfect, UPI, June 27 1994: https://www.upi.com/Archives/1994/06/27/Novell-completes-deal-for-WordPerfect/2659772689600/ - Corel to buy WordPerfect from Novell, Deseret News, January 31 1996: https://www.deseret.com/1996/1/31/19222369/corel-to-buy-wordperfect-from-novell/ - Novell, Inc. v. Microsoft Corp., 731 F.3d 1064 (10th Cir. 2013): https://www.leagle.com/decision/infco20130923057 - Corel Corporation announces transaction creating two independent companies, 2026: https://www.parallels.com/newsroom/news/press-releases/20260226-corel-announcement/ --- # Who Owns Sound Blaster Now? Creative Labs in 2026 URL: https://404memoryfound.com/posts/what-happened-to-creative-labs-sound-blaster.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-04 Updated: 2026-09-20 Topics: Hardware, Gaming **Summary:** Yes, Sound Blaster is still made, and Creative Technology still owns it. The Singapore company, which trades in the United States as Creative Labs, is listed on the Singapore Exchange as C76 and reported US$67.4 million in sales for the year to 30 June 2025, against a market value of about US$1.6 billion at its 1994 peak. Sim Li Ern, a nephew of founder Sim Wong Hoo, became chief executive in January 2026. **Key facts:** - Founded: 1981 in Singapore, by Sim Wong Hoo and Ng Kai Wa - Status today: Still trading; listed on the Singapore Exchange as C76 - Owner today: Independent public company; CEO Sim Li Ern since January 2026 - Revenue: US$67.4 million in FY2025, against a US$1.6 billion market value in 1994 - Price today: Sound Blaster AE-9 card, $349.99 from Creative ## PC sound was one beeping speaker until 1987 Through the mid 1980s an IBM compatible PC had exactly one audio device: the small internal speaker soldered to the motherboard. It made square wave tones and little else. Programmers found tricks to squeeze crude music out of it, but next to a Commodore (https://404memoryfound.com/posts/what-happened-to-commodore-computers.html) Amiga or an Atari (https://404memoryfound.com/posts/who-owns-atari-now.html) ST, machines that shipped with real sound chips, the PC sounded broken. The fix came from Quebec City. Ad Lib, founded in 1987 by Martin Prevel, a former music professor, sold the Ad Lib Music Synthesizer Card built around the Yamaha YM3812 FM synthesis chip, better known as the OPL2. Sierra On-Line supported it, other studios followed, and by 1990 "Ad Lib compatible" was printed on game boxes as the standard. Ad Lib had one blind spot. The card synthesized music but could not play back recorded digital audio: no sampled voices, no recorded gunshots, no speech. That gap was the opening. ## How the Sound Blaster took the market from Ad Lib Creative Technology started in 1981 as a computer repair shop in Singapore, founded by Sim Wong Hoo and Ng Kai Wa. It had been selling audio add-ons since 1987, including the Creative Music System, without much impact outside Asia. That changed in November 1989 at Comdex in Las Vegas, where Creative showed a card codenamed "Killer Kard" and sold as the Sound Blaster 1.0. It did everything the Ad Lib card did, using the same OPL2 chip, and added an 8-bit digital to analog converter for sampled audio, a game port that doubled as a MIDI interface, and a microphone input. One card replaced three. The booth was small and the queue was not. "We sold one Sound Blaster every four minutes," Sim Wong Hoo told Custom PC, describing three cashiers working a 300 square foot stand. Creative then did the unglamorous work: free cards to studios, a development kit people could actually use, real technical support. Because the Sound Blaster was Ad Lib compatible, a developer targeting it gave up nothing, and any game that used digital audio handed Sound Blaster owners speech and effects that Ad Lib owners simply could not hear. By 1991 the boxes said "Sound Blaster compatible" instead. Ad Lib answered with the Ad Lib Gold in 1992 and filed for bankruptcy the same year. When id Software shipped Doom (https://404memoryfound.com/posts/how-doom-changed-pc-gaming-internet.html) in 1993, it shipped into a market Creative had already defined. ## The peak years: Nasdaq, the AWE32 and EAX In June 1992 Creative launched the Sound Blaster 16, the first card in the line with 16-bit sampling at 44.1 kHz, the same resolution as a compact disc. It became the card every multimedia PC advertised, and it is the model most people still picture when they hear the name. The money followed. Creative listed on Nasdaq in 1992, the first Singapore company to do so, added a Singapore Exchange listing in 1994, and reached a market value of roughly US$1.6 billion that year. The hardware kept getting better. The Sound Blaster AWE32 arrived in March 1994 with an E-mu EMU8000 wavetable synthesizer, so MIDI music stopped sounding like a cheap organ and started sounding like instruments. On 11 August 1998 the Sound Blaster Live! shipped with the EMU10K1 processor and hardware Environmental Audio Extensions, which let a game place a footstep behind the player and make a corridor echo differently from an open hall. For a few years the sound card mattered to a gaming PC almost as much as the 3dfx Voodoo card (https://404memoryfound.com/posts/what-happened-to-3dfx-voodoo-graphics-card.html) sitting next to it. ## What killed the add-in sound card Two decisions did it, and neither of them was Creative's. The first came from Intel. In 1997 Intel Architecture Labs published AC'97, a codec standard that let motherboard makers put competent 16-bit audio directly on the board. Onboard sound was not as good as a Sound Blaster. It was free, it was already installed, and for anyone playing MP3s it was enough. Intel High Definition Audio replaced AC'97 in 2004 and closed the gap further. The second came from Microsoft. Windows Vista, released in 2007, shipped a rewritten audio stack with no direct path from DirectSound to the audio driver. Hardware acceleration for DirectSound and DirectSound3D was gone, and EAX went with it, because EAX was built on DirectSound3D. Creative released ALchemy in 2007 to translate old DirectSound3D calls into OpenAL so legacy games kept their effects, and pushed developers toward OpenAL EFX instead. The patch worked. The market never came back. Needing that translation layer is part of why running 90s PC games on Windows 11 (https://404memoryfound.com/posts/run-90s-pc-games-windows-11.html) takes more setup than it should. ## What Creative did after the sound card Creative moved into portable players with the Creative Zen line and lost that market to the iPod (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html), but it held a patent. In 2006 Apple paid Creative US$100 million for a paid up license to the Zen patent, which covered the hierarchical menu used to pick a track on a player. "This settlement resolves all of our differences with Creative, including the five lawsuits currently pending between the companies," Steve Jobs said in the announcement. Creative joined Apple's Made for iPod program the same day. In 2007 Creative voluntarily delisted from Nasdaq, with 31 August 2007 the last day of trading, citing the cost of US reporting obligations. Its shares have traded only in Singapore since. The audio work carried on under the Sound Blaster and Super X-Fi names, moving from ISA and PCI cards to PCIe cards, USB digital to analog converters, headsets and speakers. Creative says more than 400 million Sound Blasters have been sold since 1989. ## Is Creative Labs still in business in 2026? Yes. Creative Labs is the name Creative Technology trades under in the United States, and the parent is still an independent, publicly listed company headquartered in Singapore, quoted on the Singapore Exchange as C76. It is a far smaller business than it was. Creative reported sales of US$67.4 million for the financial year ended 30 June 2025, up about 7 percent on the year before, with a net loss of US$10.5 million. Revenue for the first half of FY2026 was US$34.2 million, down 8.6 percent year on year, with the net loss narrowed to US$1.24 million. The top job has changed hands repeatedly since Sim Wong Hoo died on 4 January 2023. Song Siow Hui retired in February 2025, Freddy Sim was appointed in May 2025 and stepped down within three months for health reasons, executive chairman Tan Jok Tin covered as interim chief executive, and in January 2026 the board named Sim Li Ern, a nephew of the founder who has worked at Creative since 2002. ### Where to find one today Creative still sells new cards directly: the Sound Blaster AE-9, the 30th anniversary flagship, lists at $349.99, and the AE-7 at $299.99. For the original ISA hardware, working Sound Blaster 16 cards ran roughly $40 to $100 at retro parts sellers in September 2026, with the SCSI versions at the top of that range. Etsy sellers list vintage Sound Blaster cards (https://www.etsy.com/search?q=sound+blaster+isa+sound+card), and condition matters more than the model number: look for a card sold as tested, with the capacitors intact. ## Frequently Asked Questions ### Does Creative still make Sound Blaster sound cards? Yes. Creative Technology still sells Sound Blaster products in 2026, including internal PCIe cards such as the Sound Blaster AE-9 and AE-7, external USB digital to analog converters, speakers and gaming headsets. The line left ISA and PCI slots behind decades ago, but the Sound Blaster name has been in continuous use since the first card shipped in 1989. ### Who owns Creative Labs now? Creative Labs is the United States trading name of Creative Technology Ltd, which no larger group owns. The company has been listed on the Singapore Exchange since 1994 under the ticker C76, delisted from Nasdaq in 2007, and in January 2026 appointed Sim Li Ern, a nephew of founder Sim Wong Hoo, as chief executive. ### Who was Sim Wong Hoo? Sim Wong Hoo co-founded Creative Technology in Singapore in 1981 with Ng Kai Wa and led it as chairman and chief executive until his death on 4 January 2023 at the age of 67. He took the Sound Blaster from its November 1989 Comdex launch to a company worth roughly US$1.6 billion by 1994. **Sources:** - Creative Technology, "Sound Blaster: 30 Years of Revolutionizing Audio" (2019): https://sg.creative.com/corporate/pressroom?id=13641 - Creative Technology Ltd, Annual Report 2025 (SGX filing): https://links.sgx.com/1.0.0/corporate-announcements/TT55FF3OWVK06BAD/863800_CreativeAnnualReport2025.pdf - Custom PC, "The Sound Blaster Story": https://www.custompc.com/retro-tech/the-sound-blaster-story - Macworld, "Apple settles Creative lawsuits for $100M" (2006): https://www.macworld.com/article/181311/creative-38.html - The Edge Singapore, "Late Creative Technology founder Sim Wong Hoo's nephew to take over as CEO" (2026): https://www.theedgesingapore.com/news/new-appointments/late-creative-technology-founder-sim-wong-hoos-nephew-take-over-ceo --- # Is Lycos Still Around in 2026? Who Owns It Now URL: https://404memoryfound.com/posts/what-happened-to-lycos-search-engine.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-04 Updated: 2026-09-20 Topics: Internet Culture, Business Blunders, Money & Tech **Summary:** Lycos still exists in 2026 as a small web portal, but it no longer runs its own search crawler, web hosting or email service. Lycos Inc. is listed as a subsidiary of the Indian company Brightcom Group, with a 56 percent stake placed in receivership by a New York court in 2018. Terra Networks paid $12.5 billion for Lycos in 2000; the last completed sale, in 2010, was $36 million. **Key facts:** - Launched: July 1994, Carnegie Mellon University, by Michael Mauldin - Peak price: $12.5 billion, Terra Networks, deal closed October 2000 - Last sale price: $36 million, Ybrant Digital, August 2010 - Owner today: Lycos Inc., a Brightcom Group subsidiary; a 56 percent stake in receivership since 2018 - Status today: Portal still online; Angelfire and Tripod closed April 24, 2026 ## How a Pittsburgh spider crawler became a public company Lycos started in July 1994 as a research project by Michael Mauldin, a computer scientist at Carnegie Mellon University in Pittsburgh. Mauldin built a web crawler alongside the university's digital library work and named it after Lycosidae, the family of wolf spiders that chase their prey instead of waiting in a web. CMGI put in roughly $2 million to spin the crawler out of the university and into a company. Bob Davis, who had no background in search, became chief executive and first employee in 1995. His read of the business was that indexing pages was the cheap part and holding an audience was the valuable part. The market agreed fast. In April 1996, about nine months after incorporation, Lycos completed what was then the fastest initial public offering from inception to offering in NASDAQ history, and closed its first day of trading with a market value of $300 million. It reached the public markets ahead of Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) and Excite (https://404memoryfound.com/posts/what-happened-to-excite-search-engine-google.html). ## How did Lycos pass Yahoo in 1999? By buying everything else. Across 1998 and 1999 Lycos assembled a portal out of other people's companies. It acquired the web host Tripod in February 1998 for a reported $58 million in stock, then WhoWhere, a people-search and free email service, for $133 million in stock. That second deal carried in Angelfire, which would outlive almost everything around it. In October 1998 Lycos paid $83 million in stock for Wired Digital, the online arm of Wired after the print magazine went to Advance Magazine Publishers. One transaction delivered the HotBot search engine, Wired News, HotWired (https://404memoryfound.com/posts/first-banner-ad-hotwired-1994-history.html), Webmonkey and Suck.com. Gamesville, Quote.com, Matchmaker.com and Raging Bull followed. The arithmetic worked for exactly one month. Media Metrix counted nearly 32 million people visiting Lycos in March 1999, or 51.8 percent of United States internet users, against 31.2 million and 50.8 percent for Yahoo. It was the first time Lycos had finished ahead of Yahoo, and the company spent the rest of its life pointing at that number. ## Why did Terra Networks pay $12.5 billion for Lycos? Terra Networks was the internet arm of Telefonica, the Spanish telecommunications group. It had subscribers across Spain and Latin America and almost no American audience. Lycos had the American audience and no international network. Each side looked like the missing half of the other. On May 16, 2000, about two months after the NASDAQ peaked in March of that year, Lycos announced definitive agreements with Terra Networks in a deal valued at $12.5 billion. Terra chief executive Juan Villalonga would chair the merged company and Davis would run it. The transaction closed in October 2000, and the company was renamed Terra Lycos. Davis had no public doubts. "Overnight, in one fell swoop, this company has jumped from strong Internet competitor to a global powerhouse," he said of the deal in 2000, in a line CFO magazine later quoted. It remains one of the largest sums ever paid for a search engine, and it bought a business whose advertising revenue was about to collapse with the rest of the sector. ## Why did Lycos lose search to Google? Because it stopped doing search. In late 2001 Terra Lycos shut down its own crawler and began taking results from FAST, the Norwegian search company Microsoft would later buy. A company named after a hunting spider had outsourced the hunting. The portal logic assumed that wrapping email, games, stock quotes and dating around a search box would hold people in place. Google, which launched in September 1998, assumed the opposite and shipped one box on a white page. Readers picked the box. Fast Company later put the Lycos share of the global search market at 0.01 percent by 2007. Lycos was not alone in the error. AltaVista (https://404memoryfound.com/posts/what-happened-to-altavista-search-engine.html) bolted a portal onto the best index of the 1990s and lost it, and Excite took the same route into bankruptcy. The difference is scale. Lycos had bought more of the portal than anyone else, so it had more to unwind when the model stopped paying. ## Who owns Lycos today? The answer is contested, which is an unusual thing to say about a search engine. Terra sold Lycos to the South Korean portal Daum Communications on August 2, 2004 for $95.4 million in cash, less than 2 percent of what Terra had paid four years earlier. That sale closed in October 2004. Daum sold it on in August 2010 to Ybrant Digital, a digital marketing firm based in Hyderabad, India, for $36 million. "Brand Lycos needs no introduction, we are excited to bring in the Lycos properties into our fold," Ybrant chairman and chief executive Suresh Reddy said in the 2010 announcement. Ybrant paid $20 million at signing and then fought Daum over the remainder. In 2018 a New York court appointed Daum, by then merged with Kakao, as receiver over the 56 percent interest in Lycos that Ybrant held, to satisfy roughly $37 million in arbitral awards. Ybrant renamed itself Brightcom Group in May 2018 and still lists Lycos Inc. as a subsidiary. ## What is left of Lycos in 2026? Less of it every month. On March 6, 2026 a notice went up on the Lycos homepage telling users of Angelfire and Tripod that "Unfortunately we will be shutting down in the next 30 days," and asking them to move their hosting elsewhere. The message was pulled from the visible page within days but stayed in the source code. Both services went dark on April 24, 2026 after months of server errors, taking a large part of the personal web of the late 1990s with them. What happened to Angelfire (https://404memoryfound.com/posts/is-angelfire-still-online.html) is the clearest measure of where Lycos stands now. Lycos Mail followed. New addresses on the lycos.com domain were switched off, and Lycos pointed remaining customers toward a third-party provider, Exact Hosting, as the service wound down through 2026. Lycos.com itself is still up, as a small portal of syndicated news and a search box whose results are supplied by someone else. The brand turned 32 this year and no longer runs a crawler, a web host or a mailbox. ## Frequently Asked Questions ### Does Lycos still exist in 2026? Yes. Lycos.com is still online in 2026 as a small portal with syndicated news and a search box that returns results from another provider. The Lycos web hosting services Angelfire and Tripod went offline on April 24, 2026, and Lycos Mail wound down through the same year. ### Who owns Lycos now? Lycos Inc. is listed as a subsidiary of Brightcom Group, the Indian company that was called Ybrant Digital until May 2018 and that bought Lycos for $36 million in August 2010. Ownership is disputed: in 2018 a New York court placed the 56 percent interest Ybrant held in Lycos into receivership under Daum, the South Korean seller, which had merged with Kakao. ### Why did Lycos lose to Google? Lycos spent 1998 and 1999 buying web hosts, email, games and stock message boards to turn its search engine into a portal, and in late 2001 it stopped running its own crawler and took results from FAST instead. Google, which launched in September 1998, competed only on the quality of search. By 2007, Fast Company put Lycos at 0.01 percent of the global search market. **Sources:** - Terra Lycos, Inc. company history: https://www.encyclopedia.com/economics/encyclopedias-almanacs-transcripts-and-maps/terra-lycos-inc - CFO magazine, 'Terra Firma?': https://www.cfo.com/news/terra-firma/684255/ - PR Newswire, 'Ybrant Buys Lycos for $36 Million' (2010): https://www.prnewswire.com/news-releases/ybrant-buys-lycos-for-36-million-100870919.html - Archive Team wiki: Angelfire shutdown timeline: https://wiki.archiveteam.org/index.php/Angelfire - Fast Company, 'How Did a $12 Billion Search Engine Sell for $36 Million?': https://www.fastcompany.com/1682840/pop-quiz-how-did-12-billion-search-engine-sell-36-million --- # Is HD DVD Still Around? Why It Lost to Blu-ray URL: https://404memoryfound.com/posts/what-happened-to-hd-dvd-format-war.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-03 Updated: 2026-09-19 Topics: Hardware, Business Blunders **Summary:** HD DVD is not around in 2026. Toshiba stopped developing, making and selling HD DVD players and recorders on February 19, 2008, less than two years after the format reached US stores, and applied to join the rival Blu-ray Disc Association in August 2009. The discs and the surviving players still work, but nothing new has been pressed or built since 2008. **Key facts:** - Launched: April 18, 2006, Toshiba HD-A1 at $499 - Status today: Discontinued since February 19, 2008; existing discs and players still work - Owner today: Toshiba is private, bought out by a Japan Industrial Partners consortium in 2023 - Cost to Toshiba: About 110 billion yen, roughly $1.1 billion, in the 2007 fiscal year - What replaced it: Blu-ray, then streaming; 4K Blu-ray US sales rose 12 percent in 2025 ## What was HD DVD, and how was it different from Blu-ray? HD DVD answered a simple physical problem. A single-layer DVD held 4.7 gigabytes, and a two-hour film in 1080p needs far more room than that. Toshiba and NEC built a disc read by a blue-violet laser at 405 nanometers rather than the 650-nanometer red laser inside a DVD player. The shorter wavelength focused on a smaller spot, so the data pits could sit closer together. The DVD Forum adopted the format as HD DVD in November 2003. Sony and the Blu-ray Disc Association, backed by Panasonic, Philips, Samsung and LG, had announced their competing disc in February 2002. Both formats used the same laser and the same video codecs. The difference was capacity and manufacturing cost. HD DVD held 15 gigabytes per layer and 30 gigabytes on a dual-layer disc. Blu-ray held 25 and 50. HD DVD kept its data layer 0.6 millimeters below the surface, exactly like a DVD, so existing pressing plants could be retooled cheaply. Blu-ray moved the layer to 0.1 millimeters and needed new equipment and tighter tolerances. The two camps tried to merge the standards in early 2005. Talks collapsed that August over whose technology would form the base layer, and the industry got a war instead of a format. ## How did the PlayStation 3 win the war for Sony? Toshiba moved first. The HD-A1, the first HD DVD player sold in the United States, reached retailers on April 18, 2006 at $499, with a launch slate of four titles from Warner Bros. and Universal. Samsung's BD-P1000, the first standalone Blu-ray player, did not ship until June 25 that year, and it listed at $999.99. Then Sony did the thing Toshiba had no answer for. The PlayStation 3 launched in North America on November 17, 2006 at $499 for the 20GB model and $599 for the 60GB model, and every single unit shipped with a Blu-ray drive inside. Buyers spent years complaining about that price. They still carried a Blu-ray player home. Microsoft's counter was the Xbox 360 HD DVD Player, an external accessory sold from November 2006 at $199 and cut to $179 by July 2007. No Xbox 360 game ever shipped on HD DVD. The drive was optional, and studios could see that Microsoft's young console business (https://404memoryfound.com/posts/what-happened-to-original-xbox-microsoft-gaming.html) was hedging while Sony had wagered an entire console generation on one disc. ## What did Toshiba pay Paramount and DreamWorks Animation? By 2007 Toshiba was buying shelf space outright. On August 20, 2007, Paramount Pictures and DreamWorks Animation both went HD DVD exclusive worldwide, a commitment that covered Paramount Vantage, Nickelodeon Movies, MTV (https://404memoryfound.com/posts/when-mtv-actually-played-music-rise-fall.html) Films and the DreamWorks Animation library. Films directed by Steven Spielberg were carved out of the deal. The New York Times reported that the studios received $150 million in promotional payments for that exclusivity, split as $50 million to Paramount and $100 million to DreamWorks Animation. Toshiba was winning on hardware price at the same time. On November 2, 2007, Walmart sold the Toshiba HD-A2 for $98.87, about a fifth of what the first player had cost eighteen months earlier. It worked on the standalone player count and nowhere that mattered. Every PS3 sold was another Blu-ray household, and Sony was moving consoles by the million while Toshiba moved players by the hundred thousand. ## What happened on January 4, 2008? Warner Bros. Entertainment was the biggest home video distributor in the world and the one major studio still releasing on both formats. On January 4, 2008, two days before CES opened in Las Vegas, it announced it would release high-definition titles on Blu-ray only. "Warner Bros.' move to exclusively release in the Blu-ray disc format is a strategic decision focused on the long term," chairman and chief executive Barry Meyer said in the 2008 announcement, warning that the window for high-definition disc could be missed "if format confusion continues to linger." The HD DVD Promotion Group cancelled the CES press event it had already booked. Within weeks Best Buy said it would steer customers toward Blu-ray, Walmart said it would stock Blu-ray only, and Netflix dropped HD DVD from a rental catalogue that was still mailing discs in red envelopes (https://404memoryfound.com/posts/netflix-dvd-mail-service-shutdown.html). Six weeks after Warner Bros. spoke, the war was settled. ## How much money did HD DVD cost Toshiba? On February 19, 2008, Toshiba announced it would stop developing, manufacturing and marketing HD DVD players and recorders, with shipments to retailers winding down by the end of March. The format had been on sale in the United States for under two years. "We carefully assessed the long-term impact of continuing the so-called 'next-generation format war' and concluded that a swift decision will best help the market develop," Toshiba president and chief executive Atsutoshi Nishida said in the company's statement that day. The bill arrived the following month. Toshiba told investors the retreat would cost roughly 110 billion yen, about $1.1 billion, in the 2007 fiscal year, Variety reported in 2008, more than double the loss the company had projected earlier. Close to a million HD DVD players had been sold worldwide by then, and more than 400 HD DVD titles had been released in the United States. On August 10, 2009, eighteen months after surrendering, Toshiba applied to join the Blu-ray Disc Association and began selling Blu-ray players of its own. ## Who owns Toshiba now, and can you still play HD DVD discs? Toshiba still exists in 2026, but not as the listed company that fought the format war. In December 2023 it left the Tokyo Stock Exchange after 74 years, bought out for about $14 billion by a consortium led by the private equity firm Japan Industrial Partners. It has been privately held ever since. HD DVD itself is finished commercially. No disc has been pressed and no player built since 2008. The hardware that survives still works: an HD-A1, an HD-A2 or an Xbox 360 HD DVD Player reads a 2007 disc exactly as it did in 2007. Nothing else reads them. A Blu-ray player cannot, and neither can a PC drive unless it is one of the rare combination units built during the war. Where to find one today. Players and discs trade secondhand only, through thrift stores, estate sales and collector listings on Etsy (https://www.etsy.com/search?q=hd+dvd). A machine that listed at $499 in April 2006 now sells as a curiosity, and a few titles are worth hunting because their HD DVD masters were never reissued on Blu-ray. Buy a tested player first: a dead laser cannot be replaced from a parts bin that stopped being made in 2008. ## Did Blu-ray actually win anything? Blu-ray took the territory and then watched it shrink. US consumer spending on physical discs fell 28 percent in the first half of 2023, to $754 million from $1.05 billion a year earlier, CNBC reported. Best Buy stopped selling DVDs and Blu-ray discs across its stores and website in 2024. Sony ended production of recordable Blu-ray discs in February 2025, closing an 18-year manufacturing run. The format did not die, though, and 2025 was better than the decline suggested. FlatpanelsHD reported that 4K Ultra HD Blu-ray sales in the United States rose 12 percent in 2025 while total US physical media spending came to $870 million, down 9.3 percent year over year, against a 23.4 percent drop the year before. The UK 4K market grew 19.5 percent, driven by boutique labels, steelbooks and collectors who want a copy nobody can remove from a library. That is the real shape of what Toshiba lost: not the mass market DVD had owned, but a premium niche. It burned $1.1 billion on a prize that became a collector's format, which is roughly what happened to Sony when it lost Betamax to VHS (https://404memoryfound.com/posts/betamax-vs-vhs-format-war.html) a generation earlier, with the roles reversed. ## Frequently Asked Questions ### Can you still play HD DVD discs in 2026? Yes, but only on hardware built before 2008. HD DVD discs play on original Toshiba players such as the HD-A1 and HD-A2, on the Xbox 360 HD DVD Player accessory, and on a few combination PC drives from that era. Blu-ray players and standard DVD drives cannot read them, and no HD DVD hardware has been manufactured since Toshiba discontinued the format on February 19, 2008. ### Why did HD DVD lose to Blu-ray? HD DVD lost because Sony put a Blu-ray drive in every PlayStation 3 from its November 17, 2006 launch, creating millions of Blu-ray households before Toshiba could sell a comparable number of standalone players. Blu-ray also held more data, 25 gigabytes per layer against HD DVD's 15. The decisive blow came on January 4, 2008, when Warner Bros. went Blu-ray exclusive and Toshiba quit six weeks later. ### How many HD DVD players were sold? Close to a million HD DVD players had been sold worldwide by the time Toshiba discontinued the format in February 2008, counting Toshiba's own machines and the Xbox 360 HD DVD Player add-on. More than 400 HD DVD titles had been released in the United States over the format's two years on sale, and none have been pressed since. **Sources:** - Toshiba, Discontinuation of HD DVD Businesses (February 19, 2008): https://www.global.toshiba/ww/ir/corporate/news/2008/02/20080219-03.html - Engadget, Warner goes Blu-ray exclusive (January 4, 2008): https://www.engadget.com/2008-01-04-warner-goes-blu-ray-exclusive.html - Variety, HD DVD flop costs Toshiba $1.1 bil (2008): https://variety.com/2008/digital/features/hd-dvd-flop-costs-toshiba-1-1-bil-1117982699/ - CNN Business, Toshiba is set to delist in Japan after 74 years (2023): https://www.cnn.com/2023/09/21/investing/japan-toshiba-stock-privatization-deal-intl-hnk - FlatpanelsHD, UHD Blu-ray returned to growth in 2025: https://www.flatpanelshd.com/news.php?subaction=showfull&id=1772040307 --- # Is Lotus 1-2-3 Still Around? IBM Killed It in 2013 URL: https://404memoryfound.com/posts/what-happened-to-lotus-1-2-3-spreadsheet.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-04-03 Updated: 2026-09-19 Topics: Software & Apps, Business Blunders **Summary:** Lotus 1-2-3 is discontinued. IBM, which bought Lotus Development Corporation in 1995 for about $3.5 billion, withdrew the spreadsheet from marketing on June 11, 2013 and ended support on September 30, 2014. The DOS releases still run in emulators, but no new license has been sold in more than a decade. **Key facts:** - Launched: January 26, 1983, Lotus Development Corporation - Price at launch: $495 - Status today: Discontinued. Withdrawn from marketing June 11, 2013, support ended September 30, 2014 - Owner today: IBM, which bought Lotus Development Corporation in 1995 for about $3.5 billion - Peak: About 70% of the spreadsheet market in 1988, on $468.5 million in company revenue ## What made Lotus 1-2-3 the first killer app? Mitch Kapor and Jonathan Sachs founded Lotus Development Corporation in Cambridge, Massachusetts, in 1982. Kapor had written VisiTrend and VisiPlot for VisiCorp and knew what buyers wanted. Sachs wrote the program. Lotus 1-2-3 shipped on January 26, 1983 at a list price of $495. The name came from its three functions in one product: spreadsheet, charting and database. VisiCalc, the 1979 Apple II spreadsheet that got there first, did one of those. Sachs wrote 1-2-3 in assembly language, straight against the Intel 8088 inside the IBM PC. Recalculations that crawled in VisiCalc finished almost instantly. For a controller closing the books, that was the whole argument. The money arrived at once. Lotus grossed $53 million in 1983 and ended its first year as the second largest software company in the world, behind Microsoft. Kapor later called the product "the Google or Facebook of its time" in an interview with The Register in 2013, and said that "People were buying PCs to learn 1-2-3." A $495 list price was serious money in 1983, which is easier to feel next to what 1990s tech cost in 2026 dollars (https://404memoryfound.com/posts/90s-tech-inflation-calculator.html). ## How big did Lotus get before Excel arrived? Lotus sold 750,000 copies of 1-2-3 in 1986 alone. By 1988 company revenue reached $468.5 million, and Lotus 1-2-3 held roughly 70 percent of a spreadsheet market worth about $500 million. Lotus was the largest independent software company in the world. An economy grew around the product: training courses, macro libraries, add-on templates and shelves of third-party manuals. Job postings asked for Lotus 1-2-3 proficiency the way they now ask for advanced Excel skills. That dominance also set the standard. Competing spreadsheets had to read .WK1 files, and they had to answer to the same keystrokes, because millions of users had years of muscle memory in the slash-key menu. Lotus did not only hold market share. It held habits, which is the harder thing to take away. Kapor stepped down as chief executive in 1986, four years after founding the company, and went on to co-found the Electronic Frontier Foundation in 1990 with John Perry Barlow and John Gilmore. Jim Manzi ran Lotus through the decade that followed, and that decade went badly. ## Why did Lotus miss the shift to Windows? Microsoft shipped Excel for the Macintosh in 1985 and Excel for Windows in 1987, well before either platform mattered to corporate buyers. Windows 3.0 arrived in 1990 and was the first version businesses actually installed. Excel was already there: graphical, mouse-driven and finished. Lotus did not ship a Windows version of 1-2-3 until 1991, and it was poorly received. It behaved like a DOS program wearing a Windows costume. Users went back to the DOS release or went to Excel. The improved Windows release in 1993 was a competent product that arrived after the decision had been made, and Excel passed Lotus 1-2-3 in spreadsheet market share around that year. The delay was arithmetic more than incompetence. Lotus 1-2-3 for DOS generated the revenue, the sales force sold DOS, and an entire training industry taught DOS. Doing the Windows version properly meant attacking the company's own book of business. Microsoft had nothing to protect. WordPerfect (https://404memoryfound.com/posts/what-happened-to-wordperfect-word-processor.html) lost its market in the same years for the same reason. By the time Windows 95 launched (https://404memoryfound.com/posts/the-night-windows-95-launched.html), Microsoft Office was the default purchase and the spreadsheet question was settled. ## What happened in the Lotus v. Borland lawsuit? In 1990 Lotus sued Borland International over Quattro Pro, which reproduced the Lotus 1-2-3 menu command hierarchy: the slash key, then Worksheet, Range, Copy, Move, File, Print, Graph, Data, System and Quit. Lotus argued the menu tree was creative expression protected by copyright. Borland argued it was a method of operation, and methods of operation are not copyrightable under 17 U.S.C. 102(b). The First Circuit Court of Appeals sided with Borland in March 1995. "We hold that the Lotus menu command hierarchy is an uncopyrightable 'method of operation,'" the court wrote in Lotus Development Corp. v. Borland International, Inc. The Supreme Court took the case and split 4 to 4 in 1996, which left the appeals ruling in force without setting national precedent. The stakes ran well past one spreadsheet. Had Lotus won, whoever arranged a set of menus first could have owned that arrangement. The ruling is the reason you can move between competing programs today without relearning where Save lives. It also marked where Lotus had ended up. A company that had out-engineered everyone in 1983 was, by 1990, trying to litigate a rival out of the market. ## Who owns Lotus 1-2-3 today? IBM has owned Lotus 1-2-3 since 1995, and it stopped selling the product more than a decade ago. IBM opened a hostile tender offer for Lotus Development Corporation on June 5, 1995 at $60 a share, raised the offer to $64 within a week, and closed the tender on July 5, 1995 with roughly 97 percent of the 47.4 million outstanding Lotus shares accepted. The equity value came to about $3.5 billion. IBM was not buying the spreadsheet. It was buying Lotus Notes, the groupware platform that large companies ran their email and workflow on. Lotus 1-2-3 came along with the paperwork. Under IBM the spreadsheet was folded into Lotus SmartSuite and steadily de-emphasized. IBM dropped the Lotus name from Notes and Domino in November 2012, withdrew Lotus 1-2-3 from marketing on June 11, 2013, and ended support on September 30, 2014 with no extensions offered. The rest of the Lotus line left IBM altogether. IBM agreed in December 2018 to sell Notes, Domino and Connections to HCL Technologies, and the deal closed in July 2019. Those products are sold today as HCL Notes and HCL Domino. The spreadsheet was not part of that sale. ## Can you still run Lotus 1-2-3 in 2026? Yes, and people do. No license has been sold since 2013, but the DOS releases run cleanly in DOSBox, and the Internet Archive holds installable disk images of several versions, including Release 2.3 and Release 4.0 for DOS. PCjs Machines boots Release 1A in a browser tab with nothing to download. There is a stranger route. In May 2022 the security researcher Tavis Ormandy patched the 1990 UNIX System V build of Lotus 1-2-3 to run natively on modern x86 Linux, without emulation, after a collector supplied a copy of the long-lost Lotus development toolkit. The Register covered it that month. What you cannot easily do is carry the files forward. Modern versions of Microsoft Excel no longer open Lotus .wk1 and .wk3 spreadsheets, so old corporate financial models sit in a format their own successor refuses to read. Converting them usually means running 1-2-3 in emulation and exporting from there. One habit outlasted the software. Excel still ships a Lotus compatibility setting for transition navigation keys, a leftover for anyone who learned to reach for the slash key first. ## Frequently Asked Questions ### Is Lotus 1-2-3 still available to buy? No. IBM withdrew Lotus 1-2-3 from marketing on June 11, 2013 and ended support on September 30, 2014, so no new license has been sold since. The DOS releases of Lotus 1-2-3 still run in emulators such as DOSBox, and the Internet Archive hosts disk images of several versions. ### Why did Lotus 1-2-3 lose to Microsoft Excel? Lotus 1-2-3 was built for DOS, and Lotus did not ship a Windows version until 1991, six years after Microsoft shipped Excel for the Macintosh in 1985. That first Windows release of Lotus 1-2-3 was poorly received, and Excel passed it in spreadsheet market share around 1993. ### How much did Lotus 1-2-3 cost when it launched? Lotus 1-2-3 launched on January 26, 1983 at a list price of $495. It grossed $53 million in that first year, which made Lotus Development Corporation the second largest software company in the world behind Microsoft. **Sources:** - History of Lotus Development Corporation, FundingUniverse: https://www.fundinguniverse.com/company-histories/lotus-development-corporation-history/ - Lotus Development Corp., Schedule 14D-9 on IBM's tender offer (SEC, 1995): https://www.sec.gov/Archives/edgar/data/0000711761/000095011295001622/0000950112-95-001622.txt - Lotus Development Corp. v. Borland International, Inc., 49 F.3d 807 (1st Cir. 1995): https://cyber.harvard.edu/people/tfisher/IP/1995_Lotus.pdf - So long Lotus 1-2-3: IBM ceases support after over 30 years of code, The Register (2014): https://www.theregister.com/2014/10/02/so_long_lotus_123_ibm_ceases_support_after_over_30_years_of_code/ - Lotus 1-2-3 turns 30: Mitch Kapor on the Google before Google, The Register (2013): https://www.theregister.com/2013/01/26/mitch_kapor_lotus_123_anniversary/ --- # Was the iPod the First MP3 Player? Not Even Close URL: https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-03 Updated: 2026-09-19 Topics: Hardware, Music & Entertainment **Summary:** The iPod was not the first MP3 player. SaeHan's MPMan reached buyers in March 1998 and the Diamond Rio PMP300 followed that September, three years before Apple announced the iPod on October 23, 2001. Apple won the category anyway with a hard drive, a store and a Windows version, then closed it: the iPod line was discontinued on May 10, 2022, and Apple has not made one since. **Key facts:** - Actually first: MPMan, SaeHan Information Systems, March 1998 - iPod launched: Announced October 23, 2001; on sale November 10, 2001 - Launch price: $399 for 5GB, Mac only - Status today: Discontinued. Apple ended the iPod line on May 10, 2022 - Price today: About $80 to $250 for a working iPod classic in 2026 ## Who built the MP3 player that beat Apple by three years? The first mass-produced portable MP3 player came from SaeHan Information Systems, a South Korean company almost nobody in the United States had heard of. The MPMan went on public display at the CeBIT trade show in Hannover on March 10, 1998, and Eiger Labs imported it to North America as the MPMan F10 for around $250 with 32MB of memory. 32MB held roughly a dozen songs, depending on the bitrate you encoded at. There was no store, no sync software worth the name, and no way to fill it that did not involve a cable and a lot of patience. It worked, and it proved the category could exist. Apple announced the iPod on October 23, 2001, three years and seven months later. Steve Jobs never claimed to have invented the portable music player. He claimed something narrower and, as it turned out, far more defensible. "Apple has invented a whole new category of digital music player that lets you put your entire music collection in your pocket," he said in the company's launch release that day. The category he meant was the one where the whole library came with you. ## Why did the Diamond Rio matter more than the MPMan? Diamond Multimedia shipped the Rio PMP300 on September 15, 1998, six months after the MPMan, at $200 with the same 32MB of flash memory. It is the player Americans actually saw in stores, and the one the record industry decided to kill. The Recording Industry Association of America sued, arguing the Rio broke the Audio Home Recording Act of 1992 because it carried no copy-management system. The Ninth Circuit decided the case on June 15, 1999, ruled against the RIAA, and treated what the Rio did as space-shifting: moving files a listener already owned from a hard drive into a pocket. That ruling is why every MP3 player sold in the United States afterward was legal, the iPod included. Diamond won the case and lost the market. The Rio proved demand without solving the part that actually hurt, which was getting music onto the device. A second wave of hard-drive players followed, led by the Creative Nomad Jukebox in 2000, and they did not solve it either. They held more songs and were still miserable to load. ## What did the 2001 iPod actually do better? The first iPod went on sale on November 10, 2001 at $399 for 5GB in a 6.5-ounce case, and it was Mac only. Apple described it as up to 1,000 CD-quality songs in an ultra-portable design. Three things separated it from everything on the shelf beside it. The first was the hard drive. 5GB was more than 150 times the Rio's 32MB, which turned the product from a playlist you curated before leaving the house into your entire collection. The second was FireWire. Filling a library over the connections rivals used took hours; the iPod did it in minutes. The third was the scroll wheel, and it is the one people still bring up. Rivals gave you a directional pad and asked you to click through a thousand songs one at a time. Apple gave you a wheel that got faster the longer you spun it. It was a small thing you did dozens of times a day, and it was the difference between a gadget you tolerated and one you reached for. The Sony Discman (https://404memoryfound.com/posts/what-happened-to-sony-discman-portable-cd-player.html) it replaced never had that problem, because it only ever held one album. ## How did iTunes and Windows support decide the fight? On July 17, 2002 Apple shipped the second-generation iPod at $399 for 10GB and $499 for 20GB, and for the first time it worked with Windows, bundled with MusicMatch Jukebox rather than iTunes. Windows ran the overwhelming majority of home computers in 2002. Until that day the iPod could only sell to Mac owners, a small slice of the market. The second move was the store. Apple opened the iTunes Music Store on April 28, 2003 with more than 200,000 songs at 99 cents each and no subscription. Jobs sold it on ownership rather than rental, promising buyers "the revolutionary rights to burn an unlimited number of CDs for personal use and to put music on an unlimited number of iPods". Apple said it sold a million tracks in the first five days. That combination is the whole answer to why the iPod won. Napster (https://404memoryfound.com/posts/napster-destroyed-music.html) had already proved people wanted digital music, and nobody had yet made paying for it easier than not paying for it. Apple did, and it owned the hardware the music played on. The next year it pushed further down the price list with the iPod mini, announced on January 6, 2004 at $249 for 4GB and the first iPod to use the click wheel. ## Is the iPod still made in 2026? No. Apple discontinued the iPod on May 10, 2022, when it stopped building the seventh-generation iPod touch, the last model standing. There is no iPod in the Apple Store in 2026, and Apple has announced no successor. The line came apart one model at a time. The iPod classic, the hard-drive model with the click wheel, was dropped in 2014 after the 1.8-inch drives it depended on went out of production and Apple said the parts were no longer available anywhere. The iPod nano and the iPod shuffle, both introduced in 2005, were discontinued together on July 27, 2017. That left the touch alone for five years, unchanged since 2019. Apple framed the ending as absorption rather than retreat. "Today, the spirit of iPod lives on," said Greg Joswiak, Apple's senior vice president of worldwide marketing, in the company's May 2022 announcement, pointing at the iPhone, the Apple Watch and HomePod. He had a point in the narrow sense, because the iPhone that finished the iPod shipped with an iPod app on its home screen. Microsoft's Zune (https://404memoryfound.com/posts/why-microsoft-zune-actually-failed.html) never got close enough to matter, and Apple's own phone did the job instead. ## Where to find one today Apple has not sold an iPod since 2022, so the second-hand market is the only route. In 2026 a working iPod classic from the fifth to seventh generation typically runs about $80 to $250 depending on capacity and condition, while a clean first-generation unit from 2001 sits closer to $200 to $500, and more with the original box and cable. Sealed and special-edition units go far higher and belong to collectors rather than listeners. Before paying, check that the battery has been replaced and that the click wheel responds all the way around. Sellers who refurbish these for a living usually list what they swapped. Rebuilt and modified units, including iPods fitted with flash storage in place of the original hard drive, are sold by independent sellers on Etsy (https://www.etsy.com/search?q=ipod+classic). ## Frequently Asked Questions ### What was the first MP3 player ever made? The first mass-produced portable MP3 player was the MPMan, built by SaeHan Information Systems of South Korea and shown at the CeBIT trade show in March 1998. Eiger Labs sold it in North America as the MPMan F10 for around $250 with 32MB of storage. The Diamond Rio PMP300, which arrived on September 15, 1998 at $200, was the first MP3 player to sell in volume in the United States. ### When did Apple stop making the iPod? Apple discontinued the iPod on May 10, 2022, when it ended production of the seventh-generation iPod touch. The iPod classic had already been dropped in 2014, and the iPod nano and iPod shuffle went in July 2017. Apple has not sold a new iPod since 2022 and has named no replacement for it. ### How much did the first iPod cost in 2001? The first iPod cost $399 when it went on sale on November 10, 2001, with a 5GB hard drive and Mac-only support. Apple advertised it as holding up to 1,000 CD-quality songs. Windows support arrived with the second-generation iPod on July 17, 2002, priced at $399 for 10GB and $499 for 20GB. **Sources:** - Apple Presents iPod (Apple Newsroom, October 23, 2001): https://www.apple.com/newsroom/2001/10/23Apple-Presents-iPod/ - Flashback 1998: Birth of the MP3 Player (Sound & Vision): https://www.soundandvision.com/content/flashback-1998-birth-mp3-player - RIAA v. Diamond Multimedia Systems, 180 F.3d 1072 (9th Cir. 1999): https://caselaw.findlaw.com/court/us-9th-circuit/1054784.html - Apple Launches the iTunes Music Store (Apple Newsroom, April 28, 2003): https://www.apple.com/newsroom/2003/04/28Apple-Launches-the-iTunes-Music-Store/ - Apple discontinues the last iPod model (CNBC, May 10, 2022): https://www.cnbc.com/2022/05/10/apple-discontinues-the-last-ipod-model.html --- # Is Craigslist Still Around in 2026? Who Owns It Now URL: https://404memoryfound.com/posts/what-happened-to-craigslist-ugliest-website-worth-billions.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-03 Updated: 2026-09-19 Topics: Internet Culture, Money & Tech **Summary:** Craigslist is still in business in 2026, running local sites in more than 570 cities in over 70 countries. The company is private and held by founder Craig Newmark and chief executive Jim Buckmaster, who has run it since November 2000. Revenue has fallen from $1.034 billion in 2018 to $302 million in 2024, its lowest total since 2013. **Key facts:** - Launched: 1995 as Craig Newmark's San Francisco email list; incorporated 1999 - Status today: Still operating, in more than 570 cities in over 70 countries - Owner today: Private; founder Craig Newmark and CEO Jim Buckmaster - Revenue: $302 million in 2024, down from $1.034 billion in 2018 (AIM Group) - Price today: Most listings free; paid job ads in big metros are the largest revenue line ## Who owns Craigslist today? Craigslist, Inc. is private. It has never sold shares to the public and it has no outside investors pushing it toward a sale. When the AIM (https://404memoryfound.com/posts/what-happened-to-aim-aol-instant-messenger.html) Group's revenue estimate made headlines in January 2019, CNBC described the company as "wholly owned by founder Craig Newmark and CEO Jim Buckmaster," running on a staff of about 50 people in San Francisco. It was not always that tidy. In 2004 a rival online marketplace bought roughly a quarter of Craigslist from a former employee, and the relationship soured. The shareholder sued in Delaware in 2008 after the Craigslist board adopted a rights plan and other measures that worked against its stake. On September 9, 2010, the Court of Chancery struck the poison pill down, with Chancellor William B. Chandler III writing that "Having chosen a for-profit corporate form, the craigslist directors are bound by the fiduciary duties and standards that accompany that form." The stake went back to Craigslist in June 2015 and the litigation ended with it. Jim Buckmaster has been chief executive since November 2000, which makes him one of the longest-serving CEOs on the American web. Craig Newmark stepped back from running the site years ago and now spends most of his time funding journalism, veterans' groups and cybersecurity work through Craig Newmark Philanthropies, which he set up in 2015. ## How much money does Craigslist make now? Less than a third of what it made at the peak. The AIM Group, a classifieds consultancy that has tracked the site since 2004, estimated Craigslist revenue at $1.034 billion in 2018, the first year it cleared a billion dollars. The next year it fell 27 percent to $760 million, the first drop AIM had ever recorded, and a 37 percent fall in job advertising did most of the damage. The slide did not stop there. AIM's 2025 report put 2024 revenue at $302 million, down another 20 percent, a third straight down year and the lowest total since 2013. What has not changed is the shape of the business. Most Craigslist listings are still free to post. The money comes from job ads in major metros, brokered apartment listings, commercial real estate and a short list of other paid categories, so a weak year in hiring is a weak year for the whole company. Even after the decline, AIM's 2025 report still ranked Craigslist first among United States horizontal classifieds sites in both revenue and traffic, ahead of Mercari and OfferUp. ## How a San Francisco email list turned into a classifieds giant Craig Newmark started sending friends in San Francisco a list of local events, apartments and jobs by email in 1995. People asked to be added, the list outgrew email, and it moved to the web in 1996. Newmark incorporated the company in 1999, right as the dot-com bubble (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html) was inflating around companies with far more funding and far less traffic. Jim Buckmaster arrived as a programmer in 2000 and took over as chief executive that November. The pair then did the opposite of what everyone else did. They did not raise venture money, did not expand the paid categories aggressively, and did not grow the staff past about 50 people. Craigslist charged for job ads in a handful of big cities and gave away everything else. That choice is what made the 2010 Delaware ruling so unusual. A for-profit Delaware corporation that openly declines to maximize its own value had no obvious precedent, and the court said so. ## What Craigslist did to newspaper classifieds American newspapers ran on classified advertising, and the collapse of that line is the closest thing the industry has to a cause of death. Newspaper Association of America figures show classified revenue peaking at $19.6 billion in 2000 and falling to about $6 billion by 2010, a drop of roughly 70 percent in a decade. By 2012 it was down to $4.6 billion. Craigslist was not the only reason. Monster.com took jobs, Realtor.com took housing and Cars.com took autos, and readers were leaving print for other reasons entirely. But Craigslist took the whole page at once and charged nothing for most of it. One academic study of 1,000 newspapers between 2000 and 2007 estimated that classified advertisers saved $5 billion by posting free on Craigslist instead of buying newspaper space. Online display advertising, which started with the first banner ad in 1994 (https://404memoryfound.com/posts/first-banner-ad-hotwired-1994-history.html), never replaced that money for newspapers. Classifieds had carried high margins and low costs. A free listing board with no sales force could not be undercut. ## Why the personals and adult sections disappeared On August 24, 2010, 17 state attorneys general wrote to Craigslist demanding that it close its adult services section, arguing the company could not keep ads for prostitution and trafficking off the site. Craigslist first covered the link with a black and white "censored" bar, then shut the section down in early September 2010. The bigger cut came eight years later. On March 23, 2018, two days after Congress passed the Fight Online Sex Trafficking Act, Craigslist deleted its personals section worldwide and posted a notice to users: "Any tool or service can be misused. We can't take such risk without jeopardizing all our other services." The new law made platforms liable for sex trafficking ads posted by users, which stripped away the protection that had let a company of 50 people host millions of listings it did not read. Craigslist did not fight it or staff up to moderate. It removed the category, which is what a company built to stay small does when the legal risk arrives. ## Why the site still looks like 1996 Craigslist is blue links on a gray page, and has been for close to thirty years. Most of its software is written in Perl, a language that was common when the site was built and rare now. The company has never rebuilt the front end around images, feeds or recommendations, and it has never run an algorithm designed to keep anyone scrolling. The clearest sign of how little that philosophy bends: Craigslist did not ship an official smartphone app until December 2019, more than eleven years after the App Store opened. It still belongs on the short list of 90s websites still online (https://404memoryfound.com/posts/90s-websites-still-online.html) in something close to their original form. The trade is visible in the numbers. Facebook Marketplace, OfferUp and Mercari offer profiles, ratings and built-in payments, and they have taken share for a decade. Craigslist offers a text box and a phone number, and for apartments, gigs, free furniture and odd local jobs, a lot of people still prefer it. ## Frequently Asked Questions ### Is Craigslist still in business in 2026? Yes. Craigslist is still operating in 2026, with local sites in more than 570 cities in over 70 countries, and it remains the largest general classifieds site in the United States by revenue and traffic. Its estimated revenue was $302 million in 2024, down from $1.034 billion in 2018. ### Who owns Craigslist now? Craigslist, Inc. is a private company owned by founder Craig Newmark and chief executive Jim Buckmaster, who has held the CEO job since November 2000. An outside marketplace held about a quarter of the company from 2004 until June 2015, when it sold the stake back and ended its Delaware litigation with Craigslist. ### When did Craigslist shut down its personals section? Craigslist removed its personals section on March 23, 2018, two days after Congress passed the Fight Online Sex Trafficking Act. The company had already closed its adult services section in September 2010 after 17 state attorneys general demanded it, and neither section has returned. **Sources:** - AIM Group: Craigslist revenue drops once more (2025 Craigslist Report): https://aimgroup.com/2025/03/12/craigslist-revenue-drops-once-more-rival-mercari-unable-to-pull-in-front/ - CNBC: Craigslist posts annual revenue of $1 billion (2019): https://www.cnbc.com/2019/01/24/craigslist-posts-annual-revenue-of-1-billion-study.html - eBay Domestic Holdings, Inc. v. Newmark, Del. Ch. (September 9, 2010): https://caselaw.findlaw.com/court/de-supreme-court/1558886.html - NPR: Craigslist shuts down personals section (March 2018): https://www.npr.org/sections/thetwo-way/2018/03/23/596460672/craigslist-shuts-down-personals-section-after-congress-passes-bill-on-traffickin - MinnPost: How Craigslist killed the newspapers' golden goose (2014): https://www.minnpost.com/business/2014/02/how-craigslist-killed-newspapers-golden-goose/ --- # Does the Nintendo Wii Still Work Online in 2026? URL: https://404memoryfound.com/posts/what-happened-to-nintendo-wii.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-03 Updated: 2026-09-20 Topics: Gaming, Hardware **Summary:** The Nintendo Wii still works in 2026 as an offline console, and every disc in its library runs without touching a server. Nintendo ended official online play on May 20, 2014, stopped selling through the Wii Shop Channel on January 30, 2019, and took its last repair request in Japan on March 31, 2020. Games already purchased can still be redownloaded, and used consoles sell for about $40 to $75. **Key facts:** - Launched: November 19, 2006 in North America at $249.99 - Units sold: 101.63 million worldwide, Nintendo's lifetime figure - Status today: Discontinued October 21, 2013; consoles still play discs offline - Online today: Nintendo's servers ended May 20, 2014; fan services Wiimmfi and WiiLink fill the gap - Price today: About $40 to $75 used in 2026, $80 to $120 boxed with Wii Sports ## Why did Nintendo build a console with last-generation graphics? The Nintendo Wii reached North American stores on November 19, 2006 at $249.99. Two days earlier Sony had launched the PlayStation 3 at $499 for the 20 GB model and $599 for the 60 GB one, and Microsoft's Xbox 360 had already been on sale for a year. On paper the Wii lost every comparison. It could not output HD video, it shipped without a hard drive, and its processor was a faster relative of the chip inside the GameCube. Cheap, familiar silicon let Nintendo sell the console at a profit from day one instead of subsidizing every unit the way Sony and Microsoft did. The strategy came from Satoru Iwata, who had run Nintendo since 2002 and worried that every console generation was shedding players rather than adding them. More buttons, longer tutorials, higher prices. His answer was to stop fighting over the same customers and go find people who had never owned a console at all. Nintendo had a patchy record with unusual hardware, from the Power Glove (https://404memoryfound.com/posts/what-happened-to-nintendo-power-glove.html) to the Virtual Boy and its red-on-black display (https://404memoryfound.com/posts/what-happened-to-virtual-boy-nintendo.html), so the bet was not a safe one. Iwata opened his 2005 Game Developers Conference keynote with a line the industry still repeats: "On my business card, I am a corporate president. In my mind, I am a game developer. But in my heart, I am a gamer." The Nintendo Wii was that sentence turned into hardware. ## How did Wii Sports sell 82.9 million copies? Nintendo bundled Wii Sports with the console in every market except Japan and South Korea. That one decision made it the best-selling game on the system, at 82.90 million copies in Nintendo's own sales data, and the best-selling game ever released on a single platform. The game held five events: tennis, baseball, bowling, golf and boxing. None of them were deep. All of them could be explained in a sentence and played with one hand. Someone who had never held a controller could bowl a strike within two minutes, and the Wii Remote's pointer did the rest of the teaching. The effect spread past living rooms. Retirement homes ran bowling leagues on it, physical therapists used it for range-of-motion work, and local news stations filed the same segment about grandparents playing video games through 2007 and 2008. Nintendo kept feeding the audience it had found. Mario Kart Wii, released in April 2008, sold 37.38 million copies and is still the best-selling Wii game never bundled with hardware. Wii Fit, which shipped with the Wii Balance Board in 2008, sold 22.67 million and created the fitness-game aisle that Ring Fit Adventure later inherited. ## Did the Wii really beat the PlayStation 3 and Xbox 360? Yes, and not narrowly. Nintendo puts lifetime Nintendo Wii shipments at 101.63 million units. Sony's PlayStation 3 finished its run at 87.4 million. Microsoft has never published a lifetime figure for the Xbox 360, which is its own sort of answer. The scale looks stranger next to what came immediately before. The GameCube sold 21.74 million units (https://404memoryfound.com/posts/why-nintendo-gamecube-lost-console-war.html) across its entire life. The Wii passed that total inside its first year on sale. Supply was the story of 2007 and 2008. Shoppers called stores in the morning to ask what had come off the truck, and retailers ran lotteries instead of stocking shelves. The Wii never took the all-time crown. Sony's PlayStation 2 still leads every home console ever built (https://404memoryfound.com/posts/playstation-2-best-selling-console-ever.html), and Nintendo's own Switch has since passed the Wii by more than 50 million units. But for one generation, the cheapest and technically weakest machine on the shelf outsold both expensive rivals. ## Why did the Wii fade after 2009? The audience Nintendo recruited turned out to be the audience that leaves. Households that bought a Wii for Wii Sports and Wii Fit had little reason to buy a third or fourth game, and by 2010 a lot of consoles were sitting under televisions in the off position. Third-party publishers never solved the machine either. A studio building for the PlayStation 3 and Xbox 360 could not move that work onto a 480p console with motion controls without rebuilding it, so most shipped cheap originals instead. The library filled with party compilations and licensed filler sold on cover art alone. Then the phone arrived. The same buyers who wanted five minutes of something simple were now carrying a device that delivered it for a dollar, with no extra hardware and no television to borrow. Nintendo had proved a huge casual market existed and then watched Apple and Google serve it more conveniently. Nintendo discontinued the original Wii and the Wii Family Edition on October 21, 2013. The stripped-down Wii Mini, which dropped internet access entirely, stayed in production until November 13, 2017. ## What happened to Wii online play and the Wii Shop Channel? Nintendo dismantled the online side in stages. WiiConnect24, which powered the Forecast Channel, the News Channel, Everybody Votes and the Wii Message Board, went dark on June 27, 2013. Nintendo Wi-Fi Connection followed on May 20, 2014, ending official matchmaking, leaderboards and online play for every Wii game that used it. The Wii Shop Channel stopped taking money on January 30, 2019. Nintendo gave the reason plainly in its closure Q&A: "Now that customers have shifted to Nintendo eShop on Nintendo Switch and Nintendo 3DS family systems, we plan to focus our efforts in those areas." Redownloads survived that shutdown. A console still signed in to its original account can pull down WiiWare and Virtual Console titles it already owns, and the Wii System Transfer Tool to Wii U still runs, though Nintendo has said both will end at a date it has not announced. Hardware support is finished. Nintendo Japan accepted its last Wii repair request on March 31, 2020, citing a shortage of parts. Anything that breaks now is a job for an independent repair shop. ## What replaced the Nintendo Wii? The Wii U arrived on November 18, 2012 and sold 13.56 million units in its entire life, less than a seventh of what the Wii managed. The name was the first problem: shoppers could not tell whether they were buying a new console or an accessory for the one they owned. The Nintendo Switch, released in March 2017, is where the Wii's ideas actually landed. Detachable controllers with motion sensing, a machine built around people playing in the same room, and a deliberate refusal to compete on raw graphics. It has shipped 155.92 million units as of March 31, 2026, which is more than the Wii and the Wii U combined and then some. Nintendo also kept mining the catalog: Wii Sports returned as Nintendo Switch Sports in 2022. ## Can you still play a Wii in 2026, and what does one cost? Every disc still works. The original RVL-001 model plays Wii discs and GameCube discs with no server involved, which covers Wii Sports, Mario Kart Wii, Super Mario Galaxy and the rest of the first-party library. The later Wii Family Edition and Wii Mini dropped GameCube support, so the model number matters when you buy. Online play has a fan replacement. Wiimmfi revives matchmaking for games that used Nintendo Wi-Fi Connection, and WiiLink, which absorbed the RiiConnect24 project in December 2023, brings back the WiiConnect24 channels including the Forecast and News Channels. Both need a softmodded console and an SD card rather than anything from Nintendo. ### Where to find one today Used Wii hardware is cheap because Nintendo built more than 100 million of them. A working console with a controller and cables typically runs $40 to $75 in 2026, and a complete boxed bundle with Wii Sports reaches $80 to $120. Stone Age Gamer (https://stoneagegamer.com/nintendo/wii/) carries Wii accessories, replacement parts and controllers, which is usually what fails first. ## Frequently Asked Questions ### Does the Nintendo Wii still work in 2026? Yes. A Nintendo Wii still boots and plays every game disc in its library in 2026, because disc-based games never needed Nintendo's servers. What no longer works is online play, which Nintendo ended on May 20, 2014, and the WiiConnect24 channels, which closed on June 27, 2013. ### Can you still buy games on the Wii Shop Channel? No. The Wii Shop Channel stopped selling WiiWare and Virtual Console games on January 30, 2019, and points cards stopped being redeemable in March 2018. Owners can still redownload titles they bought before that date, but Nintendo has warned that the redownload service will end at an unannounced future date. ### How much is a used Nintendo Wii worth in 2026? A working Nintendo Wii with a controller and cables typically sells for $40 to $75 in 2026, and a complete boxed console bundled with Wii Sports reaches $80 to $120. Prices stay low because Nintendo shipped 101.63 million Wii consoles between 2006 and 2013, making it one of the most common used consoles on the market. **Sources:** - Nintendo Support: Wii Shop Channel Discontinuation: https://en-americas-support.nintendo.com/app/answers/detail/a_id/27560/~/wii-shop-channel-discontinuation - Nintendo IR: Dedicated Video Game Sales Units: https://www.nintendo.co.jp/ir/en/finance/hard_soft/index.html - Game Developer: Satoru Iwata's 'Heart of a Gamer' keynote at GDC 2005: https://www.gamedeveloper.com/business/video-satoru-iwata-s-heart-of-a-gamer-keynote-at-gdc-2005 - Video Games Chronicle: Nintendo ends Wii repair support after 13 years: https://www.videogameschronicle.com/news/nintendo-ends-wii-repair-support-after-13-years/ - Engadget: Nintendo's Wii Shop Channel shuts down today: https://www.engadget.com/2019-01-30-wii-shop-channel-shuts-down.html --- # Do They Still Make Netbooks? Why the Eee PC Died URL: https://404memoryfound.com/posts/what-happened-to-netbooks-laptops.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-03 Updated: 2026-09-10 Topics: Hardware, Business Blunders **Summary:** Nobody makes netbooks in 2026. The category started with the Asus Eee PC in October 2007 at United States prices of $300 to $400, peaked at 38.8 million units shipped in 2010, and ended in 2012 when Asus and Acer stopped building them. Chromebooks, low-cost Windows 11 laptops and tablets took the job, with budget Chromebooks selling under $200 in 2026. **Key facts:** - Launched: October 2007, Asus Eee PC, $300 to $400 in the US - Peak year: 2010, 38.8 million netbooks shipped (ABI Research) - Status today: Dead category; no major brand sells a netbook in 2026 - What replaced it: Chromebooks, cheap Windows 11 laptops, tablets - Cheapest replacement: Chromebooks under $200 (PCWorld, August 2026) ## What replaced netbooks, and what do those cost in 2026? No major brand sells a machine called a netbook in 2026. The word left the shelves with the last Asus Eee PC, and three product categories absorbed everything netbooks used to do. Chromebooks took the cheap laptop slot. PCWorld's Chromebook guide, updated August 28, 2026, puts mid-range models in the "$400 to $600 range" and picks the Asus Chromebook CX15 for anyone who needs a laptop "for under $200". PCWorld associate editor Ashley Biancuzzo calls that machine "extremely inexpensive" in the same 2026 guide. Low-cost Windows 11 laptops took the second slot. They ship with full-size screens and the same class of budget processor a netbook used, but without the licensing cap that once dictated how much screen and memory a cheap Windows machine was allowed to have. Tablets took the third. Apple's iPad went on sale in 2010 and sold 14.8 million units that first year, according to ABI Research figures reported by The Register in 2011. Reading, video and email, which is most of what people actually did on a netbook, moved to a tablet or a phone. ## What did the Asus Eee PC actually launch at in 2007? The category began with one machine. Asus set United States pricing for the Eee PC at the end of October 2007, and Engadget published the lineup on October 29, 2007: three models at $300, $350 and $400. The top model, the Eee PC 4G, went on sale first, on November 1, 2007. The heavily promoted $199 version was missing from that US price list. Laptop Magazine reviewed the 4G on October 25, 2007 and gave it four stars out of five with an Editors' Choice award. It ran Linux rather than Windows, weighed about two pounds, and had a 7-inch screen, small enough that reviewers kept reaching for book comparisons instead of laptop ones. That price is the whole story of why it sold. By Laptop Magazine's own account in October 2007, a weekend circular was advertising a full-size Windows Vista laptop with a 14.1-inch screen for $399, so the Eee PC was asking laptop money for something a fifth of the weight. Anyone who wanted a small computer before that was carrying a BlackBerry (https://404memoryfound.com/posts/what-happened-to-blackberry-smartphone.html) or the last of the PDAs like the Palm Pilot (https://404memoryfound.com/posts/what-happened-to-palm-pilot-pda.html), and neither one had a real keyboard and a desktop browser. ## Why did Microsoft keep Windows XP alive for netbooks? Netbooks landed while Microsoft was pushing Windows Vista, and Vista would not run on hardware that cheap. Linux shipped on the first Eee PC instead. That is the part that worried Microsoft: an entire new PC category selling well with no Windows license attached to it. So Microsoft extended Windows XP (https://404memoryfound.com/posts/windows-xp-wallpaper-bliss-story.html) for what it called ultra low-cost PCs, and fenced the offer in. Engadget reported the terms on May 10, 2008. A machine qualified only with a screen no larger than 10.2 inches, no more than 80GB of storage, no more than 1GB of RAM, a processor no faster than 1GHz, and no touchscreen. Engadget put the license itself at $26 in developing nations and $32 everywhere else. Those limits existed to stop cheap XP machines from eating Vista sales, and they froze the netbook in amber. A netbook could not grow a larger screen or more memory without losing its cheap copy of Windows, so the category shipped near-identical machines for two years while ordinary laptops kept getting better and cheaper. ## How big did the netbook boom get before it broke? Very big, very fast. IDC counted 4.5 million netbooks shipped in the first quarter of 2009, a sevenfold jump on the same quarter of 2008 and 8 percent of all PC shipments worldwide, PCWorld reported on May 3, 2009. IDC expected 22 million for the full year, double the 2008 total. IDC's definition explains what was being counted: laptops with screens between 7 and 12 inches running low-power processors such as Intel's Atom. Atom was built for exactly this, a small, slow, power-sipping x86 chip that made a $300 laptop possible and also guaranteed it would never feel fast. Acer chief executive Gianfranco Lanci told the same story from the factory floor, expecting to ship 10 million to 12 million netbooks in 2009 after just over 5 million in 2008. IDC research analyst Jay Chou explained the appeal to PCWorld in 2009: "Vendors are waking up to the fact that people respond to so-called 'good-enough' computing." The high-water mark came the next year. ABI Research put worldwide netbook shipments at 38.8 million units in 2010, roughly 10 percent above 2009, as reported by The Register in October 2011. That was the top, and nothing after it went up. ## What killed netbooks: the iPad, the Chromebook, or Intel? All three, and the fall took about two years. After that 38.8 million peak in 2010, ABI Research was forecasting only 32 million netbooks for 2011. Shipments went from 8.4 million units in the first quarter of 2011 to 7.3 million in the second, a 13.1 percent drop in three months, The Register reported that October. Tablets passed netbooks in that same quarter. ABI counted 13.6 million tablets shipped in the second quarter of 2011 against those 7.3 million netbooks. Jeff Orr, ABI's consumer research chief, forecast that netbook shipments would "fall year on year from 2011 through 2016". Intel felt it on the income statement. Atom revenue fell 32 percent year on year to $269 million in the third quarter of 2011, The Register noted in the same piece. The chip that defined the category was now shrinking with it. Google applied the other squeeze. Chromebooks went on sale in the United States in June 2011, announced at Google I/O the previous month, as Engadget covered in May 2011. They used the same low-power Intel parts but ran a browser-first operating system with no Windows license to buy and no 10.2-inch ceiling to respect. The idea netbooks had promised, a cheap laptop for web work, shipped again without the handicap. The end was administrative rather than dramatic. The Register reported in December 2012 that Asus and Acer, the last two companies still building netbooks, were finished with them. HP, Samsung, Sony and Toshiba had already left. Five years after the Eee PC, nothing was left to buy. Where to find one today. Netbooks were treated as disposable and most went to recycling, but early Eee PC units still turn up with collectors, usually with a dead battery and a storage drive too small for any current operating system. Sellers list them among Etsy's Asus Eee PC listings (https://www.etsy.com/search?q=asus+eee+pc). ## Frequently Asked Questions ### Why did netbooks fail? Netbooks failed because the rules that made them cheap also stopped them from improving. Microsoft's ultra low-cost PC terms for Windows XP, reported by Engadget in May 2008, capped a qualifying netbook at a 10.2-inch screen, 1GB of RAM and a 1GHz processor. Once tablets and full-size budget laptops caught up in 2011, buyers had no reason to keep accepting the compromise. ### What happened to the Asus Eee PC? The Asus Eee PC created the netbook category in October 2007 at United States prices of $300 to $400, per Engadget's October 29, 2007 pricing report. Asus sold the line for five years and then retired it. The Register reported in December 2012 that Asus and Acer, the last two netbook makers, had both stopped production, and no new Eee PC has been sold since. ### Is a netbook the same as a Chromebook? No. A netbook was a small Windows XP or Linux laptop sold between 2007 and 2012 with a 7-inch to 12-inch screen and a low-power Intel Atom processor, which is how IDC defined the category in 2009. A Chromebook runs Google's ChromeOS, first went on sale in June 2011, and comes in normal laptop sizes, with budget models under $200 and mid-range models in the $400 to $600 range as of PCWorld's August 2026 guide. **Sources:** - Engadget: Asus lets loose US pricing, launch details for Eee PC (October 29, 2007): https://www.engadget.com/2007-10-29-asus-lets-loose-us-pricing-launch-details-for-eee-pc.html - Engadget: Microsoft's XP for low-cost PCs defines some boundaries (May 10, 2008): https://www.engadget.com/2008-05-10-microsofts-xp-for-low-cost-pcs-defines-some-boundaries.html - PCWorld: Netbook Shipments Jump in First Quarter, IDC data (May 3, 2009): https://www.pcworld.com/article/528856/netbook_shipments_jump.html - The Register: Netbook shipments slump, ABI Research data (October 21, 2011): https://www.theregister.com/2011/10/21/netbook_shipments_slump/ - PCWorld: Best Chromebooks 2026 (updated August 28, 2026): https://www.pcworld.com/article/608636/best-chromebooks.html --- # Is the PS2 Still the Best-Selling Console in 2026? URL: https://404memoryfound.com/posts/playstation-2-best-selling-console-ever.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-02 Updated: 2026-09-19 Topics: Gaming, Hardware **Summary:** The PlayStation 2 is still the best-selling console ever made, with 160 million units sold, a total Sony confirmed in November 2024. Sony stopped building the console in January 2013, and the closest challenger is the Nintendo Switch at 155.92 million units by mid-2026. You can still play PS2 games in 2026, but a PlayStation 5 will not read a PS2 disc. **Key facts:** - Launched: March 4, 2000 in Japan; October 26, 2000 in North America at $299 - Units sold: 160 million, confirmed by Sony in November 2024 - Status today: Discontinued in January 2013; still the best-selling console ever - Closest challenger: Nintendo Switch, 155.92 million units by mid-2026 - Price today: $75 to $150 for a tested used console in 2026 ## Why did Sony ship only half the consoles it promised? The PlayStation 2 went on sale in Japan on March 4, 2000 at 39,800 yen, and in North America on October 26, 2000 at $299. The American launch was meant to be the largest in console history, and Sony had told retailers to expect one million units. In late September 2000, that allocation was cut to 500,000. The bottleneck was the Graphics Synthesizer chip, which Sony was moving to a smaller die and could not turn out in working volume fast enough. Retailers canceled preorders they had already taken. Stores were cleaned out on launch day. Resale prices on auction sites ran past $1,000 for a machine that listed at $299, and Sony spent the holidays pushing roughly 100,000 units a week into North America. Sony Computer Entertainment America announced in January 2001 that it had finally gotten more than a million consoles onto North American shelves by the end of 2000, which was the original launch target hit about two months late. ## Why did Japan put a game console under export control? In April 2000, Japan's Ministry of International Trade and Industry placed the PlayStation 2 under the Foreign Exchange and Foreign Trade Control Law, the regime that covers dual-use technology. The worry was the console's floating-point performance, which officials considered adaptable to missile guidance. The rule bit in a strange place. Approval was required for controlled goods worth more than 50,000 yen, and a PlayStation 2 cost 39,800 yen, so one console could leave the country in a suitcase and two could not. The military commentator Kensuke Ebata told the Asahi Shimbun in 2000 that "there are so many items that have technology for civilians that can also be used for military purposes," and counted the console among them. A Sony Computer Entertainment official told the same newspaper, "We have mixed feelings because our efforts to produce a game console of the highest quality have resulted in legal restrictions." Demand never noticed, and no other home console has been treated as arms-control paperwork. ## Did people buy the PS2 as a DVD player? Plenty did. In 2000 a standalone DVD player still sold for $400 or more, and the PlayStation 2 played DVDs out of the box at $299. For many households in the United States it became the first DVD player in the building, and the games came along with it. That was the design, not an accident. Ken Kutaragi, who ran Sony's hardware program, had built the machine as a media hub rather than a toy. "You can communicate to a new cybercity. This will be the ideal home server," he told Newsweek in 2000, in the same interview where he reached for a comparison to The Matrix. The first PlayStation had run a smaller version of the same play, arriving as a CD machine when cartridges were the standard. The DVD move landed harder because the format was young, dedicated players were expensive, and Sony owned a film studio with a direct interest in getting disc players into living rooms. A console that doubled as the cheapest way into a new video format reached buyers who were not shopping for a console at all. ## How badly did the Dreamcast, GameCube and Xbox lose? Badly. The Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html) Dreamcast launched first, in September 1999, and sold 9.13 million units before Sega discontinued it in March 2001 and left the hardware business for good (https://404memoryfound.com/posts/why-sega-stopped-making-consoles-dreamcast.html). Nintendo's GameCube (https://404memoryfound.com/posts/why-nintendo-gamecube-lost-console-war.html), released in 2001, finished at about 21.7 million. Microsoft's original Xbox (https://404memoryfound.com/posts/what-happened-to-original-xbox-microsoft-gaming.html), also a 2001 machine, sold about 24 million. All three together account for roughly 55 million units against the PlayStation 2's 160 million. The install base then fed itself. No publisher could justify skipping a platform that large, so the PlayStation 2 ended up with more than 4,000 released titles, the deepest library any console has had. Guinness World Records lists Grand Theft Auto: San Andreas as the system's best seller at 17.33 million copies. Software outlived the hardware. Pro Evolution Soccer 2014, the last new retail game for the system, shipped on November 8, 2013, ten months after Sony had stopped building consoles. ## Can you still play a PS2 in 2026? Yes, with two caveats. Online play is finished: PlayStation 2 multiplayer ran on each publisher's own servers rather than one central service, and the last official server, for Final Fantasy XI, closed on March 31, 2016, with Sony's DNAS authentication system following on April 4, 2016. Any game that had to sign in through DNAS went offline that week. The second caveat is the disc drive on newer hardware. A PlayStation 5 will not read a PlayStation 2 disc. Sony's official route is the Classics Catalog in PlayStation Plus Premium, which runs a curated set of emulated PS2 titles with save states, rewind and trophy support, and Sony was still adding and reworking titles there through 2026. That catalog holds dozens of games, not thousands. Where to find one today: a tested console runs about $75 to $150 in 2026, and boxed slimline units go past $200, according to 2026 price guides. Buy from a shop that tests hardware and says so in writing, because the laser and the disc tray are what fail first on a machine this old. Stone Age Gamer (https://stoneagegamer.com/) stocks tested consoles, controllers, memory cards and cables. Budget for a component or RGB cable too, since composite video on a modern television is the usual reason an old console looks worse than anyone remembers. ## Is the Nintendo Switch about to take the record? It is close enough to matter. Sony's public count sat near 155 million for years, the figure from fiscal reporting that ended in March 2012, until November 26, 2024, when Sony updated its PlayStation history timeline to 160 million and accounted for the consoles built and sold before production ended in January 2013. The Nintendo Switch reached 155.92 million units by mid-2026, about 4.08 million short, and it is still on sale alongside its successor. Nintendo Life reported in May 2026 that price increases could slow the Switch down before it closes the gap. What cannot repeat is the situation the PlayStation 2 walked into. In 2001 a console was how most people played games at home, there was no serious mobile market, and one machine could take nearly the whole mainstream audience. Attention now splits across PC, phones, subscriptions and two console families at once. Whoever takes the number eventually, nothing will win a generation by that margin again. ## Frequently Asked Questions ### How many PS2 consoles were sold? Sony sold 160 million PlayStation 2 consoles between the launch in March 2000 and the end of production in January 2013. Sony left its public count at roughly 155 million for over a decade and only confirmed the final 160 million total in November 2024, a number that keeps the PlayStation 2 the best-selling console ever made. ### When did Sony stop making the PS2? Sony ended PlayStation 2 production in January 2013, almost 13 years after the console went on sale in Japan on March 4, 2000. The last new retail game for the system, Pro Evolution Soccer 2014, arrived on November 8, 2013, ten months after the final consoles were built. ### Can you play PS2 games on a PS5? A PlayStation 5 cannot read a PlayStation 2 disc. The only official route in 2026 is the Classics Catalog in PlayStation Plus Premium, which offers a curated set of emulated PS2 titles with save states and trophies. Anything outside that catalog needs original PlayStation 2 hardware. **Sources:** - Sony Interactive Entertainment: expanded company timeline: https://sonyinteractive.com/en/our-company/expanded-company-timeline/ - CNN, PlayStation 2 makes its North American debut (October 26, 2000): https://www.cnn.com/2000/TECH/computing/10/26/ps2.main.story/ - Newsweek, Here Comes PlayStation 2 (2000): https://www.newsweek.com/here-comes-playstation-2-156589 - Deseret News, Export limits put on video game (April 17, 2000): https://www.deseret.com/2000/4/17/19502394/export-limits-put-on-video-game-br-playstation2-could-be-used-for-weaponry/ - VGChartz, Switch vs PS2 sales comparison (July 2026): https://www.vgchartz.com/article/468794/switch-vs-ps2-sales-comparison-july-2026/ --- # Who Owns CompuServe Now? Yahoo Runs the Portal URL: https://404memoryfound.com/posts/what-happened-to-compuserve-online-service.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-02 Updated: 2026-09-19 Topics: Internet Culture, Software & Apps **Summary:** CompuServe still exists in 2026, but only as compuserve.com, a news portal operated under an AOL Media copyright and owned by Yahoo, which funds managed by Apollo Global Management bought in September 2021. The original dial-up service shut down on June 30, 2009, and the last CompuServe forums closed on December 15, 2017. What outlived the company is the GIF, released by CompuServe in 1987. **Key facts:** - Status today: No online service; compuserve.com runs as a news portal - Owner today: Yahoo, owned by Apollo Global Management funds since September 2021 - Founded: 1969, Columbus, Ohio, as a Golden United Life Insurance subsidiary - Peak subscribers: 3 million, April 1995 - Shut down: Dial-up service June 30, 2009; forums December 15, 2017 ## Is CompuServe still in business in 2026? Not as an online service. CompuServe exists in 2026 only as a domain. Type compuserve.com and you get a news portal with world, business, sports and entertainment sections, refreshed from wire copy. The ownership chain runs through four companies. H&R Block bought CompuServe in 1980 and sold it in February 1998. AOL took the consumer service, Verizon bought AOL in 2015 and Yahoo in 2017, and funds managed by Apollo Global Management completed their purchase of Yahoo in September 2021. CompuServe sits inside that Yahoo business today, and the portal's pages carry an AOL Media LLC copyright. The pages are not even CompuServe-branded. In 2026 the news section at compuserve.com/news loads under the page title "Netscape News", because it runs on the template built for Netscape (https://404memoryfound.com/posts/is-netscape-still-around.html)'s leftover web properties. Email is the one thread of continuity. Addresses ending in @compuserve.com still deliver, but they sit on AOL's mail servers, and the people who hold them sign in through the same login page that serves AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html) mail. ## How did an insurance company build the first online service? CompuServe was incorporated in 1969 in Columbus, Ohio, as Compu-Serv Network, a subsidiary of Golden United Life Insurance. The Ohio historical marker erected at its old headquarters in 2024 calls it "the first major online information services provider." John R. Goltz was its first president. Jeffrey Wilkins, son-in-law of Golden United founder Harry Gard Sr., replaced him as chief executive inside the first year and ran the company through its consumer expansion. For a decade the business was timesharing: renting mainframe cycles to corporations that could not justify buying their own machines. At night those mainframes sat idle, and the idle capacity became the consumer product. CompuServe launched it on September 24, 1979, as MicroNET, sold through Radio Shack (https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html) at $5 an hour plus the cost of the phone call. What subscribers got was a text menu. Special Interest Groups, shortened to SIGs, held message boards and file libraries on everything from ham radio to programming languages. On February 21, 1980, CompuServe released CB Simulator, written by company executive Sandy Trevor and generally recorded as the first public, commercial multi-user chat program. ## What did an hour on CompuServe actually cost? It was metered, and metered hard. On the early 1980s evening rate, CompuServe charged $6 an hour at 300 baud and $12 an hour at 1200 baud. Daytime rates doubled, because the mainframes were still running insurance work and the pricing was set to keep consumers off them. At 300 baud, roughly 30 characters a second, $6 an hour bought text that arrived slower than a person reads. A 400K download took about 40 minutes and cost about $8 at the evening rate, which is why file libraries were browsed carefully rather than grazed. That user base skewed to people who could justify the bill: engineers, lawyers, doctors and programmers. It was a professional service that happened to have chat rooms. Prices then collapsed under competition. Through the early 1990s CompuServe's hourly rate fell from over $10 an hour to $1.95 an hour, and AOL moved the whole market to a flat monthly fee, which a metered service could not answer. H&R Block, which had bought CompuServe in 1980, spent those years advertising in computer magazines and bundling trial accounts with modems. ## Why does every GIF trace back to Columbus, Ohio? On June 15, 1987, CompuServe released the Graphics Interchange Format. The engineering lead was Steve Wilhite, and the problem he was solving was money as much as bandwidth: subscribers paying by the hour needed color images that did not take twenty minutes to arrive. GIF used Lempel-Ziv-Welch compression, supported 256 colors, and rendered identically on the Apple II, the Commodore (https://404memoryfound.com/posts/what-happened-to-commodore-computers.html) 64 and the IBM PC, which mattered on a service whose subscribers ran all three. Animation was in the specification from the start, although almost nothing used it for another decade. Wilhite spent the rest of his life defending the pronunciation. After collecting a Webby lifetime achievement award in 2013 he told The New York Times, "The Oxford English Dictionary accepts both pronunciations. They are wrong. It is a soft g." He died in 2022 at the age of 74. The Ohio marker singles the format out among everything the company shipped: "One especially successful CompuServe innovation was the GIF graphics format." It is the only CompuServe invention still in daily use worldwide. ## How did AOL end up with 2.6 million CompuServe subscribers? By 1994 the American online market had settled into three services: CompuServe, Prodigy (https://404memoryfound.com/posts/what-happened-to-prodigy-online-service.html) and America Online. CompuServe had the oldest and deepest content. AOL had flat-rate pricing and a mailing list, and that turned out to be the larger market. The open web finished the argument. Once Netscape Navigator (https://404memoryfound.com/posts/how-netscape-lost-the-browser-war.html) put the internet in front of ordinary buyers in late 1994, a walled service billing by the hour had to explain why its forums were worth more than an unmetered web. CompuServe bolted internet access onto its own client, slowly, and kept losing money. The exit, announced in September 1997 and closed in February 1998, was a three-way trade rather than a straight sale. WorldCom bought all of CompuServe for $1.2 billion in stock and kept the network infrastructure, which was what it wanted. It then handed the CompuServe Information Service and its 2.6 million consumer subscribers, along with $175 million in cash, to AOL, and took AOL's ANS Communications business-access division in exchange. Bertelsmann paid AOL a further $75 million to keep its half of their European joint venture. H&R Block sold its WorldCom stock on February 2, 1998, and reported net proceeds of $1,032,699,000 from the sale in its annual report that year. ## When did the CompuServe forums finally go dark? In two stages, eight years apart. AOL kept the original dial-up product running as CompuServe Classic until June 30, 2009, when it stopped operating as an internet service provider after thirty years. The remaining subscribers were pushed toward CompuServe 2000. The forums outlived the service that created them. They ran on, thinly populated but genuinely active, until November 14, 2017, when a notice appeared at the top of every board: "We regret to inform you that the Forums will be removed from the CompuServe service effective December 15, 2017." The owner at that point was Oath, the Verizon unit that held AOL and Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html). The same notice said the rest of the site, meaning news, weather and webmail, would carry on as normal. That is exactly what happened, and it is why the domain is still serving headlines in 2026 while the community that made the name is gone. The Columbus headquarters at 5000 Arlington Centre Boulevard housed the company from 1973 to 2009. In 2024 the Ohio History Connection and Leadership Upper Arlington put a marker on the site, which is now the closest thing CompuServe has to a monument. ## Frequently Asked Questions ### Is CompuServe still around in 2026? CompuServe exists in 2026 only as compuserve.com, a news portal. The CompuServe online service stopped operating as an internet provider on June 30, 2009, and its forums were removed on December 15, 2017. Email addresses ending in @compuserve.com still work, handled on AOL's mail servers. ### Who owns CompuServe now? CompuServe belongs to Yahoo. AOL acquired the CompuServe service from WorldCom in February 1998, Verizon bought AOL in 2015 and folded it together with Yahoo, and funds managed by Apollo Global Management completed their acquisition of the combined Yahoo business in September 2021. The compuserve.com portal carries an AOL Media LLC copyright. ### Did CompuServe really invent the GIF? Yes. CompuServe released the Graphics Interchange Format on June 15, 1987, with engineer Steve Wilhite leading the work, so that subscribers paying by the hour could download color images quickly. GIF is the only CompuServe invention still in general use, and Wilhite died in 2022 at the age of 74. **Sources:** - WorldCom is buying CompuServe (Associated Press, Deseret News, 1997): https://www.deseret.com/1997/9/8/19332889/worldcom-is-buying-compuserve/ - H&R Block Inc., Form 10-K for fiscal 1998 (SEC EDGAR): https://www.sec.gov/Archives/edgar/data/0000012659/000095012498004013/0000950124-98-004013.txt - Remember CompuServe forums? They're about to die (The Register, 2017): https://www.theregister.com/2017/11/15/compuserve_to_kill_off_its_forums/ - CompuServe World Headquarters Ohio Historical Marker (Ohio History Connection): https://www.hmdb.org/m.asp?m=246815 - Apollo Funds Complete Acquisition of Yahoo (Apollo Global Management, 2021): https://www.apollo.com/insights-news/pressreleases/2021/09/apollo-funds-complete-acquisition-of-yahoo-161530593 --- # Who Owns Gateway Computers Now? Acer Still Does URL: https://404memoryfound.com/posts/what-happened-to-gateway-2000-cow-box-computer.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-02 Updated: 2026-09-19 Topics: Hardware, Business Blunders **Summary:** Gateway computers are still sold in 2026, but Gateway has not been a company since 2007, when Acer bought it for $710 million. Acer owns the trademark, and since September 2020 the Gateway name has appeared on budget laptops and tablets sold exclusively at Walmart. Every Gateway Country Store closed on April 9, 2004. **Key facts:** - Founded: September 5, 1985, on a cattle ranch near Sioux City, Iowa - Owner today: Acer Inc., which bought Gateway in 2007 for $710 million - Status today: A brand only: Walmart-exclusive laptops and tablets since September 2020 - Stores left: Zero. The last 188 Gateway Country Stores closed April 9, 2004 - Peak: $1.11 billion in revenue in 1992, about 25,000 employees in 2000 ## How did two guys on an Iowa cattle ranch start a PC company? Gateway 2000 opened for business on September 5, 1985, in a farmhouse on the Waitt family cattle ranch outside Sioux City, Iowa. Ted Waitt was 22. He started the company with his brother Norm Waitt Jr. and his friend Mike Hammond, on a $10,000 bank loan that his grandmother, Mildred Smith, backed with her own certificate of deposit. The pitch was plain. Skip the computer store, call a salesperson, describe what you needed, and the machine got built to that order and shipped to your door. In its first four months the company grossed about $100,000. By the end of 1986 it had passed $1 million in revenue and had traded its original name, the TIPC Network, for Gateway 2000. Being nowhere near Silicon Valley was part of the math. Space in Iowa was cheap, payroll was cheaper, and the savings showed up in the price quoted over the phone. Revenue went from $12 million in 1988 to $70.6 million in 1989, $275 million in 1990 and $626 million in 1991. In 1992 it hit $1.11 billion, and Gateway 2000 was being written up as the fastest-growing company in the United States. In December 1993 the company went public on Nasdaq under the ticker GATE at $15 a share. The Waitt family kept the large majority of the stock. ## Why did Gateway ship its computers in cow-spotted boxes? The black and white Holstein spots went onto the shipping cartons in 1991 and stayed there for the rest of the company's independent life. Ted Waitt did not draw them. "The cow spot was actually developed by a graphic designer. I can't take a claim for that," he told Inc. magazine in 1991. The pattern did two jobs at once. It made Gateway 2000 look like Iowa rather than like Compaq (https://404memoryfound.com/posts/compaq-lost-pc-business-hp-merger.html) or IBM, which mattered when buyers were being asked to send money to a mail-order company they had never heard of and then wait. And two-color printing on corrugated cardboard cost less than full color, so the cheapest packaging Gateway could buy was also the most recognizable. It became the whole brand. Cow-print mugs, notepads, binders and stress toys went out to new owners in welcome kits, and those leftovers now outnumber the surviving computers on collector sites. The spotted box on a porch in 1996 told the neighbors what had just arrived without anyone reading a label. ## What were Gateway Country Stores, and when did they close? Gateway started opening Gateway Country Stores in the late 1990s. They were not stores in the sense that CompUSA (https://404memoryfound.com/posts/is-compusa-still-around.html) was a store. You could not carry a computer out. They were showrooms with demo machines and staff who took orders, and the machine you configured was built at a Gateway factory and shipped to your house. The idea addressed the one real weakness in mail order, which was that plenty of families still wanted to touch a computer before spending two thousand dollars on it. For a few years it worked. The chain peaked at just over 320 locations in 2001. Then the arithmetic turned. The stores carried rent, fixtures and staff at exactly the moment buyers stopped needing convincing that ordering a PC online was safe. Gateway closed about 30 stores early in 2001 and another 80 in 2003. On April 1, 2004 it announced that the remaining 188 stores would close on April 9, cutting roughly 2,500 retail jobs. Every one of them shut on schedule, and Gateway has had no stores of its own since. ## What went wrong for Gateway after 2000? Gateway 2000 dropped the 2000 from its name in October 1998, the same year it moved its headquarters out of North Sioux City, South Dakota to San Diego, California. It moved again in 2001, to Poway. The company that had sold Midwestern plainness as its whole personality was now run from Southern California. The peak was short. Gateway employed about 25,000 people in 2000. The dot-com collapse and a brutal PC price war followed, and the 2001 restructuring cut the workforce from roughly 24,600 to about 14,000. Dell was running the same direct model at greater scale and lower cost, and the PC itself had turned into a commodity where the cheapest acceptable box won. Gateway's answer was to buy its way down-market. On March 11, 2004 it completed the purchase of eMachines, a budget PC maker, for $262 million in cash and stock. eMachines chief executive Wayne Inouye took over as chief executive of Gateway, and Ted Waitt stayed on as chairman. The combination sold more boxes at thinner margins and never restored the company's independence. ## How did Acer end up buying Gateway for $710 million? On August 27, 2007, Acer and Gateway announced that Acer would acquire all outstanding Gateway shares for $1.90 each, a deal worth about $710 million. It was the largest acquisition in Acer's history to that point and it was explicitly about the United States market, where Acer was weak. "This strategic transaction is an important milestone in Acer's long history," J.T. Wang, chairman of Acer, said in the joint press release. Gateway's chief executive framed it as a fit rather than a rescue: "We believe our complementary geographical and product mixes, and our mutual focus on the consumer market makes Acer an outstanding partner for Gateway," Ed Coleman said in the same release. The deal closed in October 2007. Acer spent that year collecting distressed PC brands on both sides of the Atlantic. It took Gateway in the United States and Packard Bell (https://404memoryfound.com/posts/packard-bell-computer-brand-today.html) in Europe, which lifted it to third place among global PC vendors. Gateway's own eMachines line came along with it. ## Is Gateway still in business in 2026? Not as a company. Gateway has been a trademark inside Acer since 2007, with no independent operations, no factories and no stores. Acer let the name go quiet after 2013 and put out no new Gateway products for years. The name came back on September 11, 2020, when Walmart announced a line of Gateway laptops and tablets sold only through Walmart, with tablets starting at $69 and laptops at $179. The cow-spotted packaging came back with them. The hardware is not built by Acer: the brand is licensed to Bmorn Technology, a Shenzhen manufacturer, which makes and sells the machines under the Gateway name. In 2026 Walmart still lists Gateway laptops, mostly entry-level and mid-range AMD and Intel notebooks. Acer handles the trademark, not the assembly line. Where to find one today. The original cow-era machines are collectors' items rather than usable computers, and the promotional items outlast the hardware by a wide margin: cow-print mugs, notepads, welcome-kit binders and the stress-toy cows that shipped to new owners in the late 1990s. Prices swing hard on condition and on whether the original spotted box survived. Etsy is the practical place to look, through its Gateway 2000 listings (https://www.etsy.com/search?q=gateway%202000). If you want to actually run software from that era, emulation on a modern PC is the cheaper route, as covered in running 90s PC games on Windows 11 (https://404memoryfound.com/posts/run-90s-pc-games-windows-11.html). ## Frequently Asked Questions ### Are Gateway laptops still made in 2026? Yes. Gateway-branded laptops and tablets have been sold exclusively at Walmart since September 2020, and Walmart still lists them in 2026. Acer owns the Gateway trademark and licenses it to Bmorn Technology of Shenzhen, which builds and sells the machines, so a Gateway laptop today is not made by the company that shipped cow boxes from Iowa. ### Why did Gateway computers come in cow boxes? Gateway 2000 adopted the black and white Holstein pattern on its shipping cartons in 1991, tying the brand to its origins on a cattle ranch near Sioux City, Iowa. The design was drawn by an outside graphic designer, not by founder Ted Waitt, and two-color printing on cardboard was cheaper than full color, so the packaging that made Gateway recognizable also cost the company less. ### When did Gateway Country Stores close? Gateway announced on April 1, 2004 that its remaining 188 Gateway Country Stores would close on April 9, 2004, eliminating about 2,500 retail jobs. The chain had peaked at just over 320 locations in 2001, with roughly 30 closed early that year and another 80 in 2003. Gateway has operated no retail stores of its own since April 2004. **Sources:** - Acer and Gateway joint press release, August 27, 2007 (SEC EDGAR): https://www.sec.gov/Archives/edgar/data/895812/000119312507190259/dex991.htm - Gateway Inc. Form 10-Q, Q1 2004 (eMachines purchase price): https://www.sec.gov/Archives/edgar/data/0000895812/000119312504083842/d10q.htm - Computerworld: Gateway to close all retail stores April 9 (2004): https://www.computerworld.com/article/1701502/update-gateway-to-close-all-retail-stores-april-9.html - Walmart corporate news: Gateway laptops and tablets launch exclusively at Walmart (September 11, 2020): https://corporate.walmart.com/news/2020/09/11/iconic-gateway-brand-launches-legen-dairy-new-line-of-laptops-tablets-exclusively-at-walmart - Encyclopedia.com: Gateway Inc. company history: https://www.encyclopedia.com/social-sciences-and-law/economics-business-and-labor/businesses-and-occupations/gateway-inc --- # Why Boo.com Failed and Who Owns the Domain Now URL: https://404memoryfound.com/posts/what-happened-to-boo-com-fashion-dot-com.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-02 Updated: 2026-09-19 Topics: Business Blunders, Internet Culture, Money & Tech **Summary:** Boo.com failed because it spent $135 million in 18 months on a fashion site most 1999 shoppers could not load, and it went into receivership on 18 May 2000, six months after launch. The brand and domain sold for about $375,000 to Fashionmall.com, then passed to Web Reservations International, now Hostelworld Group plc, which still holds boo.com. There is no fashion store at the address: the travel site run there closed in October 2010. **Key facts:** - Launched: 3 November 1999, London, in 18 countries at once - Founders: Ernst Malmsten, Kajsa Leander and Patrik Hedelin - Capital burned: About $135 million in 18 months - Status today: Dead since 18 May 2000; no store at boo.com - Owner today: Hostelworld Group plc, formerly Web Reservations International ## What was Boo.com supposed to be? Boo.com was an online store for designer sportswear and street fashion, founded in 1998 by three Swedes: Ernst Malmsten, Kajsa Leander and Patrik Hedelin. They arrived with a record. Malmsten and Leander had started the Swedish online bookstore Bokus.com in 1997 and sold it the following year to the Swedish group KF Media. The plan for Boo.com was to skip the slow country-by-country rollout entirely. The site would open in 18 countries at once, in seven languages, taking payment in 18 currencies, with prices set for each local market. The technology was the pitch. Shoppers could drag a garment onto a 3D model, zoom in and rotate it. A cartoon assistant called Miss Boo offered suggestions while you browsed. Nothing else in 1999 retail looked like it, and nothing else in 1999 retail cost as much to build. Investors bought the story. J.P. Morgan, Goldman Sachs, Bernard Arnault of LVMH and the Benetton family put money in across three rounds totalling roughly $135 million, including a $101 million round in July 1999. Before a single item had been sold, Boo.com had spent about $25 million on advertising and public relations. ## What happened on launch day, 3 November 1999? Boo.com opened on 3 November 1999 after months of missed launch dates. About 50,000 people visited on the first day. Roughly four in every thousand of them placed an order, a conversion rate of 0.25 percent. The reasons were not subtle. The site leaned on JavaScript and Flash, which corporate firewalls blocked and older browsers handled badly. It did not work on Macintosh computers at all, which cut out a large share of the design and media crowd the brand was aimed at. Then there was the bandwidth. The 3D viewer and the Miss Boo animations had to download before a shopper saw anything, and most of the audience in 1999 was on a dial-up modem. Our dial-up speed comparison (https://404memoryfound.com/posts/dialup-speed-then-vs-now.html) puts a 56k line next to a 2026 connection and shows the gap. A store built for broadband, served over a line that slow, is not a store. It is a waiting room. Boo.com kept rebuilding the site after launch. Each rebuild cost money the company was already spending faster than it could raise it. ## How fast did Boo.com burn $135 million? Fast, and in public. By October 1999, a month before the store opened, Boo.com had around 400 employees across eight offices, with staff in Amsterdam, Munich, New York, Paris and Stockholm as well as London. The headquarters and the call centre sat on Carnaby Street in London, one of the most expensive addresses in British retail, rather than somewhere cheaper. That single choice became the shorthand for how the company spent. Through the first and second quarters of 2000, cash was leaving at more than $10 million a month. Revenue never came close to matching it. The founders were funding a global logistics and technology operation out of venture capital while the shop itself converted a quarter of one percent of its visitors. The arithmetic was not unusual for the period. Pets.com (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html) and Webvan (https://404memoryfound.com/posts/what-happened-to-webvan-grocery-delivery.html) ran the same equation in the United States, spending ahead of demand on the assumption that the next round would always arrive. What set Boo.com apart was the speed. The $135 million was gone in 18 months, and the store had been open for only six of them. ## What happened on 18 May 2000? Boo.com ran out of road. Malmsten spent the final weeks trying to raise a bridge round from backers who had already watched the Nasdaq slide through March and April of 2000. "Unless we raise $20 million by midnight, boo.com is dead," Malmsten said on 18 May 2000, in a line recorded in Dave Chaffey's case study of the collapse for Smart Insights. Around half the money was pledged. It was not enough and it was not in time. KPMG was appointed that day to wind the company up. More than 400 staff and contractors lost their jobs, roughly 200 of them at the Carnaby Street office and the rest across Stockholm, Paris, New York and Munich. Many had gone unpaid. The next morning the Financial Times ran the headline "Boo.com collapses as investors refuse funds" and called the company Europe's first big internet casualty. Malmsten wrote his own account, "Boo Hoo: A Dot.com Story from Concept to Catastrophe", published in 2001. "By writing a book, I hoped to come to terms with the overwhelming sadness and emptiness that follows such a great loss," he wrote in it. ## Who owns Boo.com today? Hostelworld Group, the hostel booking company listed in London, holds the boo.com domain. No fashion store has traded at that address since May 2000. The wreckage was split in two at the liquidation. Bright Station, the British technology company run by Dan Wagner, bought Boo.com's e-commerce software from KPMG for 250,000 pounds, about $372,500 at the time. Fashionmall.com, a New York fashion portal, bought the brand, the domain and the Miss Boo character for about $375,000, a figure Forbes filed in May 2000 under the headline "Boo.com Assets Sell for Spooky Price: $375,000". Fashionmall.com ran a thin version of the site for several years, then sold the domain to Web Reservations International, the Dublin company behind Hostelworld. It relaunched boo.com in May 2007 as a travel planning site with reviews and destination listings. In November 2009 the private equity firm Hellman and Friedman bought Web Reservations International outright, and the travel version of Boo.com was shut down in October 2010. An official told PhocusWire the company wanted to "channel its resources" into its existing brands. Web Reservations International became Hostelworld Group plc and listed on the London Stock Exchange on 2 November 2015, valuing the group at 245 million euros. The boo.com name has sat on its books ever since, unused. ## Was Boo.com right about online fashion? About the idea, yes. About the year, no. Two companies founded in 2000, months after Boo.com closed, built what Boo.com had described. ASOS launched in June 2000 selling affordable fashion to young British shoppers. Net-a-Porter launched the same year selling luxury labels. Both started narrow, in one market, with plain product photography and pages that loaded. The difference was not vision. It was sequencing. Boo.com spent $25 million telling the world about a store before the store worked, opened in 18 countries at once, and shipped a 3D viewer the average 1999 connection could not carry. ASOS and Net-a-Porter waited for broadband and spent on inventory and logistics instead. The shape of the failure keeps recurring, which is why the case is still taught. Amazon came within sight of the same cliff (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html) in 2000 and survived because it had revenue and a bond issue behind it. Boo.com had neither. It had a story, $135 million and six months of trading. ## Frequently Asked Questions ### Why did Boo.com fail? Boo.com failed because it spent $135 million in 18 months building a global fashion site that most shoppers in 1999 could not use. The store opened on 3 November 1999 needing JavaScript, Flash and a 3D product viewer, would not run on Macintosh computers, and crawled over dial-up connections. When investors refused a final $20 million, Boo.com went into receivership on 18 May 2000. ### How much money did Boo.com lose? Boo.com raised and spent about $135 million between 1998 and May 2000, backed by J.P. Morgan, Goldman Sachs, Bernard Arnault of LVMH and the Benetton family. Roughly $25 million of it went on advertising and public relations before the site opened. At liquidation the remains sold for under $1 million in total: about $375,000 for the brand and domain, and 250,000 pounds for the technology. ### Is Boo.com still a website? No. Boo.com has not sold clothes since May 2000. The domain passed from Fashionmall.com to Web Reservations International, which ran boo.com as a travel site from May 2007 until October 2010. That company is now Hostelworld Group plc, listed on the London Stock Exchange since November 2015, and it still holds the boo.com name. **Sources:** - Forbes: Boo.com Assets Sell for Spooky Price: $375,000 (2000): https://www.forbes.com/2000/05/30/mu5.html - Smart Insights: Boo.com case study, a classic example of failed e-business strategy: https://www.smartinsights.com/digital-marketing-strategy/online-marketing-mix/boo-com-case-study-a-classic-example-of-failed-ebusiness-strategy/ - PhocusWire: Curse of Boo.com strikes again, travel version closes: https://www.phocuswire.com/Curse-of-Boo-com-strikes-again-travel-version-closes - TechCrunch: Boo.com Got Sold Again! (2009): https://techcrunch.com/2009/11/13/boo-com-got-sold-again/ - Hostelworld Group plc IPO pricing announcement (2015): https://www.hostelworldgroup.com/~/media/Files/H/Hostelworld-v2/reports-and-presentations/hostelworld-group-ipo-pricing-announcement.pdf --- # Motorola Razr V3: 130 Million Sold, No Signal Left URL: https://404memoryfound.com/posts/motorola-razr-v3-coolest-phone-ever-made.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-01 Updated: 2026-09-09 Topics: Hardware, Business Blunders **Summary:** The Motorola Razr V3 cannot be used as a phone in the United States anymore. It was a 2G-only GSM handset, and T-Mobile switched off the last major US 2G network on August 3, 2026, after AT&T dropped 2G in 2017 and Verizon in 2020. The name survives under Lenovo, which sells folding Razrs from $799 to $1,499 in 2026, while original V3s trade second-hand from about $28. **Key facts:** - Launched: November 16, 2004, Motorola, exclusive to Cingular Wireless - Launch price: $500 with a two-year contract (2004) - Units sold: About 130 million across all V3 variants - Status in 2026: Unusable in the US; last 2G network shut off August 3, 2026 - Owner today: Lenovo, which sells folding Razrs from $799 to $1,499 ## Can you still use a Motorola Razr V3 in 2026? No, not as a phone in the United States. The V3 was a 2G quad-band GSM handset with no 3G radio, so it depends on networks that every major US carrier has now switched off. AT&T retired its 2G service in 2017. Verizon shut its 2G network down in 2020. T-Mobile held on the longest, partly for legacy alarm panels and trackers and partly for tourists arriving with old handsets, and then killed the last major US 2G GSM signal shortly after 9 a.m. Eastern on August 3, 2026. That date is the practical end of the line for the original Razr. A V3 in 2026 still powers on, still lights up that etched keypad, still snaps shut with the same sound. It just cannot register on a network, which means no calls, no texts and no 911. Buyers treat it as an object now, not a phone. ## How many Razr V3s did Motorola actually sell? About 130 million, counting every V3 variant over the roughly four years it stayed in production. Motorola never published an audited lifetime total, so that number is a tally assembled after the fact, the one Wikipedia carries and the tech press repeats. Treat it as the accepted figure rather than a company-certified one. The climb is easier to trust because it came in public milestones. Motorola had moved 750,000 units by the end of 2004, weeks after the US launch. The count passed 50 million by July 2006 and reached roughly 100 million by late 2007. For scale, no other clamshell has come close, and in the United States the V3 and its variants were the best-selling handset of 2005, 2006 and 2007. Three straight years at number one was not normal then and has not been repeated often since. ## Why did a $500 flip phone sell like that? Motorola launched the V3 in the US on November 16, 2004, exclusively through Cingular Wireless, at $500 with a two-year contract. That was the most expensive phone in Motorola's range at the time, and the exclusivity meant anyone on another carrier had to wait or switch. What justified the price was the body. The V3 measured about 13 millimeters thick and weighed 95 grams, which contemporary coverage described as a half-inch-thick clamshell in a market full of plastic bricks. The chassis was anodized aluminum. The keypad was a flat etched metal sheet rather than the usual rubber nubs, backlit so the digits glowed through it. The trade-offs were real. The camera was 0.3 megapixels, storage was minimal, the browser was a basic WAP client, and there was no headphone jack, so audio ran through the mini-USB port. None of that mattered to the people buying it. The V3 sold on how it looked and felt, the same way the Nokia 3310 (https://404memoryfound.com/posts/what-happened-to-nokia-3310.html) sold on how much abuse it survived. ## What killed the original Razr? Success, mostly. Motorola spent the back half of the decade re-releasing the same phone. The V3i added a better camera and a memory card slot. The V3x and V3xx bolted on 3G. The Razr2 arrived in 2007 as the actual sequel, three years after the original, with the same hinge and the same silhouette. While that was happening, the rest of the industry moved to software. BlackBerry (https://404memoryfound.com/posts/what-happened-to-blackberry-smartphone.html) owned corporate email. Nokia (https://404memoryfound.com/posts/does-nokia-still-make-phones.html) pushed Symbian. Apple shipped the iPhone in 2007, and the iPhone 3G took the US sales crown from the Razr line in the second half of 2008. The V3 was discontinued that same year. Motorola had built the most popular phone in America and had nothing queued behind it, and the phone division spent the following years shrinking. The company whose flip phone had been as much of a status object as the iPod (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html) became an also-ran in a market Apple and Samsung were splitting between them. ## Is the Razr back? The foldables, 2019 to 2026 The name came back, under different ownership. Motorola is a Lenovo subsidiary now, and in November 2019 the Lenovo-owned company revived the Razr as a vertically folding smartphone, priced at $1,499 and sold in the US through Verizon, with pre-orders opening December 26, 2019. Prices fell as the folding hinge got cheaper to build. By 2025 the US lineup was three phones: the Razr at $699, the Razr+ at $999 and the Razr Ultra at $1,299. Then they went back up. The 2026 lineup, announced April 29, 2026, with pre-orders on May 14 and shipping May 21, moved to $799 for the Razr, $1,099 for the Razr+ and $1,499 for the Razr Ultra. The top model costs exactly what the first foldable Razr cost in 2019. Reviewers noted that the 2026 phones are close to spec-identical to the 2025 ones, with higher price tags attached. ## Where to find one today Original V3s are everywhere on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html).com and they are cheap. Checked in September 2026, listings ran from roughly $28 for a plain used handset to a few hundred dollars, with pristine and boxed examples near the top of that range and one refurbished unit in as-new condition asking $33.50 plus shipping. Two things to watch. Many listings are described as seller refurbished, which usually means a replacement outer shell or a fresh battery rather than an untouched 2004 unit, so read the description if originality matters to you. And nothing you buy will make a call in the US, whatever the listing says about being unlocked, because the networks it needs are gone. Complete boxes with the mini-USB charger and manual carry most of the premium. A loose handset with a tired battery is a $30 desk ornament, and for a lot of buyers that is exactly the point. ## Frequently Asked Questions ### Does the Motorola Razr V3 still work in 2026? It powers on and the hardware works, but it cannot connect to a US carrier. The V3 is 2G GSM only, and the last major US 2G network, T-Mobile's, was shut off on August 3, 2026, after AT&T dropped 2G in 2017 and Verizon in 2020. ### How much is a Motorola Razr V3 worth today? Used V3s on eBay.com started around $28 in September 2026, and clean or boxed examples ran into the low hundreds. Condition and completeness set the price, since none of them work as phones anymore. ### Who owns Motorola now? Lenovo. The Chinese computer maker owns Motorola's phone business and sells the modern folding Razr line, which in 2026 runs from $799 for the base Razr to $1,499 for the Razr Ultra. **Sources:** - NBC News: Motorola's hyped Razr released via Cingular (November 2004): https://www.nbcnews.com/id/wbna6504736 - Wikipedia: Motorola Razr V3: https://en.wikipedia.org/wiki/Motorola_Razr_V3 - Fortune: Motorola Razr foldable phone price and release date (November 13, 2019): https://fortune.com/2019/11/13/motorola-razr-folding-phone-price-preorder-date-release-date/ - 9to5Google: Motorola Razr (2026) specs, price and release date: https://9to5google.com/2026/04/29/motorola-razr-2026-specs-price-release-date/ - TechEBlog: T-Mobile turns off the last major US 2G network (August 2026): https://www.techeblog.com/t-mobile-turns-off-2g-network/ --- # Why Google Reader Died and What Replaced It in 2026 URL: https://404memoryfound.com/posts/what-happened-to-google-reader-rss.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-04-01 Updated: 2026-09-18 Topics: Software & Apps, Internet Culture **Summary:** Google Reader is gone. Google retired the RSS reader on July 1, 2013, less than eight years after launching it through Google Labs, and has never brought it back. Feedly, Inoreader, NewsBlur and NetNewsWire absorbed the readers who stayed with RSS, and Feedly's paid tier runs about $6 a month in 2026. **Key facts:** - Launched: October 7, 2005, through Google Labs - Built by: Chris Wetherell, from a prototype codenamed Fusion - Shut down: July 1, 2013, data deleted July 15, 2013 - Status today: Gone. No revival announced as of September 2026 - Price today: Feedly free for 100 sources, Pro about $6 a month annually ## How did Google Reader start, and who built it? Google Reader came out of Google's 20 percent time, the policy that let engineers spend a day a week on their own projects. The engineer behind it was Chris Wetherell, who had been building feed tools around Blogger since the early 2000s. The prototype was codenamed Fusion. The team floated names like Transmogrifier and Reactor before landing on Reader, and Google shipped it publicly on October 7, 2005, through Google Labs, the company's holding pen for experiments. The first day did not go smoothly: the site crashed under its own launch traffic. What survived that launch was plain by design. White background, no trending panel, no recommendation engine. You added a site's RSS feed and everything that site published showed up in order, newest first. Keyboard shortcuts let heavy users move through hundreds of items without a mouse: J and K to move between articles, S to star, V to open the original page. That was the whole product, and it was enough. Within a couple of years Reader had pulled the RSS audience away from Bloglines and NewsGator and become the default way to read the web on the web's own terms. ## What made Google Reader's shared items so hard to replace? In 2007 Google added shared items. One click pushed an article into a public feed that your contacts could subscribe to. There were no likes, no follower counts and no reshare metrics attached to it. The result was a recommendation system made of people rather than math. Instead of leaning on a link aggregator like Digg (https://404memoryfound.com/posts/what-happened-to-digg-social-news.html), readers got their next article from a friend who happened to read closely in one narrow field. The volume was low and the hit rate was high. It also picked up a use Google never planned for. In Iran, where the government blocked news sites one at a time, Reader's shared items circulated material the state media would not carry. Blocking Reader meant touching the same Google login that Gmail and Search ran on, which made it an awkward target. That is the feature people still ask about, and it is the one no replacement has matched. Feedly, Inoreader and NewsBlur all read feeds well. None of them carry the same network of readers who were already inside the product. ## How did Google+ turn Reader into a maintenance project? In October 2011 Google rebuilt Reader's interface and stripped the sharing features out of it. The shared items page went away. Notes went away. A +1 button pointing at Google+ (https://404memoryfound.com/posts/why-google-plus-actually-failed.html) took their place, which meant the networks people had spent four years building inside Reader had nowhere to go. A small protest turned up outside a Google office in Washington over it. By then the product was already in maintenance mode, meaning critical bugs got fixed and nothing else got built. Brian Shih, who ran Reader as product manager until he left Google in 2011, wrote on Quora in 2013 that "I'm pretty sure Reader was threatened with de-staffing at least three times before it actually happened." Chris Wetherell read the same signal from the outside. He told GigaOm in 2013 that "when they replaced sharing with +1 on Google Reader, it was clear that this day was going to come." Reader was not competing with Feedly for resources at that point. It was competing with Google+, and Google+ was the company strategy. ## What did Google say when it pulled the plug? On March 13, 2013, Google published "A second spring of cleaning", a post by senior vice president Urs Holzle that listed eight products being retired. Reader was on the list. The explanation was one sentence: "While the product has a loyal following, over the years usage has declined. So, on July 1, 2013, we will retire Google Reader." Declining is relative. In March 2013 Reader's most popular feeds each carried more than 24 million subscribers. The product was smaller than its peak and still larger than most of what has replaced it. The response was quick and loud. A petition on Change.org passed 100,000 signatures in under two days and stood above 116,000 by that Saturday. At least eight related petitions went up alongside it. Google did not move the date. Reader stopped serving feeds on July 1, 2013. Google left a two-week window to pull subscription lists out through Takeout, then deleted the data on July 15, 2013. There was no recovery path after that. ## Where did Reader's users actually go? Feedly, a small reader at the time, picked up more than 500,000 new users in the 48 hours after the announcement and roughly 3 million by early April 2013. NewsBlur, run largely by one developer, Samuel Clay, went from about 1,500 users to more than 60,000 and spent weeks fighting its own servers. The Old Reader and Inoreader took similar waves. Most Reader users went nowhere. They did not pick a replacement, they simply stopped using RSS and read whatever the default feed on Twitter or Facebook put in front of them. That is the part the migration numbers hide. The feeds themselves were not entirely lost. Archive Team, the volunteer preservation group, pulled feed history out of Reader's API before the shutoff and uploaded roughly 8,800 GB of it to the Internet Archive, where it still sits. Reading a specific feed out of those archives takes technical work, but the material is public. ## Is Google Reader coming back in 2026? No. Google Reader has been shut since July 1, 2013, and Google has never announced a revival, a successor or a reopening of the API. Searches for the product still run high, and the answer has not changed in thirteen years. The product it was cleared out of the way for did not last either. Google+ closed for consumers on April 2, 2019, after Google said more than 90 percent of user sessions on it lasted under five seconds. Reader outlived its own replacement in usefulness and lost anyway. What Reader left behind is a reflex. Every Google launch now draws the same response, because the list of shutdowns kept growing after 2013: Google Wave, Google Buzz, Google Inbox, Google Play Music, Picasa, Google Glass (https://404memoryfound.com/posts/why-google-glass-actually-failed.html). Reader is the one people name first, because it worked, people used it, and Google closed it anyway. ## What does an RSS reader cost in 2026? Less than most people assume, and often nothing. Feedly's free tier covers 100 sources with no AI features. Feedly Pro runs about $6 a month billed annually in 2026, and the Pro+ tier with AI filtering is about $99 a year. The free and open options are stronger than they were in 2013. NetNewsWire is open source and free on Mac and iOS. Miniflux is open source and self-hosted. NewsBlur still runs a free tier with a paid upgrade, and Inoreader sits between the free clients and Feedly on features. RSS itself never stopped working. It is still how every podcast in the world is distributed, and most news sites still publish a feed whether or not they advertise it. What changed after 2013 is visibility: browsers dropped their built-in feed readers, the orange feed icon disappeared from address bars, and RSS became something you have to go looking for rather than something you trip over. ## Frequently Asked Questions ### Why did Google shut down Google Reader? Google said in March 2013 that usage of Google Reader had declined and that it wanted to support fewer products. Former Reader product manager Brian Shih and creator Chris Wetherell both pointed instead at Google+, the social network Google reorganized its products around starting in 2011, which left Reader in maintenance mode with no staff to defend it. ### When did Google Reader shut down? Google Reader stopped working on July 1, 2013. Google announced the closure on March 13, 2013, and gave users until July 15, 2013 to export their subscriptions through Google Takeout before the data was deleted permanently. ### What is the best Google Reader alternative in 2026? Feedly is the closest mainstream successor to Google Reader, with a free tier for 100 sources and a Pro tier at about $6 a month billed annually in 2026. Inoreader suits heavier users, NewsBlur offers a free tier alongside paid accounts, and NetNewsWire and Miniflux are free and open source. **Sources:** - Google, 'A second spring of cleaning' (March 2013): https://blog.google/company-news/inside-google/company-announcements/a-second-spring-of-cleaning/ - Forbes: Google Reader's last product manager on the shutdown (2013): https://www.forbes.com/sites/alexkantrowitz/2013/04/08/google-readers-last-product-manager-calls-its-shutdown-a-missed-opportunity/ - Om Malik: Google Reader creator Chris Wetherell reflects: https://om.co/gigaom/chris-wetherll-google-reader/ - The Next Web: Save Google Reader petition passes 100,000 signatures: https://thenextweb.com/google/2013/03/15/petition-to-save-google-reader-passes-100000-signatures-but-dont-expect-google-to-reverse-its-decision - Archive Team: Google Reader preservation project: https://wiki.archiveteam.org/index.php/Google_Reader --- # Is the Nokia N-Gage Still Around? Who Owns It Now URL: https://404memoryfound.com/posts/what-happened-to-nokia-ngage-gaming-phone.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-01 Updated: 2026-09-19 Topics: Gaming, Hardware, Business Blunders **Summary:** The Nokia N-Gage is not still around. Nokia discontinued the handheld in November 2005 after shipping about 3 million units against a 6 million target, and the N-Gage software service that replaced it closed in September 2010. Nokia Corporation still owns the N-Gage trademark in 2026, but no N-Gage hardware, store or service exists. **Key facts:** - Launched: October 7, 2003, $299.99 in the United States - Units sold: About 3 million by 2007, against a 6 million target - Status today: Discontinued; hardware ended November 2005, service closed September 2010 - Owner today: Nokia Corporation, which still holds the N-Gage trademark - Price today: About $60 to $290 used, depending on condition ## Why did Nokia build a phone that played games? In 2002 Nokia sold more mobile phones than anyone on earth. The Nokia 3310 (https://404memoryfound.com/posts/what-happened-to-nokia-3310.html), launched in 2000, moved 126 million units before it was retired in 2005, a record for a single handset at that point. Nokia's problem was what came next. Handheld gaming was a Nintendo business, and the Game Boy Advance would finish its run at 81.51 million units. Nokia had carriers, factories and shelf space in every country that mattered. What it did not have was a game platform. The answer was the N-Gage, announced in November 2002 and pitched to the press as a game deck rather than a phone. Nokia signed real publishers instead of budget studios: Activision, Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html), EA and Ubisoft all committed titles. "The Nokia N-Gage brings a whole new level of interactive gaming with its innovative features," said Ilkka Raiskinen, then head of Nokia's Entertainment and Media unit, in the February 2003 announcement of Activision's support. On paper it was the strongest launch line-up any new handheld had assembled since the original Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html). Nokia was not guessing about demand for portable games. It was guessing about what that machine should look like. ## What went wrong at the October 2003 launch? The N-Gage reached US shelves on October 7, 2003 at $299.99. A Game Boy Advance SP cost $99 that fall. Nokia was asking three times the price of the market leader for an unproven platform from a company that had never shipped a game console. Two design decisions did the damage. The earpiece and microphone sat on the narrow edge of the device, so taking a call meant holding the whole slab sideways against your head. The internet named the pose sidetalking within days, and the device became the taco phone. A parody site, sidetalkin.com, collected reader photos of people pressing laptops, pizza boxes and keyboards to their ears, and it outlived the product: the last picture went up in 2008. The second problem was the game slot. Swapping a game on the original N-Gage meant powering the device down, pulling off the back cover, removing the battery, changing the card, then reassembling the whole thing. Nokia effectively conceded the point a year later, when it sold the replacement model on its hot-swappable card slot as a headline feature. On a Game Boy Advance the same job took three seconds. ## How badly did the N-Gage actually sell? Nokia announced on October 23, 2003 that it had sold 400,000 N-Gage units in two weeks. That number counted units pushed into the retail channel, not units bought by people. Arcadia Research, an independent market research firm, put actual US sell-through at under 5,000 in the same period. Retailers moved faster than Nokia did. Within roughly two weeks of launch, GameStop (https://404memoryfound.com/posts/is-gamestop-still-in-business.html) and Electronics Boutique were bundling three games with a $100 instant rebate, which brought the package to $199.99 against the $299.99 launch price. A platform that needs a third off in its first month is not one anyone is fighting over. The rest followed. Nokia had said it wanted 6 million N-Gage units in three years. By November 2005 it had sold about a third of that, and Anssi Vanjoki, who ran Nokia's multimedia division, put it bluntly: "I am not happy. I said we needed to sell six million in three years, and we sold one-third of that." Nokia had shipped around 3 million decks in total by 2007. ## Did the N-Gage QD fix any of it? Nokia tried. The N-Gage QD arrived in May 2004, smaller and rounder, with the earpiece moved to the front so calls no longer required the sideways pose, and with the card slot moved so games could be changed without taking the battery out. It started at $99 with a carrier contract and roughly $199 unsubsidized. It was the device the N-Gage should have been in 2003, and it landed seven months too late. The name was already a punchline and the reviews were already written. Then Nintendo shipped the DS in November 2004 with two screens, a touchscreen and every major publisher behind it. The DS finished at 154.02 million units. The N-Gage finished at about 3 million. Nokia discontinued the hardware in November 2005. Close to 60 games were released for the original N-Gage across its three years, a library smaller than a single strong year on a Nintendo handheld. ## Is the N-Gage still around in 2026, and who owns it? No. There has been no N-Gage hardware since November 2005 and no N-Gage service since 2010. Nokia made one more attempt in between: a software platform, also called N-Gage, that put games on Nokia's existing Symbian smartphones. It went live on April 3, 2008 for the N81, N82 and N95. Apple's App Store opened three months later, and on October 30, 2009 Nokia announced there would be no more N-Gage games. Online support ended in September 2010, with 49 titles released. The name still has an owner. N-Gage is a registered trademark of Nokia Corporation, which kept its patents and brands when it sold the phone business to Microsoft for about $7.2 billion in a deal that closed on April 25, 2014. Nokia is a networks company in Espoo, Finland today, and the Nokia name on handsets (https://404memoryfound.com/posts/does-nokia-still-make-phones.html) has been licensed out ever since. HMD Global stopped making Nokia-branded smartphones at the start of 2025 and now builds only Nokia feature phones under that licence. None of it covers N-Gage. The games outlived the company that made them. EKA2L1, an open source Symbian and N-Gage emulator, runs both the 2003 cartridge games and the 2008 service titles on Windows, macOS, Linux and Android. ### Where to find one today A working N-Gage is a shelf item now rather than a phone anyone carries. A 2025 buying guide put tested N-Gage QD units with replaced batteries at $140 to $190, clean Classic units have sold for as much as $290 in 2026, and rough or untested ones change hands for around $60. Handsets, boxes and loose game cards turn up on Etsy (https://www.etsy.com/search?q=nokia+n-gage). ## What Nokia got right about mobile gaming Strip out the taco jokes and the pricing and Nokia's thesis holds up. The device in your pocket would become the machine most people play games on, and that is exactly what happened. It just happened without Nokia. The N-Gage also shipped ideas that are now standard. N-Gage Arena, the online service Nokia built for it in 2003, handled matchmaking, downloadable extras and global high score tables years before phone gaming had anything comparable. Bluetooth multiplayer was in the box at launch. Digital distribution, the whole point of the 2008 relaunch, is now how essentially every mobile game is sold. Nokia read the map correctly and drove off the road. By 2008, when the idea was finally ready, Apple and Android were already taking the company apart. ## Frequently Asked Questions ### Is the Nokia N-Gage still around in 2026? No. Nokia discontinued the Nokia N-Gage handheld in November 2005, and the N-Gage software service that replaced it stopped taking new games in October 2009 and closed its online features in September 2010. Nokia Corporation still owns the N-Gage trademark in 2026, but there is no N-Gage hardware, store or service. ### How many Nokia N-Gage units were sold? Nokia shipped about 3 million Nokia N-Gage units across the original 2003 model and the 2004 N-Gage QD, against a stated target of 6 million in three years. Nokia's claim of 400,000 units in the first two weeks of October 2003 counted stock sent to retailers; Arcadia Research put actual US sales under 5,000 in that period. ### Why was the Nokia N-Gage called the taco phone? The Nokia N-Gage, released on October 7, 2003, put its earpiece and microphone on the narrow side edge, so making a call meant holding the device sideways against your face like a taco. The pose was nicknamed sidetalking, and the parody site sidetalkin.com collected photos of it until 2008. **Sources:** - Nokia Reveals Overall N-Gage Sales To Date, Game Developer: https://www.gamedeveloper.com/game-platforms/nokia-reveals-overall-n-gage-sales-to-date - Nokia gloomy about N-Gage, NBC News, 2005: https://www.nbcnews.com/id/wbna9897772 - N-Gage price descends again, GameSpot, 2003: https://www.gamespot.com/articles/n-gage-price-descends-again/1100-6083629/ - Activision Set to Develop Titles for Nokia N-Gage, Activision investor relations, 2003: https://investor.activision.com/news-releases/news-release-details/activisionr-set-develop-titles-nokia-n-gagetm-mobile-game-deck - EKA2L1, Symbian OS and N-Gage emulator: https://eka2l1.github.io/ --- # Is Excite@Home Still Around? AT&T Took the Network URL: https://404memoryfound.com/posts/what-happened-to-excite-at-home-broadband.html Author: Dana Reyes (404 Memory Found) Published: 2026-04-01 Updated: 2026-09-19 Topics: Business Blunders, Internet Culture, Money & Tech **Summary:** Excite@Home is not still around. The company filed for Chapter 11 bankruptcy on September 28, 2001, and AT&T bought its broadband network for $307 million, less than three years after the stock peaked at $128.34 a share and a $35 billion valuation. The Excite portal outlived the company and still loads at excite.com in 2026, run by Ask Media Group for IAC. **Key facts:** - Founded: @Home Network, 1995, by Milo Medin, William Randolph Hearst III and three cable operators - Peak: $128.34 a share and about $35 billion in Q1 1999; 4.1 million broadband subscribers - Status today: Gone. Chapter 11 on September 28, 2001; network switched off December 1, 2001 - Owner today: excite.com is an IAC property, operated by Ask Media Group; search by System1 - Network buyer: AT&T, $307 million in cash, later merged into Comcast ## How did @Home Network become a $35 billion company? @Home Network started in 1995 on a bet that cable television wiring, not the telephone line, was the fastest way into an American living room. Milo Medin built the engineering side. William Randolph Hearst III ran it as the first chief executive. The money came from Kleiner Perkins and from three cable operators that already had the lines in the ground: Tele-Communications Inc., Comcast and Cox Communications. The division of labor was clean. The cable companies installed the modems and billed the customer. @Home Network ran the network, the portal and the support. Subscribers got a connection that stayed on all day and moved data at rates no telephone modem could reach, which mattered enormously to anyone used to waiting on a 28.8k modem (https://404memoryfound.com/posts/dialup-speed-then-vs-now.html) to load a single page. The company went public in 1997 and the stock did what broadband stocks did in the late 1990s. By the first quarter of 1999, At Home Corporation, as the filings called it, traded at $128.34 a share and carried a market capitalization of roughly $35 billion. It had about 330,000 broadband subscribers at the time, which works out to more than $100,000 of market value per customer. ## Why did @Home pay $6.7 billion for a losing search portal? Excite was a search engine and web portal of the same 1995 vintage, competing with Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html), Lycos (https://404memoryfound.com/posts/what-happened-to-lycos-search-engine.html) and AltaVista (https://404memoryfound.com/posts/what-happened-to-altavista-search-engine.html) for the job of being the first page anyone saw in the morning. It had the traffic. It did not have the profit. By December 1998, Excite was negotiating to sell itself to Yahoo for somewhere between $5.5 billion and $6 billion. The merger documents At Home Corporation later filed with the Securities and Exchange Commission describe what happened next: on December 19, at the prompting of Kleiner Perkins, @Home Network chairman and chief executive Thomas Jermoluk met Excite chairman and chief executive George Bell, and the two worked out a rival deal for Excite and its debt. The acquisition closed on January 19, 1999 at about $6.7 billion in stock, one of the largest mergers between two internet companies up to that point. The combined business took the public name Excite@Home while the shares kept trading as At Home Corporation. The theory was vertical integration. Own the pipe, own the homepage at the end of it, and sell advertising against subscribers you were already billing. Excite's separate life as a search engine (https://404memoryfound.com/posts/what-happened-to-excite-search-engine-google.html) ran on its own track and ended on its own terms. ## What did the dot-com crash do to the numbers? Subscriber growth was never the problem. Excite@Home went from roughly 330,000 broadband customers in early 1999 to more than 3.7 million by August 2001, and peaked at 4.1 million across the United States, Canada, Japan, Australia and the Benelux countries. Demand for cable broadband was real and it was compounding. Revenue was the problem. At Home Corporation booked about $616 million in sales for 2000 and lost $7.44 billion that year, against a $1.5 billion loss in 1999. Online advertising, the half of the model that was supposed to pay for the network buildout, went down with the startups that had been buying it. Each quarter of 2001 was worse than the one before. Excite@Home lost $832.6 million in the first quarter, $346.3 million in the second and $271.3 million in the third, when sales fell to $138.4 million from $169.9 million a year earlier. A company once valued at $35 billion was burning more cash in a single quarter than it collected in revenue across a full year. ## Why did AT&T refuse to save its own investment? Excite@Home never owned the cable. It rented access to lines that belonged to its partners, and after AT&T absorbed Tele-Communications Inc. the largest of those partners was also the company's controlling shareholder. A restructuring on August 28, 2000 lifted AT&T's voting control from 56 percent to 74 percent. By mid-2001, AT&T held 74 percent of the votes against a 23 percent equity stake. That is an awkward position for everyone involved: enough control to steer the company, not enough ownership to lose much if it died. AT&T was also building its own broadband business and had no lasting reason to keep paying a middleman for access to its own wires. When Excite@Home needed cash in 2001, AT&T did not provide it. Excite@Home filed for Chapter 11 protection in the U.S. Bankruptcy Court for the Northern District of California on September 28, 2001, and announced the filing on October 1. Chief executive Patti Hart called it "a tool to protect the value of the broadband business," and the same announcement set out a deal to sell that business to AT&T for $307 million in cash. That is close to one percent of the 1999 peak valuation. ## How did 850,000 customers lose service overnight? The creditors did not think $307 million was enough. Their leverage was the network itself, and they used it. In court papers, Excite@Home accused them of wanting to "play a 'game of chicken' in which the threat of a blackout is used to extort" the cable companies into paying more for service. The court let it happen. On November 30, 2001, Judge Thomas Carlson ruled that Excite@Home could reject its master distribution agreements with the cable operators, saying the contracts, which executives put at around $6 million a week, were "clearly burdensome." Negotiations with AT&T ran into Saturday morning and broke down. Service to about 850,000 AT&T customers went dark on December 1, 2001. Comcast and Cox Communications had already begun moving their own subscribers onto networks they controlled. The remaining 1,350 employees were laid off over the following months. The estate kept litigating for years. In May 2005, AT&T agreed to pay $340 million to settle claims brought by the At Home Corporation bondholders' liquidating trust, with Comcast reimbursing half of that amount; the trust put its total recovery at about $400 million once reserved funds were released. ## Who owns Excite today, and is the site still up? Excite@Home the company does not exist in 2026. The two assets it held went in opposite directions. The network went to AT&T and became AT&T Broadband. On December 19, 2001, AT&T announced a $72 billion merger of that unit with Comcast, covering more than 22 million subscribers, and the deal closed in November 2002. The coaxial plant @Home Network helped light up in the late 1990s is part of what Comcast sells as Xfinity today, which is the closest thing to a happy ending in this story. The portal changed hands more often. Excite was sold to iWon in 2002, iWon's business was folded into Ask Jeeves, and Ask Jeeves was acquired by IAC. In 2026, excite.com still loads. It is operated by Ask Media Group on behalf of Ask Applications, an IAC business, and its search results are supplied by System1. There is no Excite crawler and no Excite index behind the box. The same parent company shut Ask.com (https://404memoryfound.com/posts/what-happened-to-ask-jeeves-search-engine.html) down in 2026, which is a fair measure of how much of that business is left. ## Frequently Asked Questions ### What happened to Excite@Home? Excite@Home, the broadband company created when @Home Network bought the Excite portal in January 1999, filed for Chapter 11 bankruptcy on September 28, 2001. AT&T agreed to buy its broadband network for $307 million, and the service was switched off on December 1, 2001, cutting off roughly 850,000 AT&T customers. ### Who owns excite.com now? excite.com is an IAC property in 2026, operated by Ask Media Group on behalf of Ask Applications. The portal passed through iWon and Ask Jeeves after Excite@Home's 2001 bankruptcy, and its search results are now supplied by System1 rather than by any Excite index of its own. ### How much was Excite@Home worth at its peak? Excite@Home traded at $128.34 a share in the first quarter of 1999, for a market capitalization of about $35 billion. Less than three years later, AT&T bought the Excite@Home broadband network out of bankruptcy for $307 million, roughly one percent of that peak value. **Sources:** - At Home Corporation press release on the Chapter 11 filing and AT&T transaction (SEC EDGAR, October 1, 2001): https://www.sec.gov/Archives/edgar/data/5907/000089882201500657/october1pressrel.txt - Excite@Home pulls the plug on AT&T cable Internet (Computerworld, December 2001): https://www.computerworld.com/article/1351764/excite-home-pulls-the-plug-on-at-t-cable-internet.html - Excite@Home company history (Encyclopedia.com): https://www.encyclopedia.com/economics/encyclopedias-almanacs-transcripts-and-maps/excitehome - AT&T Reaches At Home Litigation Settlement (Comcast Corporation, May 2005): https://corporate.comcast.com/news-information/news-feed/att-reaches-at-home-litigation-settlement - AT&T Broadband to Merge with Comcast Corporation in $72 Billion Transaction (Comcast Corporation, December 2001): https://corporate.comcast.com/news-information/news-feed/att-broadband-to-merge-with-comcast-corporation-in-72-billion-transaction --- # Why the Sega Saturn Failed and What It Costs Now URL: https://404memoryfound.com/posts/what-happened-to-sega-saturn-console.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-01 Updated: 2026-09-19 Topics: Gaming, Hardware, Business Blunders **Summary:** The Sega Saturn failed because Sega surprise-launched it in North America on May 11, 1995 at $399, stocked at only a handful of retailers, on the same day Sony priced the PlayStation at $299. It sold 9.26 million units worldwide against 102.49 million for the PlayStation, and Sega dropped it in the United States in 1998. A working Saturn costs about $180 to $280 in 2026. **Key facts:** - Launched: November 22, 1994 in Japan; May 11, 1995 in North America - Launch price: $399 in North America, 44,800 yen in Japan - Units sold: 9.26 million worldwide, 1.83 million in North America - Status today: Discontinued; dropped in the US in 1998, sold in Japan until 2000 - Price today: $180 to $280 for a working console with controllers, 2026 ## What happened at E3 1995? On May 11, 1995, at the first Electronic Entertainment Expo in Los Angeles, Sega (https://404memoryfound.com/posts/is-sega-still-making-games.html) of America president Tom Kalinske told the room that the Sega Saturn was not shipping in the fall. It was already on shelves. Sega had quietly sent 30,000 consoles to a short list of chains, among them Toys R Us (https://404memoryfound.com/posts/is-toys-r-us-still-in-business.html), Babbage's, Electronics Boutique and Software Etc., priced at $399. The decision came from Tokyo. Sega chairman Hayao Nakayama wanted the Saturn in stores before Sony's first console arrived. "Nakayama was so concerned over Sony launching a hardware platform that he wanted to beat them to the punch," Kalinske told Fast Company in 2025. Kalinske argued against the plan and was overruled. Sony answered the same day. Sony Computer Entertainment America president Steve Race walked to the podium, said "$299," and walked off without reading his prepared remarks. The PlayStation reached American stores on September 9, 1995, a hundred dollars under the Saturn. The 32-bit generation was effectively settled in those few hours, four months before most buyers had even seen the two machines side by side. ## Why did American retailers drop Sega? Retail in 1995 ran on notice. Chains committed floor space, ordered stock and booked holiday advertising months ahead of a launch. Sega gave its partners none of that, and the stores left out of the 30,000-unit shipment found out from the same press conference everyone else watched. Walmart, Best Buy and KB Toys (https://404memoryfound.com/posts/kb-toys-stores-left.html) were among the chains passed over, and several of them cut back or dropped Sega hardware in response. That left Sega paying for national advertising for a console large parts of the country could not walk in and buy, while the retailers who could have pushed it had a reason not to. The damage outlasted the console. Kalinske resigned on July 15, 1996, effective that September, after six years running Sega of America, and said afterward that losing the authority to make decisions for the western market was why he left. Sega's American arm never got that independence back, and when the company returned in 1999 it was negotiating shelf space with buyers who remembered 1995. The same head-office pattern runs through the 32X, the stopgap add-on Sega sold alongside it (https://404memoryfound.com/posts/what-happened-to-sega-32x-add-on.html). ## What made the Saturn so hard to program for? The Saturn ran two Hitachi SH-2 processors. On paper that beat the PlayStation's single CPU. In practice the two chips shared a bus and could not reach memory at the same moment, so any developer who wanted the second processor doing useful work had to hand-schedule the traffic between them. Sega's own star developer said so in public. "I don't think all programmers have the ability to program two CPUs," Yu Suzuki, head of Sega AM2, told Next Generation magazine in 1995. Most, he said, could get only about one and a half times the speed of a single SH-2, and roughly one programmer in a hundred could get close to double. Outside studios did that math and picked the cheaper machine. The PlayStation was easier to build 3D games on, better documented, and inside a year it had the bigger installed base to sell into. The Saturn's 2D hardware genuinely was the better of the two, which is why Capcom's fighting games ran cleaner on it, but 2D was the part of the market that was shrinking in 1996. ## How many Sega Saturns were sold worldwide? 9.26 million, across the console's entire life. Japan took 5.80 million of that, North America 1.83 million and Europe 1.1 million. The Japanese start had been strong: roughly 170,000 units on the first day, November 22, 1994, at 44,800 yen, with the Virtua Fighter port selling at close to one copy per console. The comparison settles the argument. Sony's original PlayStation sold 102.49 million units worldwide before production ended in March 2006, more than eleven times the Saturn's lifetime total. Sega stopped selling the Saturn in the United States in 1998, when the Dreamcast took over, and kept it on the shelf in Japan until 2000. Six games were available at the American launch, among them Virtua Fighter as the pack-in, Daytona USA, Panzer Dragoon, Clockwork Knight and Worldwide Soccer. The library collectors now chase, including Nights into Dreams, Panzer Dragoon Saga and Radiant Silvergun, arrived after most American buyers had already committed to a PlayStation. ## Is the Sega Saturn still supported in 2026? No. Sega has not built hardware since 2001, and there is no repair program, no warranty and no first-party service for the Saturn. Anything keeping one running in 2026 is second-hand parts or emulation. Sega has also re-released very little of the library: Panzer Dragoon Saga has never appeared on a modern platform, and the closest thing to an official return is the Panzer Dragoon II Zwei remake announced in 2025 for PlayStation 5, Xbox, Switch and PC. Prices track how few were built. A working gray Model 2 with controllers runs roughly $180 to $280 in 2026, worn units land nearer $120 to $150, and refurbished listings at large retailers sit around $270. Discs are the expensive half of the hobby, and the reason is the same 1.83 million figure: North America simply never absorbed enough Saturns to leave a deep used market behind. Where to find one today: a retro specialist is the safer route, because a Saturn's internal battery and its laser both age badly and a seller who tests consoles will tell you which one is tired. Stone Age Gamer stocks Saturn hardware and accessories (https://stoneagegamer.com/sega/saturn/). Budget for a replacement CR2032 battery, and expect a boxed console to cost noticeably more than a loose one. ## Did the Saturn end Sega's console business? It started the ending. The Saturn cost Sega its retail relationships, its third-party support in North America and the cash it needed to fund the next machine properly. Bernie Stolar, who took over Sega of America in 1996, said the quiet part in print while the console was still for sale: "Saturn is not our future," he told Electronic Gaming Monthly in its September 1997 issue. The Dreamcast followed on November 27, 1998 in Japan and September 9, 1999 in North America. It launched well. It did not last: Sega announced on January 31, 2001 that it would discontinue the Dreamcast and leave hardware, and production ended that March. Sammy took a controlling stake in 2003, and the two companies merged into Sega Sammy Holdings in 2004. Sega is still trading as a publisher, shipping games on PlayStation, Xbox, Nintendo and PC through studios including Atlus, Creative Assembly and Sports Interactive. That is the long tail of the surprise launch: a company that stopped making consoles after the Dreamcast (https://404memoryfound.com/posts/why-sega-stopped-making-consoles-dreamcast.html) and now sells its catalog on the platforms that beat it. For the back catalog without the original hardware, the current ways to play Genesis games (https://404memoryfound.com/posts/sega-genesis-games-how-to-play-2026.html) cover most of the pre-Saturn era. ## Frequently Asked Questions ### How much is a Sega Saturn worth in 2026? A working Sega Saturn with controllers sells for roughly $180 to $280 in 2026, with worn consoles nearer $120 to $150 and refurbished units at large retailers around $270. Boxed models and scarce discs such as Panzer Dragoon Saga cost far more, because Sega sold only 9.26 million Saturns worldwide between its 1994 debut and the end of Japanese sales in 2000. ### Why did the Sega Saturn fail? The Sega Saturn failed largely because of its North American launch on May 11, 1995: Sega shipped it early and unannounced to a handful of retailers at $399, Sony priced the PlayStation at $299 the same day, and the console's dual Hitachi SH-2 processors were hard for outside studios to program. Chains left out of the launch pulled back on Sega, and the company dropped the Saturn in the United States in 1998. ### How many Sega Saturn units were sold? The Sega Saturn sold 9.26 million units worldwide over its life: 5.80 million in Japan, 1.83 million in North America and 1.1 million in Europe. Sony's original PlayStation, its direct rival from 1995, sold 102.49 million units before Sony ended production in March 2006. **Sources:** - Sega Saturn, Wikipedia: https://en.wikipedia.org/wiki/Sega_Saturn - How Sega's surprise Saturn launch backfired, Fast Company (2025): https://www.fastcompany.com/91363906/sega-surprise-saturn-launch-backfired-changed-gaming-forever - 30 Years Ago, Sega Took Its Biggest Gamble With Saturn And Failed, Time Extension (2025): https://www.timeextension.com/news/2025/05/30-years-ago-sega-took-its-biggest-gamble-with-saturn-and-failed - How A Series Of Bad Decisions Led To The Sega Saturn Failure, Game Informer: https://gameinformer.com/b/features/archive/2017/07/03/gi-classic-the-saturn-spiral.aspx - Sega Saturn console price guide, PriceCharting: https://www.pricecharting.com/game/sega-saturn/sega-saturn-console --- # Why the Apple Newton Failed and What It Costs Now URL: https://404memoryfound.com/posts/what-happened-to-apple-newton.html Author: Marcus Vale (404 Memory Found) Published: 2026-04-01 Updated: 2026-09-09 Topics: Hardware, Business Blunders, Then vs Now **Summary:** The Apple Newton was Apple's handheld computer line, launched on August 2, 1993 at $699 and canceled by Steve Jobs on February 27, 1998. It failed because the handwriting recognition did not work at launch, the price stayed high, and a $299 PalmPilot did the same job reliably. Its maker, Apple, earned $309 million in fiscal 1998 after losing $1.05 billion the year before, then slipped back to a $25 million loss in fiscal 2001. **Key facts:** - Launched: August 2, 1993, Apple Computer, $699 - Status: Discontinued February 27, 1998, no successor until the iPhone - Units sold: About 50,000 in the first three months; Apple never published a lifetime total - Maker today: Apple Inc., which returned to profit in fiscal 1998 with $309 million - Price today: $541 for a complete MessagePad 2100 at auction in 2023 ## Has Apple been profitable ever since the Newton died? The MessagePad was made by Apple, and the answer people keep searching for is yes, with one exception. Apple lost $1.05 billion in fiscal 1997. In fiscal 1998, the year it shut the Newton down, it earned $309 million on $5.94 billion in sales. Fiscal 1999 brought $601 million and fiscal 2000 brought $786 million. Then came fiscal 2001, when Apple lost $25 million on revenue of $5.36 billion as the PC market fell apart. That loss is the part people leave out when they say Apple has been profitable since 1998. The turn itself was real. Two years before that $309 million, in fiscal 1996, Apple lost $816 million, and it spent the middle of the decade burning through cash it did not have (https://404memoryfound.com/posts/when-apple-almost-went-bankrupt-1997.html). Fiscal 1998 was also the year of the first iMac (https://404memoryfound.com/posts/how-imac-g3-saved-apple-killed-beige-box.html). The Newton was cut in the same twelve months. ## Why did the Apple Newton fail? Apple announced the concept at the Consumer Electronics Show on January 7, 1992, showed a prototype that May, and put the first MessagePad in buyers' hands on August 2, 1993 at $699. The promise was that the device would read ordinary handwriting, no training, no special alphabet. At launch it did not. The recognizer guessed at words, and it guessed wrong often enough that the failure became the product's public identity within weeks. Garry Trudeau spent a full week of Doonesbury in August 1993 on the joke, including a panel where "Catching on?" came back as "Egg freckles." The Simpsons later ran its own version, turning "Beat up Martin" into "Eat up Martha." The sales figures tell the rest. Apple moved about 50,000 MessagePads in the first three months and promoted the number as a strong start. Writing about the device years later in Time, Harry McCracken judged it a disappointment against what Apple had expected to sell. Prices never came down to meet the market either. The last model, the MessagePad 2100, started at $1,000. ## What the Newton actually got right The hardware was not the weak part. The original MessagePad ran an ARM 610 at 20 MHz with 640 KB of RAM, which was a serious amount of computing for something that fit in a jacket pocket in 1993. The MessagePad 2000 arrived in March 1997 with a 162 MHz StrongARM and 5 MB of RAM, and the 2100 followed in November 1997 with 8 MB. The recognition problem was also fixed, just too late to matter. Newton OS 2.0 replaced the licensed engine with one Apple wrote itself, called Rosetta, refined again in Newton 2.1. It read printed letters rather than cursive, and it worked. Reviewers and owners were still rating Newton 2.1 recognition ahead of the alternatives ten years after it shipped. The software ideas held up too. Newton kept data in shared stores that any application could read, so a note containing a name and a time could offer to become a calendar entry on its own. That kind of cross-application awareness is something phone software rebuilt from scratch more than a decade later. ## How a $299 rival took the market Palm (https://404memoryfound.com/posts/is-palm-still-a-company.html)'s answer was to stop trying to read natural handwriting at all. Its Graffiti system made the user learn a simplified stroke alphabet, one motion per letter, inside a fixed box. It was less ambitious than what Apple attempted and it was reliable from the first day, which turned out to be the trade buyers wanted. Price finished the argument. The PalmPilot (https://404memoryfound.com/posts/what-happened-to-palm-pilot-pda.html) sold for $299, well under half the Newton's $699 launch price and under a third of what the MessagePad 2100 asked. A category Apple had spent years defining was taken by a cheaper device that did fewer things without embarrassing its owner. ## Is the Apple Newton still around, and what does one sell for? No. Steve Jobs canceled the Newton hardware and software platform on February 27, 1998, which ended the MessagePad line and the keyboard-equipped eMate 300 that Apple had been selling to schools. There has been no Newton device and no Newton OS release since, and Apple did not ship another handheld computer until the iPhone. What a Newton costs today depends almost entirely on whether the box and the accessories survived. RR Auction sold a MessagePad 2100 with its original box, power adapter, stylus, software disc, two battery packs, serial cable, fax modem card and keyboard for $541 in August 2023, against a pre-sale estimate of $200. A boxed MessagePad 120 with its accessories, documentation and a copy of the game Cogito brought $740 in August 2021. Where to find one today: eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html).com carries MessagePads more or less continuously, and loose units with no box, no stylus and no power supply go for a fraction of those auction results. Complete sets are what carry the price, and a corroded battery pack is the most common thing wrong with a listing. ## The ARM shares that outlasted the product Apple needed a processor that could run all day on batteries, so it went in with Acorn Computers on the chip venture that became ARM. The Newton was the first Apple product to ship with the result, and the chip company turned out to be worth far more than the handheld. Apple sold 18.9% of the ARM shares it held in 1998 for a gain before foreign taxes of about $24 million. In fiscal 1999 it sold 163 million shares for roughly $245 million in net proceeds and a pre-tax gain of about $230 million. In fiscal 2000 it sold 45.2 million more for about $372 million and a gain of about $367 million, and still held 34.8 million shares valued at $383 million at the end of that year. Those two years of sales alone brought in around $617 million, at a time when Apple's entire fiscal 1998 profit was $309 million. The architecture Apple went shopping for because a 1993 handheld needed to last a day on batteries now runs every iPhone and every Mac built on Apple silicon. ## Frequently Asked Questions ### How much did the Apple Newton MessagePad cost? The first MessagePad shipped on August 2, 1993 at $699. The final model, the MessagePad 2100, started at $1,000 when it arrived in November 1997. A PalmPilot, by comparison, sold for $299. ### Why did Steve Jobs kill the Newton? Jobs canceled the Newton platform on February 27, 1998, not long after returning to a company that had lost $1.05 billion in fiscal 1997 and $816 million the year before that. Cutting product lines was the recovery plan, and a slow-selling four-figure handheld was an obvious place to start. ### What is an Apple Newton worth today? Complete examples do best at auction. RR Auction sold a fully accessorized MessagePad 2100 for $541 in August 2023 and a boxed MessagePad 120 for $740 in August 2021. Loose units on eBay.com, with no box or accessories, sell for considerably less. **Sources:** - Apple Computer, Inc. Form 10-K, fiscal year 2000 (SEC EDGAR): https://www.sec.gov/Archives/edgar/data/0000320193/000091205700053623/a2032880z10-k.txt - Apple Reports Fourth Quarter Profit of $66 Million (October 17, 2001): https://www.apple.com/newsroom/2001/10/17Apple-Reports-Fourth-Quarter-Profit-of-66-Million/ - Harry McCracken, 'Newton, Reconsidered', Time: https://time.com/archive/7235122/newton-reconsidered/ - MessagePad, Wikipedia: https://en.wikipedia.org/wiki/MessagePad - Apple Newton MessagePad 2100, RR Auction lot detail: https://www.rrauction.com/auctions/lot-detail/347712206735029-apple-newton-messagepad-2100/ --- # Is the Original Xbox Still Supported in 2026? URL: https://404memoryfound.com/posts/what-happened-to-original-xbox-microsoft-gaming.html Author: Marcus Vale (404 Memory Found) Published: 2026-03-31 Updated: 2026-09-17 Topics: Gaming, Hardware, Business Blunders **Summary:** The original Xbox has not been supported since Microsoft stopped making it in 2006, and the console's Xbox Live service shut down on April 15, 2010. Microsoft still owns the Xbox brand, 39 original Xbox games run on Xbox Series X and Series S through backward compatibility, and a free fan-run service called Insignia has put 199 of the old online games back on the internet. Working consoles sell for roughly $60 to $100 in 2026. **Key facts:** - Launched: November 15, 2001 in the US, at $299 - Units sold: About 24.65 million worldwide; production ended in 2006 - Status today: Discontinued; its Xbox Live service was switched off on April 15, 2010 - Owner today: Microsoft, which still runs the Xbox brand - Price today: About $60 to $100 used in 2026, near $275 complete in box ## How four DirectX engineers talked Microsoft into a console In 1998, four members of Microsoft's DirectX team, Seamus Blackley, Kevin Bachus, Ted Hase and Otto Berkes, started building a games machine out of PC parts after hours. Sony was about to ship the PlayStation 2, and Sony was selling it as a living room computer rather than a toy. The obstacle was internal. Microsoft in 1999 was a Windows company, and a rival team out of the WebTV group (https://404memoryfound.com/posts/what-happened-to-webtv-internet-television.html) wanted a cheap set-top box running a cut-down copy of Windows. The DirectX plan called for custom silicon, a built-in hard drive and an Ethernet port, none of which looked like Windows at all. So the engineers described it as a Windows device anyway. "We absolutely, intentionally hoodwinked [Bill Gates]," Blackley told GamesRadar in 2021, saying he had told Gates in writing, many times, that the team was building a Windows console. By the time senior management understood what the hardware actually was, the project was too far along to stop. Microsoft had already tried to buy Nintendo (https://404memoryfound.com/posts/microsoft-tried-buy-nintendo.html) and been turned down, so building the box was the only route left into the living room. ## Why did Microsoft lose money on every Xbox it sold? The Xbox reached US stores on November 15, 2001 at $299, a year after the PlayStation 2 and three days before the Nintendo GameCube (https://404memoryfound.com/posts/why-nintendo-gamecube-lost-console-war.html). Inside the case was a 733 MHz Intel Pentium III, an Nvidia graphics chip, 64 MB of memory, a hard drive and an Ethernet port. No other console of that generation shipped with a hard drive and a network port as standard. Those parts cost more than the sticker. Digital Trends' history of the console puts the build cost near $425 a unit against the $299 price, which is a loss of more than $100 before a single game changed hands. The plan was the usual console plan: give away the hardware, take a licensing fee on every game sold for it. That plan needs volume, and the volume never arrived. The Xbox sold about 24.65 million units worldwide before Microsoft ended production in 2006, against more than 160 million for the PlayStation 2 (https://404memoryfound.com/posts/playstation-2-best-selling-console-ever.html), a lifetime figure Sony finally confirmed in 2024. Microsoft's losses on the first Xbox ran past $4 billion. ## What did Halo and Xbox Live actually change? Microsoft bought Bungie in June 2000 and pulled Halo, then in development for Mac and PC, onto the Xbox as a launch title. Halo: Combat Evolved went on to sell about 6.4 million copies on the console, which works out to roughly one Xbox owner in four buying the same game. Xbox Live followed on November 15, 2002, exactly one year after the console. Microsoft sold it as a $49.95 starter kit with a headset and a one-year subscription, and it was broadband only, with no dial-up option at a time when most American homes still had a modem. One account carried a single gamertag, one friends list and system-level voice chat across every game that supported it. The payoff was Halo 2 on November 9, 2004. NBC News reported that week that the game sold 2.4 million copies and took $125 million in its first 24 hours, more than the opening day of any film to that point. Friends lists, party chat, matchmaking and downloadable map packs all became normal because they worked here first. ## Why did the Xbox fail in Japan? The Xbox launched in Japan on February 22, 2002 and sold 123,929 units over its opening weekend, a respectable start that fell apart quickly. Buyers reported that the drive was scratching their discs, and some Japanese shops pulled the console off sale while Microsoft sorted it out. Sales never recovered. Microsoft had shipped 190,092 units in Japan by the end of March 2002, and the console finished its life there under 500,000 units, a rounding error next to what Sony and Nintendo were doing at home. The reasons stack up: a machine the size of a small suitcase, the original controller that Japanese players found unusable, and a library built around American shooters and sports games. Microsoft spent heavily on Japanese studios and still could not move the numbers. The Xbox brand has never recovered in that market, and every Xbox since has sold worse in Japan than the one before it. ## Can you still play an original Xbox online in 2026? Not on Microsoft's servers. Larry "Major Nelson" Hryb announced the end of the service on February 5, 2010, saying the change "will allow us to continue evolving the LIVE service with new features and experiences that fully harness the power of Xbox 360." Xbox Live for the original console went dark on April 15, 2010, taking Halo 2 matchmaking with it. It came back without Microsoft. Insignia, a free fan-built reverse engineering of the old Xbox Live servers, has been running since 2020 and reopened Halo 2 matchmaking in a public beta in March 2024. As of August 2026 it had more than 34,000 registered users and supported 199 of the 381 games that used Xbox Live, with friends lists, clans, voice chat and leaderboards working again. Offline, the disc in your hand still boots. Microsoft's backward compatibility library also lists 39 original Xbox titles that run on Xbox One and Xbox Series X and Series S, a list frozen in November 2021 when licensing and technical limits ended new additions. Microsoft said in 2026 that the program will restart for the brand's 25th anniversary. ## Where to find one today A working console with a controller and cables runs about $60 to $100 in 2026, according to BGR's survey of the resale market this year. Complete boxed systems average near $275, and sealed units go for several times that. Check that the machine has not sat unused for a decade with its internal clock capacitor leaking, which is the failure that kills these boards. For the parts that wear out, the S-type controllers and the HD-capable video cables the console needs on a modern television, Stone Age Gamer stocks original Xbox gear (https://stoneagegamer.com/xbox/xbox/). ## Frequently Asked Questions ### How much is an original Xbox worth in 2026? A used original Xbox from 2001 sells for roughly $60 to $100 in 2026 with a controller and cables, and complete boxed examples average close to $275, according to BGR's 2026 look at the resale market. Sealed consoles are collector items and go for several hundred dollars or more. ### Does Xbox Live still work on the original Xbox? Microsoft's Xbox Live service for the original Xbox was switched off on April 15, 2010, and it has not returned. Online play on the 2001 console now runs on Insignia, a free fan-run replacement that supported 199 of the 381 Xbox Live games as of August 2026, including Halo 2. ### How many original Xbox consoles were sold? The original Xbox sold about 24.65 million units worldwide between its US launch on November 15, 2001 and the end of production in 2006. That is a fraction of the PlayStation 2's confirmed lifetime total of more than 160 million, and Microsoft lost over $4 billion on the console. **Sources:** - Xbox (console), Wikipedia: launch, hardware, sales and Japan figures: https://en.wikipedia.org/wiki/Xbox_(console) - Xbox Wire: Xbox LIVE being discontinued for Original Xbox consoles and games, February 5, 2010: https://news.xbox.com/en-us/2010/02/05/xbox-live-being-discontinued-for-original-xbox-consoles-and-games/ - NBC News: 'Halo 2' reports $125 million in first-day sales, November 2004: https://www.nbcnews.com/id/wbna6456214 - GamesRadar: Seamus Blackley on tricking Bill Gates into approving the original Xbox: https://www.gamesradar.com/seamus-blackley-on-tricking-bill-gates-and-the-jedi-mind-tricks-that-were-needed-to-make-the-original-xbox/ - Insignia: the fan-run Xbox Live replacement for the original Xbox: https://insignia.live/ --- # Is AOL Time Warner Still a Company? What Is Left URL: https://404memoryfound.com/posts/aol-time-warner-merger-worst-deal-in-history.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-31 Updated: 2026-09-18 Topics: Business Blunders, Internet Culture, Money & Tech **Summary:** AOL Time Warner is not a company any more. The business AOL and Time Warner formed in January 2001 dropped the AOL name in 2003, wrote down $99 billion in 2002 and spun AOL off in 2009. In 2026 AOL is being sold to the Italian software firm Bending Spoons for about $1.5 billion, while the old Time Warner studios and networks sit inside Warner Bros. Discovery. **Key facts:** - Announced: January 10, 2000, valued at about $182 billion - Status today: Dissolved. AOL was spun off in 2009 and the name is gone - Owner today: AOL: Bending Spoons. Time Warner assets: Warner Bros. Discovery - Worst year: A loss of about $99 billion in 2002 - Price today: AOL sold for about $1.5 billion in the deal agreed in October 2025 ## Why was a dial-up company able to buy Time Warner? On January 10, 2000, America Online announced it was acquiring Time Warner in a stock deal valued at about $182 billion, the largest merger anyone had assembled to that point. The arithmetic only worked because of the bubble. AOL had roughly 30 million dial-up subscribers and a market capitalization near $163 billion, more than the company that owned CNN, HBO, Warner Bros., Time magazine and the second largest cable system in the United States. Internet companies in 1999 and 2000 were priced on subscriber growth and page views rather than profit, and AOL was the largest of them. Steve Case, AOL's chairman, was holding the most valuable currency on the market, and he spent it on assets that produced real cash. That timing is the part of the deal that still looks smart. The Federal Trade Commission cleared the transaction in December 2000 and it closed in January 2001. AOL shareholders took 55 percent of the combined company and Time Warner shareholders took 45 percent. A company that mailed out free trial CDs now controlled a film studio, a cable news network and a magazine business. ## What was the broadband plan the deal depended on? Every dollar of that $182 billion rested on one assumption: AOL's dial-up customers would move to high-speed internet over Time Warner's cable lines and stay AOL customers while they did it. Time Warner owned Road Runner, one of the largest cable internet services in the country. On a slide the handoff looked clean. AOL brings the audience and the brand, Time Warner Cable brings the wire, and the subscriber never has to leave. It never happened. The cable division already sold internet access directly and carried its own revenue targets. Putting AOL in front of those customers meant handing a middleman the relationship and a share of the margin, so the cable executives had no reason to cooperate, and nobody above them forced the issue. Broadband arrived anyway, through cable operators and telephone DSL, and the reason to pay AOL every month left with it. Dial-up subscribers did not convert to an AOL-branded cable product. They canceled. The synergy that justified the price required divisions of one company to act against their own numbers, which is close to the standard reason large mergers disappoint. ## How did one company lose $99 billion in a single year? Goodwill did it. When AOL paid for Time Warner in overvalued stock, the amount above the book value of the assets sat on the balance sheet as goodwill, and accounting rules force a write-down once that premium stops being defensible. By 2002 it was not defensible. The dot-com collapse (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html) had taken the market with it, AOL's advertising revenue had stopped growing, and subscribers were leaving faster than the company could replace them. Time Warner registered a loss of $99 billion for 2002, the largest annual loss an American company had reported at that point. No cash left the building that year. The write-down was the company stating in its own filings that what it bought in 2000 was worth roughly $99 billion less than what it paid, which is a more exact definition of a bad deal than any headline. The damage reached individuals too. The Hollywood Reporter has put Ted Turner's losses from the merger at about $8 billion, roughly 80 percent of his wealth; Turner had sold Turner Broadcasting into Time Warner in 1995 and became the combined company's largest individual shareholder. The AOL name came off the corporate letterhead in 2003. ## Who owns AOL and Time Warner in 2026? Neither name describes a working company any more, and the two halves went in opposite directions after the split. Time Warner spun AOL off as an independent public company in 2009. Verizon bought AOL in 2015 for $4.4 billion and folded it in with Yahoo. Apollo Global Management bought that whole media unit from Verizon for $5 billion in 2021 and ran it under the Yahoo name (https://404memoryfound.com/posts/who-owns-yahoo-now.html). In October 2025 Apollo agreed to sell AOL to Bending Spoons, an Italian software company, for about $1.5 billion. What the buyer gets is a news portal, a games page and free email accounts (https://404memoryfound.com/posts/does-aol-still-exist-today.html), not the subscription business the merger was built on. The Time Warner side changed hands just as often. AT&T bought it in 2018 and renamed it WarnerMedia, then spun it back out in 2022 to merge with Discovery as Warner Bros. Discovery, which is where CNN, HBO and the Warner Bros. studio sit today. Paramount Skydance agreed in 2026 to buy Warner Bros. Discovery at $31.00 a share, a transaction worth about $111 billion. A group of state attorneys general sued to block it, and Paramount agreed in July 2026 not to close before a trial is held or June 1, 2027, whichever comes first. ## What did the executives say once it was over? Gerald Levin, the Time Warner chief executive who drove the deal, went on CNBC in January 2010, ten years after the announcement, and took the blame on camera. "I presided over the worst deal of the century, apparently," he said. He went further in the same appearance. "I'm really very sorry about the pain and suffering and loss that was caused. I take responsibility," Levin told CNBC, then listed the people he said were not at fault: the board, his colleagues at Time Warner, the bankers and lawyers, and Steve Case. Jeff Bewkes, running Time Warner by 2010, was blunter about the deal itself. He called the January 2000 merger "the biggest mistake in corporate history" in an interview with the Daily Telegraph that year, as Forbes reported. Bewkes had already acted on that opinion rather than just voicing it. Getting AOL out of the company was among his first significant moves as chief executive, and the spinoff closed in 2009. ## Was it really the worst deal in corporate history? On the two measures that travel best, it holds the title. No merger has combined a bigger announced price with a faster collapse in the value of what was bought, and the man who signed it and the man who cleaned it up both said so in public, by name, on the record. The more useful reading is narrower. AOL did not fail because the internet failed. It failed because the company's product was access, access was about to get cheaper and faster from other suppliers, and the merger's entire defense against that was a cable division with no incentive to help. The same shape keeps returning. A company with an inflated share price buys a business with real assets, promises that the two will fit together, and then discovers that the fit required cooperation nobody was paid to provide. The assets usually survive the parent. The promise rarely does. What survived is the case study. Business schools still teach the January 2000 announcement, and the $99 billion write-down of 2002 is still the number that ends the argument. ## Frequently Asked Questions ### Does AOL Time Warner still exist? No. AOL Time Warner dropped the AOL from its name in 2003, and Time Warner spun AOL off as a separate public company in 2009. As of 2026, AOL is being sold to Bending Spoons for about $1.5 billion, and the former Time Warner networks and studios belong to Warner Bros. Discovery, which Paramount Skydance has agreed to buy. ### How much money did the AOL Time Warner merger lose? Time Warner recorded a loss of about $99 billion for 2002, two years after the AOL Time Warner merger was announced, the largest annual loss an American company had reported at that point. Most of it was a write-down of goodwill created by the January 2000 deal, which was valued at about $182 billion. ### Who owns AOL now? Apollo Global Management agreed in October 2025 to sell AOL to Bending Spoons, an Italian software company, for about $1.5 billion. Apollo had bought AOL and Yahoo from Verizon for $5 billion in 2021, and Verizon had bought AOL for $4.4 billion in 2015, six years after Time Warner spun it off as an independent company. **Sources:** - AOL-Time Warner merger announced, January 10, 2000 (HISTORY): https://www.history.com/this-day-in-history/january-10/aol-time-warner-formed - CNBC excerpts: David Faber talks with former AOL Time Warner CEO Gerald Levin (CNBC, 2010): https://www.cnbc.com/2010/01/06/cnbc-excerpts-cnbcs-david-faber-talks-with-former-aoltime-warner-ceo-gerald-levin-in-a-cnbc-original-marriage-from-hell-the-breakup-of-aol-time-warner-tonight-wednesday-january-6th-at-10pm-etpt.html - The Biggest Mistake in Corporate History (Forbes, 2010): https://www.forbes.com/sites/trevorbutterworth/2010/09/30/the-biggest-mistake-in-corporate-history/ - AOL to be sold to Bending Spoons for roughly $1.5B (Axios, 2025): https://www.axios.com/2025/10/29/aol-bending-spoons-deal - Paramount agrees to delay WBD acquisition to as late as June 2027 amid legal challenge (CNBC, 2026): https://www.cnbc.com/2026/07/24/paramount-wbd-merger-delay.html --- # Are LAN Parties Still Around? QuakeCon Sold Out URL: https://404memoryfound.com/posts/golden-age-of-lan-parties.html Author: Marcus Vale (404 Memory Found) Published: 2026-03-31 Updated: 2026-09-18 Topics: Gaming, Hardware **Summary:** LAN parties are still around in 2026, but the biggest one stopped asking people to bring the computers. QuakeCon ran its 30th anniversary bring-your-own-computer hall at the Gaylord Texan Resort in Grapevine, Texas from August 6 to 9, 2026, and its BYOC seats sold out. DreamHack announced on May 26, 2026 that it was ending its own BYOC hall after 32 years, because fewer than one ticket holder in ten still brought a machine. **Key facts:** - First LAN convention: August 1996, La Quinta Inn, Garland, Texas, about 40 people - Biggest ever: DreamHack Winter 2013, 22,810 visitors (Guinness World Records) - Status today: Still running: QuakeCon held its BYOC hall August 6 to 9, 2026 - Price today: QuakeCon 2026 tickets ran $15 for a fan pass to $400 for Elite BYOC - What changed: DreamHack ended its BYOC hall on May 26, 2026, after 32 years ## Why dial-up made hauling a desktop PC worth it The reason people drove computers across town was arithmetic, not sentiment. In early 2000, the Pew Research Center counted 3 percent of American adults with a broadband connection at home. Dial-up was what nearly everyone else used, and it peaked at 41 percent of adults in April 2001 before sliding back down to 3 percent by August 2011. A 56k modem could hold a deathmatch together for a handful of players. It could not hold sixteen, and it could do nothing about the lag that arrived the moment someone in the house picked up the phone. Anyone curious about what those dial-up speeds actually felt like (https://404memoryfound.com/posts/dialup-speed-then-vs-now.html) can run the numbers against a modern connection and wince. A local network erased the problem in one move. Machines plugged into a hub or a switch answered each other in a fraction of the time a phone line needed, and file copies ran at speeds no home connection matched for another decade. The catch was physical. The network could not travel to the players, so the players moved the hardware: tower cases, CRT monitors heavy enough to want two people on them, and a crate of power strips, patch cables and adapters that never came home in the order it left. ## How a hotel meeting room in Garland became QuakeCon The first QuakeCon took place in August 1996 in a conference room at a La Quinta Inn in Garland, Texas, minutes from id Software's offices. About 40 people came. Jim Elson, known online as H2H and connected to the Dallas gaming scene, and Yossarian Holmberg, a computer consultant from Waterloo, Ontario who went by yossman, had floated the idea to regulars of the #quake IRC (https://404memoryfound.com/posts/what-happened-to-irc-chat.html) channel, and the regulars said yes. What they built was improvised: a small network strung across a hotel meeting room, mostly Quake (https://404memoryfound.com/posts/what-happened-to-quake-online-fps.html) and Doom (https://404memoryfound.com/posts/how-doom-changed-pc-gaming-internet.html), and a tournament whose prizes were T-shirts. On the eve of the last day the id Software team turned up to say hello, which is the moment a weekend among strangers became an annual obligation. Then it compounded. QuakeCon 1997 drew an estimated 650 people to a Holiday Inn in Plano, Texas. In 1998 it counted 800 attendees, 300 of whom hauled a computer. In 1999 it reached 1,100 attendees and 500 BYOC seats. By 2002 the event passed 3,200 attendees with 1,300 BYOC seats, which is roughly where a LAN party stops being a party and becomes logistics: power distribution, cooling, and a network crew that works while everyone else plays. ## How big did LAN parties actually get? Big enough to need a record category. Guinness World Records lists the largest LAN party as DreamHack Winter 2013 in Sweden, with 22,810 visitors, and no gathering has taken that title back in the years since. DreamHack is the Swedish half of the story. It ran a bring-your-own-computer hall for 32 years, a tradition that traces back to 1994, and it grew from a gathering of schoolmates into a festival that filled exhibition halls with rows of tables, borrowed extension cords and the heat of several thousand machines. Scale changed what the weekend felt like. At that size the network itself was the attraction: a closed pipe where a patch, a map pack or a mod arrived in seconds, at a moment when downloading the same file at home took a night. Players slept in shifts on the floor beside their rigs because a seat left empty was a seat someone else would take. Tournaments for Counter-Strike (https://404memoryfound.com/posts/what-happened-to-counter-strike-half-life-mod.html), StarCraft (https://404memoryfound.com/posts/what-happened-to-starcraft-birth-of-esports.html) and Quake drew crowds who stood behind the players to watch, which is the shape competitive gaming still uses today. ## Is DreamHack still holding LAN parties in 2026? Not the part that made it a LAN party. On May 26, 2026, DreamHack published a notice titled "Goodbye, BYOC" that ended the bring-your-own-computer hall across its festivals. The reason it gave was attendance. The number of visitors bringing their own rigs, DreamHack wrote, "has steadily decreased year over year, making up less than 10% of our total ticket holders in 2025." The festival itself continues. DreamHack said "The ocean of screens isn't going anywhere," and the space the BYOC hall used is going to free play machines, tournaments and exhibitor stands instead. Read plainly, that is a brand deciding that most of its audience now wants to show up empty-handed and sit down at a computer somebody else set up. It is a fair reading of the market. A visitor who brings nothing can fly in. A visitor who brings a desktop is committing to a car, a hotel and an hour of cable management at each end of the trip. After 32 years, the second group had shrunk to under a tenth of the room. ## What a LAN party costs and looks like in 2026 QuakeCon went the other direction. Its 30th anniversary event ran from August 6 to 9, 2026 at the Gaylord Texan Resort and Convention Center in Grapevine, Texas, with the BYOC hall open around the clock for tournaments, free play, tabletop games and meetups. Tickets ran from $15 for a fan pass to $400 for an Elite BYOC seat, and the BYOC seats sold out. QuakeCon describes the appeal in its own words: "For three decades, QuakeCon has been the place where friendships are forged over frag counts, sleep schedules disappear, and the glow of RGB lights up the night." That is the same pitch the 40 people in Garland responded to in 1996, with better carpet. The format survived a scare. QuakeCon ran online only in 2020, 2021 and 2022 before the in-person BYOC hall returned in 2023 at the same Grapevine venue. Below it sits a layer of volunteer-run events, including LANFest and LAN All Night, that still book halls and sell bring-your-own-computer seats. What has changed is the cargo. The tower and the CRT have mostly given way to a gaming laptop and a backpack, which is why the weekend is easier to attend and harder to build an identity around. ## Frequently Asked Questions ### Do LAN parties still exist in 2026? Yes, LAN parties still exist in 2026, though fewer of them require you to bring a computer. QuakeCon held its bring-your-own-computer hall at the Gaylord Texan Resort in Grapevine, Texas from August 6 to 9, 2026 and sold out its BYOC seats, while DreamHack ended its own BYOC hall on May 26, 2026 after 32 years and kept the rest of its festival running. ### What was the biggest LAN party ever? DreamHack Winter 2013 in Sweden is the biggest LAN party ever recorded. Guinness World Records lists it as the largest LAN party, with 22,810 visitors, and that mark still stood in 2026, twelve years after the event. ### Why did LAN parties decline? LAN parties declined because home internet caught up with them. Only 3 percent of American adults had home broadband in early 2000, so a local network was the only practical way to run a large, low-latency game, and dial-up did not fall away until the 2010s. By 2025, fewer than 10 percent of DreamHack ticket holders still brought a computer to the show. **Sources:** - QuakeCon official site (Bethesda Softworks): https://quakecon.bethesda.net/en/ - QuakeCon history, QuakeWorld QWiki: https://www.quakeworld.nu/wiki/QuakeCon - DreamHack, 'Goodbye, BYOC', 26 May 2026: https://dreamhack.com/stockholm/updates/2026/05/26/goodbye-byoc/ - Guinness World Records, Largest LAN party: https://www.guinnessworldrecords.com/world-records/largest-lan-party - Pew Research Center, Trends in broadband adoption: https://www.pewresearch.org/internet/2010/08/11/trends-in-broadband-adoption/ --- # Is Kozmo.com Still Around? The Brand's Fate in 2026 URL: https://404memoryfound.com/posts/what-happened-to-kozmo-com-delivery.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-31 Updated: 2026-09-18 Topics: Business Blunders, Internet Culture **Summary:** Kozmo.com no longer exists as a delivery service. The company shut down on April 11, 2001, firing roughly 1,100 employees, and Los Angeles grocer Yummy Foods, LLC bought the name and trademark in January 2013. A 2018 relaunch as an online warehouse club did not survive either, and the kozmo.com domain has been dark since 2023. **Key facts:** - Status today: Defunct since April 11, 2001; the domain is dark - Owner today: Yummy Foods, LLC (Yummy.com), Los Angeles, since January 2013 - Founded: March 1998 in New York by Joseph Park and Yong Kang - Money raised: More than $250 million; a $280 million loss cited at closing - 1999 results: $3.5 million revenue, $26.3 million net loss (SEC Form S-1) ## What did Kozmo.com actually deliver? Kozmo.com launched in New York in March 1998. Its founders, Joseph Park and Yong Kang, were investment bankers in their twenties who quit banking to build a courier service for the web. The offer was one thing, repeated everywhere: free delivery inside an hour, no minimum order, no fee. The catalog was built for impulse buying, with videos, games, DVDs, music, magazines, books, food, basics and Starbucks coffee. A customer could order a single rental and a bag of candy at 9pm and a messenger on a bike would bring both at no extra cost. The network behind that promise was human rather than algorithmic. Kozmo.com ran its markets out of small distribution centers, typically about 10,000 square feet each according to the company's Form S-1 filed with the Securities and Exchange Commission in March 2000. There were no smartphones, no GPS routing and no live tracking. Orders came in through a desktop web browser because that was the only screen anyone had. By mid-2000 the service covered more than ten cities, among them New York, San Francisco, Boston, Seattle, Chicago, Houston, Atlanta, Portland, Washington D.C. and Los Angeles, with more than 3,300 people on the payroll. ## How much money did Kozmo.com burn? The company's own filing is the clearest record of the problem. The March 2000 Form S-1, prepared for a $150 million public offering, reported 1999 revenue of $3.5 million against a net loss of $26.3 million. That is roughly seven and a half dollars lost for every dollar taken in. Money kept arriving anyway. Amazon bought a 32 percent stake with a $60 million investment in 2000, a vote of confidence from the one internet retailer that survived the crash intact (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html). The largest single commitment was not a delivery expense at all. In February 2000 Kozmo.com signed a five-year co-marketing agreement with Starbucks worth $150 million, which placed Kozmo drop boxes for video returns in up to 500 coffee shops. Kozmo.com paid about $15 million into that deal before ending it in March 2001. By the summer of 2000 the company was spending more than $30 million a month. Total funding passed $250 million. When the doors closed, management cited a $280 million loss. The arithmetic of free one-hour delivery on a $4 rental never changed: volume made the hole deeper rather than shallower, which is the trap that also swallowed Webvan and its automated grocery warehouses (https://404memoryfound.com/posts/what-happened-to-webvan-grocery-delivery.html) in the same year. ## What went wrong in the final twelve months? The plan was to go public, and the plan died first. On August 18, 2000, Kozmo.com withdrew its $150 million offering. The Nasdaq had peaked on March 10, 2000, and investors who had funded growth without profit were no longer interested in the pitch. Plans to reach as many as 30 cities were shelved. What followed was retreat. In December 2000 the company placed a $1.99 delivery fee on orders under $30 while keeping delivery free above that line, which was a plain admission that the founding promise did not pay for itself. Markets were cut and staff were let go in waves. Leadership changed too. Joseph Park resigned as chief executive during 2000 and left the chairman's post in January 2001. Gerry Burdo took over a company with no public market, no new private money and a burn rate it could not reverse fast enough. On April 11, 2001, Kozmo.com stopped all operations, shut the website and dismissed about 1,100 employees. Just Food reported that the timing was chosen so the company could still pay severance to most of them. Burdo's statement that week was defensive and precise: "Given more time and more hospitable market conditions, Kozmo would have succeeded in rounding the corner and would have continued to grow." He also pointed to "some decisions made early in the company's development," which put part of the blame on the founders. ## Who owns the Kozmo.com name today? The delivery company has been gone since 2001, but the name kept trading. In January 2013 the Kozmo.com domain and trademark were bought by Yummy Foods, LLC, the Los Angeles company behind the grocery service Yummy.com, which still delivers groceries in about 30 minutes across parts of Los Angeles in 2026. None of the original founders or staff went with the brand. Yummy Foods relaunched Kozmo in March 2018, and not as a one-hour courier. The new version was an online warehouse club selling in bulk, available only in Los Angeles at first, with plans to expand that year. That second life did not hold. The Kozmo.com site went offline in 2023, and as of September 2026 the domain does not resolve at all. There is no app, no store and no delivery service operating under the name. Yummy.com, the parent, is the part that survived. Barnaby Montgomery, who runs Yummy.com and bought the Kozmo brand, was blunt with Grocery Dive in 2019 about what he had acquired. The original, he said, "had a flawed assortment of things you didn't really need." ## Why did DoorDash work when Kozmo.com did not? Kozmo.com was right about demand and wrong about everything that makes demand profitable. People did want a movie and snacks brought to the door on a weeknight. The company simply could not serve that want for less than it charged, which was nothing. The delivery platforms that came later started from the opposite end. They charge delivery fees, raise prices at peak times, set minimum order sizes and take a cut from the merchant rather than owning the inventory. They pay couriers per job instead of carrying thousands of salaried staff across eighteen locations. The technology gap matters just as much. A smartphone turns an order into three taps with the address already filled in, and dispatch software clusters orders by geography and routes them in real time. Kozmo.com had dispatchers, radios and a checkout page built for a 56k modem. The lesson is not that delivery was a bad idea. It is that a real market and a viable business are different things, the same gap that sank Pets.com and its sock puppet (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html) while the underlying demand for online pet supplies kept growing. ## Frequently Asked Questions ### What happened to Kozmo.com? Kozmo.com, the New York one-hour delivery company founded in March 1998, ceased all operations on April 11, 2001 and dismissed roughly 1,100 employees. It had raised more than $250 million and cited a $280 million loss when it closed, after withdrawing a $150 million public offering in August 2000. ### Why did Kozmo.com fail? Kozmo.com failed because free one-hour delivery with no minimum order lost money on nearly every transaction. Its Form S-1 reported 1999 revenue of $3.5 million against a $26.3 million net loss, and by mid-2000 the company was spending more than $30 million a month. When the public markets closed in 2000, there was no funding left to cover the gap. ### Who owns Kozmo.com now? Yummy Foods, LLC of Los Angeles, the company behind the grocery delivery service Yummy.com, bought the Kozmo.com domain and trademark in January 2013 and relaunched the brand as an online warehouse club in March 2018. That relaunch has since ended, the site went offline in 2023, and the kozmo.com domain does not resolve in 2026. **Sources:** - Kozmo.com Inc., Form S-1, U.S. Securities and Exchange Commission (2000): https://www.sec.gov/Archives/edgar/data/0001075749/000091205700012562/0000912057-00-012562.txt - USA: Closure of Kozmo.com, CEO blames slump in market and earlier decisions, Just Food (2001): https://www.just-food.com/news/usa-closure-of-kozmo-com-ceo-blames-slump-in-market-and-earlier-decisions/ - Kozmo.com withdraws its $150 million IPO, CNN Money (2000): https://money.cnn.com/2000/08/18/deals/kozmo/ - Inside Kozmo.com's online grocery comeback, Grocery Dive (2019): https://www.grocerydive.com/news/inside-kozmocoms-online-grocery-comeback/552949/ - Kozmo.com is back from the dead, kind of, TechCrunch (2018): https://techcrunch.com/2018/03/21/kozmo-com-is-back-from-the-dead-kind-of/ --- # Is Circuit City Still in Business? Zero Stores Left URL: https://404memoryfound.com/posts/what-happened-to-circuit-city-electronics.html Author: Marcus Vale (404 Memory Found) Published: 2026-03-31 Updated: 2026-09-18 Topics: Hardware, Business Blunders **Summary:** Circuit City as shoppers knew it is gone. The chain filed for Chapter 11 on November 10, 2008 and liquidated its last 567 US stores on March 8, 2009, ending about 34,000 jobs. The name still exists as an online-only brand owned by Circuit City Corporation, run by Ronny Shmoel, who bought the trademark from Systemax in 2016, but there are no Circuit City stores. **Key facts:** - Status today: Online-only brand. No stores since March 8, 2009 - Owner today: Circuit City Corporation, led by Ronny Shmoel (bought the brand from Systemax in 2016) - Stores left: Zero. The last 567 closed in March 2009 - Founded: 1949 in Richmond, Virginia, as Wards, by Samuel Wurtzel - Peak: About 700 stores and $12.43 billion in net sales in fiscal 2007 ## Who owns Circuit City today? The Circuit City name belongs to Circuit City Corporation, a privately held company run by retail investor Ronny Shmoel, who bought the brand and the trademark rights from Systemax in 2016. Systemax had picked up the name and the CircuitCity.com domain for about $14 million in the 2009 bankruptcy sale and ran it as an online storefront for several years before selling it on. Shmoel relaunched the brand online in 2018. There is no chain behind it. Circuit City in 2026 is a website and a licensing play, not a retailer with its own buildings, and not one of the 567 stores that closed in 2009 has reopened under the name. In November 2023 the company announced a Series A funding round and a shop-in-shop program it calls "Powered by Circuit City," with JCPenney named as the first partner. A month later, its Delaware holding company filed paperwork indicating it wanted to raise $25 million. The stated plan has been the same since 2016: kiosks first, then boutiques inside other retailers, then freestanding stores of its own. The freestanding stores have not arrived. ## How did a barber's chair start Circuit City? In 1949 Samuel Wurtzel, a New York importer, was on vacation in Richmond, Virginia, and went in for a haircut. The barber mentioned that the South's first commercial television station, WTVR, was about to go on the air in the city. Wurtzel worked out that people who could suddenly receive local broadcasts would need sets to watch them on. He moved his family to Richmond and opened a store called Wards. The name was an acronym of the family: W for Wurtzel, A for his son Alan, R for his wife Ruth, D for his son David, and S for Samuel himself. Alan Wurtzel took over as chief executive in 1972 and spent the decade building something larger than an appliance shop. The company pioneered the electronics superstore: a big showroom, deep inventory, and salespeople paid on commission who could walk a customer through the difference between two television sets. The format worked well enough that the chain took the Circuit City name in 1984 and spent the next fifteen years as the largest specialty electronics retailer in the United States. ## Why did Best Buy pass Circuit City? Best Buy introduced a format it called Concept II in 1989. It removed commissioned salespeople, moved the stock onto the sales floor, and let shoppers pull a box off the rack and carry it to the register. Suppliers disliked it. Shoppers did not. Circuit City kept its commissioned floor staff, which was a real advantage for a decade and then became a cost line that Best Buy simply did not carry. The second decision was worse. In July 2000 Circuit City walked away from major appliances to free up floor space for small electronics. It had been the second-largest appliance retailer in the country in 1999, behind only Sears (https://404memoryfound.com/posts/who-owns-sears-now-stores-left.html), with close to $1.6 billion in appliance sales that year. The housing boom that followed went to Home Depot, Lowe's and eventually Best Buy. None of this looked fatal at the time. Jim Collins put Circuit City in "Good to Great" in 2001 as one of eleven companies that made the leap from good to great, noting that between 1982 and 1999 it delivered "cumulative stock returns 22 times better than the market." By fiscal 2008, Best Buy was reporting roughly $40 billion in revenue against Circuit City's $11.6 billion. ## What happened when Circuit City fired 3,400 workers? On March 28, 2007, Circuit City dismissed 3,400 store employees for being paid too much. Not the weakest performers: the highest earners, which in a commissioned environment meant the most experienced people on the floor, roughly 9 percent of an in-store workforce of about 40,000. They received severance and were told they could reapply after ten weeks at the lower market rate. The company called it a wage management initiative. "All companies at one time or another need to go through and make sure their cost structure works with market conditions," spokesman Jim Babb told The Washington Post in 2007. It did not work. By May 1, 2007, the same paper was reporting that analysts considered the cuts to be backfiring, with large flat-panel and projection televisions singled out as the products that needed knowledgeable staff to sell. Deutsche Bank analyst Mike Baker said the labor change had hurt service levels and allowed Best Buy to "take share." The one thing Circuit City had that Best Buy did not was the person on the floor who knew the answer, and the company had just walked that person out of the building. ## When did Circuit City close its stores? Philip Schoonover, chief executive since 2006, resigned in September 2008 with the company unsold and losing money. Board member James Marcum took over as acting chief executive. On November 3, 2008, Circuit City said it would close 155 stores and cut 17 percent of its workforce. Seven days later it filed for Chapter 11 in the US Bankruptcy Court for the Eastern District of Virginia. No buyer materialized. On January 16, 2009, the court approved a full liquidation of the remaining 567 stores. "We are extremely disappointed by this outcome," Marcum said in a statement reported by CNNMoney that day. "We were unable to reach an agreement with our creditors and lenders to structure a going-concern transaction in the limited timeframe available." The last stores turned off the lights on March 8, 2009, and about 34,000 jobs went with them. Congress held a hearing on it that year under the title "Circuit City Unplugged: Why Did Chapter 11 Fail to Save 34,000 Jobs?" The short answer was that a retailer whose vendors and lenders stop extending credit has nothing left to restructure. Fewer than two years separated the March 2007 layoffs from the last day of trading. ## What is left of the chain in 2026? In the United States, nothing operational. The stores were re-leased to other retailers, and the long low buildings with the red trim are still easy to spot in strip malls that have changed tenants twice since. The Canadian side outlived the parent. InterTAN Canada ran the chain as The Source by Circuit City, and Bell Canada bought it as a going concern for C$135 million on July 1, 2009. Bell rebranded those stores as Best Buy Express in 2024, which means the last physical descendant of Circuit City now carries the name of the company that beat it. The collapse was not unique, only fast. Fry's Electronics (https://404memoryfound.com/posts/what-happened-to-frys-electronics.html) shut all 31 of its stores in 2021, and RadioShack (https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html) went through two bankruptcies before ending up as a name licensed to other sellers. Circuit City got there first and got there hardest: the largest specialty electronics chain in America through the 1990s, with about 700 stores and $12.43 billion in net sales in fiscal 2007, and then nothing at all two years later. ## Frequently Asked Questions ### When did Circuit City go out of business? Circuit City filed for Chapter 11 bankruptcy protection on November 10, 2008, and a court approved the liquidation of its remaining 567 US stores on January 16, 2009. The last Circuit City stores closed on March 8, 2009, ending about 34,000 jobs. ### Is Circuit City coming back? Circuit City has been an online-only brand since its relaunch in 2018, owned by Circuit City Corporation under Ronny Shmoel, who bought the name from Systemax in 2016. The company announced a shop-in-shop program called "Powered by Circuit City" in November 2023 with JCPenney as its first partner, but no freestanding Circuit City store has opened anywhere since the chain liquidated in 2009. ### Why did Circuit City fail? Circuit City lost ground to Best Buy after quitting major appliances in July 2000, then fired 3,400 of its highest-paid store staff on March 28, 2007, a cut that analysts told The Washington Post that May was backfiring. The credit crisis of 2008 finished a company that had already given away its main advantage, and Circuit City filed for bankruptcy that November. **Sources:** - Circuit City Cuts 3,400 'Overpaid' Workers, The Washington Post (2007): https://www.washingtonpost.com/wp-dyn/content/article/2007/03/28/AR2007032802185.html - Circuit City's Job Cuts Backfiring, Analysts Say, The Washington Post (2007): https://www.washingtonpost.com/wp-dyn/content/article/2007/05/01/AR2007050101623.html - Circuit City seeking to liquidate, CNNMoney (2009): https://money.cnn.com/2009/01/16/news/companies/circuit_city/index.htm - Circuit City Unplugged: Why Did Chapter 11 Fail to Save 34,000 Jobs? US House Judiciary Committee hearing (2009): https://www.govinfo.gov/content/pkg/CHRG-111hhrg47924/html/CHRG-111hhrg47924.htm - Circuit City Announces Series A Funding Round with Plans for 'Powered by Circuit City' Partnerships, Business Wire (2023): https://www.businesswire.com/news/home/20231114272070/en/Circuit-City-Announces-Series-A-Funding-Round-with-Plans-for-Exclusive-Powered-by-Circuit-City-Partnerships --- # Did Amazon Almost Go Bankrupt in the Dot-Com Crash? URL: https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-31 Updated: 2026-09-18 Topics: Business Blunders, Internet Culture, Money & Tech **Summary:** Amazon came close to a funding crisis in 2000 but never filed for bankruptcy. Its shares fell more than 90% from their 1999 high, and a Lehman Brothers credit analyst predicted it would run out of cash within four quarters, yet a $672 million bond sold to European investors in February 2000 carried the company to its first profitable quarter at the end of 2001. Amazon still trades on the Nasdaq and its market value passed $3 trillion for the first time in August 2026. **Key facts:** - Status today: Still public as AMZN; market value passed $3 trillion in August 2026 - Run by today: Andy Jassy, the executive who built Amazon Web Services - Worst of the crash: Shares fell more than 90%, from a 1999 high near $113 to $5.51 in late 2001 - The lifeline: $672 million convertible bond sold to European investors, February 2000 - First profitable quarter: Q4 2001: $5 million net income, one cent a share, on $1.12 billion of sales ## Why Amazon looked like every other doomed dot-com Between 1995 and 2000, venture money poured into anything with a .com in the name, and Amazon fit the pattern exactly. Jeff Bezos founded the company in 1994, took it public on the Nasdaq on May 15, 1997 at $18 a share, and then spent three years losing money on purpose. The internal mantra was "get big fast." Amazon built warehouses, hired hard, and pushed out of books into music, DVDs, electronics and toys. Sales reached $2.76 billion in 2000, up from $1.64 billion in 1999, and the losses grew alongside them. That was also the strategy at Pets.com (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html), at Webvan (https://404memoryfound.com/posts/what-happened-to-webvan-grocery-delivery.html) and at Kozmo.com (https://404memoryfound.com/posts/what-happened-to-kozmo-com-delivery.html). All three are gone. From the outside in early 2000, nothing visibly separated Amazon's cash burn from theirs. Then the market turned. The Nasdaq Composite peaked on March 10, 2000 and did not find a bottom until October 2002. Amazon's share price went from a 1999 high near $113 to $5.51 in late 2001, a fall of more than 90%. ## The $672 million bond that landed weeks before the crash In February 2000, a month before the Nasdaq peaked, Amazon sold $672 million of convertible subordinated notes to European investors. The coupon was 6.9% and the conversion terms were loose, both of which were worse for Amazon than what it had paid on earlier debt. The market was already hardening. The timing mattered more than the price. After March 2000, the window for a loss-making internet company to raise several hundred million dollars closed almost completely. Companies that had not already raised did not raise. The money showed up on the balance sheet precisely when it was needed. Amazon ended 2000 with $1.1 billion in cash and marketable securities, up from $706 million at the end of 1999, and Bezos credited the increase to the early 2000 euroconvert financing in his letter to shareholders. Bezos made a long run of good decisions in this period. This one also involved luck. Amazon extended its runway at close to the last moment the market was willing to fund it. ## What did Lehman Brothers analyst Ravi Suria get wrong? On June 22, 2000, Ravi Suria, a convertible bond analyst at Lehman Brothers, published a note saying Amazon's credit was "extremely weak and deteriorating" and that the company would run out of cash within four quarters. The San Francisco Chronicle called the report a must-read for anyone holding the stock. Amazon shares fell 19% to 20% in a single session, depending on which account you read, and the convertible bonds fell with them. Suria was not careless. His arithmetic on the burn rate held up, and his central worry, that Amazon could not yet fund itself from its own operations, was the right worry to have in mid-2000. What he missed was the slope. Amazon's pro forma operating loss ran at 26% of sales in the fourth quarter of 1999. A year later it was 6%. The losses were shrinking against a growing business, and the February cash gave that trend enough time to finish. Suria was right about the fragility and wrong about the ending, which is roughly the verdict on dot-com analysis in 2000 generally. ## How Amazon reached its first profit in the fourth quarter of 2001 Bezos opened his 2000 letter to shareholders with one word. "Ouch. It's been a brutal year for many in the capital markets," he began, and then spent the rest of the letter arguing that the business was in better shape than the share price suggested. He had numbers for the claim. Amazon served 20 million customers in 2000, up from 14 million in 1999. He quoted Benjamin Graham on the market being a voting machine in the short term and a weighing machine in the long term, and added: "We're a company that wants to be weighed, and over time, we will be." At the same time the company cut. It set a public target of pro forma operating profitability by the fourth quarter of 2001 and hit it, reporting pro forma operating profit of $59 million for that quarter against a $60 million loss a year earlier, a swing of $119 million. On a straight accounting basis the win was tiny. Net income was $5 million, one cent a share, on $1.12 billion of sales. It was still the first profitable quarter in Amazon's history, and it arrived about six quarters after Suria said the cash would be gone. ## Is Amazon still in business in 2026, and who runs it? Yes. Amazon is still a public company, still trades on the Nasdaq under AMZN, and is now one of the most valuable companies in the world. Its market capitalization passed $3 trillion for the first time in August 2026. Jeff Bezos no longer runs it day to day. Andy Jassy, the executive who built Amazon Web Services, is chief executive, and the cloud division that did not exist during the crash now carries much of the company's profit. The retail business that nearly ran out of money in 2000 is only part of the story. The recovery was slow in a way that gets edited out of the retelling. Amazon shares did not reach new all-time highs again until late 2009, roughly eight years after the bottom. ## Why Pets.com and Webvan died while Amazon did not The bond bought Amazon time. What it did with the time is the rest of the answer. Books were a good fit for mail order: standardized, easy to ship, and impossible for any physical store to stock in full. CDs and DVDs behaved the same way, and so did most of what Amazon added next. Each new category ran through the same warehouses and the same software, which got cheaper per order as volume rose. Pets.com was shipping heavy bags of pet food that cost more to deliver than customers were paying. Webvan was building refrigerated grocery logistics for a market that would not support the cost for another two decades. Kozmo.com sent snacks across Manhattan by bicycle and lost money on close to every trip. None of that was a cash-flow problem that scale would fix. That was the business. Amazon's spending looked identical from the outside and was not. The dot-com collapse (https://404memoryfound.com/posts/what-happened-to-theglobe-com-dot-com-ipo.html) was not evidence that the internet did not work. It was evidence that most of the companies built on it had no path to positive unit economics, and that telling those two cases apart in real time is genuinely hard. ## Frequently Asked Questions ### Did Amazon almost go bankrupt? Amazon never filed for bankruptcy, but it came close enough to a funding crisis that the question was taken seriously. In June 2000 a Lehman Brothers credit analyst publicly predicted Amazon would run out of cash within four quarters, and a $672 million bond sold to European investors in February 2000 is the main reason it did not. ### How much did Amazon stock fall in the dot-com crash? Amazon shares fell more than 90% during the dot-com crash, from a 1999 high near $113 to $5.51 in late 2001. The stock did not return to new all-time highs until late 2009, about eight years after the bottom. ### When did Amazon first make a profit? Amazon reported its first profitable quarter in the fourth quarter of 2001: net income of $5 million, or one cent a share, on sales of $1.12 billion. Consistent profitability took several more years after that first one-cent quarter. **Sources:** - Amazon.com 2000 Letter to Shareholders: https://s2.q4cdn.com/299287126/files/doc_financials/annual/00ar_letter.pdf - Amazon.com Announces 4th Quarter Profit (January 2002): https://press.aboutamazon.com/2002/1/amazon-com-announces-4th-quarter-profit-exceeds-sales-and-profit-objectives-lower-prices-for-customers-drove-sales-and-profits - Amazon.com Form 10-K405 for fiscal year 2001, SEC EDGAR: https://www.sec.gov/Archives/edgar/data/1018724/000103221002000059/d10k405.htm - Scathing Report of Amazon Is a Must-Read for Stock Owners, San Francisco Chronicle, 2000: https://www.sfgate.com/business/networth/article/scathing-report-of-amazon-is-a-must-read-for-2750932.php - Amazon tops $3 trillion market cap, CNBC, August 2026: https://www.cnbc.com/2026/08/03/amazon-amzn-stock-market-cap-earnings.html --- # Is the PSP Still Supported? Sony's Handheld in 2026 URL: https://404memoryfound.com/posts/what-happened-to-sony-psp-handheld.html Author: Marcus Vale (404 Memory Found) Published: 2026-03-31 Updated: 2026-09-10 Topics: Gaming, Hardware **Summary:** The Sony PSP is not supported in 2026. Sony stopped making it, closed the handheld's store in July 2021, and now sells PSP downloads only through the PS3 and PS Vita stores, which close starting July 2027. Sony's last published sales figure for the PSP is more than 76.4 million units, against 154.02 million for the Nintendo DS. **Key facts:** - Launched: December 12, 2004 in Japan; March 24, 2005 in the US - US launch price: $250 for the PSP Value Pack (2005) - Units sold: More than 76.4 million (Sony, as of March 31, 2012) - Status today: Discontinued, no Sony support; PSP downloads run through the PS3 and PS Vita stores until July 2027 - Cost today: About $80 to $205 loose depending on model (September 2026) ## How many PSPs did Sony actually sell? Sony's corporate sales page still lists the PSP at "More than 76.4 million (As of March 31, 2012)" and has never replaced that line with a final lifetime total. Software on the platform stopped at more than 331.0 million units on the same date. Bigger numbers float around, usually somewhere near 80 million, but Sony has not published them, so 76.4 million is the figure that traces back to the company itself. Sony kept shipping PSPs after 2012 and simply stopped updating the page. The scale problem shows up when you put the handheld next to Sony's own hit. The same page credits the PlayStation 2 (https://404memoryfound.com/posts/playstation-2-best-selling-console-ever.html) with more than 160 million units. The PSP, Sony's most successful portable by a wide margin, did not reach half of that. None of which made the PSP a flop. It outsold every handheld anyone had launched against Nintendo up to that point, including Sega's Game Gear (https://404memoryfound.com/posts/what-happened-to-sega-game-gear-handheld.html). It just never got close to the machine it was built to beat. ## Why did the Nintendo DS win the handheld war? Nintendo's investor relations site puts lifetime Nintendo DS hardware sales at 154.02 million units as of June 30, 2026. That is roughly twice Sony's published PSP figure, from a console that launched in North America a few months earlier and cost a lot less. Price did most of the damage. When Sony confirmed the US launch, the Associated Press reported the PSP Value Pack at $250 while the Nintendo DS sold for $150. Nintendo had also already moved more than 2.8 million units worldwide by early February 2005, before the PSP reached an American store shelf. Sony was not really aiming at the same buyer. "It has gaming at its core, but it's not a gaming device. It's an entertainment device," Kaz Hirai, then president of Sony Computer Entertainment America, told NBC News in 2005. Analysts liked the logic. "When it comes to entertainment, Sony has advantages over other players in the market," P.J. McNealy of American Technology Research said in the same 2005 report. The trouble is that Nintendo was selling something people already understood. Two screens, a stylus, cheap batteries, and a software library aimed at anyone who had ever owned a Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html). Sony was selling a portable media platform in the same year the iPod (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html) became the default answer for portable media. ## What was UMD, and why did PSP movies fade? UMD stood for Universal Media Disc, a proprietary 1.8 GB optical disc Sony designed specifically for the PSP. Games shipped on it, and so did films. The first million US Value Packs came with a UMD copy of Spider-Man 2 from Sony Pictures, which tells you how central the movie plan was. Sony was open about wanting the format to spread. In February 2005 Hirai said the company was in discussions with other movie studios about supporting UMD for future releases. Studios tried it, then drifted away, and the discs became a novelty section that retailers quietly shrank. The clearest verdict came from Sony itself. When the PSP Go arrived, it had no UMD drive at all: everything was downloaded to internal flash storage. A hardware company does not remove its own format from its own product unless the format has stopped earning its place. UMD movies also had a structural problem. A film on UMD played on exactly one device, at 4.3 inches, and could not be ripped, lent, or watched on a television without extra hardware. A DVD cost less and played anywhere. ## Which PSP models did Sony release? There were five. The PSP-1000 is the original, the heavy one, with the 4.3-inch 16:9 screen Sony built the whole pitch around. The PSP-2000, sold as Slim and Lite, cut weight and thickness and added video output. The PSP-3000 improved the screen and put a microphone in the chassis. Then Sony got strange. The PSP Go dropped the UMD drive, shrank the body, hid the controls behind a sliding screen, and sold only downloads. It is the rarest of the five today and the most expensive second-hand. The PSP-E1000 came last, a stripped-down budget model sold in Europe with no Wi-Fi and a single speaker. American buyers mostly never saw it, which is part of why the PSP lineup gets remembered as four machines rather than five. All five play UMD games except the PSP Go, and all five are long out of production. Sony has not manufactured a PSP in over a decade. ## Can you still buy PSP games in 2026? Yes, for now, and not on the handheld. Sony gives the machine no support, no firmware work, and no official repair path, so everything left runs through two other consoles on a clock. Sony's support site explains the sequence. The PlayStation Store on the PSP was first closed in 2016, and then, "Starting July 6th, 2021, you'll no longer be able to perform searches or make in-game purchases" on the handheld. Downloads of content you already bought still work through the Download List on the device. Buying is the part people get wrong. Sony's own notice says "You'll still be able to purchase and play PSP content that is available on the PS3 and PS Vita stores," so PSP games did not vanish from sale in 2021. They moved to two other consoles. That window is closing too: Sony has confirmed that the PlayStation Store on PS3 and PS Vita closes starting July 2027, after which previously purchased content can still be downloaded "for the foreseeable future" but nothing new can be bought. The PS Vita, Sony's follow-up handheld, never got a proper obituary either. Sony's sales page carries a single footnote on the subject: "Sales data on PlayStation Vita are not disclosed." It is the only PlayStation system the company declines to put a number on. What keeps the PSP running in 2026 is the homebrew scene. Custom firmware turns the handheld into an emulator box, a media player, and a way to load game backups from a memory card, and it has been actively maintained by hobbyists for longer than Sony supported the machine officially. That community, not Sony, is the reason a 2004 handheld still has new software in 2026. ## What does a used PSP cost today? PriceCharting's tracked sales in September 2026 put a loose PSP-1000 in black at $79.99 and a complete one at $153.07. A loose PSP-2000 in black runs $89.93, a PSP-3000 sits at $134.95, and a Piano Black PSP Go reaches $204.30 loose, which is more than the machine cost new at several points in its life. Condition drives most of that spread. The analog nub wears out, the battery swells, and the UMD door snaps, so a cheap listing is often a repair project. Limited editions run higher again, and games are the affordable part of the hobby. Where to find one today: most tested PSP systems and refurbished shells trade in the $80 to $200 range depending on model. Sellers who service the batteries and nubs before shipping list on Etsy's Sony PSP console listings (https://www.etsy.com/search?q=sony+psp+console). ## Frequently Asked Questions ### How many PSPs were sold compared with the Nintendo DS? Sony's last published figure for the PSP is more than 76.4 million units as of March 31, 2012, and the company never issued a final lifetime total. Nintendo reports 154.02 million Nintendo DS units as of June 30, 2026, so the DS outsold Sony's handheld by roughly two to one. ### When did the PSP store shut down? Sony first closed the PlayStation Store on the PSP in 2016, and on July 6, 2021 it removed the handheld's remaining store searches and in-game purchases. PSP titles stayed on sale through the PS3 and PS Vita stores, which Sony has confirmed will close starting July 2027. ### What PSP models are there? Sony released five PSP models between 2004 and the end of production: the original PSP-1000, the slimmer PSP-2000, the PSP-3000 with its improved screen, the download-only PSP Go with no UMD drive, and the budget PSP-E1000 sold in Europe. All except the PSP Go take UMD discs. **Sources:** - NBC News (AP): U.S. release date set for PlayStation Portable, February 3, 2005: https://www.nbcnews.com/id/wbna6908316 - Sony Interactive Entertainment: Business Data and Sales, cumulative hardware unit sales: https://sonyinteractive.com/en/corporate/data.html - Nintendo: Dedicated Video Game Sales Units: https://www.nintendo.co.jp/ir/en/finance/hard_soft/index.html - PlayStation Support: Important notices regarding PlayStation products and services: https://www.playstation.com/en-us/support/important-notice/ - PriceCharting: PSP console and game prices: https://www.pricecharting.com/console/psp --- # Is Webvan Still Around? The $830 Million Collapse URL: https://404memoryfound.com/posts/what-happened-to-webvan-grocery-delivery.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-31 Updated: 2026-09-18 Topics: Business Blunders, Money & Tech **Summary:** Webvan does not exist. The online grocery service closed every market on July 9, 2001 and filed for Chapter 11 after burning roughly $830 million in about two years. Amazon later ended up with the Webvan name and the webvan.com address, and founder Louis Borders now runs a robotics company chasing the same idea. **Key facts:** - Launched: June 1999, San Francisco Bay Area - Founder: Louis Borders, co-founder of the Borders bookstore chain - Status today: Closed. Shut down July 9, 2001 and liquidated in Chapter 11 - Owner today: Amazon holds the Webvan name and the webvan.com address - Total burned: About $830 million in roughly two years ## Who was Louis Borders and what did he promise? Webvan was the second big retail idea from Louis Borders, who co-founded the Borders bookstore chain (https://404memoryfound.com/posts/who-owns-borders-now-bookstore.html) in 1971. He had already built a national chain on top of inventory software, and he was convinced the same trick would work on food. The promise was not a website bolted onto a supermarket. Webvan would own the whole chain: purpose-built automated warehouses, refrigerated vans split into temperature zones, and 30-minute delivery windows the customer picked. Groceries cannot travel in a parcel truck, so Webvan built its own everything. The company opened to the public in the San Francisco Bay Area in June 1999, out of one distribution center in Oakland. That building ran about 330,000 square feet, roughly seven times the floor of a supermarket, with more than five miles of conveyor belts and separate zones for frozen, chilled and dry goods. Customers who tried it generally liked it. What never arrived was the order volume the building had been sized for. ## How much did Webvan spend before it knew the model worked? In July 1999, weeks after opening that first warehouse, Webvan signed an agreement with the engineering firm Bechtel to build up to 26 more distribution centers over three years. The prospectus filed with the Securities and Exchange Commission told investors the program was expected to exceed $1 billion. Read the order of events again. Webvan committed to a national build-out before it had a single full quarter of operating data from the one warehouse it had. Bechtel finished four. The money kept arriving anyway. Webvan went public in November 1999 at $15 a share and raised $375 million, on top of the venture capital that Benchmark Capital, Sequoia Capital, Softbank and Goldman Sachs had already committed. By the close of the first day of trading the market was valuing a grocery service in the billions after five months of selling groceries in one metropolitan area. By the end, Webvan had burned roughly $830 million. That number is the whole story compressed: the company never lacked capital, it just ran a business that could absorb any amount of it. ## Why did the Oakland warehouse never fill up? The Oakland distribution center was engineered to handle 8,000 orders a day. Supermarket News reported that Webvan's oldest and strongest market was averaging about 2,160 orders a day, roughly a quarter of what the building was built for. A warehouse at a quarter capacity costs almost exactly what a full one costs. Conveyors, refrigeration, software, rent and salaried staff do not scale down with demand. Webvan needed somewhere around 3,000 orders a day in Oakland just to cover that site, and it did not get there and stay there. Delivery made the arithmetic worse. Grocery retail runs on thin margins, so a typical basket earned Webvan very little gross profit before a driver spent 30 to 45 minutes on the road with it. The company would not charge a serious delivery fee because it was afraid of losing the orders it already had. That is the trap. Raise the fee and volume falls. Keep delivery free and every order loses money. Webvan chose free and tried to outrun the loss with scale, which only meant losing money faster. ## What did the HomeGrocer merger do to the losses? By mid-2000 the market for money-losing internet companies had closed and Webvan's stock was sliding. The response was to get bigger. In June 2000 Webvan agreed to buy HomeGrocer.com, a West Coast rival, in an all-stock deal valued at about $1.2 billion, and completed it that September. The stated logic was consolidation: one competitor fewer, more delivery density, shared overhead. What Webvan actually inherited was another company's underused sites, a second technology stack and a second cash burn, at the exact moment it needed to stop spending. Investors saw it immediately. Shares in both companies fell on the day the deal was announced. Webvan then spent months converting HomeGrocer's markets to its own brand and systems, paying for integration while orders in its existing cities went sideways. By spring 2001 the stock traded below a dollar and Webvan was closing markets instead of opening them. George Shaheen, who had left the top job at Andersen Consulting to run Webvan, was gone before the end. ## Who owns the Webvan name today? Webvan announced on July 9, 2001 that it was closing every market, laying off about 2,000 remaining employees and filing for Chapter 11 protection in Delaware, where it was incorporated. Across the full collapse it let go just under 4,500 people. Chief executive Robert Swan pointed at the final quarter in the shutdown statement reported by Computerworld in 2001: "our order volume declined considerably during the second quarter, accelerating our need for capital." The warehouses and vans went to liquidation for a fraction of what they cost, and the brand was sold with the rest. Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html) ended up with the Webvan name and the webvan.com address, which it ran for a period as an Amazon-family storefront for shelf-stable groceries. There is no Webvan service in 2026 and no company operating under that name. Amazon also hired former Webvan operations people, and the grocery business they helped build became Amazon Fresh. Louis Borders is still working on it. He founded and runs HDS Global, a private robotics company developing fully automated fresh fulfillment. "COVID-19 has revealed to consumers how supermarkets are unnecessary middlemen, between their families and the fresh goods they need," he said in a Business Wire announcement in 2020. The service has not opened to the public. ## What did Webvan get right about grocery delivery? The vision held. Americans order groceries online in volume now, and the thing Webvan described exists: somebody picks your order and drives it to your door inside a window you chose. What changed is the capital. Instacart built its version on top of stores that already existed, sending shoppers into other companies' aisles. Walmart and Kroger turned their own supermarkets into fulfillment sites. Amazon bought Whole Foods and acquired a national cold chain rather than pouring one. Everyone who made it work borrowed the infrastructure. Timing did the rest of the damage. In 1999 broadband was rare, smartphones did not exist, and there was no on-demand pool of drivers to hire by the hour. Webvan had to manufacture the demand and the supply at the same time, with its own money. It was not alone in that. The same last-mile arithmetic destroyed Pets.com (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html) and Kozmo.com (https://404memoryfound.com/posts/what-happened-to-kozmo-com-delivery.html) in the same stretch of years. Webvan simply did it at the largest scale anyone attempted. ## Frequently Asked Questions ### What happened to Webvan? Webvan, the online grocery service that launched in the San Francisco Bay Area in June 1999, closed all operations on July 9, 2001 and filed for Chapter 11 bankruptcy protection in Delaware. It burned roughly $830 million in about two years and let go just under 4,500 people across the collapse. Its automated warehouses and delivery vans were liquidated, and the service never came back. ### Who owns Webvan now? No company operates as Webvan in 2026. Amazon ended up with the Webvan name and the webvan.com address after the 2001 bankruptcy and used the address for a period as an Amazon storefront selling shelf-stable groceries. Amazon runs its own grocery delivery under the Amazon Fresh name instead, so Webvan survives only as a dormant trademark. ### Who founded Webvan and what is he doing now? Webvan was founded by Louis Borders, who co-founded the Borders bookstore chain in 1971. After Webvan's 2001 bankruptcy he went back to the same problem with HDS Global, a private robotics company building fully automated fresh-grocery fulfillment, which he was still leading in 2025. **Sources:** - Webvan Group, Inc., Form 424B1 prospectus, U.S. Securities and Exchange Commission, 1999: https://www.sec.gov/Archives/edgar/data/0001092657/000089161899004914/0000891618-99-004914.txt - Webvan Group, Inc., International Directory of Company Histories (Encyclopedia.com): https://www.encyclopedia.com/economics/encyclopedias-almanacs-transcripts-and-maps/webvan-group-inc - Webvan Vanguard, Supermarket News: https://www.supermarketnews.com/grocery-operations/webvan-vanguard - Struggling online grocer Webvan shuts down, Computerworld, 2001: https://www.computerworld.com/article/1337313/struggling-online-grocer-webvan-shuts-down.html - Webvan Founder, Louis Borders, Unveils eCommerce Model of the Future, Business Wire, 2020: https://www.businesswire.com/news/home/20200514005082/en/Webvan-Founder-Louis-Borders-Unveils-eCommerce-Model-of-the-Future --- # Why the Nintendo GameCube Lost the Console War URL: https://404memoryfound.com/posts/why-nintendo-gamecube-lost-console-war.html Author: Marcus Vale (404 Memory Found) Published: 2026-03-30 Updated: 2026-09-17 Topics: Gaming, Hardware, Business Blunders **Summary:** The Nintendo GameCube lost the sixth-generation console war because it could not play DVDs, its 1.5 GB proprietary disc and thin installed base drove publishers to Sony, and its family image cost it older buyers. Nintendo shipped 21.74 million GameCube consoles against more than 155 million PlayStation 2 units and ended production in 2007. The GameCube still exists as a Nintendo Classics library on Nintendo Switch 2, where Expansion Pack members have had GameCube games since June 5, 2025. **Key facts:** - Launched: September 14, 2001 in Japan, November 18, 2001 in North America at $199.99 - Units shipped: 21.74 million worldwide, plus 208.58 million games - Status today: Out of production since 2007; the library runs on Nintendo Switch 2 - Owner today: Nintendo, which launched GameCube Nintendo Classics on June 5, 2025 - Price today: About $50 to $130 for a loose console in 2026, around $200 boxed ## Why couldn't the GameCube play DVDs in 2001? Nintendo built the Nintendo GameCube around a proprietary 8 cm disc developed with Matsushita, the parent company of Panasonic. Each disc held about 1.5 GB against the 4.7 GB of a standard DVD, and no living room DVD deck could read one. The reasoning was piracy and money. A non-standard disc was much harder to copy than a PlayStation 2 or Xbox game, and it kept Nintendo clear of DVD Forum licensing fees. The cost landed somewhere else: the GameCube could not play movies at all. That mattered more in 2001 than it sounds now. The PlayStation 2 had arrived in North America in October 2000 as one of the cheapest DVD players on the market, and the Xbox added playback through a remote accessory. Parents buying one machine for the family television had an easy answer, and it was not the purple cube. Japan got the exception. The Panasonic Q, a GameCube variant with a full DVD drive, launched on December 14, 2001 at 39,800 yen and was discontinued on December 18, 2003 after selling fewer than 100,000 units. Collectors now pay four figures for the one version nobody wanted at the time. ## Did the $199.99 launch price ever close the gap? The GameCube reached Japan on September 14, 2001 and North America on November 18, 2001 at $199.99. That was a full $100 under the PlayStation 2 (https://404memoryfound.com/posts/playstation-2-best-selling-console-ever.html) and the Xbox, which had both launched at $299. Nintendo kept cutting. The price dropped to $149 in May 2002 and to $99 in September 2003, twenty two months after launch. Nothing in the generation was cheaper for longer. The hardware was not the weak link either. The GameCube ran an IBM PowerPC processor and an ATI graphics part, and studios that pushed it got results that stood up against anything on the Xbox. Factor 5 shipped Star Wars Rogue Squadron II: Rogue Leader as a launch title and used it as a technical showcase for what the machine could do. Price and silicon were never what Nintendo was short of. Shelf appeal was. The launch software said family before it said anything else. Luigi's Mansion, which went on to sell 3.33 million copies, put Mario's brother and a vacuum cleaner in front of a market that was buying Grand Theft Auto III and Halo: Combat Evolved. ## What broke the GameCube's relationship with publishers? The damage started on the Nintendo 64. Cartridges cost more to manufacture and held far less than CDs, and the biggest Japanese publishers moved their headline series to Sony through the late 1990s. The GameCube's switch to optical discs was meant to undo that. It did not. Installed base decided the question instead. As the PlayStation 2 pulled away, publishers built for the platform with the audience, and the 1.5 GB disc gave them one more reason to skip a port. Rockstar never brought Grand Theft Auto III, Vice City or San Andreas to the GameCube, and that series defined the generation. The sharpest cut came from Capcom. The publisher had lined up a set of GameCube exclusives known as the Capcom Five to give the console a grown-up library, with Resident Evil 4 as the headline. On November 1, 2004, with the GameCube version still two months from its January 2005 release, Capcom announced Resident Evil 4 for PlayStation 2 as well, pointing to requests from fans, market conditions and shareholders. Buyers who had picked up a GameCube for that one game took the point. ## How many GameCubes did Nintendo actually sell? Nintendo shipped 21.74 million GameCube consoles worldwide, 12.94 million of them in the Americas, 4.77 million in other regions and 4.04 million in Japan, along with 208.58 million games. Production ended in 2007. The comparison is unkind. The PlayStation 2 passed 155 million units, with Sony citing more than 160 million, and it is still the best selling home console ever made. The original Xbox (https://404memoryfound.com/posts/what-happened-to-original-xbox-microsoft-gaming.html), Microsoft's first attempt at hardware, finished around 24 million. Nintendo came third in a field of three. The software numbers tell a kinder story. Super Smash Bros. Melee sold 7.41 million copies, Mario Kart: Double Dash 6.88 million, Super Mario Sunshine 5.91 million, The Legend of Zelda: The Wind Waker 4.43 million and Metroid Prime 2.84 million. For a machine with a fraction of the PlayStation 2 audience, Nintendo's own games sold at a rate nobody else in the generation matched. ## What did Nintendo change after the GameCube? Volume came first. At Nintendo's E3 2004 press conference the company sent Reggie Fils-Aime on stage with a line that outlived the console: "My name is Reggie. I'm about kicking ass. I'm about taking names. And we're about making games." The GameCube still finished the generation in third place. Behind him, Satoru Iwata had already decided the race was not worth running. Speaking to Kikizo at that same E3 in 2004, Nintendo's president said, "I don't think the next-generation consoles that Sony and Microsoft are considering have a future." That is the GameCube's real legacy. The Nintendo Wii (https://404memoryfound.com/posts/what-happened-to-nintendo-wii.html) that followed in 2006 stopped chasing Sony and Microsoft on graphics, raw power and third-party support, sold to people who had never owned a console, and outsold the GameCube several times over. Nintendo has not competed on specifications since. ## Is the Nintendo GameCube still worth buying in 2026? The console has been out of production since 2007, but the library came back. GameCube games launched as Nintendo Classics on June 5, 2025, the day the Nintendo Switch 2 went on sale, starting with The Legend of Zelda: The Wind Waker, F-Zero GX and Soulcalibur II. The catalogue has grown to close to ten titles since. There are two catches. The GameCube library needs a Nintendo Switch Online + Expansion Pack membership, which costs $49.99 a year for an individual plan, and it runs only on Switch 2, not on the original Switch. Nintendo also sells a wireless GameCube controller for Switch 2 at $64.99, released on the same June 2025 date. Where to find one today: a loose GameCube console in 2026 runs roughly $50 to $130 depending on color and condition, and a complete boxed unit in good shape starts around $200. Stone Age Gamer (https://stoneagegamer.com/nintendo/gamecube/consoles/) stocks tested consoles, controllers and memory cards, which is worth the premium over an untested machine with no return. Look for the DOL-001 model, the early revision that kept the digital video output Nintendo dropped later. ## Frequently Asked Questions ### How many GameCube consoles were sold? Nintendo shipped 21.74 million GameCube consoles worldwide between the September 2001 launch and the end of production in 2007, along with 208.58 million games. That left the GameCube third in its generation, behind the PlayStation 2 at more than 155 million units and the original Xbox at roughly 24 million. ### Can you still play GameCube games in 2026? Yes. GameCube games run on Nintendo Switch 2 through the Nintendo Classics library, which launched on June 5, 2025 and needs a Nintendo Switch Online + Expansion Pack membership at $49.99 a year. Original GameCube hardware built between 2001 and 2007 still works, and used consoles sell for about $50 to $130 loose in 2026. ### Why did the GameCube lose to the PlayStation 2? The GameCube lost to the PlayStation 2 on three counts: it could not play DVD movies, its 1.5 GB proprietary disc and small installed base pushed publishers toward Sony, and its family image cost it older buyers. The GameCube sold 21.74 million units between 2001 and 2007 against more than 155 million for the PlayStation 2. **Sources:** - GameCube, Wikipedia: https://en.wikipedia.org/wiki/GameCube - Nintendo GameCube - Nintendo Classics for Nintendo Switch 2, Nintendo: https://www.nintendo.com/us/store/products/nintendo-gamecube-nintendo-classics-switch-2/ - Resident Evil 4 coming to PS2, GameSpot (2004): https://www.gamespot.com/articles/resident-evil-4-coming-to-ps2/1100-6111911/ - Watch Nintendo's Satoru Iwata predict the future of consoles 20 years ago, Digital Trends: https://www.digitaltrends.com/gaming/satoru-iwata-e3-2004-restored-interview/ - Here's How Much Your Nintendo GameCube Is Worth In 2026, BGR: https://www.bgr.com/2250091/how-much-nintendo-gamecube-is-worth-2026/ --- # Is eBay Still Around in 2026? 135 Million Buyers URL: https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-30 Updated: 2026-09-17 Topics: Internet Culture, Business Blunders, Money & Tech **Summary:** eBay still exists. It has traded on the Nasdaq under the ticker EBAY since September 1998, no single person owns it, and the company reported 135 million active buyers and $79.6 billion of gross merchandise volume for 2025. The site began in September 1995 as AuctionWeb, a Labor Day weekend side project whose first sale was a broken laser pointer for $14.83. **Key facts:** - Launched: September 1995, as AuctionWeb, by Pierre Omidyar - First sale: A broken laser pointer, $14.83 - Status today: Still operating, public on Nasdaq as EBAY - Owner today: Public shareholders; no single owner - 2025 scale: $79.6 billion in goods sold, 135 million active buyers ## What was the first item ever sold on eBay? Pierre Omidyar wrote the code for a site he called AuctionWeb over Labor Day weekend in September 1995 and ran it from his home in San Jose, California. To find out whether the thing worked at all, he listed a laser pointer that had stopped working a couple of weeks after he bought it. It sold for $14.83. Omidyar emailed the winning bidder to make sure there had been no misunderstanding about the word broken. The buyer, a Canadian named Mark Fraser, wrote back that the condition was the point: "I'm a collector of broken laser pointers." Fraser identified himself publicly in 2015, around the company's twentieth anniversary, and gave a more practical version of his reasoning. He traveled constantly for work, he wanted a pointer for presentations, working ones cost far more than he was willing to pay, and he was confident he could repair a broken one himself. Either version makes the same point. There was a buyer for an object with no retail value, and the only thing missing was a place for the two of them to meet. That was a different idea from the first online purchases (https://404memoryfound.com/posts/first-online-purchase-pizza.html) of the early 1990s, which were ordinary store transactions moved onto a wire. ## Did eBay really start with Pez dispensers? For years the standard telling involved candy. Omidyar, it went, built the site so his fiancee, Pam Wesley, could trade Pez dispensers with other collectors. The version ran in BusinessWeek, The Wall Street Journal, the San Francisco Chronicle and the San Jose Mercury News, and people still repeat it. It was invented. Mary Lou Song, eBay's first public relations manager, made it up in 1997 because reporters would not write about the real one. Adam Cohen quoted her reasoning in The Perfect Store, his 2002 history of the company: "Nobody wants to hear about a thirty-year-old genius who wanted to create a perfect market." What they wanted, she judged, was a man who had done something for his fiancee. The fake story worked exactly as intended. Collectors read it, recognized themselves in it, and showed up to trade. It is worth knowing which story is which, because the Pez version suggests eBay was built for a hobby, and the laser pointer version suggests what Omidyar actually built: a market where strangers set the price of anything, including things nobody had ever priced before. ## How did AuctionWeb start charging for listings? AuctionWeb ran free for its first months. In early 1996 Omidyar's internet provider told him the traffic had outgrown a personal account and he would have to pay business rates, so he added a listing fee and a commission on completed sales to cover the difference. The fees came in faster than the bill did. That detail separates eBay from almost every company founded in the same window. eBay took a cut of transactions that other people paid for, packed and shipped. It held no inventory, ran no warehouses and employed no drivers, so the cost of serving one more seller was close to nothing. The contrast with its contemporaries is sharp. Pets.com (https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html) bought dog food, stored it and shipped it at a loss. Webvan (https://404memoryfound.com/posts/what-happened-to-webvan-grocery-delivery.html) built refrigerated warehouses before it had the orders to fill them. eBay charged a few cents to list and a small percentage to sell, and the model held up for thirty years without a rewrite. ## How big did the Beanie Baby trade get on eBay? Ty Inc. retired its stuffed animal designs deliberately, which turned a five dollar toy into a speculative asset and sent every buyer looking for a resale market. eBay was the resale market. The scale is easy to underrate. By 1997, Beanie Baby auctions on eBay came to roughly $500 million, more than 6 percent of the company's entire business, according to Mental Floss's account of the mania. During the 1998 holiday season there were stretches when 7 percent of everything listed on the site was a Beanie Baby. eBay's 1998 filings with the Securities and Exchange Commission named the volatility of that one category as a risk to the business, which is an unusual thing for a technology company to tell its investors. The users mattered more than the money. Beanie Baby speculators were not early adopters. They were suburban parents, retirees and hobbyists who learned to use the web specifically to buy and sell stuffed animals, and who then discovered they could buy and sell everything else. When the Beanie Babies bubble (https://404memoryfound.com/posts/beanie-babies-bubble-what-happened.html) collapsed, the toys went to landfill and the accounts stayed. ## How much did eBay's stock jump on its first day? eBay went public on the Nasdaq on September 24, 1998, priced at $18 a share. It closed the first day at $47.375, a gain of 168 percent, and CNNMoney wrote it up that evening as the return of the internet IPO after a summer of flat offerings. What sat underneath the pop was unusual for 1998. eBay's own year-end announcement reported net revenues of $47.4 million for 1998, an increase of 724 percent over the $5.7 million it took in during 1997, and net income of $2.4 million. A profitable internet company was close to a contradiction in terms that year. Meg Whitman had been hired as chief executive in March 1998, and her main contribution was not technical. She formalized the feedback system, the two-way rating both sides of a sale leave for each other. In a market where you mailed a money order to a stranger for an item you had never seen, a public score for past behavior did the work that a store's reputation does offline. Every marketplace built since, from ride hailing to short term rentals, runs on some version of it. ## Who owns eBay today, and is it still in business? eBay is still in business, and no one person owns it. The company has been publicly traded on the Nasdaq under the ticker EBAY since its September 1998 listing, and its largest holders are institutional index managers such as Vanguard and BlackRock rather than any individual or parent company. Jamie Iannone has been chief executive since April 2020. The 2025 numbers are not those of a fading site. eBay reported $79.6 billion in gross merchandise volume for the full year, up 7 percent, revenue of about $11.1 billion, and 135 million active buyers worldwide as of December 31, 2025. What changed is the merchandise. eBay lost general shopping to Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html) years ago and stopped fighting for it, concentrating instead on collectibles, trading cards, refurbished electronics, car parts and secondhand fashion, the categories where buyers want a specific object rather than a generic one. It also no longer owns the two acquisitions people remember. PayPal, bought in October 2002, was spun off as a separate public company in July 2015 and was worth more than eBay on its first day of independent trading. Skype (https://404memoryfound.com/posts/skype-shut-down-what-replaced-it.html), bought in 2005 for a use case that never materialized, was written down in 2007, sold in 2009 and eventually shut down by Microsoft. ## Frequently Asked Questions ### When did eBay start? eBay started in September 1995, when Pierre Omidyar launched it from his home in San Jose, California under the name AuctionWeb. The site was renamed eBay before it went public on the Nasdaq on September 24, 1998, at $18 a share. ### Who owns eBay? No single person or company owns eBay. It has been publicly traded on the Nasdaq under the ticker EBAY since September 1998, and its shares are held mostly by institutional investors such as Vanguard and BlackRock. Jamie Iannone has been chief executive officer since April 2020. ### What was the first item sold on eBay? The first item sold on eBay was a broken laser pointer, which went for $14.83 in September 1995 on the site then called AuctionWeb. The buyer, a Canadian named Mark Fraser, told founder Pierre Omidyar that he collected broken laser pointers, and identified himself publicly in 2015. **Sources:** - eBay Inc. Reports Fourth Quarter and Full Year 2025 Results: https://investors.ebayinc.com/investor-news/press-release-details/2026/eBay-Inc--Reports-Fourth-Quarter-and-Full-Year-2025-Results/default.aspx - eBay Inc. Announces Fourth Quarter and 1998 Year End Financial Results: https://investors.ebayinc.com/investor-news/press-release-details/1999/EBay-Inc-Announces-Fourth-Quarter-and-1998-Year-End-Financial-Results/default.aspx - CNNMoney: eBay, return of the IPO (September 24, 1998): https://money.cnn.com/1998/09/24/technology/ebay/ - TIME: This Is the First Item Ever Sold on eBay: https://time.com/4018841/first-item-sold-listed-ebay/ - Mental Floss: How Playground Rumors and Artificial Scarcity Created Beanie Baby Mania: https://www.mentalfloss.com/fun/toys/how-playground-rumors-and-artificial-scarcity-created-beanie-baby-mania --- # Is RadioShack Still in Business? 2026 Store Count URL: https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html Author: Marcus Vale (404 Memory Found) Published: 2026-03-30 Updated: 2026-09-09 Topics: Hardware, Business Blunders **Summary:** RadioShack is still in business in 2026, but as a brand and a website rather than a chain. Unicomer Group of El Salvador has owned the RadioShack name worldwide since May 2023. No company-owned RadioShack stores are left in the United States after the March 2017 bankruptcy, and the US presence is now more than 170 authorized dealers in 40 states. **Key facts:** - Founded: 1921, in the United States - Peak size: 5,087 US company stores plus 2,099 dealer stores (Dec 1999) - Bankruptcies: February 5, 2015 and March 8, 2017 - Owner today: Unicomer Group, El Salvador, since May 2023 - Status today: Website plus 170 plus US dealers; no company-owned US stores ## Is RadioShack still in business in 2026? RadioShack still exists, but almost nothing about it matches the chain Americans grew up with. The name, the website and the product line belong to Unicomer Group, a retailer based in El Salvador that bought the global brand in May 2023. There is no RadioShack chain left to walk into in the United States. What exists is radioshack.com, a wholesale operation that sells into other retailers, and a network of independent stores licensed to hang the sign. Modern Retail reported in May 2026 that the brand works with more than 170 resellers across 40 states, most of them independent shops or small regional chains. About 60 percent of revenue now comes from business to business sales, and the rest is split between the company's own site and online marketplaces. The catalog runs to roughly 600 products, built around batteries, power adapters, memory and portable speakers, and about a third of sales come from vintage styled goods such as turntables and AM/FM radios. RadioShack is alive as a brand and finished as a chain. That is a strange answer, but it is the accurate one. ## Who owns RadioShack now? Unicomer Group owns it, through an affiliate called Global Franchising Corporation. Unicomer announced the deal on May 9, 2023, buying RadioShack's intellectual property and domains across about 70 countries, a list that includes the United States, Canada, Europe and China. This was not a stranger walking in off the street. Unicomer took the RadioShack franchise in El Salvador in January 1998 and spent the next 25 years running RadioShack stores across Central America, South America and the Caribbean. The 2023 purchase turned the franchise it already operated into ownership of the name itself, covering what the company called more than 2,000 points of sale globally. The previous owner was Retail Ecommerce Ventures, a US firm that collected distressed retail names and picked up RadioShack in 2020. Under Unicomer the brand was relaunched as a product line in August 2024, which is why RadioShack batteries and cables now show up in stores that have no other connection to the old chain. So the short answer to who owns RadioShack: since 2023, a Latin American retail group, not an American one. ## When did RadioShack close, and how many stores are left? RadioShack did not close on a single day. It closed twice, in two bankruptcies filed two years apart. RadioShack Corporation filed for Chapter 11 on February 5, 2015 in the US Bankruptcy Court for the District of Delaware. On April 1, 2015 it completed the sale of 1,743 company-owned stores and their inventory to General Wireless Inc., an affiliate of the investment firm Standard General, and to Sprint Solutions. The bankruptcy court approved the auction result on March 31, 2015. The price came to about $160.7 million, made up of $47.6 million in cash and a $113.0 million credit bid. The rescue lasted two years. General Wireless Operations, the company that then owned the RadioShack brand, filed its own Chapter 11 on March 8, 2017 with roughly 1,740 stores still trading. It said it would close about 200 of them immediately and weigh its options on the remaining 1,300. Very few survived that review, and March 2017 is the date most people mean when they ask when RadioShack closed. As for stores left in 2026: none owned by the company in the United States. The American footprint is the 170 plus authorized dealers, and the retail that still carries the name day to day sits in Central America, South America and the Caribbean. ## What did Tandy Corporation do with RadioShack? Tandy is the reason RadioShack ended up on every corner. Charles Tandy ran a Fort Worth leather goods company, bought the small Boston electronics chain in the 1960s, and rebuilt it around a format almost nobody else wanted: tiny stores, everywhere, stocked with the parts and accessories other retailers could not be bothered to carry. The scale that came out of that is easy to underrate now. Tandy Corporation's Form 10-K for 1999 counted 5,087 company-owned RadioShack stores in the United States on December 31 of that year, averaging about 2,300 square feet each, alongside a separate network of 2,099 dealer and franchise stores. That is more than 7,000 outlets selling batteries, connectors, antennas and phone cords. Tandy also put RadioShack in the computer business. The TRS-80 arrived in the late 1970s and sold through the same stores that sold speaker wire, which dropped a leather goods company into the middle of the personal computer market at the same moment Commodore (https://404memoryfound.com/posts/what-happened-to-commodore-computers.html) was chasing the same customers. The parent company eventually took the store's name. The corporate filer registered with the SEC as Tandy Corporation is the same filer that submitted the 2015 bankruptcy paperwork as RadioShack Corporation. ## Why did the corner store model stop working? Two forces hit at once. The parts and repair business shrank as consumer electronics became sealed, cheap and disposable, and the accessory business moved to online sellers who did not need a storefront on every commercial strip. RadioShack's answer was cell phones. Carrier commissions were the richest thing a 2,300 square foot store could sell, so phones took over the shelves and the staff, and a parts retailer turned itself into a middleman in a market run by a handful of carriers. That bet ran straight into the 2015 restructuring, which handed 1,743 stores to General Wireless and Sprint and put Sprint counters inside them. By the time the second filing landed in March 2017, Fortune noted that both companies had been struggling. The squeeze was not unique to RadioShack. Circuit City (https://404memoryfound.com/posts/what-happened-to-circuit-city-electronics.html) and Fry's Electronics (https://404memoryfound.com/posts/what-happened-to-frys-electronics.html) ran out of road in the same stretch of years, and neither of them was carrying thousands of small strip mall leases while it happened. ## What happened to all the abandoned RadioShack stores? Search for an abandoned RadioShack and you get a whole genre of photograph: a dim storefront with the sign still up, shelves half stocked, a handwritten notice taped to the glass. That look comes from how fast 2017 moved. A chain running about 1,740 stores announced immediate closures at 200 of them and put the other 1,300 under review, so leases lapsed in waves instead of in an orderly wind-down. The buildings were an awkward size. A 2,300 square foot box at the front of a strip mall is too small for a supermarket or a big box tenant and too plain to convert cheaply, so plenty of them simply sat. Some RadioShack signs never went dark at all, and that is the part people misread. The dealer stores were always separate businesses. Tandy counted 2,099 of them in 1999, independently owned shops licensed to sell under the name, and they were not company-owned assets in either bankruptcy. A working RadioShack in a small town in 2026 is usually one of those, which is also why the current owner's US strategy is to sign up more of them rather than build stores. For collectors, what is left is the merchandise. Old catalogs, store signage and Tandy-era hardware trade steadily on Etsy (https://www.etsy.com/search?q=radioshack+vintage) and at swap meets, where condition drives the price, so compare several listings before paying an asking price. ## Frequently Asked Questions ### Does RadioShack still sell electronic components? Not in any real sense, at least not yet. A search of radioshack.com for a basic part such as a resistor returns nothing, and the current range is built around batteries, power adapters, memory, speakers and vintage styled audio. The brand's commercial lead told Modern Retail in May 2026 that soldering supplies, components for repair and restoration, and DIY kits are on the roadmap as the catalog grows from roughly 600 products toward 1,000. ### Who founded RadioShack? RadioShack began operations in 1921 in Boston, started by brothers Theodore and Milton Deutschmann to supply amateur radio operators, and the name comes from the term for the room that housed a ship's radio gear. The man who built the chain most Americans remember was Charles Tandy, whose Fort Worth company bought the small Boston business in the 1960s and turned it into a national format. ### How many RadioShack stores are left? In the United States, none are company-owned. The chain ended with the 2017 bankruptcy, and the US presence today is more than 170 authorized dealers and resellers across 40 states. Outside the US the name is still on real stores, mostly in Central America, South America and the Caribbean, part of what Unicomer described in 2023 as more than 2,000 points of sale worldwide. **Sources:** - Tandy Corporation, Form 10-K for fiscal 1999 (SEC EDGAR): https://www.sec.gov/Archives/edgar/data/0000096289/000009628900000005/0000096289-00-000005.txt - RadioShack Corporation, Form 8-K on the Chapter 11 sale, April 2015 (SEC EDGAR): https://www.sec.gov/Archives/edgar/data/0000096289/000119312515117783/d902146d8k.htm - Fortune: RadioShack files for bankruptcy again, March 8, 2017: https://fortune.com/2017/03/08/radioshack-bankrupt-chapter-11/ - Unicomer Group acquires the global RadioShack brand, May 9, 2023: https://grupounicomer.com/en/unicomer-group-acquires-the-global-radioshack-brand/ - Modern Retail: Inside RadioShack's audio-fueled B-to-B revival, May 7, 2026: https://www.modernretail.co/operations/inside-radioshacks-audio-fueled-b-to-b-revival/ --- # Is Friendster Still Around? It Came Back in 2026 URL: https://404memoryfound.com/posts/what-happened-to-friendster-social-network.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-30 Updated: 2026-09-17 Topics: Internet Culture, Business Blunders **Summary:** Friendster still exists in 2026, but not as the site anyone remembers. The original social network deleted every user profile on May 31, 2011 and shut its services down on June 14, 2015. The name returned on April 29, 2026, when Friendster Labs Inc. relaunched it as an iOS app with no ads and no algorithm, where the only way to add a friend is to tap two iPhones together in person. **Key facts:** - Launched: Founded 2002 by Jonathan Abrams, public beta March 2003 - Status today: Relaunched April 29, 2026 as an iOS-only app - Owner today: Friendster Labs Inc., run by programmer Mike Carson - Peak size: 115 million registered users in 2008, over 90% of traffic from Asia - Sold for: $39.5 million to MOL Global in December 2009 ## Who owns Friendster today? The Friendster name belongs to Friendster Labs Inc., a company set up by Mike Carson, a programmer based in Philadelphia who noticed the lapsed domain resolving again in 2023 and bought it. The app he built launched on April 29, 2026, and it runs on iOS only. There are no ads, no algorithmic feed, no follower counts and no suggested strangers. The rule that got the relaunch written up everywhere is the friending mechanic: the only way to add someone is to tap two iPhones together, in person, using Apple's NameDrop gesture. Connections also decay. If two people stop seeing each other in real life, the link between them weakens on its own. Carson told Cybernews in 2026 that the motive was corrective. "Today I feel that social networks foster a lot of negativity, but I remembered Friendster as being a really positive and enjoyable experience," he said. None of the old data came back with the name. Profiles written in 2003 were destroyed in 2011 and no public archive of them survives, so the 2026 app is a new network wearing an old label. ## Why did Jonathan Abrams turn down $30 million from Google? Jonathan Abrams founded Friendster in 2002 and opened the public beta in March 2003. By the fall of that year, still in beta, close to three million people had signed up. Google noticed and offered roughly $30 million to buy the company. Abrams said no. His investors, who included Peter Thiel and Google's own early backer K. Ram Shriram, argued there was too much money being left on the table. In October 2003, Kleiner Perkins Caufield and Byers and Benchmark Capital put about $13 million into the company at a reported $53 million valuation, well above what Google had offered for all of it. The decision was not obviously wrong at the time. Friendster had the growth curve, the network effects and the only proof anywhere that ordinary people would put their real names and real photographs on the web and come back daily to look at them. It belongs in the same file as the deal where Yahoo passed on buying Google (https://404memoryfound.com/posts/yahoo-rejected-google.html): a sensible call in the moment that aged into a punchline. ## Why did Friendster pages take 40 seconds to load? Friendster kept its social graph in a relational database. That is fine for looking up one person and poor at answering the question the entire product was built around: how am I connected to this stranger? Every new member did not add one row. They added thousands of possible paths the database had to walk in real time. Under load the math stopped working. Friendster pages took as long as 40 seconds to appear. In 2003, with most American households on dial-up or early DSL, 40 seconds was not a slow page. It was a closed tab. The engineering team tried the obvious repairs through 2003: spreading MySQL across more servers, adding load balancers, tuning queries. What it could not do was replace the foundation while the site was still growing and rivals were launching. Management made it worse by shipping features, which added load, and then by slowing user acquisition instead of buying capacity. ## Why did Friendster delete the Fakesters? By mid-2003 users had filled Friendster with profiles for things that were not people: celebrities, pets, bands, cities, Jackalope, God, Beer and more than a dozen Homer Simpsons. Friendster called them Fakesters and treated them as a threat to the premise of the site. Abrams was blunt about it in Salon in August 2003. "Fake profiles really defeats the whole point of Friendster," he said, explaining that when thousands of people link themselves to one popular Fakester, the connection map stops meaning anything. He was right about the mechanic and wrong about the fix. Friendster hired support staff and purged the accounts, and the purge turned into whack-a-mole: deleted profiles came straight back as duplicates. What it also did was punish the most creative users on the platform at the exact moment MySpace was inviting them to customize everything (https://404memoryfound.com/posts/rise-and-fall-of-myspace.html). MySpace had launched in August 2003, months after the Friendster beta, with much the same feature set plus editable HTML profiles and music. ## Where did the 115 million users go? Abrams was removed as chief executive in April 2004 and Tim Koogle, the former Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) chief executive, came in on an interim basis. Replacements followed replacements. Facebook had launched that February, restricted to Harvard students, with an architecture built for social graphs and a founder nobody was about to remove. Friendster kept growing anyway, just not where its investors lived. By 2008 it reported more than 115 million registered users, and over 90% of its traffic came from Asia, led by the Philippines, Indonesia, Malaysia and Singapore. In the United States and Europe the argument was already over. In December 2009 the Malaysian internet payments company MOL Global bought Friendster for $39.5 million. After roughly $13 million in secured debt, fees and executive bonuses, shareholders received $26.4 million, less than Google had offered six years earlier. MOL then sold 18 of Friendster's social networking patents to Facebook for $40 million, transferred on June 7, 2010, which recovered the entire purchase price on paper. ## What happened to the old profiles and photos? In April 2011 Friendster told users that all profiles, photos, messages and friend connections would be deleted on May 31 of that year. In June it reopened as a social gaming site. Eight years of user-generated content was erased on schedule, and unlike the volunteer scramble that saved much of GeoCities (https://404memoryfound.com/posts/history-of-geocities-websites.html), no large-scale rescue of Friendster pages happened before the deadline. The gaming version lasted four years. On June 14, 2015 the site and its services shut down indefinitely, replaced by a notice saying the company was taking a break. The stated reason was demand: "the online gaming community did not engage as much as we had hoped for," Friendster wrote, citing the state of the social gaming industry. The corporate entity was not formally wound up until 2018. That is the part of the story the 2026 relaunch cannot undo. A Facebook account opened in 2008 still holds its photos. A Friendster account opened in 2003 holds nothing, because the company that owned it decided storage was a cost rather than a promise. ## Frequently Asked Questions ### Is Friendster still around in 2026? Friendster is around in 2026, but only as a new product. The original social network deleted every user profile on May 31, 2011 and shut its services down on June 14, 2015. The name relaunched on April 29, 2026 as an iOS app from Friendster Labs Inc., with no ads, no algorithm, and friends added only by tapping two iPhones together in person. ### Who founded Friendster and who owns it now? Jonathan Abrams founded Friendster in 2002 and opened it in beta in March 2003, and he was removed as chief executive in April 2004. The company was sold to the Malaysian payments firm MOL Global in December 2009 for $39.5 million. The name is now held by Friendster Labs Inc., the company behind the 2026 app, run by the programmer Mike Carson. ### Why did Friendster fail? Friendster failed for three reasons that compounded between 2003 and 2005. Its relational database could not answer social graph queries at scale, so pages took as long as 40 seconds to load. Its 2003 purge of the joke profiles known as Fakesters turned its most active users against it. And the removal of founder Jonathan Abrams in April 2004 left product decisions to a rotating management team while MySpace and Facebook grew. **Sources:** - Salon: Faking out Friendster (2003): https://www.salon.com/2003/08/14/fakesters/ - TechCrunch: Friendster Valued At Just $26.4 Million In Sale (2009): https://techcrunch.com/2009/12/15/friendster-valued-at-just-26-4-million-in-sale/ - VentureBeat: Facebook buys Friendster's social networking patents (2010): https://venturebeat.com/2010/08/04/facebook-friendster-patents - Cybernews: The programmer reviving Friendster (2026): https://cybernews.com/news/friendster-revival-no-ads-algorithm-nostalgic/ - Friendster on the Apple App Store: https://apps.apple.com/us/app/friendster/id6760240416 --- # Is Fry's Electronics Still Around? All 31 Stores Closed URL: https://404memoryfound.com/posts/what-happened-to-frys-electronics.html Author: Marcus Vale (404 Memory Found) Published: 2026-03-29 Updated: 2026-09-16 Topics: Hardware, Business Blunders **Summary:** Fry's Electronics no longer exists. The San Jose chain shut all 31 of its remaining stores on February 24, 2021, without a bankruptcy filing or a closing sale, and handed its assets to creditors that April. What survives of the family business is First Electronic Bank, a Utah industrial bank, while the themed buildings are being demolished or converted. **Key facts:** - Status today: Closed since February 24, 2021 - Stores left: Zero, in any state - Founded: May 17, 1985, Sunnyvale, California - Peak size: 34 stores in nine states, 2019 - What is left: First Electronic Bank, Salt Lake City ## What happened on the night of February 24, 2021? Employees at Fry's Electronics found out the chain was finished at roughly the same time customers did. On the evening of February 23, 2021, word spread that every remaining store would not reopen. By the early hours of February 24 the frys.com storefront had been replaced with a single letter. The notice was blunt. Fry's Electronics said it had made "the difficult decision to shut down its operations and close its business permanently as a result of changes in the retail industry and the challenges posed by the Covid-19 pandemic." It counted the damage in one line: nearly 36 years, nine states, 31 stores. There was no Chapter 11 filing, no going-out-of-business sale and no last weekend for anyone to walk the aisles one more time. The company called it an orderly wind down. On April 2, 2021 it entered a general assignment for the benefit of creditors, and Hilco Global took over liquidating the inventory and the real estate. That silence is part of why the closure still feels unfinished. Retailers that die in bankruptcy court leave a paper trail of filings, hearings and creditor lists that reporters can read for years afterward. Fry's Electronics, privately held for its entire life, left a letter. ## Why were the shelves empty two years before the end? Anyone who visited after 2018 saw it coming. Aisles that once held rows of graphics cards, cases and power supplies had bare metal shelving instead. On September 10, 2019, The Mercury News reported the empty shelves across most locations and asked whether the chain was about to fold. Fry's Electronics had an explanation ready. It was moving suppliers to a consignment model, in which vendors keep ownership of the goods and get paid only after a sale. The company said 245 vendors had agreed to the new terms and that stock would be back within weeks. In November 2019, company spokesman Manuel Valerio told The Dallas Morning News, "We are not liquidating, or planning to close any stores." The one exception, he said, was Palo Alto, which had lost its lease. The other 34 stores in nine states would have product again over the next several weeks. Consignment only works when suppliers trust the retailer holding their goods. Distributors and manufacturers began refusing shipments over unpaid invoices, and vendors that had already placed products on consignment started collecting them back. The pivot meant to cut risk drained the stores instead, and four more locations closed before the end came. ## Who owns Fry's Electronics today? Nobody operates it. Fry's Electronics, Inc. stopped trading in February 2021, has not reopened under new ownership, and the name has not been licensed onto a replacement site the way some dead retail brands have. There is no Fry's store, no catalog and no working storefront at frys.com. The family business did not vanish, though. Forbes reported in December 2023 that the operation lives on as First Electronic Bank, a state-chartered, FDIC-insured industrial bank in Salt Lake City that partners with fintech lenders. Utah's rules let those partners charge rates that are capped in other states, and Forbes put some of the resulting loans as high as 180 percent interest. The money moved in the opposite direction to the stores. Forbes reported that First Electronic Bank's net income rose from $1.7 million in 2020 to $10.6 million in 2021, then to $25.8 million in 2022, over the same stretch that the retail chain was being liquidated. So the answer splits in two. As a place to buy a hard drive, Fry's Electronics has been dead since 2021. As a balance sheet, the Fry family's capital kept working, in consumer credit rather than consumer electronics. ## What happened to the themed buildings? Every location had its own theme, and the buildings outlived the business. The Burbank store, opened in 1995 at Hollywood Way and Vanowen, had a flying saucer crashed through its facade, built by movie prop designer Eric Christensen into a Googie building from 1962. Woodland Hills ran on Alice in Wonderland. Fountain Valley was ancient Rome. Anaheim was a NASA space center. Campbell was Egyptian. Burbank is gone. LaTerra Development began demolition in April 2025 to clear the eight-acre site for Burbank Aero Crossings, two seven-story buildings holding 862 apartments above about 9,700 square feet of ground-floor shops and restaurants. San Jose went the other way. Supermicro bought the 19.7-acre former headquarters at 550 East Brokaw Road in February 2024 for roughly $80 million, after years of leasing space there. City planners approved a 333,400-square-foot manufacturing and warehouse building that reuses the old retail shell and keeps the store's Mayan-themed facade, with the industrial wing built behind it. The result is the strangest monument in Silicon Valley: a fake stone temple that now fronts a server plant. The theme survived the retailer that paid for it. ## How did a grocery family end up selling motherboards? Fry's Electronics came out of a supermarket. Charles Fry built a chain of 41 Fry's Food Stores across California and Arizona through the 1960s and sold it in 1972 for about $14 million, a deal company histories credit to Dillon Companies. John Fry had managed the computer systems for his father's supermarkets. On May 17, 1985, he opened a 20,000-square-foot store in Sunnyvale, California with his brothers Randy and David and a fourth partner, Kathryn Kolder. The pitch was one-stop shopping for the high-tech professional: integrated circuits and test equipment on the same floor as software, televisions and snacks. The snack aisle was the supermarket habit showing through. Chips, candy and soda sat a few steps from the components, because a shopper who stays longer buys more, and the floor plan was built to keep people wandering past things they had not come for. It worked for two decades. The chain reached 34 stores in nine states by 2019, spread across California, Texas, Arizona, Georgia, Illinois, Indiana, Nevada, Oregon and Washington, and became the default parts counter for anyone building a PC on the West Coast. ## Why did Fry's lose to Amazon and Best Buy? The pandemic took the blame in the closing letter, but the chain had been losing for years before 2020. The shutdown "represents the end of an era and a sad day for its army of loyal shoppers," Neil Saunders, managing director of GlobalData, said in February 2021. He argued the elaborate store themes had turned from an attraction into a burden as the shelves thinned out. Fry's Electronics never built a real online business. Its website stayed an afterthought while Newegg and Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html) took the component buyers who had made the chain necessary in the first place. Best Buy spent the 2010s rebuilding around price matching, vendor shop-in-shops and Geek Squad services. Fry's kept running the 1995 format. The rivals it outlived show how narrow the window was. Circuit City (https://404memoryfound.com/posts/what-happened-to-circuit-city-electronics.html) went down in 2009, CompUSA (https://404memoryfound.com/posts/is-compusa-still-around.html) had already stopped being a national chain, and RadioShack (https://404memoryfound.com/posts/what-happened-to-radioshack-electronics.html) shrank to a brand attached to a website. Big-box electronics retail had no room left for a company that treated the internet as optional. Jefferies analyst Jonathan Matuszewski estimated in 2021 that Best Buy stood to pick up more than $400 million in sales from the Fry's exit. Micro Center absorbed part of the enthusiast trade. Neither one replaced the experience of walking into a building with a spaceship stuck in the front of it. ## Frequently Asked Questions ### When did Fry's Electronics close? Fry's Electronics closed every remaining store on February 24, 2021, after nearly 36 years in business. All 31 stores across nine states shut on the same day, with no bankruptcy filing and no closing sale, and the company entered a general assignment for the benefit of creditors on April 2, 2021. ### Why did Fry's Electronics go out of business? Fry's Electronics blamed changes in the retail industry and the Covid-19 pandemic in its February 2021 closing notice, but the chain had been failing since at least 2019, when a shift to a consignment inventory model left shelves bare and suppliers stopped shipping over unpaid invoices. It never built an online business capable of competing with Amazon and Newegg. ### Are any Fry's Electronics stores still open? No. Zero Fry's Electronics stores have operated anywhere since February 24, 2021, and the brand has not been revived. The former San Jose headquarters at 550 East Brokaw Road was bought by Supermicro in February 2024, and the Burbank store with the crashed flying saucer was demolished in April 2025. **Sources:** - Fry's Electronics abruptly goes out of business, closes stores (Retail Dive, 2021): https://www.retaildive.com/news/frys-electronics-abruptly-goes-out-of-business-closes-stores/595614/ - Fry's Electronics says it's not going out of business (The Dallas Morning News, 2019): https://www.dallasnews.com/business/retail/2019/11/15/frys-electronics-says-its-not-going-out-of-business/ - How Fry's Electronics Pivoted From Retail To Backing 180%-Interest Consumer Loans (Forbes, 2023): https://www.forbes.com/sites/brandonkochkodin/2023/12/05/how-frys-electronics-pivoted-from-selling-tech-to-backing-consumer-loans-with-180-interest/ - LaTerra bulldozes Burbank site of crashed UFO for apartments (The Real Deal, 2025): https://therealdeal.com/la/2025/04/16/laterra-bulldozes-iconic-site-of-crashed-ufo-in-burbank/ - Supermicro moves in on old Fry's, turning San Jose icon into new tech hub (Hoodline, 2026): https://hoodline.com/2026/01/supermicro-moves-in-on-old-fry-s-turning-san-jose-icon-into-new-tech-hub/ --- # Does AltaVista Still Exist? Yahoo Closed It in 2013 URL: https://404memoryfound.com/posts/what-happened-to-altavista-search-engine.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-29 Updated: 2026-09-17 Topics: Internet Culture, Software & Apps **Summary:** AltaVista no longer exists. Yahoo shut the search engine down on July 8, 2013, and altavista.com now redirects to Yahoo Search. The name and domain belong to Yahoo Inc., which has been majority-owned by Apollo Global Management funds since September 2021. **Key facts:** - Launched: December 15, 1995, by Digital Equipment Corporation - Status today: Closed on July 8, 2013; altavista.com redirects to Yahoo Search - Owner today: Yahoo Inc., majority-owned by Apollo Global Management funds since 2021 - Last sale price: $140 million to Overture Services in 2003, down from $2.3 billion in 1999 - Index at launch: About 16 million pages and 300,000 queries on day one ## Who owns AltaVista today? AltaVista has not worked as a search engine since 2013, and the name belongs to Yahoo Inc. Yahoo picked it up in 2003 by buying Overture Services, the company that had bought AltaVista earlier that year, and it has held the brand ever since. Yahoo itself changed hands in September 2021, when funds managed by the private equity firm Apollo Global Management completed their purchase of the business from Verizon. That deal made Apollo the majority owner of Yahoo, and with it the owner of what is left of AltaVista: a domain name and a trademark. Typing altavista.com into a browser does not produce an error. It sends you to Yahoo Search. There is no archive of old AltaVista results behind it, no separate index, and no way to run the kind of query the site accepted in 1997. The redirect is the whole of the estate. We traced every owner Yahoo has had (https://404memoryfound.com/posts/who-owns-yahoo-now.html) in more detail separately. ## How a chip demo turned into the first full-text web search engine AltaVista started inside Digital Equipment Corporation, the Massachusetts computer maker usually called DEC. In 1995, researchers at DEC's Western Research Laboratory in Palo Alto, California were trying to prove what the company's Alpha processor could do. Benchmark scores were not convincing customers. The demonstration three of them settled on was to index the entire World Wide Web and let anyone search it. Paul Flaherty proposed the project, Louis Monier wrote the crawler that went out and collected the pages, and Michael Burrows wrote the indexer that made them searchable in a fraction of a second. AltaVista opened to the public on December 15, 1995 with about 16 million pages in its index, an enormous number for the time. It handled 300,000 queries on its first day. Within a year it was fielding tens of millions of requests a day, far past anything DEC's marketing department had planned for a hardware demonstration. That is the part of the story that still gets told. The rest of it is a company handing a working product to owners who each wanted it to be something else. ## What made AltaVista better than a directory in 1996? Before AltaVista, finding a page usually meant browsing a directory. Yahoo, the biggest name on the web then, paid people to sort sites into categories by hand. A site that nobody had filed was effectively invisible. AltaVista searched the text of the pages themselves. You could look for an exact phrase, combine terms with AND, OR and NOT, restrict results to one language, and get an answer back in about a second. For research that was a different class of tool, and the rival crawlers of the day, Lycos (https://404memoryfound.com/posts/what-happened-to-lycos-search-engine.html) among them, covered far less of the web. The interface mattered as much as the index. AltaVista was a search box on a nearly empty page, and the results came back as a plain list. Marking the engine's thirtieth anniversary in December 2025, Tom's Hardware described it as the site that pioneered "the fast and uncluttered search results model users loved, before Google stole its clothes and its thunder." It shipped extras that were genuinely new at the time, including free machine translation under the Babel Fish name and search operators that most competitors could not handle. ## Why did Compaq turn AltaVista into a portal? Compaq bought DEC in January 1998 for $9.6 billion, the largest acquisition the computer industry had seen, and AltaVista came along with the servers. Compaq (https://404memoryfound.com/posts/compaq-lost-pc-business-hp-merger.html) wanted DEC's enterprise business and its engineering staff. It did not particularly want a search engine, and it had no settled view of what one was worth. What Compaq did know was that Yahoo was the most valuable property on the web, and that Yahoo was a portal: news, mail, shopping, weather and horoscopes on a single page. So it rebuilt AltaVista in that shape. News feeds, shopping links, free email and advertising modules went onto the home page, and the search box, the only thing most visitors had come for, got pushed down the screen. The pages got busier and slower at the exact moment a competitor was going the other way. Google put one box on a white page and nothing else. People who wanted a search engine went to the site that still behaved like one. ## How much did AltaVista sell for, and to whom? Compaq did not keep it long. On June 29, 1999 it sold 83 percent of AltaVista to CMGI, an internet holding company, in a stock and notes deal valued at about $2.3 billion, and kept 17 percent for itself. CMGI intended to take AltaVista public. The offering never happened. The dot-com market broke first, CMGI's own shares fell with it, and AltaVista went through layoffs and a run of chief executives. The portal strategy was abandoned and the site went back to being a search engine, by which point the audience had already gone. In February 2003, Overture Services agreed to buy AltaVista's business for $140 million, $60 million of it in cash and the rest in Overture stock. That is a drop of about 94 percent from the CMGI price in under four years. "Adding AltaVista is an important step forward in expanding our capacity and commitment to search," said Ted Meisel, Overture's chief executive, on a conference call reported by InternetNews in 2003. Later the same year Yahoo bought Overture, and AltaVista changed hands for the fourth and final time. ## What finally killed AltaVista? Nothing dramatic. Yahoo kept altavista.com running for a decade as a second-tier property while its own search moved onto other technology. The index behind the box stopped being AltaVista's. The extras went one at a time, Babel Fish among them, handed off to Microsoft's translation service. On June 28, 2013, Yahoo posted a list of products it was closing. AltaVista was on it, with a date of July 8 and one line of instruction: "Please visit Yahoo Search for all of your searching needs." The engine went dark that morning, seventeen years and seven months after it opened. The explanation the obituaries reached for is roughly the right one. AltaVista had the better index in 1996 and the worse owners from 1998 onward. DEC treated it as an advertisement for processors, Compaq treated it as a Yahoo imitation, CMGI treated it as a stock offering, and Google treated search as the business itself. One of those four readings was correct, and it is the only one still trading. ## Frequently Asked Questions ### Does AltaVista still exist in 2026? No. AltaVista stopped working as a search engine on July 8, 2013, when Yahoo closed it. The altavista.com address still resolves, but it redirects to Yahoo Search rather than to anything AltaVista built, and no version of the original index is available to search. ### Who owns AltaVista now? The AltaVista name and domain belong to Yahoo Inc., which acquired them in 2003 by buying Overture Services. Yahoo has been majority-owned by funds managed by Apollo Global Management since September 2021, so Apollo controls whatever commercial value the AltaVista brand still carries. ### Why did AltaVista lose to Google? AltaVista lost to Google because the owners it had between 1998 and 2003 spent those years rebuilding it as a Yahoo-style portal instead of improving search. Compaq buried the search box under news, shopping and email modules, users left for Google's single box, and by the February 2003 sale to Overture Services AltaVista was worth $140 million against the $2.3 billion CMGI had paid in 1999. **Sources:** - Tom's Hardware: Search pioneer AltaVista's star shone bright 30 years ago (2025): https://www.tomshardware.com/tech-industry/big-tech/search-pioneer-altavistas-star-shone-bright-with-a-clean-and-minimal-ui-30-years-ago-engine-lost-momentum-after-multiple-ownership-changes-and-the-embrace-of-the-web-portal-trend - The Washington Post: CMGI Buys AltaVista From Ailing Compaq (June 30, 1999): https://www.washingtonpost.com/archive/business/1999/06/30/cmgi-buys-altavista-from-ailing-compaq/29b441ce-accf-41bf-aa97-8b01af7897f5/ - InternetNews: Overture to Buy AltaVista (February 2003): http://www.internetnews.com/IAR/article.php/1587171/Overture+to+Buy+AltaVista.htm - Associated Press via Salon: Yahoo shuts down Internet relic AltaVista (July 8, 2013): https://www.salon.com/2013/07/08/yahoo_shuts_down_internet_relic_altavista_ap/ - Apollo Global Management: Apollo Funds Complete Acquisition of Yahoo (September 2021): https://www.apollo.com/insights-news/pressreleases/2021/09/apollo-funds-complete-acquisition-of-yahoo-161530593 --- # Who Owns LimeWire Now? BabyDoge Bought It in 2026 URL: https://404memoryfound.com/posts/what-happened-to-limewire-file-sharing.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-27 Updated: 2026-09-15 Topics: Software & Apps, Music & Entertainment **Summary:** LimeWire the file sharing program no longer exists. A federal court disabled it on October 26, 2010, and its owner settled with the record labels for $105 million in May 2011. The LimeWire name is now owned by the crypto project BabyDoge, which acquired it on September 14, 2026 and runs it as an AI and storage platform with no file sharing. **Key facts:** - Launched: August 2000, Lime Group LLC, on the Gnutella network - Shut down: October 26, 2010, by federal court injunction - Settlement: $105 million paid to the record labels, May 2011 - Owner today: BabyDoge, since September 14, 2026 - Status today: Brand active as an AI and storage platform; file sharing dead since 2010 ## How LimeWire put every song on one search bar Mark Gorton, a Wall Street trader who had founded the firm Tower Research Capital, released LimeWire in August 2000 through his company Lime Group LLC. The software ran on Gnutella, a decentralized network with no central index. Every user's shared folder became part of the library, and a search hopped from computer to computer until it found matches. That design was the whole point. Napster (https://404memoryfound.com/posts/who-owns-napster-now.html) kept a central server listing every file, which gave the record labels one machine to aim a court order at. Gnutella had no such machine. When Napster lost in court (https://404memoryfound.com/posts/napster-destroyed-music.html) in 2001, its users scattered to LimeWire, Kazaa (https://404memoryfound.com/posts/what-happened-to-kazaa-p2p-wars.html) and BearShare, and LimeWire took the biggest share of them. Using it took three steps: type a song title, pick a result, wait. No account, no card, no copy protection. The company sold a paid tier called LimeWire PRO with faster downloads and more search results, which is how it made money while the free version did the damage the labels later put in front of a judge. ## How many people actually used LimeWire? The number everyone repeats is that LimeWire sat on one computer in three worldwide. That is a misreading of a single study. Digital Music News, BigChampagne and PC Pitstop scanned about 1.67 million Windows PCs over twelve months to 2007 and found LimeWire on 36.4 percent of the machines that had any file sharing client installed, against 11.3 percent for the runner-up, uTorrent. The share of all computers was much lower. In September 2007 the same scan found LimeWire on 17.8 percent of every machine polled that month, as TorrentFreak pointed out at the time. That is roughly one computer in six, which is still extraordinary for a program no shop sold and no manufacturer preinstalled. LimeWire's own public figure counted users rather than installs. In the weeks before the shutdown the company said it had about 50 million monthly users. However you measure it, LimeWire was the largest file sharing program in the world for most of the decade, and the record labels knew exactly where it ranked. ## Why LimeWire downloads were so risky On Gnutella anyone could share anything and name it anything. A search for a hit single returned a dozen results and nothing in the interface told you which one held the song. Files ending in .mp3.exe installed adware, browser hijackers and keyloggers, and the person who clicked usually had no idea until the pop-ups started arriving every few seconds. Some of the junk was put there on purpose. Record labels paid firms to seed the networks with decoys: tracks that played for thirty seconds and cut to noise, or looped the same bar for four minutes, so that finding a clean copy took long enough to feel like work. The industry called the tactic spoofing and ran it for years across Gnutella and its rivals. Then there was quality. An MP3 on LimeWire arrived at whatever bitrate a stranger had ripped it at, with the artist misspelled and the album field blank, which is why a generation spent evenings retagging files in Winamp (https://404memoryfound.com/posts/what-happened-to-winamp-mp3-player.html). The real problem with LimeWire was never only the law. Most of what you downloaded was simply wrong. ## Why a federal judge shut LimeWire down in 2010 Thirteen record companies led by Arista Records sued Lime Group LLC, Lime Wire LLC and Mark Gorton personally in 2006. The case, Arista Records LLC v. Lime Group LLC, ran for four years in the Southern District of New York. On May 11, 2010, Judge Kimba M. Wood granted summary judgment to the labels. She found that LimeWire had induced copyright infringement: the company knew that the overwhelming majority of traffic was infringing, it had marketed itself to former Napster users, and it had built no meaningful filter. The court held Gorton and Lime Group liable for that inducement, not just the software company. Two weeks after the ruling, Gorton told The New York Times that he had expected the labels to license the subscription service he was planning rather than sue him into the ground. "Perhaps I was naive," he said. "If I knew when the lawsuit started what I know now about the music industry, maybe we would have done something different." The injunction landed on October 26, 2010. Anyone who opened the program that week got a notice reading: "This is an official notice that LimeWire is under a court-ordered injunction to stop distributing and supporting its file sharing software." ## The damages claim the court called absurd The labels then asked for statutory damages of up to $150,000 for each of roughly 11,200 recordings they said had been infringed, multiplied by every user who had shared each one. Reported badly, that became the $72 trillion headline that still circulates. What the labels actually filed was a per-infringer theory whose arithmetic ran into the trillions. Judge Wood rejected it in March 2011. An award calculated that way, she wrote, would be "more money than the entire music recording industry has made since Edison's invention of the phonograph in 1877," and the absurdity of that result was reason enough to throw the theory out. The labels were left with one statutory award per recording. The damages trial opened in May 2011 and never reached a verdict. Gorton and the labels settled for $105 million, closing the case five years after it was filed. A statement from Lime Wire's law firm said only that "Lime Wire and its founder, Mark Gorton, are pleased that this case has concluded." A month after the shutdown, an unauthorized build called LimeWire Pirate Edition appeared, stripped of any dependency on the company's servers. LimeWire's own lawyers had the site hosting it taken down. The code lives on as an open source Gnutella client called WireShare. ## Is LimeWire still around in 2026? The file sharing program is gone and has been since 2010. The name is not. Two Austrian brothers, Julian and Paul Zehetmayr, bought LimeWire's intellectual property in 2021 and relaunched the brand in May 2022 as a music NFT marketplace, then steered it toward AI content tools and decentralized storage paid for with a token called LMWR. Ownership changed hands again this month. On September 14, 2026, the crypto project BabyDoge announced it had acquired LimeWire on undisclosed terms. Abel Czupor took over from the Zehetmayr brothers and described BabyDoge's role as a "DAO-style backer" rather than a parent company, with the platform keeping LMWR and reporting about 8 million users. None of that is file sharing. The LimeWire of 2026 shares a logo and a name with the program that frightened the record industry, and nothing else. Anything calling itself LimeWire in a search result for free music downloads today is neither the original nor safe to install. ## Frequently Asked Questions ### Who owns LimeWire now? LimeWire has been owned by the crypto project BabyDoge since September 14, 2026, when it acquired the brand on undisclosed terms from Julian and Paul Zehetmayr, the Austrian brothers who bought LimeWire's intellectual property in 2021 and relaunched it in 2022. Abel Czupor now leads the platform, which sells AI content tools and storage rather than file sharing software. ### What replaced LimeWire after it shut down in 2010? After the court disabled LimeWire's file sharing on October 26, 2010, most users moved to BitTorrent clients, and then to subscription streaming: Spotify reached the United States on July 14, 2011, two months after the LimeWire case settled. Paid streaming, not enforcement, is what ended casual music piracy for ordinary listeners. ### Is downloading from LimeWire illegal? The original LimeWire was legal software, but a federal court ruled in May 2010 that Lime Wire LLC induced copyright infringement, and downloading copyrighted music without a license was and remains illegal in the United States. LimeWire's file sharing has been disabled since October 26, 2010, so there is nothing to download from it, and the LimeWire brand sold in 2026 is an unrelated BabyDoge product. **Sources:** - Arista Records LLC v. Lime Group LLC, 715 F. Supp. 2d 481 (S.D.N.Y. 2010): https://www.leagle.com/decision/infdco20100526c41 - NPR, Court Order Shuts Down LimeWire's File-Sharing Software (2010): https://www.npr.org/sections/therecord/2010/10/27/130859566/court-order-shuts-down-limewire-s-file-sharing-software - Computerworld, RIAA request for trillions in LimeWire copyright case is 'absurd,' judge says (2011): https://www.computerworld.com/article/1541572/riaa-request-for-trillions-in-limewire-copyright-case-is-absurd-judge-says.html - TorrentFreak, LimeWire Not as Popular as Recent Reports Suggest (2007): https://torrentfreak.com/limewire-not-as-popular-071228/ - crypto.news, BabyDoge acquires LimeWire as platform plans creator focused push (2026): https://crypto.news/babydoge-acquires-limewire-as-platform-plans-creator-focused-push/ --- # Is ICQ Still Around? VK Shut It Down in June 2024 URL: https://404memoryfound.com/posts/what-happened-to-icq-instant-messenger.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-27 Updated: 2026-09-15 Topics: Software & Apps, Internet Culture **Summary:** ICQ is no longer running. VK, the Russian company that bought it from AOL in 2010, switched the messenger off on June 26, 2024 after twenty-eight years and sent users to VK Messenger. VK still owns the ICQ name, and a volunteer project called NINA now runs free servers that the old ICQ clients can log into. **Key facts:** - Launched: November 1996, by Mirabilis in Tel Aviv - Status today: Shut down on June 26, 2024, after 28 years - Owner today: VK Company Limited, formerly Mail.ru Group - Peak: Around 100 million users in 2001 - Price today: Free on NINA's unofficial ICQ servers ## What was ICQ and who built it? ICQ was a free instant messenger written by Mirabilis, a small Israeli company set up in 1996 by Yair Goldfinger, Sefi Vigiser, Amnon Amir and Arik Vardi. Arik's father, the investor Yossi Vardi, put up the money. The first client went out as a free download in November 1996. The name is a phonetic spelling of the phrase I Seek You, and the problem it solved sounds trivial now. In 1996 you could only chat with people inside the same walled service. CompuServe (https://404memoryfound.com/posts/what-happened-to-compuserve-online-service.html) members talked to CompuServe members. AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html) members talked to AOL members. Everyone else had IRC (https://404memoryfound.com/posts/what-happened-to-irc-chat.html), which assumed you already knew what a server and a channel were. ICQ ignored the walls. Download the client, take a number, add anyone you liked anywhere on the internet. It shipped with a contact list, online and away status, offline messages that waited for you to come back, file transfers and a searchable directory of users. Every one of those features is in the messenger on your phone right now, which is the strongest argument for treating ICQ as the original rather than an also-ran. ## Why did AOL pay $287 million for a free chat app? By the spring of 1998, America Online had a problem. Its own AOL Instant Messenger (https://404memoryfound.com/posts/what-happened-to-aim-aol-instant-messenger.html) mattered only inside AOL's service. ICQ had everyone else, and it was spreading with no advertising at all. So AOL bought Mirabilis outright. AOL's filing with the Securities and Exchange Commission records that the purchase closed on June 5, 1998, that AOL paid $287 million in cash for 100 percent of the Mirabilis assets, and that it agreed to contingent payments of up to $120 million over three years tied to growth targets. That is about $407 million if everything paid out, and at the time it was the largest sum ever paid for an Israeli software company. The same filing counts more than 12 million registered ICQ users at the moment of sale. "The acquisition of ICQ technology dramatically increases our presence on the Web," AOL chairman and chief executive Steve Case said in the 1998 announcement. Mirabilis stayed in Tel Aviv, still run by its founders, and ICQ kept its own name and its own client. ## What made an ICQ number worth bragging about? Every account got a Unique Identification Number, the UIN. It was your address on the network, independent of the computer you sat at, the internet provider you paid or the country you lived in. Type the number and the password anywhere and your contact list followed you. That is ordinary in 2026 and it was not ordinary in 1996. Because the numbers were handed out in order, the length of yours announced when you had arrived. Six-digit UINs belonged to people who signed up in the first year. Nine-digit numbers marked you as late. Short numbers were traded, gifted and sold, years before anyone thought of a handle as property. The other signature was the alert sound, a two-note chime that played the moment a message landed. It is still the first thing people name when ICQ comes up, and it is the reason the shutdown in 2024 made headlines in countries where the service had been dead for a decade. ## How did ICQ lose the messaging war? ICQ kept growing after the sale. Its peak was around 100 million users in 2001, which made it one of the most widely installed programs on earth. Then three things went wrong at once. AOL owned two messengers and never chose between them. AOL Instant Messenger was the default for American members. ICQ was everything outside the United States. The engineering attention went to the American product, and ICQ drifted. The client also got heavy. A program that had been small and quick filled up with news tickers, mini games, greeting cards, a browser toolbar and advertising. Meanwhile MSN Messenger (https://404memoryfound.com/posts/what-happened-to-msn-messenger.html) was free, plain, tied to a Hotmail (https://404memoryfound.com/posts/why-everyone-had-hotmail-account.html) address and, on Windows XP, already sitting on the machine when the user first switched it on. Then phones arrived. ICQ's identity was a number you had to memorise and a client you had to install. WhatsApp's identity was the phone number already in your pocket, and it was drawn for a phone screen from its first release. By the time ICQ had a mobile app worth using, the contacts had moved and there was nothing to come back for. ## Who owns ICQ today? VK Company Limited, the Russian internet group formerly called Mail.ru Group, owns the ICQ name in 2026 and has owned it since 2010. AOL sold the service on. AOL's 2010 filing with the Securities and Exchange Commission records a securities purchase agreement signed on April 28, 2010 with Digital Sky Technologies Limited, and a price of $187.5 million in cash. AOL had paid $287 million twelve years earlier, so the messenger sold for well under what it cost before inflation is counted at all. Digital Sky Technologies became Mail.ru Group, and Mail.ru Group later renamed itself VK Company Limited after VKontakte, its largest product. Under that ownership ICQ stopped being a world product and became a Russian one. It was rebuilt more than once, gaining video calls, stickers and a redesigned mobile app, and it held a real audience in Russia and its neighbours long after the rest of the world had forgotten the chime. That audience was not enough. ## Can you still log into ICQ in 2026? Not on VK's servers. In late May 2024 a notice went up on the ICQ website: "ICQ will stop working from June 26." It gave no explanation and sent people to VK Messenger for friends and VK WorkSpace for colleagues. The service went dark on June 26, 2024, twenty-eight years after the first download, and VK has run nothing under the name since. Unofficial servers kept the protocol alive. NINA, a volunteer project that also rebuilds AOL Instant Messenger and classic AOL, runs ICQ servers that the original clients connect to: point an old client at login.oscar.nina.chat on port 5190 and it logs in. NINA covers ICQ 2000a through the 8.x releases, describes its ICQ support as a beta still under work, and charges nothing. A separate MIT-licensed project, Open OSCAR Server, reimplements the same protocol in Go for anyone who would rather run their own. NINA does the same job for Yahoo Messenger (https://404memoryfound.com/posts/does-yahoo-messenger-still-exist.html), which is how most of these dead messengers are reachable at all. ## Frequently Asked Questions ### When did ICQ shut down? ICQ shut down on June 26, 2024, twenty-eight years after its first release in November 1996. VK, the Russian company that had owned the messenger since 2010, posted a notice in late May 2024 saying the service would stop working and directing users to VK Messenger and VK WorkSpace. ### Who owns ICQ now? VK Company Limited, formerly Mail.ru Group, has owned ICQ since Digital Sky Technologies bought it from AOL for $187.5 million under an agreement signed in April 2010. VK still holds the ICQ brand in 2026, but it has operated no ICQ service since June 26, 2024. ### What does ICQ stand for? ICQ is not an abbreviation. Mirabilis named the messenger in 1996 with a phonetic spelling of the phrase I Seek You, which described what the program did: find other people anywhere on the internet and message them directly, without both sides paying for the same online service. **Sources:** - America Online Inc., Form 8-K on the Mirabilis acquisition (1998), SEC EDGAR: https://www.sec.gov/Archives/edgar/data/0000883780/000088378098000023/0000883780-98-000023.txt - AOL Inc., Form 8-K on the sale of ICQ to Digital Sky Technologies (2010), SEC EDGAR: https://www.sec.gov/Archives/edgar/data/0001468516/000119312510156846/d8k.htm - ICQ, Encyclopaedia Britannica: https://www.britannica.com/technology/ICQ - Venerable ICQ messaging service to end operations in June, The Register (2024): https://www.theregister.com/2024/05/27/icq_end_of_service/ - ICQ connection guides, NINA: https://nina.chat/connect/icq/ --- # Does Ask Jeeves Still Exist? Ask.com Closed in 2026 URL: https://404memoryfound.com/posts/what-happened-to-ask-jeeves-search-engine.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-27 Updated: 2026-09-09 Topics: Internet Culture, Business Blunders **Summary:** Ask Jeeves no longer exists. The site was renamed Ask.com in February 2006, stopped running its own search engine in 2010, and its owner IAC shut it down on May 1, 2026. Garrett Gruener and David Warthen founded the company in Berkeley in June 1996, and IAC bought it for $1.85 billion in July 2005. **Key facts:** - Founded: June 1996 in Berkeley, California, by Garrett Gruener and David Warthen - Launched: April 1997, natural-language question search - Sold for: $1.85 billion to IAC, July 2005 - Owner: IAC, through its Ask Media Group division - Status: Closed. Ask.com shut down on May 1, 2026 ## Does Ask Jeeves still exist in 2026? No. The search engine that launched as Ask Jeeves, and spent its last twenty years as Ask.com, shut down on May 1, 2026. Visitors who typed the address that week got a farewell page instead of a search box. It opened with a single line: "Every great search must come to an end." The owner was IAC, the New York holding company Barry Diller assembled, which had run the site through a division called Ask Media Group. IAC explained the decision in one sentence: "As IAC continues to sharpen its focus, we have made the decision to discontinue our search business, which includes Ask.com." The closing announcement ended a longer fiction. Ask.com had not run a search engine of its own since 2010. For its final sixteen years it was a front end, taking results and paid listings from somebody else while the brand coasted on a butler most of its visitors had never seen. ## Who founded Ask Jeeves, and when? Garrett Gruener, a venture capitalist, and David Warthen, a software developer, started it in Berkeley, California, in 1996. The company was incorporated in California in June 1996 and reincorporated in Delaware in June 1999, just before it went public. Warthen was still chief technology officer at the IPO, and the prospectus told investors that losing him would hurt the business. The service opened to the public in April 1997. The pitch was one sentence long: type your question the way you would say it out loud, and a cartoon butler would go find the answer. The name came from the unflappable valet in P.G. Wodehouse's comic novels, a character built entirely around knowing things his employer did not. Underneath, it was less magic than staffing. Human editors mapped common questions to good pages, the software matched your phrasing against those templates, and anything it could not match fell through to an ordinary keyword search. That was still a mercy in 1997, when rivals like AltaVista (https://404memoryfound.com/posts/what-happened-to-altavista-search-engine.html) expected you to know Boolean operators to get a decent result. It worked. Ask Jeeves handled around 3,000 questions a day at the start. By April 1999 it was close to a million a day. ## The $14 IPO and the $1.85 billion sale to IAC Ask Jeeves sold 3,000,000 shares at $14.00 each in a prospectus dated June 30, 1999, with an over-allotment option on another 450,000. It traded on the NASDAQ as ASKJ from July 1999 until the ticker was retired in the acquisition six years later. The dot-com crash hit it as hard as it hit Excite (https://404memoryfound.com/posts/what-happened-to-excite-search-engine-google.html) and Lycos (https://404memoryfound.com/posts/what-happened-to-lycos-search-engine.html), but Ask Jeeves did something the other two never managed. It survived, and it went and bought real technology. On September 18, 2001 it acquired Teoma for more than $1.5 million, giving it a crawler and ranking system of its own instead of a licensed index and a database of scripted answers. That turnaround is what IAC paid for. IAC bought Ask Jeeves in July 2005 for $1.85 billion. In February 2006 the new owners renamed the site Ask.com and dropped the butler entirely, on the theory that a cartoon servant made the product look like a toy next to Google's blank white page. ## Why did Ask Jeeves fail against Google? Scale, and the cost of keeping up with it. Crawling and indexing the whole web got more expensive every year, and Ask was paying that bill out of a much smaller advertising business than Google, Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) or Microsoft. The math stopped working on November 9, 2010. Ask.com announced it was getting out of algorithmic search, cutting 130 engineers in Edison, New Jersey and Hangzhou, China, and handing its web results to an outside provider. At that point Ask held roughly 2 percent of the United States search market. Google held about 65 percent. Diller, who had told the world in 2005 that four or five search players could thrive together, gave the blunt version in 2010: "We've realized in the last few years you can't compete head on with Google." Ask.com kept a search box, but the thing behind it was rented. The natural-language idea was not the problem. Asking a machine a full question and getting a written answer is now the default behavior of every AI assistant on the market. Ask Jeeves was simply doing it in 1997 with editors and pattern matching, and it never found a way to make that scale into a business that could pay for a web-sized index. ## Is the Jeeves butler coming back? He is not. The butler had a good run in the mainstream: he floated as a balloon in the Macy's Thanksgiving Day Parade every year from 2000 through 2004, which is more cultural reach than most search engines ever managed. He was retired in February 2006 with the rename. Then he came back, briefly and only in Britain, when Ask brought him out of retirement in April 2009 as a redesigned CGI character for the UK site. That revival faded too, and he stayed gone. The farewell page in 2026 gave him a last mention, saying "Jeeves' spirit endures." With the search business closed and the domain retired, there is no product left for him to front. ## What happened to Ask.com at the end? Money, specifically Google's money. For years IAC's Search segment ran on a services agreement under which Google supplied most of the paid listings shown on Ask Media Group's pages. IAC's annual report for 2025 stated plainly that the Services Agreement "expires by its terms on March 31, 2026." Ask.com closed one month after that date. The site had reached 100 million users a month worldwide in 2012, but a decade of Google algorithm changes, followed by AI answer engines that gave people the written response Ask Jeeves had promised in 1997, left very little underneath the brand. What survives is the archive and the phrase. "Ask Jeeves" still gets typed into Google every month by people who assume it is out there somewhere, which is its own kind of afterlife. ## Frequently Asked Questions ### Who owns Ask Jeeves now? IAC has owned the brand since July 2005, when it bought Ask Jeeves for $1.85 billion, and it ran the site through its Ask Media Group division. IAC discontinued the whole search business in 2026 and closed Ask.com on May 1 of that year, so the brand is owned but no longer operated. ### Was Ask Jeeves before Google? Yes. Ask Jeeves was incorporated in June 1996 and its question service opened to the public in April 1997, before Google existed as a company. By the time IAC bought Ask Jeeves in 2005, though, Google had long since passed it, and Ask was down to roughly 2 percent of United States search by 2010. ### When did Ask Jeeves come out? The service launched in April 1997, about ten months after the company was incorporated in California in June 1996. It went public on the NASDAQ in July 1999 at $14.00 a share under the ticker ASKJ. **Sources:** - Ask Jeeves, Inc., Form 424B4 IPO prospectus, June 30, 1999 (U.S. Securities and Exchange Commission): https://www.sec.gov/Archives/edgar/data/1054298/0000950149-99-001225.txt - IAC Inc., Form 10-K for the year ended December 31, 2025 (U.S. Securities and Exchange Commission): https://www.sec.gov/Archives/edgar/data/1800227/000162828026009997/iaci-20251231.htm - PCWorld: Ask.com Gives Up on Search, Site Refocuses on Q&A Service: https://www.pcworld.com/article/498530/rip_ask_com.html - PPC Land: Ask.com closes after nearly 30 years as IAC exits the search business: https://ppc.land/ask-com-closes-after-nearly-30-years-as-iac-exits-the-search-business/ - TechCrunch: Farewell, Jeeves: Ask.com shuts down: https://techcrunch.com/2026/05/02/farewell-jeeves-ask-com-shuts-down/ --- # Is TiVo Still Around in 2026? Yes, but Not the DVR URL: https://404memoryfound.com/posts/what-happened-to-tivo-dvr-pioneer.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-27 Updated: 2026-09-16 Topics: Hardware, Then vs Now **Summary:** TiVo still exists in 2026, but not as a box you can buy. The brand belongs to Xperi Inc., which ships it as TiVo OS on smart televisions rather than as a recorder. TiVo stopped selling DVR hardware on October 1, 2025, 26 years after the first unit reached buyers on March 31, 1999. **Key facts:** - Launched: March 31, 1999, built by Philips Electronics: $499 for 14 hours, $999 for 30 hours - Service price in 1999: $9.95 a month, $99 a year, or $199 for a lifetime subscription - Peak subscriptions: 4,418,000 on January 31, 2006 - Owner today: Xperi Inc., through TiVo Platform Technologies LLC - Status today: DVR hardware discontinued October 1, 2025; TiVo OS ships on smart TVs ## What did the first TiVo actually do in 1999? Philips Electronics and TiVo announced the first personal television system on March 31, 1999. The product was a Philips receiver with a hard drive inside it, recording broadcast television as digital files instead of tape. Two configurations shipped: a 14-hour model at $499 and a 30-hour model at $999. The hardware was only half the purchase. Buyers also paid for the TiVo Personal Television Service, which pulled program listings down overnight and drove the season passes and the thumbs-up recommendations. TiVo priced that at $9.95 a month, $99 a year, or $199 for a lifetime subscription attached to the box. "Today marks the beginning of a dramatic revolution in the world of television," said Mike Ramsay, TiVo's president and chief executive, in the launch announcement in 1999. That held up better than most launch-day copy. Pausing live television was genuinely new in 1999, and the VCR (https://404memoryfound.com/posts/betamax-vs-vhs-format-war.html) it replaced had spent two decades failing to make timer recording simple for anyone. ## How big did TiVo get before it started shrinking? TiVo came out of Silicon Graphics. Mike Ramsay and Jim Barton, engineers who had worked on Time Warner's Full Service Network interactive television trial, incorporated the company as Teleworld in 1997 and renamed it TiVo before the launch. Growth then came from a partner rather than from shop shelves. DirecTV sold co-branded satellite receivers running TiVo software, and those households counted on TiVo's books. In the results it filed for the quarter ended January 31, 2006, TiVo reported 4,418,000 cumulative subscriptions, the highest number the company ever posted. The split inside that figure explains everything that followed. Only 1,572,000 were TiVo-owned subscriptions, meaning people who had bought a TiVo themselves. The other 2,846,000 arrived through DirecTV. When DirecTV moved to a DVR platform of its own, the larger part of TiVo's subscriber base sat inside a company that no longer had a reason to keep it. ## Why did cable companies beat TiVo at its own product? A standalone TiVo asked an American household to buy a second box, pay a second monthly fee, and wire it into a cable service they were already paying for. Comcast, Time Warner Cable and Cox Communications answered with recording built into the set-top box they already rented out, charged as a line on the existing bill. The cable DVRs were worse products. The guides were slower, the recommendation features were missing, and none of them matched what TiVo had spent years refining. That did not decide it. A cable DVR arrived with one phone call, no separate purchase and no second subscription, and convenience beat quality. TiVo's response was to license its software to the operators rather than fight them for the same living room. A few operator deals followed, but the operators kept the customer relationship and the billing, and TiVo slid into being a supplier to the industry that had taken its category away from it. ## How did one patent keep TiVo alive? TiVo held US Patent No. 6,233,389, known as the "Time Warp" patent, covering the trick of writing a broadcast stream to a disk while playing back from that same disk. In 2004, TiVo sued EchoStar Communications, the parent company of Dish Network, for infringing it. The case ran for years through trial, appeal and a contempt finding. It ended on April 29, 2011, when Dish Network and EchoStar agreed to pay TiVo $500 million: $300 million up front and $200 million in six equal annual installments between 2012 and 2017. The deal also licensed the Time Warp patent to Dish Network for the remaining life of the patent. That single payment was worth more than the hardware business produced in any comparable stretch, and it quietly changed what TiVo was. Boxes kept shipping, but the patent portfolio had become the valuable asset, and every owner TiVo has had since bought the company partly for it. ## Who owns TiVo today? TiVo is owned by Xperi Inc. The brand sits inside TiVo Platform Technologies LLC, which Xperi describes in its own announcements as a wholly owned subsidiary. TiVo has not been an independent company since 2016. The chain of ownership is short. Rovi Corporation, a guide-data and patent licensing firm, bought TiVo in 2016 and took the TiVo name for the combined business. That company merged with Xperi in 2020. In October 2022, Xperi split in two: the patent licensing arm became Adeia, and the product side, TiVo included, kept the Xperi name. What Xperi sells under the TiVo name in 2026 is software. TiVo OS ships preinstalled on televisions, and at CES 2026 the company said smart TVs powered by TiVo were available across 15 European countries through manufacturing partners representing 17 brands, among them Panasonic, Sharp and Vestel. "Our mission is to simplify and enrich television for everyone," said Geir Skaaden, Xperi's chief product and services officer, at that show. ## Why did TiVo stop making DVRs in 2025? The last TiVo DVR, the TiVo Edge, launched in 2019, and nothing replaced it. On October 1, 2025, TiVo stopped selling physical DVR products online and through its agents, closing 26 years of TiVo hardware. "TiVo no longer manufactures hardware, and our remaining inventory is now depleted, though we will continue to offer support for the products going forward," the company told Variety in October 2025. Owners keep their recordings and their service. There is simply nothing new to buy. The cause was the end of the category rather than the end of the company. Time-shifting was the entire purpose of a DVR, and Netflix (https://404memoryfound.com/posts/netflix-dvd-mail-service-shutdown.html) and the streaming services behind it made recording a broadcast pointless for most viewers. The living-room box as a business went the same way before: WebTV (https://404memoryfound.com/posts/what-happened-to-webtv-internet-television.html) showed in the 1990s how fast one of these can stop making sense. ### Where to find one today TiVo no longer sells hardware, so the Philips-built units from 1999 and the peanut-shaped remote are collector pieces rather than working products. They surface alongside other 1990s television gear on Etsy (https://www.etsy.com/search?q=tivo). Buy one as an object, not as a recorder: TiVo supports the boxes it sold, but a 1999 unit that listed at $499 new has no guaranteed route back to a program guide. ## Frequently Asked Questions ### Does TiVo still exist in 2026? TiVo still exists in 2026, but as software rather than as a recorder. The brand is owned by Xperi Inc. through TiVo Platform Technologies LLC, and TiVo OS ships preinstalled on smart televisions. TiVo stopped selling DVR hardware on October 1, 2025. ### Can you still buy a new TiVo DVR? No. TiVo stopped selling DVR hardware and accessories on October 1, 2025, and the last model it released was the TiVo Edge in 2019. TiVo said in October 2025 that its remaining inventory was depleted and that it would keep supporting the boxes already in customers' homes. ### Who invented TiVo, and when did the first one ship? TiVo was created by Mike Ramsay and Jim Barton, two engineers from Silicon Graphics who incorporated the company as Teleworld in 1997. The first TiVo reached buyers on March 31, 1999, built by Philips Electronics and priced at $499 for a 14-hour recorder and $999 for a 30-hour model. **Sources:** - TiVo and Philips Deliver First Personal Television System, press release, March 31, 1999: https://www.aes-media.org/historical/html/recording.technology.history/tivo2.html - TiVo Inc. Form 8-K, fourth quarter and fiscal year 2005 results (SEC EDGAR, February 2006): https://www.sec.gov/Archives/edgar/data/0001088825/000119312506048383/dex991.htm - TiVo, DISH Network and EchoStar Announce Half-Billion Dollar Settlement of Patent Litigation (2011): https://ir.echostar.com/news-releases/news-release-details/tivo-dish-network-and-echostar-announce-half-billion-dollar - Variety: TiVo Stops Selling DVRs, Exiting the Hardware Business After 26 Years (October 2025): https://variety.com/2025/tv/news/tivo-stopped-selling-dvrs-exits-hardware-business-1236552025/ - Xperi: TiVo OS Expands Entertainment Horizons with New Features, Content Partnerships, and Growing Device Footprint at CES 2026: https://investor.xperi.com/news/news-details/2026/TiVo-OS-Expands-Entertainment-Horizons-with-New-Features-Content-Partnerships-and-Growing-Device-Footprint-at-CES-2026/default.aspx --- # Is BlackBerry Still in Business? Yes, but No Phones URL: https://404memoryfound.com/posts/what-happened-to-blackberry-smartphone.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-27 Updated: 2026-09-16 Topics: Hardware, Business Blunders **Summary:** BlackBerry is still in business in 2026, but it has not sold a phone since 2016. BlackBerry Limited is a publicly traded software company in Waterloo, Ontario, that builds the QNX operating system used inside cars and sells secure communications to governments and defense agencies. It reported revenue of $549.1 million in fiscal 2026, against $19.9 billion at the handset peak in fiscal 2011. **Key facts:** - Founded: 1984 in Waterloo, Ontario, as Research In Motion - Status today: Active software company; no BlackBerry phones since 2016 - Owner today: Publicly traded, no parent company (NYSE: BB, TSX: BB) - Revenue today: $549.1 million in fiscal 2026, down from $19.9 billion in fiscal 2011 - Old phones: Dead. Legacy services were switched off on January 4, 2022 ## How a Waterloo pager company took over corporate email Research In Motion was founded in 1984 in Waterloo, Ontario, by Mike Lazaridis and his school friend Douglas Fregin. For its first decade the company did contract engineering on wireless data: point of sale terminals, barcode readers and radio modems. Jim Balsillie joined as co-chief executive in 1992 and took over sales and finance while Lazaridis ran engineering. The product that made the company was the BlackBerry 850, a two way pager released in 1999. It could not place calls. What it did was push email to a device in your pocket the moment the message arrived, onto a keyboard small enough to type on with two thumbs. The other half of the business mattered more than the hardware. BlackBerry Enterprise Server let a corporate IT department hand out hundreds of devices, encrypt the traffic and wipe a lost handset remotely. RIM was not really selling phones to people. It was selling a managed, secure email system to companies, and the handsets came with it. Wall Street adopted it first, then law firms, then anyone with a corporate expense account. ## How big did BlackBerry get at its peak? Big enough that the habit got its own word. Webster's New World College Dictionary named "crackberry" its word of the year for 2006, a term that covered both the device and the person who could not put it down. The money peaked in fiscal 2011, the year that ended in February 2011, when Research In Motion reported revenue of $19.9 billion, up 33 percent on the year before. BlackBerry was the default smartphone of corporate North America and standard issue in Washington and on Wall Street. BlackBerry Messenger did something the company never planned. BBM was free, instant and tied to a PIN instead of a phone number, and it turned a business tool into a teenage social network in Britain, Indonesia and Nigeria. Trading BBM PINs was how a generation swapped contact details before WhatsApp existed. The peak was also the trap. Every part of RIM was built around physical keyboards, compressed data and corporate IT buyers at the exact moment the market moved to touchscreens, app stores and consumers choosing their own phones. Nokia (https://404memoryfound.com/posts/does-nokia-still-make-phones.html) and Palm (https://404memoryfound.com/posts/is-palm-still-a-company.html) were caught by the same shift in the same years. ## What did RIM say when the iPhone launched? Steve Jobs introduced the iPhone on January 9, 2007. According to "Losing the Signal", the 2015 account of the company by Jacquie McNish and Sean Silcoff, Lazaridis pulled Balsillie in front of a computer to watch the webcast and told him: "These guys are really, really good. This is different." Balsillie was unimpressed. His reply, reported by Forbes in 2015, was "we'll be fine". The technical objections were all accurate in 2007. The first iPhone had no physical keyboard, no 3G radio, no app store and no enterprise email management. RIM executives told the authors of that book that the device was not secure and that it drained its battery. Every one of those complaints was true, and none of them mattered. What RIM evaluated was a handset. What Apple had shipped was a platform. The App Store opened in July 2008, Android phones arrived the same year, and within two product cycles the question of what a phone was for had changed from email to everything. ## Why BlackBerry 10 could not stop the slide RIM's first touchscreen answer was the BlackBerry Storm in November 2008, a screen that physically clicked when you pressed it. It satisfied neither touchscreen buyers nor keyboard loyalists. The PlayBook tablet followed in April 2011 and shipped without native email, calendar or contacts apps, which meant a BlackBerry tablet needed a BlackBerry phone tethered to it to read mail. In December 2011 RIM warned that it would take a charge of roughly $485 million on unsold PlayBook inventory. Lazaridis and Balsillie both stepped down as co-chief executives in January 2012 and were replaced by Thorsten Heins. The company renamed itself BlackBerry and launched BlackBerry 10 alongside the Z10 and Q10 handsets on January 30, 2013. The software was decent and the timing was fatal. BlackBerry 10 arrived with roughly 70,000 apps against more than 700,000 each on Apple's App Store and Google Play. Developers had no reason to build for it and buyers had no reason to switch. BlackBerry left the handset business in 2016 and licensed the name to TCL instead. In the fourth quarter of 2016, Gartner counted 207,900 BlackBerry OS handsets sold out of 432 million smartphones worldwide, a share that rounded to 0.0 percent. ## Who owns BlackBerry today? Nobody in particular. BlackBerry Limited is a public company listed in New York and Toronto under the ticker BB, still headquartered in Waterloo, Ontario, and run since December 2023 by chief executive John Giamatteo. There is no parent company and no private buyer. What it sells has changed completely. The Cylance security business, bought in 2018 for $1.4 billion, went to Arctic Wolf for $160 million in a sale that closed in February 2025. Two divisions are left. QNX builds the embedded operating system inside car dashboards, instrument clusters and driver assistance systems, and BlackBerry said in December 2025 that Counterpoint Research had found QNX software running in more than 275 million vehicles. Secure Communications sells encrypted messaging and crisis alerting to governments and defense agencies. In fiscal 2026 BlackBerry reported revenue of $549.1 million, made up of $268.0 million from QNX and $258.9 million from Secure Communications, with a QNX royalty backlog of $950 million. On the earnings call Giamatteo said: "The turnaround is complete, and the BlackBerry story is now a growth story." Guidance for fiscal 2027 is $584 million to $611 million. It is a real business, and it is under 3 percent of the fiscal 2011 peak. ## Do old BlackBerry phones still work? Not as phones. BlackBerry switched off the services behind BlackBerry OS 7.1 and earlier, BlackBerry 10 and the PlayBook on January 4, 2022. Devices running those systems lost data, calls, text messages, BBM and, over carrier networks, emergency calling. A BlackBerry OS handset now joins the Motorola Razr V3 (https://404memoryfound.com/posts/motorola-razr-v3-coolest-phone-ever-made.html) in the category of phones with no network left to reach. The exceptions are the Android handsets TCL built under licence, the KEYone of 2017 and the Key2 of 2018. Those run Google's software rather than BlackBerry's own, so they never depended on the servers that went dark in 2022. TCL's licence ended in 2020. A startup called OnwardMobility picked it up in August 2020 and promised a 5G BlackBerry with a physical keyboard, missed every deadline it set and shut down in February 2022 without shipping a device. Nobody has held the licence since. Where to find one today. A BlackBerry in 2026 is a desk object rather than a working phone, and what a handset costs depends entirely on the model, the condition and whether the battery still takes a charge. Sellers list Bold, Curve and Pearl units, often with the original box and charger, on Etsy (https://www.etsy.com/search?q=blackberry+phone). Check the model number in the listing and ask the seller whether it powers on, because nothing running BlackBerry OS will connect to a network again. ## Frequently Asked Questions ### Does BlackBerry still make phones? No. BlackBerry stopped making its own handsets in 2016 and licensed the name to TCL, which sold Android powered BlackBerry phones until that licence ended in 2020. A startup called OnwardMobility held the licence from August 2020 and closed in February 2022 without shipping a device. BlackBerry Limited has sold no handsets of any kind since. ### Who owns BlackBerry now? BlackBerry Limited is not owned by another company. It is a publicly traded firm listed on the New York Stock Exchange and the Toronto Stock Exchange under the ticker BB, based in Waterloo, Ontario, and led by chief executive John Giamatteo since December 2023. Its Cylance security unit was sold to Arctic Wolf in a deal that closed in February 2025. ### When did BlackBerry phones stop working? BlackBerry shut down the services behind BlackBerry OS 7.1 and earlier, BlackBerry 10 and the PlayBook on January 4, 2022. From that date those devices lost data, phone calls, text messages, BBM and emergency calling over carrier networks. The Android based BlackBerry phones TCL built, such as the KEYone and Key2, were not affected. **Sources:** - BlackBerry (BB) Q4 2026 Earnings Call Transcript, The Motley Fool: https://www.fool.com/earnings/call-transcripts/2026/06/02/blackberry-bb-q4-2026-earnings-transcript/ - Research In Motion Reports Year-End and Fourth Quarter Results for Fiscal 2011: https://www.globenewswire.com/news-release/2011/03/24/1430467/0/en/Research-In-Motion-Reports-Year-End-and-Fourth-Quarter-Results-for-Fiscal-2011.html - BlackBerry's Famous Last Words At 2007 iPhone Launch, Forbes (2015): https://www.forbes.com/sites/parmyolson/2015/05/26/blackberry-iphone-book/ - BlackBerry's share of the global smartphone market reaches 0%, Global News (Gartner data): https://globalnews.ca/news/3253615/blackberry-market-share-zero/ - BlackBerry 10 and BlackBerry OS Services End of Life FAQ, BlackBerry: https://www.blackberry.com/us/en/support/devices/end-of-life --- # Is Digg Still Around in 2026? Yes, as an AI Site URL: https://404memoryfound.com/posts/what-happened-to-digg-social-news.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-27 Updated: 2026-09-16 Topics: Internet Culture, Business Blunders **Summary:** Digg still exists in 2026. The site runs at digg.com as an AI news aggregator, owned by founder Kevin Rose and Reddit co-founder Alexis Ohanian, who bought it in March 2025. Their first attempt at a Reddit-style relaunch opened to the public on January 14, 2026 and was shut down two months later after a flood of automated accounts. **Key facts:** - Launched: December 5, 2004, by Kevin Rose and three co-founders - Status today: Live at digg.com as an AI news aggregator, in alpha since May 2026 - Owner today: Kevin Rose and Alexis Ohanian, who bought Digg in March 2025 - Peak valuation: $164 million in 2008, per Forbes - Sold for: $500,000 to Betaworks in July 2012 ## How did Digg decide what the internet read? Digg went live at digg.com on December 5, 2004, built by Kevin Rose with co-founders Owen Byrne, Ron Gorodetzky and Jay Adelson. Rose was already a recognizable face from TechTV (https://404memoryfound.com/posts/what-happened-to-techtv-screen-savers-channel.html), and the site he put up had one rule that separated it from every editor-run news page of the time. Users submitted links. Other users voted them up by digging them, or pushed them down by burying them. Whatever the crowd pushed hardest landed on the front page. No editor picked it, no publisher paid for it, and no ranking team overrode it. The effect on publishers was immediate. A front page slot could send hundreds of thousands of readers at a blog in an afternoon, which is why newsrooms, marketers and search-engine consultants watched the Digg queue the way an earlier generation of readers watched their RSS subscriptions (https://404memoryfound.com/posts/what-happened-to-google-reader-rss.html). Getting dugg was a traffic event with a name. That mechanic also created the problem that would define the rest of the company's history. The value of Digg sat entirely in the judgment of the people using it. The software was replaceable. The crowd was not, and the crowd had no contract with the company. ## Did Google really try to buy Digg for $200 million? Yes, and the popular version of the story has the ending backwards. TechCrunch reported in July 2008 that Google and Digg were in final negotiations, past the term sheet stage, at a price of "around $200 million". Days later, on July 26, 2008, the same outlet reported that Google had walked away during due diligence. Digg did not turn Google down. Google left. Sources cited by TechCrunch at the time pointed at something that surfaced in technical due diligence, with one suggesting the issue was closer to a personality mismatch between the two leadership teams. The difference matters, because the myth that Rose rejected a fortune out of arrogance has followed Digg for almost twenty years. The truth is less dramatic and more ordinary: a large acquisition fell apart in the final weeks, as large acquisitions often do. Anyone who has read the story of Yahoo passing on Google in 1998 (https://404memoryfound.com/posts/yahoo-rejected-google.html) knows how quickly a near miss hardens into folklore. Digg raised a venture round instead. Forbes later reported that the 2008 financing valued the company at $164 million, the high-water mark it never came close to again. ## What did the Digg v4 redesign break in 2010? On August 25, 2010, Digg launched version 4. The redesign dropped the bury button, reweighted the algorithm toward major publishers, and gave brands and outlets automatic feeds into the site. Users read that as a transfer of power. Burying was how the community rejected spam and headline bait, so removing it took away the only lever ordinary members had. Publisher feeds took the front page, the one thing members had controlled since 2004. The protest that followed was unusually literal. Members began submitting Reddit links to Digg and voting them to the top, so that Digg's own front page pointed at its competitor for days. The episode got its own name on the site, the Digg Revolt. The numbers arrived a month later. TechCrunch, citing comScore in October 2010, put Digg's worldwide unique visitors at 18.4 million in August 2010 and 12.8 million in September, a loss of 5.6 million people in a single month. Page views fell harder over the same stretch, from 155 million to 46 million. Nothing about the product had broken. The audience had simply decided the site was no longer theirs. ## How did a $164 million company sell for $500,000? Once the front page lost its regulars, the advertising case went with them. Digg had raised roughly $45 million from investors including Greylock Partners, Marc Andreessen and Ron Conway, and those investors were funding a company whose traffic was now moving in one direction. On July 12, 2012, Forbes reported that Digg had been sold to the New York investment studio Betaworks for $500,000, under the headline that the site had once been worth $164 million. Digg's leadership disputed the figure at the time, saying the deal was larger than the number in circulation, but no revised price was ever confirmed publicly. Betaworks rebuilt Digg as a small editorial aggregator, a human-curated links page with a handful of staff rather than a voting platform. It survived that way for six years. In April 2018 the advertising technology company BuySellAds acquired a majority stake, with Betaworks and earlier shareholders keeping a holding in the joint venture. For most of the next seven years Digg was a working website that almost no one argued about. The brand outlived the community by more than a decade, which is also roughly what happened to MySpace after Facebook (https://404memoryfound.com/posts/rise-and-fall-of-myspace.html). ## Who owns Digg today? Kevin Rose and Alexis Ohanian own Digg. On March 5, 2025, the two announced they had bought the site back from BuySellAds for an undisclosed sum, with funding from True Ventures, where Rose is a partner, and Ohanian's firm Seven Seven Six. Rose took the chairman role and Justin Mezzell became chief executive. The pairing was the part everyone noticed. Ohanian co-founded Reddit, the site that absorbed Digg's users in 2010, and he was now paying to revive the platform Reddit beat. The rebuilt Digg ran invite-only through most of 2025 and reached about 67,000 accounts before opening to everyone on January 14, 2026. It shipped as a mobile-first app with paid moderation tools, public moderation logs and former Reddit moderators hired as advisers. That version is gone. What runs at digg.com in September 2026 is the third design in two years: an aggregator that ranks news rather than hosting communities, launched in alpha at di.gg in May 2026 and built around artificial intelligence coverage first, with other subjects to follow if the format holds. ## Why did the 2026 relaunch collapse in two months? Bots. On March 14, 2026, two months after the public beta, Mezzell posted that Digg was shutting the open beta down, cutting staff and pulling the app, citing an "unprecedented bot problem" alongside what he called the "brutal reality of finding product-market fit in an environment that has fundamentally changed". In the same post, reported by Engadget in March 2026, Mezzell described the scale of it: "We knew bots were part of the landscape, but we didn't appreciate the scale, sophistication, or speed at which they'd find us." Digg said it had banned tens of thousands of accounts and brought in outside vendors before giving up on the beta. Part of the draw was mechanical. Digg still carried real link authority with Google after twenty years, so a fresh submission queue was worth attacking on day one. Search spammers and automated accounts arrived faster than moderation could scale. Rose returned full time in April 2026 and shipped the AI aggregator the following month. TechCrunch reported in May 2026 that the new Digg tracks what about 1,000 of what it calls the most thoughtful voices in artificial intelligence are linking to, then ranks those stories by what is climbing fastest. It is the 2004 idea with the crowd replaced by a shortlist, which is one honest answer to a web where anyone can rent ten thousand voters. ## Frequently Asked Questions ### Is Digg still around? Yes. Digg is still online at digg.com in September 2026, running as an AI news aggregator that launched in alpha at di.gg in May 2026. The Reddit-style version of Digg that opened to the public on January 14, 2026 was shut down on March 14, 2026. ### Who owns Digg now? Digg is owned by its founder Kevin Rose and Reddit co-founder Alexis Ohanian, who bought it from the advertising technology company BuySellAds on March 5, 2025 for an undisclosed price. Rose is chairman and Justin Mezzell is chief executive. Earlier owners were Betaworks, which bought Digg in July 2012, and BuySellAds, which took a majority stake in April 2018. ### Why did Digg lose to Reddit? Digg lost its audience after the version 4 redesign of August 25, 2010 removed the bury button and gave major publishers priority on the front page. Members treated it as a loss of control, and comScore figures reported by TechCrunch show Digg's worldwide unique visitors falling from 18.4 million in August 2010 to 12.8 million in September. Reddit, which launched within a year of Digg, absorbed most of the users who left. **Sources:** - TechCrunch: Google Walks Away From Digg Deal (July 2008): https://techcrunch.com/2008/07/26/google-walks-away-from-digg-deal/ - TechCrunch: Digg's Big Drop, comScore traffic figures (October 2010): https://techcrunch.com/2010/10/26/digg-big-drop/ - Forbes: Digg, Once Worth $164 Million, Sold To Betaworks For $500K (July 2012): https://www.forbes.com/sites/jeffbercovici/2012/07/12/digg-once-worth-164-million-sold-to-betaworks-for-500k/ - Engadget: Digg shuts down for a hard reset because it was flooded with bots (March 2026): https://www.engadget.com/social-media/digg-shuts-down-for-a-hard-reset-because-it-was-flooded-with-bots-153848094.html - TechCrunch: Digg tries again, this time as an AI news aggregator (May 2026): https://techcrunch.com/2026/05/11/digg-tries-again-this-time-as-an-ai-news-aggregator/ --- # Is Second Life Still Around in 2026? Yes, 23 Years Later URL: https://404memoryfound.com/posts/what-happened-to-second-life-virtual-world.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-27 Updated: 2026-09-16 Topics: Internet Culture, Software & Apps **Summary:** Second Life is still running in 2026, 23 years after Linden Lab launched it on June 23, 2003. Linden Research, Inc. is owned by an investor group led by Randy Waterfield and Brad Oberwager, and founder Philip Rosedale came back as chief technology officer in October 2024. Daily concurrency runs in the mid-30,000s, and Linden Lab put the in-world economy at roughly $650 million a year in December 2024. **Key facts:** - Launched: June 23, 2003, by Linden Lab - Status today: Still running in 2026; daily concurrency in the mid-30,000s - Owner today: Linden Research, Inc., bought by an investor group led by Randy Waterfield and Brad Oberwager - Price today: Free to join; Premium costs $119.88 a year as of July 8, 2026 - Peak economy: $567 million in 2009, about 25 percent of the US virtual goods market ## What was Second Life supposed to be in 2003? Philip Rosedale started Linden Lab in San Francisco in 1999 to build virtual reality hardware: goggles, motion rigs, the whole kit. By 2000 the team had accepted that consumer bandwidth and consumer graphics cards could not carry it, and the hardware plan was shelved. What replaced it was software. Linden Lab built a persistent three-dimensional world that ran on an ordinary desktop PC and released it as Second Life on June 23, 2003. That was a year before Gmail, two years before YouTube and three years before Twitter. There was no established category to put it in. Linden Lab insisted it was not a game. There was no score, no levels and no way to win. Users were called Residents rather than players. They bought virtual land, built objects with in-world tools, and kept the intellectual property rights to whatever they made, which is the decision that turned a sandbox into a market. Rosedale has been blunt about how early all of it was. "I started Second Life in 1999, a decade before cloud computing and two decades before AI," he said in Linden Lab's October 2024 announcement of his return to the company. ## How big did the Second Life economy actually get? Between 2006 and 2009, Second Life stopped being a curiosity and became a press event. Reuters opened a virtual news bureau inside the world in 2006 and staffed it with correspondent Adam Pasick, who filed under the avatar name Adam Reuters. IBM held meetings in-world. Toyota built showrooms, Coca-Cola ran promotions, and universities put up campuses. The traffic was real without ever being enormous. Concurrent users inside Second Life peaked at 71,159 on September 27, 2008, a figure the platform has never matched since. The money was the more startling part. VentureBeat reported that the Second Life economy grew 65 percent during 2009 to reach $567 million, which at the time was roughly 25 percent of the entire United States virtual goods market. Residents cashed out $55 million of real money that year. MySpace (https://404memoryfound.com/posts/rise-and-fall-of-myspace.html), still the social network everyone was writing about when that peak arrived, had nothing equivalent to the Linden dollar and no way for its users to earn from it. ## Why did the corporations leave Second Life? September 2008 was the month Second Life hit its concurrency peak and the month Lehman Brothers collapsed. Virtual land was an experimental marketing line item, and experimental marketing line items were the first thing cut. Reuters is the clearest case. Its Second Life bureau stopped filing on September 30, 2008, and the closure was not announced publicly until March 2009. Most other brands wound down quietly rather than issue a statement at all. Linden Lab felt it directly. In June 2010 the company laid off 30 percent of its staff and said it would rebuild Second Life around a browser instead of a downloaded client. Two weeks later chief executive Mark Kingdon resigned, and Rosedale, who had handed over the CEO job in March 2008, came back as interim chief executive. What survived the cut was not the corporate build-out. It was the Residents who had already been there for the clubs, the roleplay, the fashion shops and the freelance building work, and who had never needed a Toyota showroom to justify logging in. ## Who owns Second Life today? Second Life is still operated by Linden Research, Inc., which trades as Linden Lab, and the company changed hands this decade. In July 2020 an investor group led by Randy Waterfield and Brad Oberwager agreed to acquire Linden Research. The deal cleared regulatory review and closed in early 2021, after which Second Life was listed among the businesses in the Waterfield Network. Oberwager serves as executive chair. Rosedale came back as well. He rejoined as a strategic advisor in January 2022, alongside an investment in Linden Lab from his own company, High Fidelity. In October 2024 he returned full time as chief technology officer and joined the board of directors. The reason he gave for coming back was not nostalgia. "We were early, but the success of Second Life to this day shows that we were not wrong," Rosedale said in that same October 2024 announcement. The company he had left in 2010 spent the intervening decade proving that a virtual world could outlast the hype cycle that built it. ## What does Second Life cost in 2026? An account is free. Paying gets you land, a weekly stipend and account perks, and those prices moved twice in 2026, in both directions. On June 15, 2026 Linden Lab cut land prices and dropped the minimum fee for buying Linden dollars from $1.49 to $0.49. On July 8, 2026 membership pricing went the other way: an annual Premium subscription rose from $99.00 to $119.88. The world is also easier to reach than it used to be. Linden Lab put the Second Life mobile app into public beta on iOS and Android in June 2024, opening it to Premium members first before widening access. After two decades as a desktop client, the world finally ran on a phone. Population in 2026 sits in a narrow band. Daily concurrency has run in the mid-30,000s, with the year's high of 48,802 recorded in early March 2026. That is a fraction of 2008. It also makes Second Life one of the few worlds of its era still running its original service rather than a revival, a list that includes Habbo Hotel (https://404memoryfound.com/posts/is-habbo-hotel-still-around.html) and not much else. ## Did Second Life really invent the metaverse? In October 2021 Facebook renamed itself Meta and committed billions of dollars to building a persistent virtual world with avatars, social spaces and a creator economy. Second Life had shipped all four of those in 2003. The figures Linden Lab has published make the point without argument. In its twentieth-anniversary release in June 2023, the company said more than 73 million accounts had been created since 2003 and that around 750,000 people were using the world each month. In December 2024, Oberwager put the lifetime cost of building Second Life at $1.3 billion and the amount paid out to creators at $1.1 billion, against an economy running at roughly $650 million a year. Linden Lab keeps 10 percent of a transaction and the creator keeps the rest. Meta's Horizon Worlds arrived with a corporate roadmap attached. Second Life arrived with almost nothing except tools and property rights, which is closer to how Neopets (https://404memoryfound.com/posts/what-happened-to-neopets-virtual-world.html) and the other long-lived online worlds actually held onto their users. The lesson is not that Linden Lab was smarter than Meta. It is that a world people build themselves is much harder to walk away from than a world handed to them finished. ## Frequently Asked Questions ### Is Second Life still active in 2026? Yes. Second Life is still operating in 2026, 23 years after Linden Lab launched it on June 23, 2003. Daily concurrency through 2026 has run in the mid-30,000s, and the high for the year was 48,802 people online at once in early March 2026. ### Who owns Second Life now? Second Life is owned and operated by Linden Research, Inc., which trades as Linden Lab. An investor group led by Randy Waterfield and Brad Oberwager agreed to buy the company in July 2020 and closed the purchase in early 2021, with Oberwager as executive chair. Founder Philip Rosedale returned as chief technology officer in October 2024. ### Can you still make real money in Second Life? Yes. Second Life Residents still sell virtual goods and services for Linden dollars and convert them into US dollars through Linden Lab's exchange. Linden Lab said in December 2024 that it had paid $1.1 billion to Second Life creators since 2003, against an economy running at about $650 million a year, and that it keeps 10 percent of each transaction. **Sources:** - VentureBeat: Second Life's economy grows 65% to $567M (2010): https://venturebeat.com/business/second-lifes-economy-grows-65-to-567m - VentureBeat: Linden Lab has spent $1.3B building Second Life and paid $1.1B to creators (2024): https://gamesbeat.com/linden-lab-has-spent-1-3b-building-second-life-and-paid-1-1b-to-creators/ - Linden Lab press release: Original Metaverse Second Life Celebrates 20th Birthday (2023): https://www.businesswire.com/news/home/20230621065980/en/Original-Metaverse-Second-Life-Celebrates-20th-Birthday - Linden Lab: Philip Rosedale Returns as CTO to Help Build the Future of Second Life (2024): https://community.secondlife.com/news/featured-news/philip-rosedale-returns-as-cto-to-help-build-the-future-of-second-life-r1547/ - Linden Lab: Second Life Pricing Updates (2026): https://community.secondlife.com/news/featured-news/second-life-pricing-updates-r11294/ --- # Are Tamagotchis Still Made in 2026? Yes, by Bandai URL: https://404memoryfound.com/posts/tamagotchi-digital-pet-that-made-us-feel.html Author: Marcus Vale (404 Memory Found) Published: 2026-03-26 Updated: 2026-09-15 Topics: Gaming, Software & Apps **Summary:** Tamagotchi still exists and Bandai still makes it. The company reported that cumulative worldwide shipments passed 100 million units as of July 31, 2025, and it is running a 30th anniversary line through 2026. The original virtual pet launched in Japan on November 23, 1996 and reached the United States on May 1, 1997 at $17.99; current models sell for $44.99 to $59.99. **Key facts:** - Launched: November 23, 1996, by Bandai in Japan - US launch: May 1, 1997, at $17.99 - Owner today: Bandai, part of Bandai Namco Holdings since 2005 - Status today: Still in production, with a 30th anniversary line in 2026 - Price today: $44.99 for Tamagotchi Paradise, $59.99 for Tamagotchi Uni ## Who invented the Tamagotchi and why was it an egg? The concept came from Akihiro Yokoi at the Japanese design studio WiZ, who wanted a pet a child could carry anywhere without keeping a real animal. He developed it with Aki Maita at Bandai, who ran the marketing and served as the public face of the toy through 1997. The name is a compound of tamago, Japanese for egg, and the Japanese pronunciation of the English word watch. The shell shape was literal. A creature hatched on a liquid crystal screen, and three buttons ran the entire relationship: feed it, clean up after it, play with it, discipline it, put it to bed. The design decision that drove everything else was the absence of a pause button. Neglect the creature for a few hours and it got sick. Neglect it longer and it died, with a small angel or a headstone on the screen and no way to undo it. Bandai put the original Tamagotchi on sale in Japan on November 23, 1996. ## How fast did Tamagotchi sell in the United States? Bandai brought the Tamagotchi to American stores on May 1, 1997, at $17.99. The price was doing real work: cheap enough to be an impulse buy at a checkout counter, expensive enough that a child treated it as a possession rather than a party favor. The launch did not behave like a toy launch. The FAO Schwarz (https://404memoryfound.com/posts/fao-schwarz-store-rockefeller-center.html) in San Francisco sold its entire stock of 3,000 units by 3 p.m. on the first day. The New York flagship moved 10,000 by noon the following day. Resellers were asking several times retail within weeks, and stores rationed units per customer. The worldwide numbers ran past anything Bandai had modeled. Cumulative shipments passed 40 million within roughly two and a half years of the 1996 Japanese launch, split close to evenly between Japan and the rest of the world. For a keychain toy with a screen the size of a postage stamp, that put it in the sales bracket of a home console. ## Why did schools confiscate Tamagotchi? The toy had no pause, so a child who left it in a locker risked coming back to a dead pet. Tamagotchi went into classrooms instead, and it beeped whenever it wanted food, attention or cleaning. Teachers were losing lesson plans to it within weeks of the American launch. Schools in Florida, Massachusetts and California banned them outright during 1997, and schools in the United Kingdom, Australia and Japan did the same. Confiscation created a second problem, because a device locked in a desk drawer would starve by the end of the day. Students set up informal rotas, handing pets to friends who had a free period. Administrators ended up negotiating with parents rather than students. A dean at Hale Middle School in Woodland Hills, California, told the Deseret News in 1997 that the school required a note from home reading "I realize they are a disruption and I will keep them at home." That is close to the policy language American schools would write for mobile phones a decade later. ## Did people really bury Tamagotchi in pet cemeteries? Yes. In January 1997, two teenagers took their dead Tamagotchi, named Sid and Arty, to a pet cemetery at Pontsmill in Cornwall, England, and buried them in small wooden coffins. The owner, Terry Squires, fenced off a section of the field for digital pets after devices began arriving by post from Switzerland, Germany, France, Canada and the United States. Online, memorial pages collected eulogies, ages at death and causes of death, in the tiled-background HTML of the period. A handful of them sit alongside the other 90s websites still online (https://404memoryfound.com/posts/90s-websites-still-online.html) today, abandoned but intact. Researchers gave the behavior a name. The Tamagotchi effect describes emotional attachment to a machine or a piece of software, and it remains the standard term for what happens when a device simulates dependency convincingly enough. The loop was crude: feed it and it grows, ignore it and it suffers. That was sufficient. The same mechanic later carried Neopets (https://404memoryfound.com/posts/what-happened-to-neopets-virtual-world.html) and most of the mobile pet games that followed. ## Who owns Tamagotchi today? Bandai owns it and still makes it. Bandai merged with Namco in 2005 to form what is now Bandai Namco Holdings, and Tamagotchi sits in the toy division. The brand has never been sold off or handed to a licensee. The line nearly ended once. Bandai over-produced through 1998, retailers shipped stock back, and the company booked a loss of about 6 billion yen, roughly $38 million at the time, on inventory it could not move. Around 2.5 million unsold units were destroyed and the original run was wound down by 1999. Connectivity brought it back. The Tamagotchi Connection added infrared trading between devices in 2004. The Tamagotchi Uni arrived in 2023 with a color screen, Wi-Fi and a shared online space, and Tamagotchi Paradise reached North America on August 1, 2025. Bandai reported cumulative worldwide shipments past 100 million units as of July 31, 2025, roughly 49 percent of them in Japan, 33 percent in the United States and 16 percent in Europe. The institutions caught up the same year. The Strong National Museum of Play inducted Tamagotchi into the World Video Game Hall of Fame on May 8, 2025. Collections manager Kristy Hisert said it "provided players with feelings of connection, caring, and customization, a respite from competition and fighting games." ## Where to find one today A new device is the cheap route. Bandai lists Tamagotchi Paradise at $44.99 and Tamagotchi Uni at $59.99, both stocked by mainstream American retailers, so nobody has to hunt a 1997 unit to play one. The vintage market is where the money goes. Loose originals from 1996 and 1997 commonly sell in the $40 to $50 range in 2026. Boxed examples run into the hundreds, and so do the Japan-only color runs that never shipped to North America. If you want a working unit rather than a shelf piece, check the battery contacts before buying, because corroded cells are the usual reason a clean-looking shell has a dead screen. Vintage units, cases and keychain hardware turn up on Etsy (https://www.etsy.com/search?q=vintage+tamagotchi). ## Frequently Asked Questions ### Are Tamagotchis still made in 2026? Yes. Bandai has kept Tamagotchi in production since 1996 and is running a 30th anniversary line through 2026. The current devices include Tamagotchi Paradise, which reached North America on August 1, 2025 at $44.99, and the Wi-Fi connected Tamagotchi Uni from 2023 at $59.99. ### How many Tamagotchis have been sold worldwide? Bandai reported that cumulative worldwide Tamagotchi shipments passed 100 million units as of July 31, 2025, a milestone it announced at the Tokyo Toy Show that August. Roughly 49 percent went to Japan and 33 percent to the United States. The first 40 million shipped within about two and a half years of the November 1996 Japanese launch. ### How much is an original 1997 Tamagotchi worth? A loose original Tamagotchi from 1996 or 1997 commonly sells for $40 to $50 in 2026, while boxed and sealed examples reach into the hundreds. The highest prices go to Japan-exclusive color runs and limited collaborations, not to the standard units that sold by the tens of millions in 1997. **Sources:** - The Japan Times: Tamagotchi hatches 100 million shipments worldwide (2025): https://www.japantimes.co.jp/business/2025/09/02/companies/tamagotchi-100-million/ - The Strong National Museum of Play: 2025 World Video Game Hall of Fame inductees: https://www.museumofplay.org/press-release/2025-world-video-game-hall-of-fame-inductees-announced/ - Mental Floss: A Brief History of the Tamagotchi: https://www.mentalfloss.com/article/642373/tamagotchi-history - Time Extension: The Tamagotchi boom cost Bandai $38 million: https://www.timeextension.com/news/2024/05/tamagotchi-boom-cost-bandai-usd38-million-and-unsold-stock-was-buried-atari-style - Bandai: official Tamagotchi brand page: https://www.bandai.com/brands/tamagotchi/ --- # Who Owns MTV Now, and Does It Still Play Music? URL: https://404memoryfound.com/posts/when-mtv-actually-played-music-rise-fall.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-26 Updated: 2026-09-15 Topics: Music & Entertainment, Internet Culture **Summary:** MTV still exists in 2026. The flagship channel broadcasts in the United States and the United Kingdom under Paramount Skydance Corporation, which took over when Skydance Media and Paramount Global completed their merger on August 7, 2025. What ended on December 31, 2025 were the music-only channels: MTV Music, MTV 80s, MTV 90s, Club MTV and MTV Live all went dark, and the main channel has not built its schedule around music videos since the 2000s. **Key facts:** - Launched: August 1, 1981, with the Buggles' Video Killed the Radio Star - Owner today: Paramount Skydance Corporation, since August 7, 2025 - Status today: On air in the US and UK; five music-only channels closed December 31, 2025 - Music videos aired: Up to 8 hours a day in 2000, about 3 hours a day by 2008 - Last daily countdown: Total Request Live, September 14, 1998 to November 16, 2008 ## Did MTV shut down on New Year's Eve 2025? No. On January 1, 2026 social media filled with posts saying MTV had gone off the air after 44 years, and Variety published a correction the same week. The flagship MTV channel kept broadcasting in the United States and the United Kingdom. What actually closed were the music-only channels. Five of them went dark in the UK on December 31, 2025: MTV Music, MTV 80s, MTV 90s, Club MTV and MTV Live. Paramount shut music-only channels in Australia, Poland, France and Brazil as well. A Paramount spokesperson told Variety in 2025 that "MTV's specialist music channels in the UK will no longer operate as linear channels," and said the company was adjusting its international pay TV portfolio as audiences moved to streaming and digital platforms. MTV and MTV2 in the United States were not part of that cut. So the answer is narrower than the headlines. MTV exists, and it is owned by Paramount Skydance Corporation, which took over when Skydance Media and Paramount Global completed their merger on August 7, 2025. The version of MTV that ran videos back to back all day stopped existing in most of the world at the end of 2025. ## What did MTV broadcast on its first day in 1981? MTV launched at 12:01 a.m. on August 1, 1981 with the line "Ladies and gentlemen, rock and roll," then footage of a space shuttle launch, then the first video: "Video Killed the Radio Star" by the Buggles. The song was not even new. It had been released as a single in 1979. The channel reached only a handful of cable systems at first, most of them in New Jersey, and record labels had to be talked into handing over videos for nothing. That argument ended on December 2, 1983, when MTV premiered Michael Jackson's "Thriller." It ran 14 minutes, was directed by John Landis, and cost roughly $500,000, part of it covered by MTV and Showtime in exchange for the rights to air it. A cable channel had become the place a record was launched rather than a place it was repeated. The first MTV Video Music Awards followed in 1984. ## Why did The Real World change MTV's economics in 1992? The Real World premiered on May 21, 1992. Mary-Ellis Bunim and Jonathan Murray put seven strangers in a New York loft, pointed cameras at them and dropped the script, borrowing the idea from An American Family, the PBS documentary series from 1973. It became the longest-running program in MTV history, and it solved a problem videos never could. A music video was free to air, but it belonged to a record label. MTV could play it, promote it and build a countdown around it, and own none of it. An unscripted series was different. MTV commissioned it, kept it, repeated it, sold it abroad and shot another season for a fraction of what scripted television cost. Every reality order that followed, from Road Rules in 1995 to Jersey Shore in 2009, ran on the same logic. Videos filled airtime. Shows built a library. ## When did music videos actually disappear from MTV? There is no single date, which is why people keep asking. MTV president Van Toffler told Billboard in 2001 that "clearly the novelty of just showing music videos has worn off," and the schedule moved with him. Videos still filled up to eight hours a day in 2000. By 2008 the average was about three hours a day, much of it before breakfast, and through the 2010s even that thinned out. The last daily countdown ended in the middle of that slide. Total Request Live launched on September 14, 1998, with Carson Daly counting down the ten most requested videos from a studio window above Times Square. MTV announced the cancellation in September 2008 and aired a three-hour finale, Total Finale Live, on November 16, 2008. It is still the longest-running live series MTV has produced. Two outside forces finished the job. Napster (https://404memoryfound.com/posts/napster-destroyed-music.html) gutted the label revenue that paid for videos, and YouTube, launched in 2005, meant nobody had to sit through an hour of programming to see one. ## What fills MTV's schedule in 2026? Clip shows and reruns, mostly. Variety reported in November 2025 that MTV had canceled Ridiculousness after 46 seasons, with already-filmed episodes still scheduled into 2026. Reruns of that one show had carried most of the channel's week for years. The one music night left is the Video Music Awards, and even that moved house. The 2025 VMAs aired on CBS for the first time on September 7, 2025, simulcast on MTV and streaming on Paramount+. More than 5.5 million people watched, up 42 percent from 3.91 million the year before, and the largest television audience the show had drawn since 2019. That number is worth reading twice. The biggest music event on a channel named Music Television now needs a broadcast network to find its crowd. ## Where did the MTV News archive go? MTV News closed in May 2023, when Paramount cut 25 percent of US staff across Paramount Media Networks, Showtime and MTV Entertainment Studios. Paramount Media Networks president Chris McCarthy pointed to "broader economic headwinds like many of our peers." The reporting outlived the newsroom by about a year. In June 2024 mtv.com/news went offline and the archive went with it, with visitors redirected to the MTV home page. Paramount said in a statement that "as part of broader website changes across Paramount, we have introduced more streamlined versions of our sites, driving fans to Paramount+ to watch their favorite shows." The Internet Archive rebuilt what it could, publishing a searchable index of 460,575 pages from mtv.com/news going back to 1996. It is the same reason the handful of 90s websites still online (https://404memoryfound.com/posts/90s-websites-still-online.html) matter more every year, and the same reason people who taped MTV off cable now pay to digitize those VHS tapes (https://404memoryfound.com/posts/how-to-digitize-vhs-tapes-2026.html). The official copy is not always the one that survives. ## Frequently Asked Questions ### Is MTV still on the air in 2026? Yes. The flagship MTV channel still broadcasts in the United States and the United Kingdom in 2026, and MTV2 still runs in the US. What closed on December 31, 2025 were five UK music-only channels, MTV Music, MTV 80s, MTV 90s, Club MTV and MTV Live, along with music-only channels in Australia, Poland, France and Brazil. ### Who owns MTV now? MTV is owned by Paramount Skydance Corporation, which was created when Skydance Media and Paramount Global completed their merger on August 7, 2025. MTV sits in the same portfolio as CBS, Nickelodeon, Comedy Central and BET, and the company trades on Nasdaq under the ticker PSKY. ### When did MTV stop playing music videos? MTV never announced a stopping date, it tapered off. Videos filled up to eight hours a day in 2000 and about three hours a day by 2008, and Total Request Live, the last daily video countdown, ended with a finale on November 16, 2008. By the 2010s MTV's schedule was built on reality and clip shows instead. **Sources:** - Britannica: MTV, history, music videos and shows: https://www.britannica.com/money/MTV - Variety: MTV Didn't Shut Down on New Year's Eve, Despite Social Media Posts (2026): https://variety.com/2026/tv/news/mtv-did-not-shut-down-despite-social-media-posts-1236621840/ - Paramount: Skydance Media and Paramount Global Complete Merger, August 7, 2025: https://www.paramount.com/press/skydance-media-and-paramount-global-complete-merger-creating-next-generation-media-company - Paramount Press Express: 2025 MTV Video Music Awards ratings, September 2025: https://www.paramountpressexpress.com/cbs-entertainment/releases/?view=111760-the-2025-mtv-video-music-awards-delivers-its-largest-television-audience-since-2019-up-42-from-last-year - Variety: Internet Archive launches repository of nearly 480,000 MTV News articles (2024): https://variety.com/2024/digital/news/mtv-news-articles-internet-archive-wayback-machine-1236058997/ --- # Is Encarta Still Available? Why Microsoft Killed It URL: https://404memoryfound.com/posts/what-happened-to-microsoft-encarta.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-26 Updated: 2026-09-09 Topics: Software & Apps, Internet Culture, Business Blunders **Summary:** Microsoft Encarta is not available in 2026. Microsoft stopped selling the discs in June 2009 and closed the MSN Encarta websites on October 31, 2009, leaving Wikipedia and free web search in its place. The software itself survives: full Encarta 2009 disc images sit on the Internet Archive for anyone willing to run them in a virtual machine. **Key facts:** - Launched: March 22, 1993, Microsoft - Launch price: $395 in 1993, later cut to $99 - Status today: Discontinued. Disc sales ended June 2009, MSN Encarta closed October 31, 2009 - Last version: Encarta Premium 2009, released August 23, 2008 - What replaced it: Wikipedia and free web search ## Is Encarta still available in 2026? No. Microsoft Encarta has not been sold or hosted by Microsoft since 2009, and nothing has taken its place under that name. The end came in stages. The shutdown was reported on March 30, 2009, when Microsoft confirmed that the reference line was finished. Sales of Microsoft Student and every edition of Encarta Premium stopped worldwide in June 2009. The MSN Encarta websites closed on October 31, 2009 in every country except Japan, where the site ran until December 31, 2009. The Encarta dictionary content lingered online a little longer and was retired in 2011. There is no subscription, no download page and no support line. Microsoft did not sell the brand to another publisher the way distressed reference titles often change hands, so no rival version has appeared either. Encarta is one of the rare cases where a market leader was simply switched off, along with other pieces of the MSN family such as MSN Messenger (https://404memoryfound.com/posts/what-happened-to-msn-messenger.html). ## Can you still download or run Encarta today? Unofficially, yes. The Internet Archive hosts full disc images of the last release. One item, Microsoft Encarta Premium 2009, was uploaded on April 30, 2024 and offers a 1.9 GB ISO plus a torrent, free to download. Several other Encarta editions in several languages sit alongside it. Getting it running is the harder half. Encarta 2009 was built for the Windows PCs of 2008, so the dependable route is a virtual machine with a period version of Windows rather than a direct install on a current desktop. Nothing about it phones home, since the servers are long gone, which means the whole encyclopedia works offline exactly as it did on release day. It also means the content is frozen in 2008, so every population figure, head of state and space mission in it stops there. Where to find one today: physical copies still turn up on eBay (https://404memoryfound.com/posts/ebay-broken-laser-pointer-to-global-marketplace.html).com, mostly the 1995 through 2004 CD editions, with sealed big-box copies listed as collector items and loose discs far more common. Prices swing with condition and packaging, and no edition is scarce enough to be treated as an investment. ## Why did Microsoft shut Encarta down? Because free won. Wikipedia was written and corrected by volunteers, updated within hours of an event, and cost readers nothing. Encarta paid editors, licensed photographs and video, pressed discs, and shipped a new edition once a year. Broadband made the annual disc look absurd. Microsoft said as much in the notice it posted for customers, which explained that "the category of traditional encyclopedias and reference material has changed." The people who built it were not surprised. Tom Corddry, who ran Microsoft's multimedia group, later put it plainly: "We knew from the beginning that we were a temporary business, eventually to be overtaken by the internet." Bundling made the ending worse. Encarta had spent 15 years being given away with new PCs to win the market, which trained a generation to expect the encyclopedia as a free extra. When a free competitor arrived on the web, there was no paying customer base left to defend. ## What was it actually like to use? Encarta was an encyclopedia you clicked instead of carried. The standard edition held roughly 50,000 articles, and the premium editions passed 62,000 by 2008. The Spanish version ran about 42,000. Across its life it appeared in English, German, French, Spanish, Portuguese, Italian, Dutch and Japanese. The articles came from licensed print stock. Microsoft signed a non-exclusive deal for the Funk and Wagnalls encyclopedia in 1989, then folded in material from Collier's Encyclopedia and the New Merit Scholar's Encyclopedia bought from Macmillan. What people remember is rarely the text. It is MindMaze, the first-person trivia game hidden inside the encyclopedia, where you unlocked doors in a castle by answering questions drawn from the articles. It is also the atlas, the timelines and the audio clips, all of which felt like a different category of object than a book. Microsoft was shipping a lot of friendly consumer software in those years, including Clippy (https://404memoryfound.com/posts/what-happened-to-clippy-microsoft.html), and Encarta was the one that stuck. ## What did Encarta do to print encyclopedias? It ended them as a mass-market business. In 1980, encyclopedia publishers took around $400 million a year in US retail sales, mostly from salespeople who came to the door. A Britannica set ran about $1,400 in 1988. Microsoft had approached Britannica first, in 1985, and been turned down, because a CD-ROM did not fit a company selling leather-bound prestige. Funk and Wagnalls said yes instead. Encarta arrived on March 22, 1993 at $395, then dropped to $99 and started shipping free with new machines, including the wave of PCs that followed the Windows 95 launch (https://404memoryfound.com/posts/the-night-windows-95-launched.html). Britannica's answer in 1994 was a CD-ROM at $995, roughly ten times Encarta's standalone price. That year Britannica sold 51,000 print sets, down from more than 100,000 four years earlier. The rest of the category went the same way. Scholastic bought Grolier, the publisher behind one of the earliest digital encyclopedias, for $400 million in 2000, by which point the printed set was a legacy product. ## What replaced Encarta, and will Microsoft bring it back? Wikipedia replaced it for practical purposes, backed by free web search for everything else. That is the answer Microsoft itself pointed at in 2009, and it is why the shutdown was covered at the time with the joke that you could read all about it on Wikipedia. Microsoft has announced no plan to revive Encarta. There has been no Encarta product of any kind since the dictionary was retired in 2011, and the company's reference work now runs through search and its AI assistants rather than a branded encyclopedia. The last version, Encarta Premium 2009, shipped on August 23, 2008 and remains the final edition. What survives is the expectation Encarta created. It taught households that reference material should be searchable, illustrated, instantly available and cheap. Wikipedia met that expectation better and for nothing, which is the whole story in one sentence. ## Frequently Asked Questions ### Can Encarta run on Windows 11? Not reliably as a direct install. Encarta Premium 2009 was written for the Windows desktops of 2008, so the usual approach is a virtual machine running a period version of Windows, with the disc image from the Internet Archive mounted inside it. ### What was the last version of Encarta? Encarta Premium 2009, released on August 23, 2008. Microsoft stopped selling it, and the Microsoft Student bundle it came in, worldwide in June 2009. ### Did Wikipedia really kill Encarta? Wikipedia was the direct cause, though not the only one. Free volunteer editing beat a yearly paid edition on both speed and breadth, and Microsoft's own habit of bundling Encarta with new PCs meant there was little paying revenue left to protect by 2009. **Sources:** - Microsoft, The History of Microsoft 1993: https://learn.microsoft.com/en-us/shows/history/history-of-microsoft-1993 - TechCrunch, Microsoft To Shutter Encarta (March 30, 2009): https://techcrunch.com/2009/03/30/microsoft-to-shutter-encarta-read-all-about-it-on-wikipedia/ - Tedium, Who Killed the Encyclopedia: https://tedium.co/2017/07/13/who-killed-the-encyclopedia/ - Internet Archive, Microsoft Encarta Premium 2009: https://archive.org/details/microsoft-encarta-premium-2009 - Wikipedia, Encarta: https://en.wikipedia.org/wiki/Encarta --- # Is RealPlayer Still Around? Yes, Free From Real.com URL: https://404memoryfound.com/posts/what-happened-to-realplayer-streaming.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-26 Updated: 2026-09-15 Topics: Software & Apps, Internet Culture **Summary:** RealPlayer still exists. RealNetworks, private since December 2022 and owned by founder Rob Glaser, still publishes RealPlayer for Windows, Android and iOS, with a free tier and RealPlayer Plus at $39.99. The company's main businesses in 2026 are SAFR computer vision and GameHouse games, not streaming media. **Key facts:** - Launched: April 1995, as RealAudio 1.0 from Progressive Networks - Founder: Rob Glaser, a former Microsoft vice president - Status today: Still published; RealPlayer 25 for Windows shipped in 2026 - Owner today: RealNetworks LLC, private and controlled by Rob Glaser since December 2022 - Price today: Free tier, Downloader Express $24.99, RealPlayer Plus $39.99 ## How did a Seattle startup make audio stream in 1995? Rob Glaser spent ten years at Microsoft and left as a vice president in 1993. In February 1994 he founded Progressive Networks in Seattle, betting that the internet would end up carrying radio and television. The obstacle was bandwidth. Most people online in 1994 dialed in at 14.4 kbps, a speed at which a three minute song took close to half an hour to download. Glaser's team stopped trying to move the whole file. They compressed audio hard, sent it in a steady trickle, and buffered a few seconds ahead so playback could start while the rest was still arriving. In April 1995 Progressive Networks shipped RealAudio 1.0 along with the RealAudio Server. It sounded like a weak AM signal, and it worked. Click a link, wait a moment, hear sound. Radio stations, news organizations and sports leagues picked it up within a year, and in 1997 the company renamed itself RealNetworks and added video to the player. ## How big did RealPlayer get at its peak? RealNetworks went public on the Nasdaq in November 1997 at $12.50 a share under the ticker RNWK. The stock rose almost 50 percent on its first day and left the company valued near $600 million, which in 1997 made it one of the more convincing internet businesses on the market. By 2000, more than 115 million people had registered a copy of RealPlayer. A Media Metrix study reported by Variety in 2001 found that in November 2000, 28 percent of United States home computer users had used RealPlayer, against 22 percent for Windows Media Player and 4 percent for QuickTime. The money followed for a while. Net revenue reached $241.5 million in 2000, up 84 percent on 1999. That was the top. The company's own annual report for the next year put 2001 revenue at $188.9 million, and it kept sliding while the audience for streaming kept growing. ## Why did so many people hate the software? The reputation was earned. The free player worked as an advertisement for the paid one: pop-ups on launch, a download page that buried the free link, and a Message Center component that loaded at Windows startup and served promotions whether or not anyone had opened the player. It also took file associations from other players, planted itself in the system tray, and left pieces behind after an uninstall. On a dial-up line the playback itself was a negotiation, with the buffer counter restarting every time the connection wobbled. Anyone who remembers 56k modems (https://404memoryfound.com/posts/dialup-speed-then-vs-now.html) remembers that counter. Then came November 1999. Privacy researcher Richard Smith found that RealJukebox, RealNetworks' music companion with roughly 13 million registered users, was sending the company a unique identifier for each installation along with details of the CDs played and the music stored on the drive. Glaser called it a "mistake", RealNetworks patched the software, hired a privacy officer and set up an outside advisory committee. The suspicion stayed. ## What did RealNetworks win from Microsoft? Windows Media Player shipped with Windows. That was the whole argument. A casual user who wanted to play a video file already had a player, and it asked for nothing in return. In December 2003 RealNetworks sued Microsoft in the United States District Court in San Jose, seeking about $1 billion and arguing that bundling Windows Media Player with Windows was an illegal use of the operating system monopoly. The European Commission reached a similar conclusion about the bundling in its decision of March 2004. On October 11, 2005 the two companies settled for $761 million: $460 million in cash against the damages claim, plus $301 million in cash and services over 18 months, together with a deal to promote RealNetworks' Rhapsody service on MSN and its games through MSN Games and Xbox. "Today we're closing one chapter and opening a new one in our relationship with Microsoft," Glaser said in the joint announcement. The payment cleared. The users did not come back. ## Why did Harmony fail against the iPod? RealNetworks' other bet was on Apple's hardware. In 2004 it launched Harmony, which reverse engineered Apple's FairPlay copy protection so that songs bought from RealNetworks' store would play on an iPod (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html). Apple's response, issued in July 2004, did not hedge: "We are stunned that RealNetworks has adopted the tactics and ethics of a hacker to break into the iPod." Apple added that "when we update our iPod software from time to time it is highly likely that Real's Harmony technology will cease to work with current and future iPods." That is what happened. The subscription business had better thinking behind it. Rhapsody arrived years before Spotify, was spun out of RealNetworks in 2010, took the Napster (https://404memoryfound.com/posts/who-owns-napster-now.html) name in 2016, and was sold to MelodyVR for $70 million in 2020. None of it ran through the player. ## Who owns RealPlayer today? RealNetworks is private, and Glaser owns it. After the share price spent 2022 under the Nasdaq's $1 minimum, Glaser bid for the stock he did not already control and closed the purchase that December at $0.73 a share (https://www.sec.gov/Archives/edgar/data/1046327/000104632722000089/ex991toformcomletionofacqu.htm). The stock was delisted. What is left is not a streaming company. RealNetworks sells SAFR, a computer vision platform for live video, and runs the GameHouse mobile games business. Its KONTXT messaging analysis unit, which the company says has blocked more than eight billion spam and scam messages over eight years, went to TransUnion: announced on February 2, 2026 and completed on April 1, 2026. RealPlayer itself is still published. The Windows build reached version 25 during 2026, with Android and iOS apps alongside it, and real.com (https://www.real.com/realplayer-pricing) lists a free tier, Downloader Express at $24.99 and RealPlayer Plus at $39.99. It is a video downloader and media library now, roughly the shape Winamp (https://404memoryfound.com/posts/what-happened-to-winamp-mp3-player.html) settled into. Glaser's forecast about streaming was correct in every part except the part with his own company in it. ## Frequently Asked Questions ### Does RealPlayer still exist in 2026? Yes. RealNetworks still publishes RealPlayer, and the Windows version reached release 25 during 2026, with Android and iOS apps as well. The basic player is free, while RealPlayer Plus costs $39.99 as a one time purchase. ### Why was RealPlayer so hated? RealPlayer, first released as RealAudio in April 1995, wore users down with launch pop-ups pushing the paid version, a Message Center that ran at Windows startup, seized file associations and a messy uninstall. The RealJukebox case of November 1999, in which the software sent listening data and a unique installation identifier back to RealNetworks, turned irritation into distrust. ### How much did Microsoft pay RealNetworks? Microsoft paid RealNetworks $761 million in October 2005 to settle the antitrust suit RealNetworks had filed in December 2003 over the bundling of Windows Media Player with Windows. The package was $460 million in cash plus $301 million in cash and services spread over 18 months. **Sources:** - RealNetworks, Inc. company history, International Directory of Company Histories: https://www.encyclopedia.com/economics/encyclopedias-almanacs-transcripts-and-maps/realnetworks-inc - RealNetworks caught secretly swiping users' jukebox data, The Register, November 1999: https://www.theregister.com/1999/11/01/realnetworks_caught_secretly_swiping_users/ - Apple vs. RealNetworks, San Francisco Chronicle, July 2004: https://www.sfgate.com/business/article/Apple-vs-RealNetworks-iPod-maker-calls-rival-a-2704847.php - Microsoft and RealNetworks Resolve Antitrust Case, Microsoft, October 2005: https://news.microsoft.com/source/2005/10/11/microsoft-and-realnetworks-resolve-antitrust-case-and-announce-digital-music-and-games-partnership/ - TransUnion Completes Acquisition of the Mobile Division of RealNetworks, April 2026: https://newsroom.transunion.com/transunion-completes-acquisition-of-the-mobile-division-of-realnetworks/ --- # How MySpace Lost to Facebook and Sold for $35M URL: https://404memoryfound.com/posts/rise-and-fall-of-myspace.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-25 Updated: 2026-09-14 Topics: Internet Culture, Business Blunders **Summary:** MySpace lost to Facebook between 2006 and 2009, after News Corporation bought it for $580 million in 2005 and tied its revenue to a $900 million Google advertising agreement that rewarded page views instead of happier users. Facebook overtook MySpace in the United States in 2009, and News Corporation sold the site for $35 million in June 2011. MySpace still exists in 2026 as a music and nostalgia site owned by the Vanderhook brothers, who say a relaunch is coming but have set no date. **Key facts:** - Launched: 2003, by Tom Anderson and Chris DeWolfe at eUniverse - Peak audience: About 70 million unique monthly visitors - Sold for: $35 million to Specific Media, June 2011 - Owner today: Viant Technology, run by Tim and Chris Vanderhook - Status today: Online as a music site, relaunch promised with no date ## What made MySpace the biggest site in America? MySpace was created in 2003 by Tom Anderson and Chris DeWolfe, two employees of the internet marketing company eUniverse, which was later renamed Intermix Media. It was not the first social network. It was the first one that handed the page layout to the user. Every profile was an editable HTML window. People pasted code into a text box and got tiled backgrounds, autoplaying songs, custom cursors and a Top 8 friends list that ran real social politics in American high schools. The pages were often unreadable. They were also unmistakably personal, which is the part no mainstream platform has matched since. Music was the second engine. Bands set up profiles, uploaded songs and reached listeners without a label in between, and a generation of American teenagers found new acts on profile pages rather than on the radio. By the summer of 2006 the combination made MySpace the most visited site in the country. Hitwise put 4.5 percent of all US internet visits on MySpace in the week ending July 8, 2006, ahead of Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) Mail, Google and MSN, although that ranking counted visits and page views rather than unique users. At its peak the site drew about 70 million unique monthly visitors. ## How did the News Corporation deal change the site? Rupert Murdoch's News Corporation bought MySpace in 2005 for $580 million. The reasoning was straightforward: the largest audience on the web, bought by a company that knew how to sell advertising against an audience. A year later, News Corporation put that audience under contract. In August 2006 its internet arm, Fox Interactive Media, made Google the exclusive search and keyword advertising provider across its sites, MySpace included. The announcement filed with the SEC spells out the terms: Google owed "guaranteed minimum revenue share payments to Fox Interactive Media of $900 million based on Fox achieving certain traffic and other commitments," paid between the first quarter of 2007 and the second quarter of 2010. Read that clause as a MySpace product manager in 2007 and the job description writes itself. The money was not attached to whether anyone enjoyed the site. It was attached to traffic and to the volume of pages served. Ads multiplied, pages that were already slow under custom HTML got slower, and any change that reduced the number of clicks in a session worked against the number the company had promised Wall Street. The site kept growing for a while anyway, which is the trap. Traffic and page views held up long enough to make the strategy look sound, and by the time the audience numbers turned, the product decisions that would have kept people there had been deferred for three years. ## Why did people switch to Facebook? Facebook offered the opposite bargain. Every profile looked the same: no autoplay, no glitter, no layout a teenager could break with one unclosed tag. Pages loaded quickly and predictably, which mattered more each month as MySpace profiles grew heavier with custom code and advertising. The News Feed did the rest. Instead of walking profile to profile to find out what friends were doing, a Facebook user got the updates collected in one place. MySpace never shipped an equivalent that users kept. Facebook also grew in the opposite direction. It started closed, on college campuses, where a university email address was the entry ticket, then opened outward from a base of real names. MySpace was going the other way, fighting spam accounts, fake profiles and a reputation problem with American parents that never attached itself to Facebook. Network effects reward the leader and punish it just as fast when the direction flips. Once a user's friends had moved, staying on MySpace meant talking to an empty room. Facebook passed MySpace among American users in 2009, and MySpace's traffic never recovered. The same reversal had already taken Friendster (https://404memoryfound.com/posts/what-happened-to-friendster-social-network.html) apart a few years earlier, with MySpace as the beneficiary. ## What did the collapse cost News Corporation? On June 29, 2011, News Corporation sold MySpace to the advertising targeting firm Specific Media for $35 million in cash and stock. That was well under the $100 million News Corporation had been asking for the site, and a fraction of the $580 million it paid six years earlier. Justin Timberlake joined the buying group as an investor and a public face for the relaunch. The sale price is the number everyone remembers, but it understates the damage. News Corporation also spent six years of management attention, a run of executive turnover, and the traffic commitments it had written into the Google agreement, on a property that ended up worth less than a mid-size office building. The music paid a price too. A later server migration destroyed the audio people had uploaded in MySpace's early years, taking with it recordings by artists who never released their work anywhere else. An independent music archive disappeared while the industry was still working out what the file-sharing era had done to it, a story covered in our look at what Napster did to the music business (https://404memoryfound.com/posts/napster-destroyed-music.html). ## Is MySpace still around in 2026? Yes, in a narrow sense. The address still resolves, the front page still publishes music news, and old profiles still render. What is gone is the social network: there is no active community, and the site functions as a music and nostalgia property rather than a place people post to. Our companion piece on what is left of the site (https://404memoryfound.com/posts/is-myspace-still-around-today.html) walks through the current state in detail. Ownership traces straight back to the 2011 sale. Specific Media grew into Viant Technology, the advertising technology company run by brothers Tim and Chris Vanderhook, and Viant has held MySpace ever since. The owners say they are not finished with it. In a 2026 documentary about the site, Tim Vanderhook said: "We are stewards of the Myspace brand at this point, and we are going to relaunch Myspace." No launch date has been announced, no product has been shown, and Tom Anderson, the first friend on every new account in the 2000s, is not part of the plan. ## What the collapse taught the platforms that came after MySpace's real legacy is not a feature. It is the proof that a dominant social network can lose its audience in about three years, from the peak in 2006 to the American crossover in 2009, and that the loss is close to irreversible once it starts. The lesson companies took from it was mostly defensive: standardize the layout, keep the feed fast, and buy or copy anything growing. The lesson users took was different. Interest in handmade personal pages never fully went away, and projects like Neocities and SpaceHey, a deliberate MySpace clone, exist because people still want a page that looks like a person made it rather than a template with a photo slot. The other lesson holds for everything that followed MySpace onto the web. A site that starts optimizing for the number it promised an advertiser, rather than for the reason people showed up, gets to keep that number for about as long as it takes users to find the next place. ## Frequently Asked Questions ### Why did MySpace lose to Facebook? MySpace lost to Facebook because it optimized for advertising volume instead of the experience. After News Corporation bought MySpace for $580 million in 2005 and signed a $900 million search and advertising agreement with Google in 2006 that depended on traffic commitments, the site filled with ads and slowed down, while Facebook offered fast, standardized profiles and the News Feed. Facebook passed MySpace among American users in 2009. ### When did Facebook overtake MySpace? Facebook overtook MySpace in the United States in 2009, two years before News Corporation sold MySpace in June 2011. MySpace had been the most visited website in the country in the summer of 2006, with about 70 million unique monthly visitors at its peak, so the reversal took roughly three years from peak to defeat. ### Who created MySpace? MySpace was created in 2003 by Tom Anderson and Chris DeWolfe, who were working at the internet marketing company eUniverse, later renamed Intermix Media. Anderson became the automatic first friend on every new MySpace account, which is why a generation of American users knew him simply as Tom. Neither founder is involved with MySpace under its current owner, Viant Technology, which bought the site as Specific Media in 2011. **Sources:** - TechCrunch: MySpace hit #1 US destination last week, Hitwise (July 11, 2006): https://techcrunch.com/2006/07/11/myspace-hit-1-us-destination-last-week-hitwise/ - News Corporation Form 8-K: Fox Interactive Media Enters Into Landmark Agreement With Google Inc. (SEC, August 7, 2006): https://www.sec.gov/Archives/edgar/data/1308161/000118143106046139/rrd126691_15374.htm - Encyclopaedia Britannica: Tom Anderson, American entrepreneur: https://www.britannica.com/biography/Tom-Anderson - NPR: News Corp. Takes Huge Loss, Selling Myspace For $35 Million (June 29, 2011): https://www.npr.org/sections/thetwo-way/2011/06/29/137509647/news-corp-takes-huge-loss-selling-myspace-for-35-million - NBC New York: Could Myspace come back? Owners share plans for social media site (2026): https://www.nbcnewyork.com/news/national-international/myspace-owners-plan-relaunch-social-media-platform/6533194/ --- # Is Google Glass Still Around? Discontinued in 2023 URL: https://404memoryfound.com/posts/why-google-glass-actually-failed.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-25 Updated: 2026-09-14 Topics: Business Blunders, Hardware **Summary:** Google Glass is no longer sold or supported. Google stopped selling the last version, Glass Enterprise Edition, on March 15, 2023, and ended support on September 15, 2023, a decade after the $1,500 Explorer Edition reached its first buyers. Google's replacement is Android XR eyewear built with Warby Parker and Gentle Monster, shown in May 2026. **Key facts:** - Launched: Explorer Edition, February 2013, $1,500 - Maker: Google (Alphabet) - Status today: Discontinued. Sales ended March 15, 2023 - Support ended: September 15, 2023 - What replaced it: Android XR eyewear with Gemini, due fall 2026 ## What did the $1,500 Explorer Edition actually buy? Google Glass reached developers in February 2013 as the Explorer Edition, at $1,500. Google's own tech specs page lists what was inside: a 640 by 360 prism display that floated above the right eye, a 5 megapixel camera that shot 720p video, 2 GB of RAM, 16 GB of flash storage with 12 GB usable, and a bone conduction speaker in place of an earpiece. A dual-core Texas Instruments OMAP 4430 ran all of it. Getting one took more than money. In February 2013 Google ran a contest under the hashtag #ifihadglass, asking applicants to say in 50 words or less what they would do with the device. Roughly 8,000 people were picked, and each still had to pay the $1,500 and collect the unit in person at pop-up events in New York, Los Angeles or San Francisco. Glass went on open sale to anyone in the United States on April 15, 2014, at the same price. A teardown by the research firm IHS that year put the bill of materials at $152.47, a gap that got quoted back at Google for years. The Cooper Hewitt, Smithsonian Design Museum now holds an Explorer Edition in its permanent collection. ## Why did one small camera cause so much trouble? The camera sat above the display, pointing forward, and that was the whole problem. A phone held up to record is visible. A camera on a face is not. Sergey Brin had already made the case for moving hardware up there. Talking about smartphones at TED in February 2013, Brin said "You're standing there just rubbing this featureless piece of glass," as CNN reported that month. The public agreed with the diagnosis and rejected the cure. In March 2013, before Glass had shipped to anyone outside the contest, the 5 Point Cafe in Seattle announced a ban. Owner Dave Meinert posted that "The 5 Point is the first Seattle business to ban in advance Google Glasses," and later told reporters that part of it was a joke and part of it was serious, because the bar does not let customers film each other. Casinos, cinemas, gyms and other bars followed. The nickname arrived fast, and Google put it in print first. Its own Glass etiquette guide, published in February 2014, told Explorers: "Don't be creepy or rude (aka, a 'Glasshole')." A company that has to write that sentence about its own product has already lost the argument. ## Was the hardware good enough to survive the backlash? No. The prism was readable indoors and washed out in daylight. Battery life collapsed under video, so Glass recorded in bursts rather than sessions. The frame warmed against the temple. Voice commands opened with the phrase "OK Glass," which meant talking to your own eyebrow in public. The tether made the pitch awkward too. Glass had Wi-Fi and Bluetooth but no cellular radio, so it leaned on a phone in your pocket. A device sold as the end of pulling out your phone still required you to carry your phone. Software never caught up. Google opened a Glassware catalogue, but the screen was one small card, the battery could not run anything demanding, and the paying audience was 8,000 people. Developers who did that math walked away. Set it beside Nintendo's Virtual Boy (https://404memoryfound.com/posts/what-happened-to-virtual-boy-nintendo.html), another head-worn device shipped before its display was ready, and the shape repeats: the demo works, the second hour does not. ## Is Google Glass still supported in 2026? No. Nothing in the Glass line is sold or updated today, and Google has not announced a return of the name. The consumer version went first. On January 15, 2015, the Glass team said it was leaving Google's X lab, writing that "we've outgrown the lab" as the project moved to its own team inside Google. The Explorer program closed that month and the $1,500 device stopped being sold. Glass then became a work tool. Glass Enterprise Edition arrived in 2017, sold through Google's partners at roughly $1,828, aimed at warehouses, factory floors and hospitals. Glass Enterprise Edition 2 followed on May 20, 2019, at $999, with a Qualcomm Snapdragon XR1 chip, an 8 megapixel camera and USB-C charging. That line ended as well. Google stopped selling Glass Enterprise Edition on March 15, 2023, and its support page names September 15, 2023 as the end of support. Existing headsets kept working, but no further updates came from Google. Alphabet still owns the brand and has not used it since. ## What replaced Google Glass? Two things, and neither of them is called Glass. The first is ordinary-looking camera eyewear. Meta sells smart glasses with Ray-Ban, built on the same Luxottica relationship Google announced in January 2014 and never turned into a shipping product. The advance was design rather than silicon: glasses that read as glasses draw none of the attention that a titanium arm and a visible prism did in 2013. The second is Google trying again under a different name. At Google I/O in May 2026, Google and Samsung showed Android XR eyewear built with Warby Parker and Gentle Monster, running Gemini for live translation, navigation and notification summaries. Google's announcement splits the line in two: audio glasses that talk in your ear, arriving in the fall of 2026, and display glasses after them. Making the display optional is the exact reverse of the 2013 bet. Glass still belongs to the small club of products that arrived with more press than purpose, next to the Segway (https://404memoryfound.com/posts/what-happened-to-segway-personal-transporter.html) and Google+ (https://404memoryfound.com/posts/why-google-plus-actually-failed.html). The difference is that this one changed how everybody else designs the category. ## Where to find a Google Glass unit today Google sells nothing in the line, so the second-hand market is the only route, and what you are buying is a display piece rather than a working computer. Support ended in September 2023 and no updates have shipped since. Explorer Edition units and their frames turn up alongside other 2010s hardware on Etsy (https://www.etsy.com/search?q=google+glass), usually untested and listed as collectables. Ask the seller whether the unit powers on and whether the charger is included, because Glass used a proprietary connector and replacements are scarce. ## Frequently Asked Questions ### Is Google Glass still available in 2026? No. Google stopped selling Glass Enterprise Edition on March 15, 2023 and ended support on September 15, 2023, and there is no consumer Google Glass on sale in 2026. The consumer Explorer Edition had already been withdrawn in January 2015. Only second-hand units exist, and they get no software updates. ### How much did Google Glass cost? The Google Glass Explorer Edition cost $1,500 when it reached developers in February 2013 and still cost $1,500 when it went on open sale in the United States on April 15, 2014. Glass Enterprise Edition sold through Google's partners at roughly $1,828 in 2017, and Glass Enterprise Edition 2 launched on May 20, 2019 at $999. ### Why was Google Glass discontinued? Google Glass was discontinued because the always-on camera made it socially unwelcome, with bans starting at the 5 Point Cafe in Seattle in March 2013, and because the hardware could not justify $1,500: a washed-out prism display, short battery life under video, and a thin app catalogue. Google withdrew the consumer version in January 2015, pivoted to enterprise, and ended that line in March 2023. **Sources:** - Google Glass tech specs (Google Glass Help): https://support.google.com/glass/answer/3064128 - Glass Enterprise Edition Announcement FAQ (Google Support, 2023): https://support.google.com/glass-enterprise/customer/answer/13417888 - Google Glass Enterprise Edition 2 announced for $999 (CNBC, 2019): https://www.cnbc.com/2019/05/20/google-glass-enterprise-edition-2-announced-price.html - Sergey Brin: Smartphones are 'emasculating' (CNN, 2013): https://www.cnn.com/2013/02/28/tech/innovation/brin-google-glass/index.html - Intelligent eyewear with Gemini is coming this fall (Google blog, 2026): https://blog.google/products-and-platforms/platforms/android/android-xr-io-2026/ --- # Is the Microsoft Zune Still Around? Why It Failed URL: https://404memoryfound.com/posts/why-microsoft-zune-actually-failed.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-24 Updated: 2026-09-13 Topics: Business Blunders, Hardware **Summary:** The Microsoft Zune is gone. Microsoft stopped making Zune hardware in 2011 and shut down the Zune online services on November 15, 2015, moving its last music subscribers to Spotify. The players still work as offline music players in 2026, and used ones average about $128 on resale listings, but nothing about the Zune is supported by Microsoft. **Key facts:** - Launched: November 14, 2006, $249.99 for 30 GB - Status today: Discontinued; hardware ended in 2011, services ended November 15, 2015 - Owner today: Microsoft, which folded the brand into Xbox Music in 2012 - Peak market share: About 2 percent of the US player market in 2009 (NPD Group) - Price today: About $128 on average across 2026 resale listings ## What was the Microsoft Zune, and when did it launch? Microsoft put the first Zune in United States stores on November 14, 2006. It cost $249.99, held 30 GB, and arrived one dollar under the price of the 30 GB iPod it was built to beat. The hardware was competitive. The Zune 30 had a three inch screen, an FM tuner and Wi-Fi, none of which the iPod offered. The Wi-Fi powered a feature Microsoft pitched as wireless sharing and the rest of the internet called squirting: send a track to another Zune owner nearby, and they could play it three times within three days before it locked. Behind the player sat Zune Marketplace and a subscription called Zune Pass, $14.99 a month for unlimited access to the catalog. In 2006, renting music instead of buying it was a strange idea. It is now how most people listen. Distribution was the first sign of trouble. The Zune shipped in the United States only. Canada followed on June 13, 2008. The United Kingdom, Australia and the rest of Europe never got the hardware at all, while Apple sold iPods in nearly every country on earth. ## Why did the Zune lose to the iPod? Pricing was the first error. A challenger facing an entrenched leader has to be much cheaper or much better. At $249.99 the Zune was neither, and switching meant walking away from a music library already bought from iTunes and locked to Apple's copy protection. The second problem was everything built around the iPod: the speaker docks, the car adapters, the headphones, the software people already had installed. The iPod was not the first MP3 player (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html), but by 2006 it had five years of accessories and habit behind it, and the Zune had none of that. Apple was not worried. Asked about Zune sharing by Newsweek in October 2006, Steve Jobs walked through the three plays in three days limit and concluded, "By the time you've gone through all that, the girl's got up and left!" The line landed because it was accurate. Sharing only worked if you could find a second Zune owner in the room. Then there was timing. On January 9, 2007, eight weeks after the Zune reached stores, Jobs introduced the iPhone. Every standalone music player on the market, Apple's own included, was suddenly living on borrowed time, and Microsoft had just committed to a five year product plan for one. ## How badly did the Zune actually sell? Microsoft never published a lifetime sales number for the Zune, which is usually an answer in itself. The figures that did surface were brutal. By May 2008, roughly 18 months after launch, the Zune had sold about 2 million units in total. That is a number Apple was clearing in weeks, and it covered two hardware generations rather than one. Market share tells the same story from another angle. NPD Group put the Zune at about 2 percent of the United States portable player market in 2009, against roughly 70 percent for the iPod, with SanDisk and Sony splitting most of the remainder. The strange part is that the best Zune came last. The Zune HD, released in September 2009, had an OLED touchscreen, a web browser and an interface reviewers preferred to the software Apple was shipping. It arrived into a market the iPhone and the iPod touch had already absorbed. Good hardware landing years late is a familiar Microsoft pattern, visible again in the original Xbox (https://404memoryfound.com/posts/what-happened-to-original-xbox-microsoft-gaming.html), which burned billions before the Xbox 360 made the category pay. ## Who owns the Zune brand now, and does the hardware still work? Microsoft owns it, and nothing is being made. On March 15, 2011, Microsoft said it would develop no new Zune hardware. On October 3, 2011, it discontinued the players outright and pointed customers toward Windows Phone. In 2012 the media brand was folded into Xbox Music, which later became Groove Music, and Microsoft closed the Groove subscription on December 31, 2017 and handed its remaining subscribers to Spotify. The hard cutoff for Zune owners came on November 15, 2015, when Microsoft switched off the Zune services. From that day there was no store, no streaming and no subscription attached to the device. A Zune still works as an offline player. Load MP3s over USB with the old Zune desktop software and it plays, the FM tuner still tunes, and the battery is the part that reliably fails first. What does not work is anything that needed to reach Microsoft: Marketplace, Zune Pass and wireless sync to the store. The Zune later turned up as a plot device in Guardians of the Galaxy Vol. 3, which is closer to a revival than anything Microsoft has announced. Where to find one today. Zunes trade steadily second hand, and condition drives everything. Listings tracked across resale sites in 2026 run from a few dollars for dead units to close to $500 for sealed originals, with an average around $128, and a refurbished Zune HD with a new battery sold for $199 in 2026. Buy one with a replaced battery or budget for the swap, and check the vintage Zune listings on Etsy (https://www.etsy.com/search?q=microsoft+zune) for players sold with their original cables and packaging. ## What did the Zune get right before anyone else? Robbie Bach, the Microsoft executive who ran the Entertainment and Devices division, gave the honest post mortem in 2012. "We just weren't brave enough, honestly, and we ended up chasing Apple," he said in remarks reported by Engadget, describing the Zune as "a chasing product" that gave nobody a reason to go out and buy it. Chasing or not, the Zune was early on several things. Zune Pass sold unlimited listening for a flat monthly fee in 2006, years before that became the normal way to pay for music. Wireless track sharing anticipated the tap-to-send features every phone has now. The social layer, mocked at the time under the slogan "Welcome to the Social," is standard in every streaming app in 2026. The most durable piece was the design. The typography first, chrome light interface Microsoft built for the Zune software and the Zune HD became the Metro design language that Windows Phone launched with in 2010, and its fingerprints are still on Windows and Xbox menus. None of it saved the product, because Microsoft was selling a device and Apple was selling a habit. Sony's MiniDisc (https://404memoryfound.com/posts/what-happened-to-minidisc-sony.html) hit a version of the same wall a decade earlier: careful engineering attached to a format the rest of the market had no reason to adopt. ## Frequently Asked Questions ### Is the Microsoft Zune still around in 2026? No. Microsoft stopped developing Zune hardware in March 2011, discontinued the players in October 2011, and shut down the Zune online services on November 15, 2015. Zune players people already own still work as offline music players in 2026, but there is no store, no subscription and no support from Microsoft. ### Why did the Microsoft Zune fail? The Microsoft Zune failed because it launched in November 2006 at $249.99, matching the iPod on price without beating the ecosystem around it, and because Apple introduced the iPhone eight weeks later and began erasing the standalone music player market. Robbie Bach, who ran the Microsoft division responsible, called it a chasing product in 2012. The Zune held about 2 percent of the United States market in 2009. ### How much is a Microsoft Zune worth today? Microsoft Zune players are cheap collectibles rather than valuable ones. Listings tracked across resale sites in 2026 average about $128, ranging from a few dollars for dead units to close to $500 for sealed originals, and a refurbished Zune HD with a new battery sold for $199 in 2026. A working battery and original accessories account for most of the difference. **Sources:** - Microsoft to Put Zune Experience in Consumers' Hands on Nov. 14 (Microsoft News Center, 2006): https://news.microsoft.com/source/2006/09/28/microsoft-to-put-zune-experience-in-consumers-hands-on-nov-14/ - Steve Jobs, Newsweek interview transcript, October 2006: https://allaboutstevejobs.com/verbatim/interviews/newsweek_2006 - Robbie Bach: I wouldn't have made Zune MP3 players, we were just 'chasing Apple' (Engadget, 2012): https://www.engadget.com/2012/05/14/robbie-bach-would-not-have-made-zune/ - Zune (Wikipedia): sales, market share, discontinuation and service shutdown dates: https://en.wikipedia.org/wiki/Zune - Microsoft Zune used and refurbished listing prices, 2026: https://www.used.forsale/microsoft-zune --- # Is Adobe Flash Still Around? Flash Player in 2026 URL: https://404memoryfound.com/posts/definitive-history-macromedia-flash.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-24 Updated: 2026-09-14 Topics: Software & Apps, Internet Culture **Summary:** Adobe Flash Player reached end of life on December 31, 2020, and from January 12, 2021 Adobe blocked Flash content from running. Adobe still owns Flash but no longer ships it: a licensed enterprise build comes from Harman International and a mainland China build from Chongqing Zhongcheng Network Technology. The authoring tool survives as Adobe Animate, which Adobe moved into maintenance mode in February 2026. **Key facts:** - Launched: August 19, 1996 as FutureSplash Animator, renamed Flash 1.0 in 1997 - Owner today: Adobe, which bought Macromedia for about $3.4 billion in 2005 - Status today: Discontinued. Flash Player ended on December 31, 2020 and was blocked from January 12, 2021 - Still shipping: Harman International for enterprise, and a 34.x mainland China build - Content preserved: More than 200,000 games and animations at Flashpoint Archive ## Where did Flash come from before Macromedia bought it? Flash did not start at Macromedia, and it did not start on the web. It started at FutureWave Software, a small San Diego company founded in 1993 by Jonathan Gay, Charlie Jackson and Michelle Welsh. Their first product was SmartSketch, a vector drawing program written for pen computers running the PenPoint operating system. Pen computing collapsed before SmartSketch found an audience. GO Corporation's PenPoint tablets went nowhere, AT&T acquired the EO Personal Communicator line and discontinued it, and FutureWave was left with a good drawing program and no market to sell it into. Gay rebuilt the program around the thing that was growing instead. He added frame-by-frame animation, ported it to Windows and Mac, and shipped it on August 19, 1996 as FutureSplash Animator. Two customers made it matter: Microsoft used it for MSN, and Disney Online used it for its subscription children's service. In January 1997 Macromedia bought FutureWave for an undisclosed sum and renamed the product Flash 1.0. ## How did Flash end up on 98% of the web? The decision that mattered was the distribution model, not the software. The Flash Player plugin was free for anyone to download. The authoring tool was not. Give away the player, sell the tool: by 2005 the Flash Player was installed on roughly 98% of internet-connected desktop computers, a penetration figure Adobe tracked through studies run by the research firm Millward Brown. The tool got more capable every year. Flash 4 in 1999 added a scripting language and MP3 streaming. Flash 5 in 2000 introduced ActionScript. Flash MX in 2002 added video, which turned out to be the most consequential feature Macromedia ever shipped. Before Flash video, watching anything online meant QuickTime, RealPlayer (https://404memoryfound.com/posts/what-happened-to-realplayer-streaming.html) or Windows Media Player, each with its own codecs and its own failures. Flash video replaced all three with one player that was already on almost every machine. When Chad Hurley, Steve Chen and Jawed Karim launched YouTube in February 2005, they built it on Flash video. Vimeo, Dailymotion and Hulu followed. Neopets (https://404memoryfound.com/posts/what-happened-to-neopets-virtual-world.html), Club Penguin (https://404memoryfound.com/posts/what-happened-to-club-penguin.html) and thousands of browser games ran on the same plugin. ## Why did Adobe pay $3.4 billion for Macromedia? On April 18, 2005, Adobe Systems announced it would buy Macromedia in an all-stock deal valued at about $3.4 billion. Macromedia stockholders received 0.69 shares of Adobe stock for each Macromedia share, which worked out to $41.86 per share against the previous Friday's close. The deal completed on December 3, 2005. The logic was tidy. Adobe owned print and design through Photoshop, Illustrator and InDesign. Macromedia owned the web through Flash, Dreamweaver and Fireworks. Together they would hold the whole pipeline. Adobe rebranded the product Adobe Flash and pushed it past the browser. Adobe AIR let Flash applications run as desktop software. Flex targeted enterprise business applications. The plan was for Flash to be the runtime for everything: websites, video, games, phones and desktops. It nearly worked. The mid-2000s web filled up with all-Flash corporate sites, the kind that opened on a loading bar and a Skip Intro button. Newgrounds (https://404memoryfound.com/posts/what-happened-to-newgrounds.html) became the center of Flash animation and browser gaming, and Homestar Runner (https://404memoryfound.com/posts/homestar-runner-still-updating.html), the web cartoon made by Mike and Matt Chapman, ran entirely on the plugin. ## Why did Steve Jobs refuse to put Flash on the iPhone? The iPhone arrived on January 9, 2007 without Flash, and Apple never added it. On April 29, 2010, Steve Jobs published an open letter titled "Thoughts on Flash" that explained why, and it did more damage to Adobe than any competitor. Jobs listed six objections: Flash was proprietary rather than open, the full-web argument was misleading because video was moving to H.264, Flash was the top cause of Mac crashes, it drained batteries, it was designed for mouse rollovers rather than touch, and open standards could do the same work. "Flash was created during the PC era," he wrote, and "New open standards created in the mobile era, such as HTML5, will win on mobile devices (and PCs too)." Critics noted Apple had a commercial motive too, since Flash would have been an application platform outside the App Store. Both things were true. Adobe spent 18 months trying to prove the technical case wrong, then stopped. On November 9, 2011 it ended development of Flash Player for mobile browsers and cut 750 jobs in a corporate restructuring. ## Who owns Flash today, and where does it still run? Adobe still owns Flash. It just does not ship it. Adobe announced the end on July 25, 2017, writing that "we will stop updating and distributing the Flash Player at the end of 2020" in a statement issued with Apple, Facebook, Google, Microsoft and Mozilla. Flash Player reached end of life on December 31, 2020, and from January 12, 2021 Adobe blocked Flash content from running in the player at all. Two licensed exceptions survive. Harman International distributes a supported Flash Player to enterprise customers outside China under an agreement with Adobe, numbered in the 50.x range. Chongqing Zhongcheng Network Technology distributes a mainland China build, which continued its own version line and was still shipping 34.x releases in 2026. That build refuses to start outside China and refuses to start inside a corporate Active Directory environment. The authoring tool outlived the plugin. Flash Professional was renamed Adobe Animate in February 2016 and is still sold. Adobe almost ended that too. In early 2026 it told customers Animate would stop being available for purchase on March 1, 2026, writing that "we are planning to discontinue supporting Animate." Animators pushed back, and in February 2026 Adobe reversed the decision and moved Animate into maintenance mode: security and bug fixes continue, new features do not. ## Can you still play old Flash games and animations? Yes, and more of them than at any point since 2020. The original plugin is not the route. Ruffle, an open-source Flash Player emulator written in Rust that runs in the browser through WebAssembly, reimplements playback without any Adobe code. Newgrounds built Ruffle into its own site, so its library of user-made animations and games loads with no plugin at all. Flashpoint Archive is the bigger effort. The volunteer project started in December 2017 and had preserved more than 200,000 web games and animations by 2026, covering Flash, Shockwave and roughly a hundred other browser technologies. It ships as a downloadable launcher that plays the files offline. The Internet Archive runs its own Flash software library with Ruffle embedded in the page. Preservation is not complete. Titles that called a server for scores, advertising or multiplayer often cannot be fully restored, and some were never captured before the sites hosting them shut down. ## What replaced Flash, and what got lost Video was the clean swap. HTML5 video with the H.264 and VP9 codecs does what Flash video did, with less battery drain and no plugin. YouTube made HTML5 its default player in January 2015 across Chrome, Internet Explorer 11, Safari 8 and newer Firefox builds, and most people never noticed the change. Applications moved as well. The rich internet applications built in Flash and Flex became JavaScript single-page apps written in React, Angular and Vue, with better accessibility and tooling. Browser games moved to HTML5 Canvas, WebGL and engines such as Phaser and Unity. What got lost was who could make things. The Flash timeline let a designer animate without writing code and publish a file that played anywhere. Modern web animation usually means JavaScript. Security sat on the other side of that ledger: Flash became one of the most exploited pieces of software on the internet, and in July 2015 Facebook chief security officer Alex Stamos wrote that "It is time for Adobe to announce the end-of-life date for Flash." Two years later, Adobe set one. ## Frequently Asked Questions ### Is Adobe Flash Player still available to download in 2026? Not from Adobe. Adobe stopped distributing Flash Player when it reached end of life on December 31, 2020, and removed the download links. Two licensed builds still exist in 2026: an enterprise version from Harman International for companies outside China, and a mainland China version from Chongqing Zhongcheng Network Technology that will not run elsewhere. For everyone else, the way to open a Flash file is the Ruffle emulator or Flashpoint Archive. ### Why did Adobe kill Flash? Adobe ended Adobe Flash because open standards caught up and the plugin became a liability. HTML5 video, CSS animation, WebGL and JavaScript covered what Flash did without a plugin, while Flash was one of the most exploited programs on the internet. Apple never allowed it on the iPhone after 2007, Adobe stopped building it for mobile browsers in November 2011, and on July 25, 2017 Adobe set the end-of-life date of December 31, 2020 alongside Apple, Facebook, Google, Microsoft and Mozilla. ### How much did Adobe pay for Macromedia? Adobe agreed on April 18, 2005 to buy Macromedia in an all-stock transaction valued at about $3.4 billion, and the deal completed on December 3, 2005. Macromedia stockholders received 0.69 Adobe shares per Macromedia share, worth $41.86 per share at the prior close. Beyond Flash, the purchase brought Adobe the Dreamweaver, Fireworks and ColdFusion product lines. **Sources:** - Adobe: Flash & the Future of Interactive Content (July 25, 2017): https://blog.adobe.com/en/publish/2017/07/25/adobe-flash-update - Macromedia Inc., Form 425 filing on the Adobe merger, April 2005 (U.S. Securities and Exchange Commission): https://www.sec.gov/Archives/edgar/data/0000913949/000110465905031942/a05-6868_6425.htm - Web Design Museum: Steve Jobs and his Thoughts on Flash (2010): https://www.webdesignmuseum.org/web-design-history/steve-jobs-and-his-thoughts-on-flash-2010 - Adobe: Adobe Animate maintenance mode FAQs: https://helpx.adobe.com/animate/kb/maintenance-mode.html - Flashpoint Archive: web game and animation preservation project: https://flashpointarchive.org/ --- # Is Winamp Still Around in 2026? Yes, and It Works URL: https://404memoryfound.com/posts/what-happened-to-winamp-mp3-player.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-24 Updated: 2026-09-14 Topics: Software & Apps, Music & Entertainment **Summary:** Winamp still exists. The free desktop player is owned by Winamp Group, a Belgian company led by Alexandre Saboundjian, and version 5.9.2 from April 2023 still installs and runs on Windows 10 and Windows 11. The company now sells artist services under the Winamp name and has a Deezer-powered streaming player planned for the first half of 2027. **Key facts:** - Status today: Active. Winamp 5.9.2 (April 2023) is still a free download - Owner today: Winamp Group, Belgium, bought from AOL in January 2014 - Price today: Desktop player free; Winamp for Creators 55 USD a year - Launched: 1997, by Justin Frankel and Dmitry Boldyrev at Nullsoft - Peak users: About 60 million in 2001 ## Who owns Winamp today? Winamp belongs to Winamp Group, a Belgian company run by chief executive Alexandre Saboundjian. The same business bought the player from AOL in January 2014 under the name Radionomy, traded as Llama Group through the 2024 source code release, and now carries the Winamp name itself. That is three owners across the player's life. Nullsoft, the two-person company that wrote it. AOL, which paid about $80 million in stock for Nullsoft in June 1999. And the Belgian buyer that has held it since 2014. The current plan was never to sell the old player. It is to turn the name into a business for musicians, and then to put a paid listening service behind it. In July 2026 the company announced that Deezer would supply the catalog and the streaming technology for a Winamp-branded subscription, with a rebuilt player due in the first half of 2027. "Twenty-five years ago we changed the way people listened to digital music, and we believe the time has come to challenge the listening experience once again," Saboundjian said in that announcement. ## How did a teenager's Windows project reach 60 million users? Justin Frankel was 18 and had left college in Arizona when he wrapped a Windows interface around an existing MP3 decoding engine in 1997. The name was literal: Windows plus AMP. He built it with Dmitry Boldyrev under a two-person company called Nullsoft. Winamp 2.0, released in September 1998, is the version people remember. It added a plug-in system, a usable playlist window and skins, and the skin archives turned a media player into something people decorated. The installer also carried an audio clip of a man announcing "Winamp, it really whips the llama's ass," a line taken from the Chicago outsider musician Wesley Willis and his song "Whip the Llama's Ass." The timing did the rest. MP3 files were spreading through IRC (https://404memoryfound.com/posts/what-happened-to-irc-chat.html) channels, FTP servers and, from June 1999, Napster (https://404memoryfound.com/posts/napster-destroyed-music.html), and every one of those files needed something to play it. Winamp had 15 million users when AOL bought Nullsoft in 1999, 25 million registered users by mid-2000, and about 60 million by 2001. No other media player on Windows was close. ## Why did AOL pay $80 million and then shut Winamp down? AOL bought Nullsoft in June 1999 for roughly $80 million in stock. On paper it fit: AOL wanted digital media, and Winamp was the most used media player on Windows. In practice the two cultures could not share a building. Nullsoft kept shipping things AOL did not want shipped. Gnutella, a decentralized file sharing protocol, appeared in 2000 and was pulled almost immediately. WASTE, an encrypted peer-to-peer tool, appeared in 2003 and met the same end. Frankel resigned in January 2004, five years after the sale. The product damage was self-inflicted. Winamp3, a 2002 rewrite on a new codebase, broke compatibility with Winamp 2 skins and plug-ins and ran slower than the version it replaced. Users refused it and kept downloading the old one. Nullsoft retreated with Winamp 5 in December 2003, merging the classic engine with the newer features and skipping version 4 entirely. It was too late. iTunes and the iPod (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html) took over music libraries, and Windows Media Player came preinstalled on every PC. On November 20, 2013, AOL announced that Winamp.com and the player downloads would close on December 20, 2013, after 15 years. ## What did Radionomy do with Winamp after 2014? Radionomy, a Belgian internet radio company, bought Winamp and the Shoutcast streaming platform weeks before that deadline, with the deal confirmed in January 2014. TechCrunch reported the price at $5 million to $10 million in cash and shares, with AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html) taking a 12 percent stake in Radionomy. AOL had paid about $80 million for Nullsoft fifteen years earlier. "Winamp is a top independent player that gives millions of people the best player functionality available," Saboundjian said when the purchase was announced, promising builds for desktop, mobile, car systems and connected devices. Most of that never arrived. A build numbered 5.8 leaked in 2018 and was then released officially, mainly as a compatibility fix for newer Windows versions. Winamp 5.9 followed in 2022 after a long rebuild of the codebase, and 5.9.2 shipped on April 26, 2023. That is still the current desktop release in September 2026, which puts the classic player more than three years past its last update. ## What happened to the Winamp open source release? In September 2024 the owner, then trading as Llama Group, put the Winamp desktop source code on GitHub. It was not open source in any accepted sense. The Winamp Collaborative License banned forking, which is both the normal way people work on GitHub and a requirement of GitHub's own terms. Developers read the repository instead of using it. The license was revised to permit forking but not distribution, which fixed little. The published tree was then found to contain third-party code from companies including Dolby, Fraunhofer and Microsoft that Llama Group had no right to redistribute. The company deleted every Winamp repository in October 2024, about three weeks after publishing them. Winamp is therefore not open source in 2026, and the code that briefly appeared carries no license anyone can build on. Community work on the classic player happens outside the official code instead: WACUP, the Winamp Community Update Project, is a separate modernization by an independent developer that patches and replaces parts of the 5.x player. ## Where can you get Winamp in 2026, and what does it cost? The legacy desktop player is still free and still offered at winamp.com, and version 5.9.2 installs and runs on Windows 10 and Windows 11. It plays local files, MP3 and FLAC included, along with Shoutcast internet radio. Milkdrop visualizations and the Winamp 2 skin format still work, which is why the skin archives have outlived several of the companies involved. What changed is that the free player is no longer the product. Winamp for Creators sells musicians a website builder, a merchandising service and a fan subscription tool called Fanzone, which reached general availability in March 2026 at 50 euros or 55 dollars a year after an introductory free period. The company said in February 2026 that more than 35,000 artists from over 130 countries had signed up. The listening side returns in 2027. Deezer supplies the catalog behind a paid Winamp subscription, and the announced player is meant to put streaming tracks next to local files, internet radio and podcasts in one window. Until it ships, the thing most people are searching for, a small free player that reads a folder of MP3s, is the April 2023 build. ## Frequently Asked Questions ### Is Winamp still available to download in 2026? Yes. Winamp 5.9.2, released on April 26, 2023, is still the current desktop build and is still free at winamp.com, where it installs on Windows 10 and Windows 11. It plays local MP3 and FLAC files and Shoutcast internet radio, and it still supports classic Winamp 2 skins and Milkdrop visualizations. ### Why did AOL shut down Winamp? AOL bought Winamp's developer Nullsoft in June 1999 for about $80 million in stock, then announced on November 20, 2013 that Winamp.com and the player would close on December 20, 2013. Winamp had lost its audience to iTunes, the iPod and Windows Media Player after the failed Winamp3 rewrite of 2002. AOL ended up selling the player to the Belgian company Radionomy in January 2014 rather than closing it. ### Is Winamp open source? No. Winamp's owner published the desktop source code on GitHub in September 2024 under a license that banned forking, then deleted the repositories about three weeks later in October 2024 after the code was found to include third-party material it could not redistribute. As of September 2026 Winamp is closed source, and the briefly published code carries no license anyone can build on. **Sources:** - Slate: The death of the last maverick tech company (2004): https://slate.com/technology/2004/11/the-death-of-the-last-maverick-tech-company.html - CNN Money: AOL to shut down Winamp on Dec. 20 (2013): https://money.cnn.com/2013/11/20/technology/winamp/index.html - TechCrunch: AOL sells Winamp and Shoutcast to Radionomy (2014): https://techcrunch.com/2014/01/14/aol-sells-winamp-and-shoutcast-for-5-10m-to-radionomy-takes-12-stake-in-belgian-digital-audio-company - The Register: Releasing Winamp source goes badly (2024): https://www.theregister.com/2024/10/16/opensourcing_of_winamp_goes_badly/ - Deezer Newsroom: Winamp announces partnership with Deezer (2026): https://newsroom-deezer.com/2026/07/winamp-announces-strategic-partnership-with-deezer-to-power-the-next-generation-of-its-iconic-music-player/ --- # Why Pets.com Failed and Where the Sock Puppet Went URL: https://404memoryfound.com/posts/why-pets-com-failed-dot-com-bubble.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-24 Updated: 2026-09-10 Topics: Business Blunders, Money & Tech **Summary:** Pets.com failed because it sold pet supplies for less than it paid for them and then bought attention with the difference, including a Super Bowl spot in January 2000 that is commonly reported at $1.2 million and $11.0 million of ABC airtime it paid for in stock. It went public at $11.00 a share on February 11, 2000, and its board approved the wind-down on November 4, 2000, 267 days later. PetSmart bought the domain in December 2000, and pets.com still redirects to petsmart.com today. **Key facts:** - Founded: Domain registered November 21, 1994; store opened November 1998; incorporated February 1999 - IPO: February 11, 2000, 7,500,000 shares at $11.00, $82.5 million raised - Shut down: Web store closed November 10, 2000; dissolved January 18, 2001 - Owner today: PetSmart, which bought the domain and trademarks in December 2000 - Status: Gone. Renamed IPET Holdings and delisted from Nasdaq in January 2001 ## Why did Pets.com lose money on every bag of dog food? The merchandise itself was under water. The IPO prospectus Pets.com filed with the SEC in February 2000 reports net sales of $5,787,000 for 1999 against a gross margin of negative $7,625,000. That is before payroll, before warehouses, before a single television spot. The company was paying more for the products than customers were paying for them. The category made that hard to fix. Dog food, cat litter and birdseed are heavy, cheap per pound, and predictable enough that shoppers already bought them on a schedule at a store two miles away. Pets.com had to warehouse the pallets, discount the goods to pull people online, and then pay a carrier to move heavy, bulky bags across the country. Every extra order deepened the hole rather than closing it. The 1999 net loss came to $61,778,000 on that $5.8 million of sales, and the loss for the first nine months of 2000 reached roughly $84.9 million. The same physics broke Webvan and its grocery delivery network (https://404memoryfound.com/posts/what-happened-to-webvan-grocery-delivery.html) and Kozmo's one-hour courier model (https://404memoryfound.com/posts/what-happened-to-kozmo-com-delivery.html) in the same 18 months. ## Where did the advertising money actually go? The Super Bowl spot is the part people remember, and it is the part that is hardest to pin down. Pets.com aired its first national commercial during the January 2000 game, a cost usually reported at $1.2 million for 30 seconds. The company's own SEC filings never break out the price of that single slot, so the figure comes from contemporary press coverage rather than from the books. The filings do show something better. In January 2000, weeks before the offering, Pets.com issued 1,102,400 shares of Series C preferred stock to an affiliate of The Walt Disney Company in exchange for $11.0 million of media advertising on ABC. The network that carried that Super Bowl was paid in equity, not cash. The spending did not stop at the game. Pets.com committed to buy at least $9 million of online advertising on the GO.com portal, and in the second quarter of 2000, its first full quarter as a public company, it spent $17 million on advertising against $8.8 million of revenue, VICE reported in 2017, citing MarketWatch. Boo.com burned through its funding on the same theory (https://404memoryfound.com/posts/what-happened-to-boo-com-fashion-dot-com.html), that awareness bought early would pay for itself later. ## Who made the sock puppet, and where is it now? The puppet was a dog with a microphone, created by the agency TBWA Worldwide and voiced by the comedian Michael Ian Black, according to The Henry Ford, which holds one in its collection. The first television ads ran in the fall of 1999. A balloon float of the puppet appeared in the 1999 Macy's Thanksgiving Day Parade, and the prospectus described it in the flat language of a securities filing: "our sock puppet is a roving advocate for the brand," Pets.com told investors in February 2000. It outlasted the company that made it. When the board voted to wind down operations, it authorized management to sell the inventory, the equipment, the URLs and "the Sock Puppet brand icon" as separate assets. In 2002, Hakan and Associates and the auto lender Bar None bought the rights through a joint venture called Sock Puppet LLC for $125,000, and put the dog back on television selling car loans to people with damaged credit under the line "everybody deserves a second chance." Today the character sits mostly in museum drawers and closets. The Henry Ford catalogs a Pets.com sock puppet made by Fun-4-All Corp., dated 1999 to 2000, as object 2017.101.1, not currently on exhibit. Where to find one today: the plush version sold in stores during the campaign still circulates among collectors, boxed and loose, through Etsy seller listings for the Pets.com sock puppet (https://www.etsy.com/search?q=pets.com+sock+puppet). ## How fast did the collapse happen after the IPO? Pets.com sold 7,500,000 shares at $11.00 on February 11, 2000, closing the offering on February 16 for $82,500,000 gross and about $75.3 million net. That followed $90.2 million of private equity raised between April and December 1999. Amazon (https://404memoryfound.com/posts/how-amazon-survived-dot-com-crash.html).com was the largest stockholder, holding 8,973,029 shares, or 25.3 percent. The stock never recovered from its first quarter. It traded as high as $14.00 in the weeks after the IPO, then between $0.81 and $0.06 in the fourth quarter of 2000. In July 2000, Pets.com hired Merrill Lynch to find a buyer or new money. More than 50 domestic and international prospects were contacted, and as the 10-K put it in 2001, "fewer than eight were even prepared to meet in person." The end came in a week. The board approved the wind-down on November 4, 2000, 255 of 320 employees were laid off by November 7, and the web store closed and stopped taking sales on November 10, 2000. Stockholders approved the liquidation plan on January 16, 2001, the company renamed itself IPET Holdings, and the Certificate of Dissolution took effect on January 18, 2001. The shares closed that day at $0.125 and were delisted from Nasdaq. ## Who owns Pets.com now? PetSmart does. Greg McLemore registered the domain on November 21, 1994, four years before the store opened, and after the liquidation the domains, trademarks and subsidiaries went to PetSmart in December 2000. Type pets.com into a browser in 2026 and the request lands on petsmart.com. The registration is still active, managed through the corporate registrar MarkMonitor. Nothing of the company survives beyond the name. IPET Holdings sold the inventory, the furniture and the warehouse equipment, terminated four of its five building leases, and had 26 full-time employees left on December 31, 2000. All of them resigned in January 2001, leaving a management firm to close the books. The idea did survive. Online pet retail is now a normal business, and Pets.com is the reference point every later entrant gets measured against, which is why VICE framed the story in 2017 around whether the company was simply early rather than wrong. ## Was Pets.com wrong, or just early? Julie Wainwright, the chief executive, has argued the second. The tools a store like that runs on did not exist yet. "Cloud computing did not exist, which means that we had to have a server farm and several IT people," she told VICE in 2017, and she said the company had to employ more than 40 engineers because nothing off the shelf could scale. The rest of the stack was missing too. Parcel rates for heavy goods, warehouse software, subscription reordering and cheap targeted advertising all arrived later. Pets.com paid for national television because there was no efficient way to reach pet owners at the moment they needed food. What that argument does not excuse is the discounting. A company selling below cost while spending $17 million a quarter on ads is not waiting for infrastructure, it is buying revenue. Pets.com was 27 months old from launch to liquidation, and roughly nine of those months were spent as a public company with a mascot more famous than its store. ## Frequently Asked Questions ### How much did the Pets.com Super Bowl ad cost? Pets.com ran its sock puppet commercial during the Super Bowl in January 2000, at a price usually reported as $1.2 million for 30 seconds. The company never disclosed the cost of that slot in its SEC filings. What the filings do show is that in January 2000 Pets.com paid a Walt Disney Company affiliate $11.0 million worth of ABC airtime in Series C preferred stock rather than cash. ### Was Pets.com founded in 1998 or 1999? Both dates are right for different things. Greg McLemore registered the pets.com domain on November 21, 1994, and the store opened to shoppers in November 1998, but Pets.com, Inc. was incorporated in California in February 1999 and reincorporated in Delaware in February 2000. Its first financial year, in the IPO prospectus, runs from February 17, 1999. ### Does Pets.com still exist, and does the website work? Pets.com stopped selling on November 10, 2000, and was legally dissolved on January 18, 2001 under the name IPET Holdings. The address still works, but it is not a store: PetSmart acquired the domain and trademarks in December 2000, and pets.com now redirects to petsmart.com. **Sources:** - Pets.com, Inc. Form 10-K for fiscal year 2000, SEC EDGAR (filed April 2, 2001): https://www.sec.gov/Archives/edgar/data/1100683/000109581101002065/f71096e10-k.txt - Pets.com, Inc. IPO prospectus, Form 424B4, SEC EDGAR (February 11, 2000): https://www.sec.gov/Archives/edgar/data/1100683/0000891618-00-000749.txt - Ernie Smith, 'Pets.com Was Decades Ahead of Its Time', VICE, January 2017: https://www.vice.com/en/article/petscom-was-decades-ahead-of-its-time/ - Pets.com Sock Puppet, 1999-2000, object 2017.101.1, The Henry Ford: https://www.thehenryford.org/collections-and-research/digital-collections/artifact/449925/ - Pets.com, Wikipedia (Super Bowl spot cost and Sock Puppet LLC rights sale): https://en.wikipedia.org/wiki/Pets.com --- # Why Google+ Failed and When Google Shut It Down URL: https://404memoryfound.com/posts/why-google-plus-actually-failed.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-24 Updated: 2026-09-14 Topics: Business Blunders, Software & Apps **Summary:** Google+ is gone. Google closed the consumer version on April 2, 2019 after two data leaks, and retired the last workplace version, Google Currents, on July 5, 2023. It failed because Google built it to solve Google's problem with Facebook rather than a problem its users had, and by 2018 Google's own engineering leadership reported that 90 percent of Google+ sessions lasted under five seconds. **Key facts:** - Launched: June 28, 2011, invitation only - Status today: Shut down. The consumer version closed on April 2, 2019 - Owner today: Google, which retired the last version, Google Currents, on July 5, 2023 - Estimated cost: About $585 million, per a Forbes estimate in 2011 - What replaced it: Spaces in Google Chat, plus Google Photos and Google Meet ## What was Google+ supposed to do for Google? Google+ opened on June 28, 2011 as an invitation-only field trial, and it was not really built for the people signing up. It was built for Google. Facebook was pulling attention and data into a space Google's crawlers could not reach, and Google's advertising business depended on people moving around the open web instead. The project ran under the internal codename Emerald Sea and was led by Vic Gundotra, a former Microsoft executive who reported to Larry Page. Social was made the company's top priority that year, and teams across search, Gmail, YouTube and Android were pushed to wire Google+ into whatever they were already building. It was not the first attempt. Google Buzz, bolted onto Gmail in February 2010, built public contact lists out of people's email and drew a Federal Trade Commission complaint within weeks. Google Wave, launched in 2009, was shut down in 2012 because almost nobody could work out what it was for. Google+ was the third try, and this time the whole company was told to care. ## Why did Circles never catch on? Circles was the feature every review led with in 2011. You dragged contacts into groups called family, work or anything else, then chose which group saw each post. Facebook had nothing that clean at the time, and the problem was real: most people do not want a manager reading what they tell their college friends. The trouble was the setup cost. Circles asked for sorting work before the first post, then asked again every time someone new showed up. A social network has to be useful in the ten seconds between opening it and getting bored, and Circles spent those ten seconds on filing. There was a deeper problem. Real social graphs overlap. One person can be a coworker, a neighbor and a friend at the same time, and a system that wants a single label per group does not describe that. Facebook later shipped its own list tools and watched them go mostly unused for the same reason. People want to post something and be done, not maintain a database of their relationships. ## How empty was Google+ really? Empty enough that Google eventually published the number itself. Announcing the shutdown in October 2018, Google vice president of engineering Ben Smith wrote that "90 percent of Google+ user sessions are less than five seconds." Nine visits in ten ended before the page had a chance to be interesting. The gap between accounts and activity is the whole story. Registrations were easy for Google to grow, because signing up for Gmail signed you up for Google+ as well. Attention was the part it could not manufacture. People opened the stream, found posts from technology journalists and Google employees, did not find their sister or their neighbors, and closed the tab. Social networks run on a loop: your people are there, so you go, so you post, so your people keep going. Google+ never closed that loop. The collapse of MySpace (https://404memoryfound.com/posts/rise-and-fall-of-myspace.html) a few years earlier showed how fast a network empties once the loop breaks, and the story of Friendster (https://404memoryfound.com/posts/what-happened-to-friendster-social-network.html) showed the same thing a decade before that. Google+ never got the loop started at all. ## Why did forcing Google+ on YouTube backfire? In November 2013, Google made a Google+ account a requirement for commenting on YouTube. The stated goal was better comments. The effect was a revolt by the people who made YouTube worth watching. The sharpest response came from Jawed Karim, who co-founded YouTube and uploaded the site's first video in 2005. He changed that video's description to read: "I can't comment here anymore, since i don't want a Google+ account." The Register reported the protest in November 2013, and it spread through creators and commenters who had never asked for a social network. Google backed off in stages. The real-name rule, which had pushed away people posting under pseudonyms, was dropped in July 2014. The YouTube requirement went in July 2015, along with the Google+ profile requirement on other Google products. By then the damage was structural. Google had spent four years teaching hundreds of millions of people that Google+ was something done to them, and that is a hard lesson to unteach. ## Is Google+ still around in 2026? No. Google+ shut down for consumers on April 2, 2019, and nothing has replaced it under that name. The last piece of it, the workplace version called Google Currents, was retired on July 5, 2023, and Google deleted the remaining Currents data that August. There is no Google+ account to sign into in 2026 and no way to create one. Two security failures set the ending date. In March 2018 Google found a bug in the Google+ People API that had exposed profile data for up to 500,000 accounts, and chose not to say so publicly at the time. When the story reached the press in October 2018, Google announced the bug and the shutdown together, citing "low usage and engagement." In December 2018 a second bug surfaced, this one exposing data from 52.5 million accounts during about six days in November. NPR reported that Google then moved the shutdown up by four months, from August 2019 to April 2019. Public posts were not all lost. Volunteers from Archive Team copied public Google+ posts and profiles to the Internet Archive before the shutdown, and anyone who ran Google Takeout before April 2019 kept a copy of their own. ## What survived Google+? More than the name suggests. Hangouts, the video calling tool built inside Google+, was pulled out as its own product and eventually became Google Meet and Google Chat. Google+ Photos was split off in May 2015 as Google Photos, which outlived the network it was born in and is now one of Google's most used consumer products. Google Currents, the workplace version, ran until July 5, 2023, when Google folded its function into Spaces in Google Chat. That is where the Google+ idea actually landed: not a public square built to fight Facebook, but group conversations inside a company's own account. The pattern is familiar to anyone who watched Google Reader disappear (https://404memoryfound.com/posts/what-happened-to-google-reader-rss.html) in 2013. Google ships, Google measures, and Google closes what does not clear the bar. At roughly $585 million by the Forbes estimate of 2011, made up of about $125 million in compensation and $460 million in related acquisitions, Google+ is the most expensive example. It is not an unusual one. ## Frequently Asked Questions ### When did Google+ shut down? Google+ shut down for consumers on April 2, 2019, almost eight years after it launched on June 28, 2011. Its developer APIs were switched off on March 7, 2019. The workplace version, renamed Google Currents, kept running until July 5, 2023, which is when Google+ ended for everyone. ### Why did Google shut down Google+? Google announced the end of Google+ in October 2018 and gave two reasons: almost nobody used it, and a bug in its API had exposed profile data for up to 500,000 accounts. Google's own announcement cited "low usage and engagement" and the "challenges involved in maintaining a successful product that meets consumers' expectations." A second bug disclosed in December 2018, affecting 52.5 million accounts, moved the shutdown up to April 2019. ### Does Google have a social network in 2026? Google runs no public social network in 2026. Google+ closed in April 2019 and Google Currents closed in July 2023, leaving Spaces in Google Chat for workplace group conversations and YouTube, which carries comments and subscriptions without being sold as a social network. Google has not launched another Facebook competitor since Google+. **Sources:** - Google+ Cost $585 Million To Build (Or What Rupert Paid For MySpace) (Forbes, 2011): https://www.forbes.com/sites/bruceupbin/2011/06/30/google-cost-585-million-to-build-or-what-rupert-paid-for-myspace/ - Jawed Karim protests the Google+ requirement for YouTube comments (The Register, 2013): https://www.theregister.com/2013/11/08/jawed_karim_you_tube_google_plus_account/ - Project Strobe: improving our third-party APIs and sunsetting consumer Google+ (Ben Smith, Google, 2018): https://www.blog.google/technology/safety-security/project-strobe/ - Google Accelerates Google+ Shutdown After 52.5 Million Users' Data Exposed (NPR, 2018): https://www.npr.org/2018/12/11/675529798/with-52-5-million-users-data-exposed-on-google-google-quickens-shutdown - Final update: Google Currents shutdown (Google Chat Help): https://support.google.com/chat/answer/14310347 --- # Why Sega Stopped Making Consoles After Dreamcast URL: https://404memoryfound.com/posts/why-sega-stopped-making-consoles-dreamcast.html Author: Marcus Vale (404 Memory Found) Published: 2026-03-23 Updated: 2026-09-12 Topics: Gaming, Business Blunders **Summary:** Sega stopped making consoles on January 31, 2001, when it discontinued the Dreamcast and became a third-party publisher. The Dreamcast sold 9.13 million units worldwide, nowhere near enough after the Sega Saturn had already cost the company its retailers and its credit with publishers. Sega still exists in 2026 inside Sega Sammy Holdings, making games for PlayStation, Xbox, Nintendo and PC, with no hardware of its own. **Key facts:** - Status today: No console hardware since 2001; Sega publishes games only - Owner today: Sega Sammy Holdings, Tokyo - Hardware exit: Announced January 31, 2001 - Dreamcast sales: 9.13 million units worldwide - Price today: About $100 to $200 for a working used Dreamcast in 2026 ## How did the Sega Saturn set up the collapse? The damage started four years before the Dreamcast shipped. On May 11, 1995, at the first E3 in Los Angeles, Sega of America told the room that the Sega Saturn was on sale that day at $399, with about 30,000 units already sitting in a short list of chains. The retailers left off that list were furious. KB Toys (https://404memoryfound.com/posts/kb-toys-stores-left.html) dropped Sega from its lineup. Developers had nothing ready because nobody had told them the date had moved. Minutes later, in the same venue, Sony priced the PlayStation $100 below the Saturn. The Saturn finished its life at 9.26 million units worldwide, against 30.75 million for the Sega Genesis a generation earlier. The money mattered, but the trust mattered more. Buyers, retail chains and publishers all came out of the Saturn years (https://404memoryfound.com/posts/what-happened-to-sega-saturn-console.html) with a reason to wait and see the next time Sega announced anything. That reflex is what the Dreamcast walked into. ## What made the Dreamcast different in 1999? The Dreamcast reached North America on September 9, 1999 at $199, the first console of its generation, a year ahead of the PlayStation 2 (https://404memoryfound.com/posts/playstation-2-best-selling-console-ever.html) and two years ahead of the Xbox and the GameCube (https://404memoryfound.com/posts/why-nintendo-gamecube-lost-console-war.html). Sega put a 56k modem in every unit. Online console play, which Xbox Live would not formalize until 2002, shipped in the box in 1999, and Phantasy Star Online turned it into something people used in 2000. The controller carried a Visual Memory Unit, a save card with its own screen and buttons. The launch itself worked. Sega reported $98.4 million in combined hardware and software sales in the first 24 hours, on roughly 225,000 consoles, a record at the time. Soul Calibur, Sonic Adventure, Crazy Taxi, Jet Set Radio and Shenmue followed. For about eighteen months the comeback looked real. Peter Moore, who ran Sega of America at the end, later told MCV/DEVELOP that "Dreamcast was on fire" and that the team "really thought that we could do it." ## Why did Electronic Arts refuse to make Dreamcast games? Electronic Arts was the largest third-party publisher in the world in 1999, and it released nothing at all on the Dreamcast. No Madden, no FIFA, no NBA Live. For a United States audience that bought sports games by the million, the gap was visible on the shelf. The cause was a negotiation that broke. Bernie Stolar, then president of Sega of America, has said that Electronic Arts chief executive Larry Probst told him "Bernie, we'll do Dreamcast games, but we want sports exclusivity." Sega had just paid about $10 million for Visual Concepts, the studio behind NFL 2K, and Stolar would not shelve it. In his account of the exchange, recounted to SEGAbits in 2013, Probst answered "No, you can't even put them on the system," and that was the end of it. Visual Concepts went on to build the 2K sports line, which was good for Sega's software business and ruinous for its hardware. The console went into its first holiday season without the one publisher American sports players were waiting for. ## What forced Sega's decision in January 2001? Three problems landed together. Sony announced the PlayStation 2 in 1999 and shipped it in Japan in March 2000 and in North America that October. It played DVDs and ran PlayStation discs, and buyers were willing to wait for it. Dreamcast sales slowed while they waited. Piracy did the rest. The Dreamcast read GD-ROM discs that nobody could copy, but Sega had also built in support for MIL-CD, a music format that let the console boot software from an ordinary CD-R. No chip, no soldering, just a burned disc. Software royalties are what pay for hardware sold below cost, and those royalties leaked. Then the accounts came due. On January 31, 2001, Sega said it would stop making the Dreamcast and become a third-party publisher, ending production that spring while letting the remaining titles ship. For the year to March 2001 the company posted a consolidated net loss of 51.7 billion yen, about $417 million. President Isao Okawa handed Sega roughly $695 million of his own Sega and CSK stock and forgave the debts it owed him. He died on March 16, 2001, at 74. ## Is Sega still making consoles in 2026? No. Sega has not built a home console since the Dreamcast and has given no sign that it intends to. The only hardware it has put its name on since is plug-and-play reissues of its own back catalogue, which are boxes of old games rather than platforms with a future. The company is otherwise in good health. Sega sits inside Sega Sammy Holdings in Tokyo and publishes on PlayStation, Xbox, Nintendo and PC. For the year to March 2026, Sega Sammy reported 487.5 billion yen in net sales, up 13.7 percent on the year before, and a net loss of 5.7 billion yen, about $31.6 million, after writedowns on the Rovio and Stakelogic acquisitions. It has also pulled back from its largest bet. In May 2026 Sega cancelled Super Game, the roughly $1 billion online blockbuster programme announced in 2021, and moved developers off free-to-play work and back onto full releases of its existing series. Twenty-five years on, Sega's answer to the hardware question is the one it gave in 2001: it makes the games and somebody else makes the box. There is more on what Sega publishes now (https://404memoryfound.com/posts/is-sega-still-making-games.html). ## What does a Dreamcast cost in 2026? Used Dreamcasts are still cheap next to most retro hardware. Price guides in 2026 put a working loose console with a controller at roughly $100 to $200, with boxed and complete examples going for two to three times a bare unit. Check the laser before buying anything: a machine with disc read errors is worth a fraction of a clean one, and a tired GD-ROM drive is the usual cause of death. Where to find one today: consoles, controllers, Visual Memory Units and the replacement parts you will eventually want are stocked by retro specialists rather than general marketplaces. Stone Age Gamer (https://stoneagegamer.com/) carries Dreamcast hardware and accessories. New software still appears as well, with the publisher VGNYsoft reissuing a run of Dreamcast indie titles for sale in GameStop (https://404memoryfound.com/posts/is-gamestop-still-in-business.html) stores in the United States in 2026. ## Frequently Asked Questions ### When did Sega stop making consoles? Sega announced on January 31, 2001 that it would discontinue the Dreamcast and leave the console hardware business, with production ending that spring. The Dreamcast, launched in North America on September 9, 1999, was Sega's last home console, and the company has made none since. ### Why did the Sega Dreamcast fail? The Dreamcast sold 9.13 million units between 1999 and 2001 against a Sony PlayStation 2 that launched in North America in October 2000 with DVD playback. Electronic Arts published nothing for it, and the MIL-CD flaw let the console run burned discs, which cut the software royalties Sega needed. Sega lost 51.7 billion yen, about $417 million, in the year to March 2001. ### Will Sega ever make a new console? Nothing suggests it. Sega has published only for other companies' platforms since it left hardware in 2001, and in May 2026 it cancelled Super Game, its roughly $1 billion push for an online blockbuster, to concentrate on full releases of its existing series. Sega Sammy Holdings has announced no console project. **Sources:** - Dreamcast, Wikipedia: https://en.wikipedia.org/wiki/Dreamcast - Sega to Halt Dreamcast Production, ABC News, 2001: https://abcnews.go.com/Technology/story?id=98977&page=1 - SEGA Memories: EA not supporting the Dreamcast, SEGAbits, 2013: https://segabits.com/blog/2013/05/20/monday-memories-ea-not-supporting-the-dreamcast/ - Peter Moore: I killed the Dreamcast, MCV/DEVELOP: https://mcvuk.com/business-news/peter-moore-i-killed-the-dreamcast/ - SEGA Cancels Super Game Project and Moves Away from Live Service, Push Square, 2026: https://www.pushsquare.com/news/2026/05/sega-cancels-super-game-project-and-moves-away-from-live-service --- # Palm Pilot Explained: The PDA That Phones Replaced URL: https://404memoryfound.com/posts/what-happened-to-palm-pilot-pda.html Author: Marcus Vale (404 Memory Found) Published: 2026-03-23 Updated: 2026-09-12 Topics: Hardware, Business Blunders **Summary:** The Palm Pilot was a pocket organizer that Palm Computing launched in March 1996 for $299, and about a million sold in the first 18 months. It is finished as a product: Palm released its last two handhelds, the Palm TX and the $99 Palm Z22, in October 2005, and smartphones took over everything a PDA did. The Palm name survives as a dormant trademark owned by TCL since 2014. **Key facts:** - Launched: March 1996, Pilot 1000, $299, Palm Computing under U.S. Robotics - Units sold: About 1 million in the first 18 months - Last handhelds: Palm TX and the $99 Palm Z22, October 2005 - Status today: Discontinued. Trademark owned by TCL, webOS owned by LG - Price today: About $15 to $25 for a working used unit on Etsy ## How did a block of wood become the Palm Pilot? Jeff Hawkins founded Palm Computing in 1992 and built the product backwards, starting with the shape. He cut a piece of scrap lumber to shirt-pocket size in his garage, glued a paper face on it showing where a screen and buttons would go, carried a chopstick for a stylus, and used the thing in meetings for months. "I walked around answering phone calls with this block of wood, and of course it didn't do anything," Hawkins said of the exercise, in an account kept by the Computer History Museum, which now holds the wooden prototype in its collection. The block produced one rule: do four things well instead of everything badly. Calendar, address book, to-do list, memo pad. The Pilot 1000 shipped in March 1996 with 128 KB of memory, a monochrome touchscreen and a cradle that synced to a PC, at $299. Palm Computing was a division of U.S. Robotics by then, and about a million units sold in the first 18 months. The PalmPilot Professional followed in March 1997 at $399 with 1 MB of memory. Where the Apple Newton (https://404memoryfound.com/posts/what-happened-to-apple-newton.html) had promised a computer in your pocket and failed at it, Palm promised a better pocket notebook, and that was the version people bought. ## Why was Palm worth $53 billion for a single day? The ownership moved faster than the products did. U.S. Robotics bought Palm Computing in 1995. 3Com bought U.S. Robotics in 1997. Hawkins and Donna Dubinsky left in 1998 to found Handspring, which built Palm OS machines called the Visor and competed with the company they had started. 3Com said in September 1999 that it would spin the handheld business out as Palm, Inc. Roughly 23 million shares, about 5 percent of the company, went public on March 2, 2000. The stock closed near $95, and The Washington Post reported the result the next morning as a "$53 Billion Welcome." 3Com kept about 94 percent of the shares. Finance professors still teach that day, because 3Com's own share price implied that the Palm stake it retained was worth more than all of 3Com put together. Palm never grew into the number. Within two years it had split itself into a hardware company and a software company, then absorbed Handspring in 2003 and renamed the hardware half palmOne. ## What did the Xerox lawsuit do to Graffiti? Palm's text entry system, Graffiti, asked users to learn a simplified alphabet of one stroke per letter. Xerox said it owned the idea. Its patent, U.S. 5,596,656, covered a method called Unistrokes developed at its research labs. Xerox sued 3Com, U.S. Robotics and Palm Computing. A district court granted summary judgment of non-infringement in June 2000, holding that Graffiti's symbols were not "graphically separated" enough to qualify as unistroke symbols. A federal appeals court sent that ruling back in 2001, and Graffiti was found to infringe. The case turned again in May 2004, when Judge Michael A. Telesca ruled the Xerox patent invalid because "prior art references anticipate and render obvious the claim," citing a 1983 Bell Labs paper by D. J. Burr and a 1984 Japanese patent as prior art. Palm won, seven years after the first filing. By then the fight was over a way of entering text that on-screen keyboards had already made beside the point. ## What actually killed the PDA? Arithmetic, mostly. Gartner counted a peak of 13.1 million handhelds shipped worldwide in 2001, then 12.1 million in 2002, 11.5 million in 2003, 9.2 million in 2004 and 7.5 million in 2005. Palm took about 33.8 percent of that market in the third quarter of 2002, up from 29.1 percent a year earlier, which meant a bigger slice of a shrinking pie. Buyers had not stopped wanting a pocket computer. They had started buying phones that did the same work and rang. Handspring's Treo 600 reached carriers in late 2003, after Handspring had merged into Palm, and it put the calendar, the contacts and the email behind a phone number. BlackBerry (https://404memoryfound.com/posts/what-happened-to-blackberry-smartphone.html) had already done that for corporate email and had the mail servers to back it up. Palm read the room and left the room. The Palm TX and the $99 Palm Z22 arrived in October 2005 and were the last handhelds the company released. Everything Palm shipped after that was a phone. ## Did Palm see the iPhone coming? Not clearly. Ed Colligan, one of the three people who built the first Pilot and by then Palm's chief executive, was asked at a Churchill Club breakfast in November 2006 whether Apple could walk into the phone business. He was widely reported as answering: "PC guys are not going to just figure this out. They're not going to just walk in." Audio released years later has him phrasing it less flatly, and he was talking about the whole PC industry rather than Apple alone, but the reading stuck. Apple announced the iPhone about two months afterward. Palm's real answer came two and a half years late. The Palm Pre launched on Sprint in June 2009 with webOS, a new operating system whose card-based multitasking and merged notifications were better than what shipped on phones at the time and were later copied widely. The problem was everything around it: one carrier, a small app catalogue, and a company with no money left to advertise against Apple and Google. ## Who owns the Palm Pilot name in 2026? TCL, and it is not using it. Hewlett-Packard bought Palm in 2010, stopped webOS hardware in 2011, sold the webOS software to LG Electronics in 2013, and sold the Palm trademark in 2014 to a holding company tied to the Chinese manufacturer TCL. TCL put out one product under the name, a 3.3 inch Android companion phone in 2018, and nothing since. Our separate piece on whether Palm is still a company (https://404memoryfound.com/posts/is-palm-still-a-company.html) traces the paperwork in detail. Hawkins did not follow the brand. He co-founded Numenta in 2005 to work on brain theory and machine intelligence, and that is still what he does. The most active piece of Palm in 2026 is webOS, running on LG television sets in millions of living rooms, where nobody thinks of it as Palm at all. Where to find one today: working Palm III, Palm V and Palm m105 units turn up steadily on Etsy (https://www.etsy.com/search?q=vintage+palm+pilot+pda), commonly listed between about $15 and $25 with a stylus and a dock. Check two things before buying. The rechargeable models often have a dead internal battery, and the cradle and stylus are easy to lose and awkward to replace. ## Frequently Asked Questions ### When did the Palm Pilot stop being made? Palm released its last two handheld organizers, the Palm TX and the $99 Palm Z22, in October 2005, and every device it launched after that was a phone. Production of the remaining Palm handheld models wound down over the following years, so a Palm Pilot bought new after 2005 was old stock rather than a current product. ### How much did a Palm Pilot cost when it came out? The Pilot 1000 launched in March 1996 at $299 with 128 KB of memory, and the PalmPilot Professional followed in March 1997 at $399 with 1 MB. Those were 1996 and 1997 prices for a device that held a calendar, a contact list, a to-do list and notes, and nothing else. ### Why did PDAs disappear? PDAs disappeared because phones absorbed them. Gartner counted 13.1 million handhelds shipped in 2001 and 7.5 million in 2005, a fall that tracks the arrival of devices such as the Treo 600 in 2003 and the BlackBerry line, which carried a calendar and contacts inside something users already had to carry. After the iPhone arrived in 2007, a separate organizer had no job left to do. **Sources:** - Computer History Museum: The PalmPilot (CHM Revolution): https://www.computerhistory.org/revolution/mobile-computing/18/321 - Computerworld: A brief history of Palm: https://www.computerworld.com/article/1522788/a-brief-history-of-palm.html - The Washington Post: 3Com's Palm IPO Gets A $53 Billion Welcome (March 3, 2000): https://www.washingtonpost.com/archive/business/2000/03/03/3coms-palm-ipo-gets-a-53-billion-welcome/24f152cb-3b74-425f-aca4-91e543f122fc/ - The Register: PalmOne overturns Xerox Graffiti patent (May 24, 2004): https://www.theregister.com/2004/05/24/palm_vs_xerox/ - The Register: Handheld sales fall in 2005 (Gartner figures): https://www.theregister.com/2006/02/03/handheld_sales_fall_in_2005/ --- # Who Owns MapQuest Now? System1 and the 2026 Surge URL: https://404memoryfound.com/posts/how-mapquest-lost-to-google-maps.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-23 Updated: 2026-09-12 Topics: Software & Apps, Business Blunders **Summary:** MapQuest still exists in 2026. The site and its apps are owned by System1, a Los Angeles advertising company that bought MapQuest from Verizon in October 2019, and MapQuest has run continuously since February 1996. Google Maps overtook it in January 2009, almost four years after launching, and now holds about two thirds of US navigation use. **Key facts:** - Launched: February 5, 1996, at mapquest.com - Sold to AOL: About $1.1 billion in stock, announced December 22, 1999 - Owner today: System1, which bought it from Verizon in October 2019 - Status today: Still live at mapquest.com, with apps for iPhone and Android - Overtaken: January 2009, when comScore put Google Maps ahead by 700,000 users ## How did MapQuest get a ten-year head start? MapQuest came out of a printing company. R.R. Donnelley's cartographic arm, the group behind the free road maps handed out at gas stations, was spun out in 1994 as GeoSystems Global Corporation. The consumer site went live at mapquest.com on February 5, 1996. What it offered sounds thin now and was remarkable then. You typed a start address and an end address and got turn by turn text directions with a printable map of the route, free, in a browser. The alternative was a Rand McNally atlas or a stranger at a gas station. The ritual that followed defined the era. You printed the pages, highlighted the turns, and set the stack on the passenger seat. MapQuest it was a verb in American English years before Google it was. By the time MapQuest.com Inc. was filing as a public company with the Securities and Exchange Commission in 1999, it was the default map of the American internet. No rival had the brand, the traffic or the address data to challenge it. ## Why did AOL pay $1.1 billion for a map site? On December 22, 1999, America Online announced it would buy MapQuest.com Inc. in an all stock deal worth about $1.1 billion, exchanging 0.31558 AOL shares for each MapQuest share. The merger closed in 2000. The timing was terrible for MapQuest. AOL was weeks away from announcing its combination with Time Warner, the worst deal in corporate history (https://404memoryfound.com/posts/aol-time-warner-merger-worst-deal-in-history.html) by most reckonings, and MapQuest became a line item inside a media conglomerate that spent the next decade tearing itself apart. Inside AOL, MapQuest worked as an advertising property rather than a technology project. Its maps were images rendered on a server, so every pan, every zoom and every new query reloaded the page and served another ad impression. That architecture made money. It also removed any internal reason to rebuild it. So MapQuest got maintained instead of reinvented, while a search company in Mountain View quietly bought mapping startups. ## What did Google Maps do that MapQuest could not? Google Maps launched on February 8, 2005, announced by one short post on the Official Google Blog. "We think maps can be useful and fun, so we've designed Google Maps to simplify how to get from point A to point B," it read. That was the whole marketing campaign. The product did one thing MapQuest's could not. You grabbed the map with the mouse and dragged it, and the map moved, with no page reload. The interface was built around asynchronous JavaScript requests, so new tiles loaded in the background while you kept scrolling. Then came the parts MapQuest never answered. Google opened a mapping API in June 2005, and real estate sites, restaurant finders and travel pages embedded Google maps by the thousand, each one an advertisement for Google's mapping. Street View arrived in the United States in May 2007, built on camera cars that drove and photographed the roads themselves. MapQuest licensed its road data from third parties. Google went out and collected its own, then kept collecting. ## When did Google Maps actually pass MapQuest? January 2009. comScore's US figures that month put Google Maps at 42.2 million unique visitors against MapQuest's 41.5 million, the first time Google had led. The gap was under a million people, and it had taken Google almost four years to close it. It never reopened. The reason was the phone rather than the desktop. Directions matter most when you are already out of the house, and by 2009 Google Maps shipped preinstalled on Android phones and was the default mapping app on the iPhone, a position it held until Apple shipped its own maps in 2012. MapQuest had mobile apps too. Being an app you have to hear about, find and download is a different business from being the app that is already on the phone. Google bought Waze in 2013 and folded its crowdsourced jams and hazard reports into the same stack. Standalone GPS units from TomTom and Garmin (https://404memoryfound.com/posts/what-happened-to-standalone-gps-tomtom-garmin-magellan.html) lost the same fight at the same time, for the same reason. ## Is MapQuest still around in 2026? Yes. mapquest.com still loads, still gives driving directions, and still has apps on iPhone and Android. The owner is System1, a Los Angeles advertising and consumer internet company that bought MapQuest from Verizon in October 2019 for an undisclosed price. Verizon owned it because it had bought AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html) in 2015. System1 has run MapQuest as a privacy pitch rather than a feature race. In October 2024 it released Private Maps by MapQuest, an app whose selling point is that it does not track the user, does not sell the data and does not serve ads against the route. That positioning paid in August 2026. When the Trump administration ordered Lake Ontario renamed Lake America and both Google Maps and Apple Maps applied the change for US users, MapQuest refused, and shipped a tool that let people label the lake whatever they wanted. Downloads jumped more than 300 percent and the app went to number one in the App Store's Navigation category. "People let us know we were making the right decision with their downloads back then, and they are downloading even more with Lake Ontario," MapQuest general manager Doug Berger told Newsweek in 2026. The scale is still tiny next to Google. Independent trackers put mapquest.com between roughly 27 million and 35 million visits a month in late 2025, depending on whose estimate you take, while Google Maps holds about two thirds of US navigation use. ## What did drivers lose when the paper printout died? The MapQuest printout forced a kind of planning that has stopped happening. You read the route before you left. You knew which direction you were heading and which highway you wanted, because you had looked at the whole thing on one page first. The trade was worth making. MapQuest had no traffic data, so a closed highway was something you found out about at the wall of brake lights. Modern navigation reroutes around the crash before you reach it, names the lane to be in, and revises the arrival time every few seconds from the position pings of millions of phones. What went with the paper was the mental map. Plenty of drivers now cannot reach a friend's house they have visited ten times without switching the app on. That is a small price for never being properly lost, and it is still a price. ## Frequently Asked Questions ### Does MapQuest still exist in 2026? MapQuest still exists in 2026. The site runs at mapquest.com with free maps and driving directions, and MapQuest apps are on iPhone and Android, including the privacy focused Private Maps app released in October 2024. MapQuest has operated continuously since February 1996 and has never shut down. ### Who owns MapQuest now? MapQuest has been owned by System1, a Los Angeles advertising and consumer internet company, since October 2019. Before that MapQuest belonged to Verizon, which got it through its 2015 purchase of AOL, and AOL had bought MapQuest in a deal announced on December 22, 1999 and valued at about $1.1 billion in stock. ### When did Google Maps pass MapQuest? Google Maps passed MapQuest in January 2009, when comScore measured 42.2 million US unique visitors for Google Maps against 41.5 million for MapQuest. Google Maps had launched on February 8, 2005, so it took almost four years to overtake a site that had led online mapping since 1996. **Sources:** - Official Google Blog, 'Mapping your way' (February 8, 2005): https://googleblog.blogspot.com/2005/02/mapping-your-way.html - U.S. Securities and Exchange Commission: MapQuest.com Inc., Form 8-K (1999): https://www.sec.gov/Archives/edgar/data/0001078284/000090256199000609/0000902561-99-000609.txt - Search Engine Land: comScore to report Google Maps now number 1 (2009): https://searchengineland.com/comscore-to-report-google-maps-now-number-1-16570 - System1 investor relations: MapQuest unveils Private Maps app (October 2024): https://ir.system1.com/news/news-details/2024/MapQuest-the-Online-Mapping-Pioneer-Is-Back-MapQuest-Unveils-Private-Mapping-App-that-Protects-Your-Privacy--Keeps-Your-Data-Away-from-Big-Tech/default.aspx - Newsweek: Who owns MapQuest? App surges after defying Trump on Lake America renaming (2026): https://www.newsweek.com/lake-ontario-america-mapquest-donald-trump-name-map-12387309 --- # Does Hotmail Still Exist? Outlook.com Runs It Now URL: https://404memoryfound.com/posts/why-everyone-had-hotmail-account.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-23 Updated: 2026-09-13 Topics: Internet Culture, Software & Apps **Summary:** Hotmail still exists, but only as an address. Microsoft bought the service on December 31, 1997, launched Outlook.com in 2012 and retired the Hotmail name in May 2013, moving 300 million accounts across. Every @hotmail.com address still sends and receives mail in 2026, and new ones can still be created. **Key facts:** - Launched: July 4, 1996, by Sabeer Bhatia and Jack Smith - Owner today: Microsoft, since December 31, 1997 - Status today: Brand retired May 2013; @hotmail.com addresses still work on Outlook.com - Price today: Free, with ads in the inbox - Free storage: 15 GB mailbox plus 5 GB Microsoft cloud storage ## What did email look like before Hotmail? Before 1996, an email address was something you rented. It came bundled with whoever sold you your internet connection, and it vanished the day you switched providers. AOL subscribers wrote from @aol.com, CompuServe (https://404memoryfound.com/posts/what-happened-to-compuserve-online-service.html) members from a string of digits, students from a .edu account that expired at graduation. Mail also lived in exactly one place. You read it in a program installed on a single computer, configured for a single provider's servers. A friend's PC, a library terminal or an office machine could not reach your inbox at all. That is the problem Sabeer Bhatia and Jack Smith set out to solve. Both were engineers in Silicon Valley, and both were blocked by a workplace firewall from reaching their personal mail during the day. Their fix was to move the inbox into the web browser itself, so any machine with a browser and a connection could open it. The service launched on July 4, 1996, a date the founders chose to advertise independence from the internet providers. The name was a pun: HoTMaiL, with the letters of HTML buried inside it. Eudora (https://404memoryfound.com/posts/what-happened-to-eudora-email-client.html) and the other desktop mail clients suddenly had a rival that needed no installation and no setup wizard. ## How did a footer turn Hotmail into the world's largest email service? Hotmail had no advertising budget. Its investors at Draper Fisher Jurvetson proposed a substitute: attach a line of promotion to the bottom of every message a Hotmail user sent. The footer read "Get your free e-mail at Hotmail," and it rode along under every birthday note, every argument and every forwarded joke. It worked because the audience was ideal. Anyone reading a Hotmail message was already using email and already curious about the sender. One signup produced dozens of impressions, and those impressions produced more signups. Marketers later named the pattern viral marketing, and Hotmail is the case study that every telling of the story begins with. The numbers moved quickly. Hotmail passed one million subscribers in under six months. By early 1999 the service had more than 30 million users and was adding roughly 150,000 a day, according to TechSpot's history of it. Microsoft had already announced in December 1998 that Hotmail was the world's largest email provider and, by its own count, twice the size of the email base at AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html). ## Why did Microsoft buy Hotmail 18 months after launch? Microsoft announced the acquisition on December 31, 1997. Its press release said Hotmail was serving "more than 9 million members worldwide" and presented the deal as a way to bolt a mass-market inbox onto The Microsoft Network. "Hotmail has been a Web-mail pioneer," said Laura Jennings, vice president of The Microsoft Network, in that announcement. She described the appeal as "a free, high-quality e-mail service that lets its members access a permanent e-mail address from any PC with an Internet connection." Microsoft did not publish a price. The transaction was a stock swap reported at a value as high as $400 million, which made a company 18 months old one of the most expensive properties on the web at the time. What Microsoft bought was not really software. It was a list of names and a habit. An email address is among the hardest things on the internet to change, because changing it means telling everyone you know. Every Hotmail account was a person who would return to a Microsoft page, see Microsoft advertising, and later sign in to MSN Messenger (https://404memoryfound.com/posts/what-happened-to-msn-messenger.html) with the same credentials. A year after the purchase, Jennings put the logic plainly: "E-mail is the most popular online activity and an important part of our portal site MSN.com." ## How did Gmail end Hotmail's lead in 2004? Google opened Gmail to the public on April 1, 2004 with one gigabyte of free storage, more than 100 times what Hotmail and Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) Mail were offering, by TechSpot's accounting. The date was improbable enough that reporters checked whether the whole thing was a prank. Storage was the argument, and Hotmail lost it. Hotmail users spent real time deleting messages to stay under a cap measured in megabytes. Gmail asked them to stop deleting anything, then added conversation threading and Google's search on top. Hotmail answered within months by raising free storage to 250 MB and lifting attachment limits, and it kept raising the ceiling for years afterward. But answering is not the same as setting the terms, and from 2004 onward Hotmail was the incumbent rather than the innovator. The branding did not help. Microsoft renamed the service MSN Hotmail, then Windows Live Hotmail, finishing that rollout in 2007. Each change asked users to learn a new name for a product they had picked precisely because they never had to think about it. ## Who owns Hotmail today? Microsoft owns Hotmail, and has since December 1997. The name is what disappeared. Microsoft launched Outlook.com on July 31, 2012 and spent the months afterward moving Hotmail users onto it. The switch finished in the first days of May 2013. Microsoft said it had moved 300 million Hotmail accounts, roughly 150 petabytes of mail, in six weeks, and that Outlook.com then held more than 400 million active accounts, as GeekWire reported at the time. Nothing was deleted in the process. An @hotmail.com address sends and receives exactly as it did, it simply does so through Outlook.com. Mail addressed to a Hotmail account still arrives, the sign-in page says Microsoft, and the interface says Outlook. The domain survives as a suffix rather than as a product. So Hotmail was not shut down, sold off or discontinued, which is what most people typing the question expect to hear. It was absorbed. The company that bought it in 1997 still runs it in 2026, under a different name, on the same domain, for a user base far larger than the one it acquired. ## What does a free Hotmail address get you in 2026? A free Outlook.com account, whatever suffix it carries, comes with 15 GB of mailbox storage for mail, contacts and calendar items, plus 5 GB of Microsoft cloud storage for attachments and files, according to Microsoft's support documentation. The two quotas are separate, and the smaller one tends to fill first. The free tier carries advertising in the inbox, which is the same bargain Hotmail struck in 1996: you pay with attention rather than money. Microsoft sells an ad-free inbox and a larger mailbox as part of a Microsoft 365 subscription. New @hotmail.com addresses are still available. Microsoft's signup flow offers hotmail.com alongside outlook.com when you create a free account, so the 1990s suffix can still be claimed in 2026 if the name you want is unused. The one thing that cannot be recovered is an address you walked away from. Microsoft's support answers are consistent on this point: an address that was once attached to a Microsoft account cannot be handed to a new one, even long after that account was closed. The sk8erboi handle is either yours or gone. Email turned out to be the most permanent identity the early web produced, which is roughly what the search engines of that era (https://404memoryfound.com/posts/yahoo-rejected-google.html) spent the following decade discovering. ## Frequently Asked Questions ### Does Hotmail still exist in 2026? Hotmail still exists as an email domain, though not as a separate product. Microsoft retired the Hotmail brand in May 2013 and migrated every account to Outlook.com, so an @hotmail.com address created in the 1990s still sends and receives mail in 2026 through the Outlook.com interface and a Microsoft account sign-in. ### Can I still create a new @hotmail.com email address? Yes. Microsoft still offers hotmail.com as an option alongside outlook.com when you create a new free Microsoft account, so a new @hotmail.com address can be registered in 2026 if the name is unused. An address that once belonged to a closed Microsoft account, however, cannot be claimed again. ### Who founded Hotmail and when did Microsoft buy it? Hotmail was founded by Sabeer Bhatia and Jack Smith, who launched it on July 4, 1996. Microsoft announced the acquisition on December 31, 1997, when Hotmail reported more than 9 million members worldwide, in a stock transaction reported at a value as high as $400 million. **Sources:** - Microsoft, "Microsoft Acquires Hotmail" (December 31, 1997): https://news.microsoft.com/source/1997/12/31/microsoft-acquires-hotmail/ - Microsoft, "MSN Hotmail, World's Largest E-Mail Provider, Surpasses 30 Million Member Milestone" (December 1, 1998): https://news.microsoft.com/source/1998/12/01/msn-hotmail-worlds-largest-e-mail-provider-surpasses-30-million-member-milestone/ - TechSpot, "What Ever Happened to Hotmail?": https://www.techspot.com/article/2376-hotmail/ - GeekWire, "Hotmail is officially toast, but Microsoft says Outlook.com is on fire" (May 2, 2013): https://www.geekwire.com/2013/microsoft-400m-outlook-users-hotmail-migration-125m-mobile/ - Microsoft Support, "Storage limits in Outlook.com": https://support.microsoft.com/en-us/office/storage-limits-in-outlook-com-7ac99134-69e5-4619-ac0b-2d313bba5e9e --- # Does GeoCities Still Exist? Where the Pages Went URL: https://404memoryfound.com/posts/history-of-geocities-websites.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-23 Updated: 2026-09-13 Topics: Internet Culture, Then vs Now **Summary:** GeoCities no longer exists. Yahoo shut the service down on October 26, 2009 and deleted roughly 38 million member-built pages, and Yahoo Japan closed the last working version on March 31, 2019. The name now sits with Yahoo Inc., owned by funds managed by affiliates of Apollo Global Management since 2021, while the surviving pages live in the Internet Archive and a 641 GB Archive Team torrent. **Key facts:** - Launched: 1994 as Beverly Hills Internet, founded by David Bohnett and John Rezner - Sold to Yahoo: January 28, 1999, about $3.57 billion in stock - Shut down: October 26, 2009 in the US; March 31, 2019 in Japan - Status today: Closed. No GeoCities page has been served from geocities.com since 2009 - Owner today: Yahoo Inc., owned by funds managed by affiliates of Apollo Global Management since September 2021 ## How Beverly Hills Internet became a city of neighborhoods The service started in 1994 as Beverly Hills Internet, a small Los Angeles company run by David Bohnett and John Rezner. It handed out free web space and did one thing no other host bothered with: it gave every page an address inside a themed neighborhood. A science fiction page lived in Area 51. A celebrity fan page went to Hollywood. Programmers took SiliconValley, families took Heartland, and money pages went to WallStreet. The address told you what you were about to open, which mattered on a web that had no search engine worth using yet. The company took the GeoCities name and the model worked fast. It signed its millionth "homesteader," the company word for a member, in October 1997. By December 1998, Media Metrix counted more than 19 million unique visitors in a month and ranked the site the third most trafficked on the web, behind only AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html) and Yahoo. That growth came from members building pages for each other and linking to their neighbors, the same loop MySpace would run on a bigger scale (https://404memoryfound.com/posts/rise-and-fall-of-myspace.html) a few years later. ## Why the first FTC internet privacy case named GeoCities In August 1998, months before Yahoo made an offer, the company settled charges brought by the Federal Trade Commission. The agency said it had collected personal details through its membership form and passed them on while telling members the data was for something else. "GeoCities misled its customers, both children and adults, by not telling the truth about how it was using their personal information," said Jodie Bernstein, director of the FTC's Bureau of Consumer Protection, in the agency's 1998 announcement. The settlement carried no admission of wrongdoing. What it carried instead was a set of rules the rest of the web later inherited: post a clear privacy notice saying what is collected and who receives it, get parental consent before collecting anything from children 12 and under, and give members a way to have their details removed from third-party databases. It was the FTC's first internet privacy case. The privacy policy link at the bottom of almost every page in 2026 traces back to a consent order about a free homepage service. ## What Yahoo paid, and the protest that followed Yahoo agreed to buy the company on January 28, 1999 in an all-stock deal valued at about $3.57 billion. Yahoo exchanged 21,640,342 of its own shares for the GeoCities stock it did not already hold. TechCrunch, looking back in 2009, put the value at closing lower, at $2.87 billion, because Yahoo shares moved between announcement and completion. The trouble started with paperwork. Yahoo rewrote the terms of service in 1999 in language that appeared to hand it the right to reuse anything a member had posted, with no payment, credit or notice. Members answered with a boycott that became known as the Haunting. They stripped their own pages back to bare backgrounds, deleted the animated GIFs and the visitor counters, and replaced the content with excerpts of the offending terms. Whole neighborhoods went grey at once. Yahoo backed down inside a week and rewrote the terms to leave rights with the people who had made the pages. A user base had renegotiated a platform contract by refusing to use the platform, five years before Facebook existed. ## Who owns GeoCities today? The name belongs to Yahoo Inc., and Yahoo is owned by funds managed by affiliates of Apollo Global Management (https://404memoryfound.com/posts/who-owns-yahoo-now.html), which completed a roughly $5 billion purchase in September 2021. There is no product left for anyone to run. The brand is a dormant asset attached to a domain that serves no pages. Yahoo announced the shutdown on April 23, 2009 and stopped taking new members that day. Traffic had already collapsed: comScore counted 11.5 million US unique visitors in March 2009, down 24 percent from 15.1 million a year earlier. The service went dark on October 26, 2009, and roughly 38 million member-built pages went with it. Yahoo kept no copy. Members were told to download their own files if they wanted them, and old geocities.com addresses now bounce to a Yahoo page explaining that the service closed. One outpost lasted a decade longer. Yahoo Japan ran its own version until March 31, 2019, blaming profitability and system problems, and that closure ended the service as a live product anywhere in the world. ## How much of it did anyone manage to save? Two rescue efforts ran at the same time. The Internet Archive ran deep crawls from July to October 2009, including sites nominated by the public, and serves those captures through the Wayback Machine as its GeoCities Special Collection. Archive Team, a volunteer group co-founded by Jason Scott in 2009, ran its own harvest from April to October 2009 with dozens of people and hundreds of machine instances. In 2010 it published the result as a single 641 GB torrent, still the largest public block of this material anywhere. Scott did not soften his view of what Yahoo had done. He told Time in 2009 that the site had been "the largest self-created folk-art collection in the history of the world," and that Yahoo "found the way to destroy the most massive amount of history in the shortest amount of time with absolutely no recourse." Artists Olia Lialina and Dragan Espenschied then built something out of that torrent. Their project One Terabyte of Kilobyte Age began posting automated screenshots of rebuilt pages on February 7, 2013, one every twenty minutes, rendered in period-correct browsers. It has since been shown in galleries and treated as an art archive rather than a backup. ## What replaced it for people who want a homepage? The direct heir is Neocities, launched in 2013 by Kyle Drake with the stated goal of reviving the free web hosting that GeoCities used to provide. A free site gets 1 GB of storage and 200 GB of bandwidth, runs no server-side code, and by May 2026 the service had hosted more than 1,612,500 sites. The bigger answer is that the audience moved. Squarespace, Wix and WordPress.com absorbed the people who wanted a presentable page, and social platforms absorbed everyone who only wanted somewhere to post. Both traded the handmade look for templates. David Bohnett, who started the company, describes the loss as one of purpose rather than design. "GeoCities was not about self-promotion. It was about sharing your interest and your knowledge," he told Gizmodo in 2022. In the same interview he said what surprised him was "how far away we've gotten from that original intent." A few pages from that era are still served at their original addresses (https://404memoryfound.com/posts/90s-websites-still-online.html), though every one of them is hosted somewhere else. ## Frequently Asked Questions ### When did GeoCities shut down? Yahoo shut GeoCities down in the United States on October 26, 2009, six months after announcing the closure on April 23, 2009. Roughly 38 million member-built pages were deleted, and Yahoo kept no copy of them. Yahoo Japan ran a separate GeoCities until March 31, 2019, which was the last working version of the service anywhere. ### Can you still see old GeoCities pages? Yes. The Internet Archive crawled GeoCities from July to October 2009 and serves those captures through the Wayback Machine as its GeoCities Special Collection. Archive Team's parallel harvest, published in 2010 as a 641 GB torrent, holds more material again. Neither copy is complete, and the pages nobody ever linked to are the ones most likely missing. ### How much did Yahoo pay for GeoCities? Yahoo announced its purchase of GeoCities on January 28, 1999 at about $3.57 billion in stock, exchanging 21,640,342 Yahoo shares for the rest of the company. TechCrunch later put the value at closing at $2.87 billion, because Yahoo's share price moved between announcement and completion. Yahoo shut the service down ten years after buying it, in October 2009. **Sources:** - Internet Site Agrees to Settle FTC Charges of Deceptively Collecting Personal Information in Agency's First Internet Privacy Case, Federal Trade Commission, 1998: https://www.ftc.gov/news-events/news/press-releases/1998/08/internet-site-agrees-settle-ftc-charges-deceptively-collecting-personal-information-agencys-first - Yahoo Quietly Pulls The Plug On Geocities, TechCrunch, 2009: https://techcrunch.com/2009/04/23/yahoo-quietly-pulls-the-plug-on-geocities - Internet Atrocity! GeoCities' Demise Erases Web History, Time, 2009: https://time.com/archive/6906661/internet-atrocity-geocities-demise-erases-web-history/ - GeoCities Special Collection 2009, Internet Archive: https://archive.org/web/geocities.php - The Founder of GeoCities on What Killed the 'Old Internet', Gizmodo, 2022: https://gizmodo.com/interview-with-geocities-founder-on-the-new-web-tiktok-1849179509 --- # Is MSN Messenger Still Around? Yes, on Escargot URL: https://404memoryfound.com/posts/what-happened-to-msn-messenger.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-23 Updated: 2026-09-13 Topics: Software & Apps, Internet Culture **Summary:** MSN Messenger no longer exists as a Microsoft product. Microsoft switched it off through the spring of 2013 and moved everyone to Skype, then retired Skype itself on May 5, 2025 in favor of Microsoft Teams. The original clients still work in 2026, but only against Escargot, a free fan-run server that rebuilt the protocol Microsoft abandoned. **Key facts:** - Launched: July 22, 1999, as MSN Messenger Service - Peak: More than 330 million active users a month, June 2009 - Shut down: April 2013 worldwide; mainland China on October 31, 2014 - Owner today: Microsoft still owns the MSN name; Escargot is independent of it - Status today: Dead at Microsoft, playable for free on the fan server Escargot ## How Microsoft forced its way into a market AOL already owned MSN Messenger Service went live on July 22, 1999, about two years after AOL Instant Messenger (https://404memoryfound.com/posts/what-happened-to-aim-aol-instant-messenger.html) and nearly three years after ICQ (https://404memoryfound.com/posts/what-happened-to-icq-instant-messenger.html). Microsoft was late and priced accordingly: the service was free, and it signed you in with the Hotmail address you already had. "Communications continues to be the cornerstone of the Internet, and instant messaging is becoming a more prevalent way for people to communicate," said Brad Chase, then a Microsoft vice president, in the company's launch announcement in July 1999. The shortcut Microsoft chose was interoperability. Version 1.0 let you message AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html) Instant Messenger contacts from inside MSN Messenger, using AOL's servers, without an AOL account. AOL read that as theft of its network, changed its protocols and cut Microsoft off. Microsoft found a way back in. AOL blocked it again. The two traded workarounds for weeks before Microsoft gave up and removed the feature, which never returned in any later version. What AOL could not block was Windows. MSN Messenger did not have to be discovered, downloaded or recommended. It was already on the machine, in a decade when almost every machine ran Windows, and that alone carried it into countries where AOL had no presence at all. ## What made MSN Messenger feel different from email The software rewarded fiddling, which is why people remember it as a place rather than a tool. Your display name was a public text field you could rewrite ten times in an afternoon, and people filled it with song lyrics, private jokes and messages aimed at exactly one person who would know. Presence was the other half. "Appear Offline" made being unreachable a deliberate act that took one click and needed no explanation, and signing out at night was normal rather than a statement. Read receipts did not exist to argue with. The feature everyone quotes arrived later than memory suggests. Nudges and winks shipped with MSN Messenger 7.0 in April 2005, alongside Xbox Live integration and handwriting input you scrawled with a mouse. A nudge shook the recipient's whole chat window and buzzed, and it was used almost exclusively to annoy people. Messenger also sat in the middle of Microsoft's consumer stack. It raised Hotmail (https://404memoryfound.com/posts/why-everyone-had-hotmail-account.html) alerts, tied into the MSN Spaces blogging service and ran small games such as Minesweeper Flags inside the chat window. For a few years Microsoft had a working social network and treated it as a utility. ## Why did Microsoft shut MSN Messenger down? The name went first. In June 2006, with version 8.0, MSN Messenger became Windows Live Messenger, filing one of the most recognized brands on the internet under a corporate umbrella that meant nothing to the people using it. MSN Messenger 7.5 was the last release to carry the original name. Usage kept climbing regardless. In June 2009 Microsoft reported more than 330 million active users a month, the highest figure the service ever posted. Then the ground shifted twice. Facebook added chat in 2008, where the contact list was already built and the switching cost was zero. WhatsApp arrived in 2009 on phones. Messenger was designed for a desk you sat down at and a session you consciously started, and that ritual was disappearing. Microsoft had bought Skype in 2011 and chose to consolidate on it. On November 6, 2012, Skype division president Tony Bates put it plainly: "We will retire Messenger in all countries worldwide in the first quarter of 2013 (with the exception of mainland China where Messenger will continue to be available)." Messenger still had more than 100 million users at the time. Skype had 280 million, and Skype was the one Microsoft had just paid for. The migration ran language by language from April 8, 2013 and finished that month. Mainland China, where the service was operated through a local partnership, kept Messenger running until October 31, 2014. ## Can you still use MSN Messenger in 2026? Yes, on a fan server. Escargot is an independent project that reimplements the MSN Messenger and Windows Live Messenger protocols, so the original clients connect to it instead of Microsoft's switched-off servers. It costs nothing, and an account needs an email address and a username. The classic MSN Messenger releases work, and Windows Live Messenger 2009, version 14.0, is the newest client Escargot supports. Nudges, custom display pictures and the display-name theater all behave the way they did, because the client is the genuine article rather than a lookalike. What you do not get back is your contact list. Everyone you want to talk to has to sign up as well, which is the same problem Messenger solved in 1999 and the reason most people only visit. Microsoft is not involved and has announced no revival of its own. The MSN name survives as msn.com, a news portal with no messaging product attached to it. ## What replaced MSN Messenger, and what replaced that? Officially Skype replaced it. In practice most Messenger users never arrived. They went to Facebook Messenger, WhatsApp or iMessage, because that is where their friends already were, and Microsoft inherited a shrinking share of a market it had once led outright. The replacement did not outlast the thing it replaced by much. Microsoft announced on February 28, 2025 that Skype would shut down (https://404memoryfound.com/posts/skype-shut-down-what-replaced-it.html), and it did on May 5, 2025. "We will be retiring Skype in May 2025 to focus on Microsoft Teams (free)," wrote Jeff Teper, Microsoft's executive vice president for apps and agents, in the announcement. Skype credentials now sign users into Teams, with chats and contacts carried across. So the line runs MSN Messenger to Windows Live Messenger to Skype to Microsoft Teams, and the only survivor of the four is a work tool with a calendar in it. The thing people actually miss, the 2005 version with the buzzing windows and the song-lyric display names, exists in exactly one place, and Microsoft does not run it. ## Frequently Asked Questions ### When did MSN Messenger shut down? MSN Messenger, renamed Windows Live Messenger in 2006, was retired language by language starting on April 8, 2013, after Microsoft announced the decision on November 6, 2012. Mainland China was the one exception and kept the service until October 31, 2014. Every user was pushed to Skype, which Microsoft then shut down itself on May 5, 2025. ### How do you get MSN Messenger working again in 2026? Install an old MSN Messenger or Windows Live Messenger client and point it at Escargot, a free fan-run server that rebuilt the protocol Microsoft switched off in 2013. Escargot supports the classic MSN Messenger versions and Windows Live Messenger 2009, version 14.0, and signing up takes an email address and a username. Microsoft offers no official way to run MSN Messenger in 2026. ### Why did Microsoft replace MSN Messenger with Skype? Microsoft bought Skype in 2011 and decided to run one messaging product instead of two. When the retirement was announced in November 2012, MSN Messenger had more than 100 million users against Skype's 280 million, so Skype won the merge. Most Messenger users moved to Facebook Messenger or WhatsApp rather than Skype, and Microsoft retired Skype in turn on May 5, 2025. **Sources:** - Microsoft Launches MSN Messenger Service (Microsoft press release, July 1999): https://news.microsoft.com/source/1999/07/21/microsoft-launches-msn-messenger-service/ - MSN Messenger: version history, rename and shutdown dates (Wikipedia): https://en.wikipedia.org/wiki/MSN_Messenger - Confirmed: Microsoft retiring Messenger, shifting to Skype (GeekWire, November 6, 2012): https://www.geekwire.com/2012/confirmed-microsoft-retiring-messenger-skype/ - The next chapter: Moving from Skype to Microsoft Teams (Microsoft 365 Blog, February 28, 2025): https://www.microsoft.com/en-us/microsoft-365/blog/2025/02/28/the-next-chapter-moving-from-skype-to-microsoft-teams/ - About Escargot, the fan-run MSN Messenger server: https://escargot.chat/about/ --- # Was Apple Really 90 Days From Bankruptcy in 1997? URL: https://404memoryfound.com/posts/when-apple-almost-went-bankrupt-1997.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-23 Updated: 2026-09-13 Topics: Business Blunders, Hardware **Summary:** Apple lost $1.04 billion in the fiscal year that ended in September 1997, and Steve Jobs later said the company was "90 days from being insolvent." It survived on a $150 million Microsoft investment announced on August 6, 1997, a product line cut to four computers, and the iMac. Apple still exists as a public company and crossed $4 trillion in market value on October 28, 2025. **Key facts:** - Status today: Apple Inc. is public on Nasdaq as AAPL; market value crossed $4 trillion on October 28, 2025 - Owner today: Shareholders. Microsoft sold its entire Apple stake by 2003 - Worst year: Fiscal 1997: a net loss of $1.04 billion, after $816 million lost in fiscal 1996 - The lifeline: $150 million of non-voting preferred stock, bought by Microsoft on August 6, 1997 - Back in profit: Fiscal 1998: net income of $309 million, the first profitable year since 1995 ## How much money was Apple actually losing in 1996 and 1997? The numbers sit in Apple's own filings and they are worse than most people remember. Apple Computer lost $816 million in the fiscal year that ended in September 1996. It lost another $1.04 billion in fiscal 1997. Two years, roughly $1.86 billion gone. Share went with the cash. Apple's worldwide share of personal computer shipments fell to 3.6 percent in 1997 from 5.7 percent a year earlier, and its share in the United States fell to 4.6 percent from 7.4 percent. Gil Amelio, chief executive from February 1996, cut about 4,100 jobs in March 1997 to slow the burn. It was not enough. The stock sat at a 12-year low and the board pushed Amelio out on July 9, 1997. Steve Jobs put the position plainly years later. Apple was "90 days from being insolvent," he told Walter Isaacson for the 2011 biography Steve Jobs. That sentence is the source of almost every 90-days headline written since. ## Why did buying NeXT matter more than the software Apple paid for? On December 20, 1996, Apple announced it would buy NeXT Software for $429 million in cash plus 1.5 million Apple shares for its founder. Apple's own accounting later recorded the total purchase price at $427 million once assumed liabilities were counted. The stated reason was the operating system. Apple had spent years trying to replace the aging Mac OS with an in-house project called Copland and had cancelled it. NeXTSTEP was finished and shipping. It became the base of Mac OS X in 2001 and of every version of macOS and iOS since. The unstated reason was the man attached to it. "I'm not just buying software, I'm buying Steve," Amelio said of the deal. He was right, and it cost him the job seven months later. NeXT also supplied the executives who ran the recovery. Avie Tevanian took over software engineering at Apple and Jon Rubinstein took over hardware, and both stayed for the decade that produced the iMac and the iPod (https://404memoryfound.com/posts/ipod-wasnt-first-mp3-player-heres-why-it-won.html). ## What did Steve Jobs cut to stop the bleeding? Jobs became interim chief executive in September 1997 and started deleting products. The tool was a two-by-two grid on a whiteboard: consumer and professional down one axis, desktop and portable across the other. Four boxes, four computers. Anything outside the grid went. Apple ended the Mac clone licensing program that let Power Computing and Motorola sell cheaper Mac compatibles into Apple's own customer base. It killed the Newton MessagePad (https://404memoryfound.com/posts/what-happened-to-apple-newton.html) in February 1998. It had already watched the Bandai Pippin, the games console Apple licensed its technology to, sell roughly 42,000 units out of the 100,000 built before the machine was discontinued in 1997. The grid then filled in: the Power Macintosh G3 and the PowerBook G3 for professionals, the iMac and later the iBook for everyone else. Cutting is the part of this story that business writers like, and it was real. On its own it would not have met payroll. Apple needed cash before the new plan could pay for itself. ## What did Microsoft actually get for its $150 million? On August 6, 1997, at Macworld Expo in Boston, Jobs announced that Microsoft would buy $150 million of Apple non-voting preferred stock. Bill Gates appeared on a screen above the stage and the room booed. The money was the smallest part of the deal. Microsoft committed to keep building Microsoft Office for the Mac for five years, which mattered because the customers Apple had left were design and publishing shops that could not walk away from Office. Apple agreed to ship Internet Explorer as the default browser on the Mac, a useful prize while Microsoft was still fighting Netscape (https://404memoryfound.com/posts/how-netscape-lost-the-browser-war.html) for the web. The two also signed a broad patent cross-license and closed their long-running dispute over the Macintosh interface. Jobs talked the crowd down with a line he repeated for years: "We have to let go of this notion that for Apple to win, Microsoft has to lose." The timing suited Microsoft as well. The United States Department of Justice was building an antitrust case, and a Microsoft with a living competitor was an easier defendant than one without. ## Which product actually ended the crisis? Apple launched the "Think Different" campaign in September 1997: black and white portraits, Albert Einstein and Mahatma Gandhi and Martin Luther King Jr. among them, and not a computer in the frame. It was aimed at Apple's own staff as much as at buyers. The money came back with hardware. Jobs introduced the iMac G3 (https://404memoryfound.com/posts/how-imac-g3-saved-apple-killed-beige-box.html) in May 1998 and it reached stores that August: one translucent Bondi blue shell, a handle moulded into the top, no floppy drive, and a shape nothing in the beige-box market could answer. The result showed up in the next annual report. Apple reported net income of $309 million for fiscal 1998, against the $1.04 billion loss of the year before, and its first profitable year since 1995. Michael Dell had told a Gartner Symposium audience in October 1997 what he would do if he ran Apple: "I'd shut it down and give the money back to the shareholders." Twelve months later Apple was making money again. ## Who owns Apple today, and what is the company worth? Apple Inc. is a public company, traded on Nasdaq as AAPL and owned by its shareholders. No founder, family or parent company controls it, and the largest holders are index fund managers. It has never filed for bankruptcy protection. The scale no longer lines up with 1997 in any useful way. Apple reported revenue of about $416 billion for fiscal 2025, which is more than a billion dollars a day, and its market value crossed $4 trillion on October 28, 2025, the third company in history to reach that mark after Nvidia and Microsoft. Microsoft is not among the shareholders. It converted its preferred stock into common shares by 2001, sold the whole position by 2003 for a reported $550 million, and missed everything after that. "I wish we'd maintained that 5% ownership," Gates wrote in his 2025 memoir Source Code. The lifeline that kept Apple alive in 1997 is now smaller than the revenue the company books before lunch. ## Frequently Asked Questions ### How close was Apple to bankruptcy in 1997? Apple lost $1.04 billion in the fiscal year that ended in September 1997, on top of $816 million lost in fiscal 1996, and its stock was at a 12-year low. Steve Jobs later told biographer Walter Isaacson that Apple was "90 days from being insolvent" at that point. Apple never actually filed for bankruptcy protection. ### Why did Microsoft invest $150 million in Apple? Microsoft bought $150 million of Apple non-voting preferred stock on August 6, 1997, and in the same agreement settled a long patent dispute, won a commitment that Internet Explorer would be the default browser on the Mac, and promised five more years of Microsoft Office for the Mac. A Microsoft facing a United States antitrust case in 1997 also had good reason to keep Apple alive as a visible competitor. ### What did Steve Jobs do first when he returned to Apple in 1997? After Apple bought NeXT in December 1996 and Gil Amelio was pushed out in July 1997, Steve Jobs became interim chief executive that September and cut the product line to four computers using a consumer-versus-professional, desktop-versus-portable grid. He also ended Mac clone licensing, killed the Newton MessagePad in February 1998, and secured the Microsoft investment announced in August 1997. **Sources:** - Apple Computer Inc., Form 10-K405 for fiscal 1998, SEC EDGAR: https://www.sec.gov/Archives/edgar/data/0000320193/000104746998044981/0001047469-98-044981.txt - Microsoft and Apple Affirm Commitment to Build Next Generation Software for Macintosh, 6 August 1997: https://news.microsoft.com/source/1997/08/06/microsoft-and-apple-affirm-commitment-to-build-next-generation-software-for-macintosh/ - Computer History Museum: NeXT, Steve Jobs' dot com IPO that never happened: https://computerhistory.org/blog/next-steve-jobs-dot-com-ipo-that-never-happened/ - CNBC: Apple crosses $4 trillion market cap for the first time, 28 October 2025: https://www.cnbc.com/2025/10/28/apple-microsoft-4-trillion-market-cap.html - Apple Newsroom: Apple reports fourth quarter results, October 2025: https://www.apple.com/newsroom/2025/10/apple-reports-fourth-quarter-results/ --- # Is AIM Still Around? AOL Instant Messenger in 2026 URL: https://404memoryfound.com/posts/what-happened-to-aim-aol-instant-messenger.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-22 Updated: 2026-09-11 Topics: Software & Apps, Internet Culture **Summary:** AOL Instant Messenger, the chat client almost everyone called AIM, no longer exists. Oath, the Verizon unit that owned AOL, closed the service on December 15, 2017 after 20 years and never launched a replacement. The AOL brand behind it has changed hands twice since and has belonged to the Italian software company Bending Spoons since January 2, 2026. **Key facts:** - Launched: May 1997, by America Online - Status today: Shut down. No official service since December 15, 2017 - Owner today: AOL brand held by Bending Spoons, since January 2, 2026 - Peak users: 36 million active users in 2001, 63 million by 2007 - What replaced it: Phone texting, Facebook messaging, later Discord and iMessage ## Who actually built AOL Instant Messenger? AOL Instant Messenger came off AOL's engineering floor rather than out of a product plan. Barry Appelman, an AOL engineer who had worked on early TCP/IP standards at IBM, built the Buddy List, the panel that showed which of your friends were signed on at that moment. "Buddy List was done without telling anybody, because we didn't have any product management then," Appelman told Mashable in 2014. "So I just decided to do it." His first version broke often enough that colleagues called it the Buggy List. By 1997 he had it working and filed a US patent for what the paperwork called "User definable on-line co-user lists." Strip the legal phrasing and that is the buddy list, and every contact panel in every messaging app since has worked the same way: a list of people, each one marked present or absent. AOL released AOL Instant Messenger in May 1997. The decision that mattered was giving the client away to anyone with an internet connection instead of only to paying AOL subscribers. A teenager on a rival dial-up service could download it, pick a screen name and talk to AOL members for free. AIM grew fastest outside AOL's own walls, which is precisely what AOL's subscription business was not built to reward. ## How big did AIM get at its peak? AIM peaked as a desktop product. TechSpot puts the high point at 36 million active users in 2001, and 63 million by 2007. Those are large numbers for software that required both people to be sitting at a computer, signed in, at the same moment. The culture around it grew faster than the feature list. Screen names worked as pseudonyms years before real-name social networks arrived. Away messages were status updates before the phrase existed: song lyrics, inside jokes, a note that dinner was happening. The buddy icon was a profile picture. Sorting buddies into groups was the first social graph most American teenagers ever edited by hand. AIM also normalized the small signals that chat apps still use. Presence, typing indicators, one-to-one windows stacked across a desktop, file transfers between friends. MSN Messenger (https://404memoryfound.com/posts/what-happened-to-msn-messenger.html), Yahoo (https://404memoryfound.com/posts/does-yahoo-messenger-still-exist.html) Messenger and ICQ all competed hard for the same users, and none of them was close to a monopoly. In American schools through the early 2000s, though, the default question was which screen name you used. ## Is AIM still in business in 2026? No. AOL Instant Messenger has not run since December 15, 2017. Oath, the Verizon unit that then owned both AOL and Yahoo, announced the shutdown on October 6, 2017 and gave users about ten weeks of notice. When the servers went dark, stored buddy lists and account data went with them. "AIM tapped into new digital technologies and ignited a cultural shift, but the way in which we communicate with each other has profoundly changed," wrote Michael Albers, Oath's vice president of communications product, in the announcement Variety reported that week. The brand behind it kept moving. Verizon later sold the AOL and Yahoo businesses to Apollo Global Management, and AOL still exists (https://404memoryfound.com/posts/does-aol-still-exist-today.html) as a going concern. On January 2, 2026, the Italian software company Bending Spoons completed its purchase of AOL Holdco I LLC for about $1.45 billion, a transaction recorded in the Form F-1 registration statement it filed with the US Securities and Exchange Commission. That filing describes AOL as an email service, news portal and search engine with roughly 8 million daily and 30 million monthly active users. None of them is AIM. The trademark now sits inside a company that sells email subscriptions. ## What actually killed AIM? Three pressures arrived at once. The iPhone shipped in 2007, and the generation coming up behind AIM was going to communicate online mainly from a phone. Once American teenagers had texting on a handset, coming home and signing on stopped being the way you reached anybody. Social networks then absorbed the function. Facebook and MySpace (https://404memoryfound.com/posts/rise-and-fall-of-myspace.html) put messaging inside the place people already spent their evenings, so a separate client and a separate screen name became an extra step rather than the point. AIM shipped mobile apps, but it was answering a question the market had already moved past. AOL's own attention was the third pressure. The company spent the 2000s inside the AOL and Time Warner merger (https://404memoryfound.com/posts/aol-time-warner-merger-worst-deal-in-history.html) and the long unwinding that followed, and its revenue came from dial-up subscriptions rather than from a free chat client whose biggest gains came from people who were not customers. AIM was strategically awkward at AOL for its entire life. The product with 63 million users in 2007 was not the product AOL was organized to defend. ## Can you still log in to an AIM screen name today? Not the original one. AOL deleted the service and its account data in December 2017, so an old screen name and password open nothing. What exists instead are independent servers that speak OSCAR, the protocol the classic AIM clients used. AIM Phoenix, run by the hobbyist group Wildman Productions, operates a private network that period AIM clients can connect to once you register a fresh account. A developer publishing as mk6i has since released Open OSCAR Server, an MIT-licensed reimplementation written in Go that anyone can host and that decades-old Windows clients connect to without modification. These are reconstructions, not a restoration. The project's own repository describes it as "Independently developed, not affiliated with or endorsed by AOL or Yahoo! Inc." and entirely non-commercial. Nobody's buddy list is waiting on these servers unless two people agree to install a 2002 client on purpose. What they preserve is the behavior: the sign-on sound, the window stack, the away message sitting there all afternoon. ICQ (https://404memoryfound.com/posts/what-happened-to-icq-instant-messenger.html) fans have the same option, because the same protocol family served both services. ## Frequently Asked Questions ### When did AIM shut down? AOL Instant Messenger shut down on December 15, 2017, after 20 years of service. Oath, the Verizon division that owned AOL at the time, announced the closure on October 6, 2017, and stored buddy lists and account data were deleted when the servers went off. ### Can you still use AIM in 2026? There is no official AOL Instant Messenger service in 2026 and there has not been one since December 2017. Enthusiast networks such as AIM Phoenix, and the open-source Open OSCAR Server that people host themselves, let classic AIM clients sign in again, but they need a new account and none of them is run by AOL. ### Why did AOL Instant Messenger fail? AOL Instant Messenger lost its users to phones and social networks. The iPhone arrived in 2007, texting and in-platform messaging took over the habit of signing on at a desktop, and AOL, whose money came from subscriptions, never turned its free chat client into a business before closing it in 2017. **Sources:** - Variety: RIP AIM: AOL Instant Messenger Is Shutting Down December 15 (2017): https://variety.com/2017/digital/news/rip-aim-1202582611/ - Smithsonian Magazine: In the 25 Years Since Its Launch, AOL Instant Messenger Has Never Been 'Away': https://www.smithsonianmag.com/innovation/in-25-years-since-its-launch-aol-instant-messenger-has-never-been-away-180980086/ - TechSpot: What Ever Happened to AIM?: https://www.techspot.com/article/2184-aim/ - Bending Spoons S.p.A., Form F-1 registration statement (U.S. SEC): https://www.sec.gov/Archives/edgar/data/0002004711/000110465926071170/tm2613674-7_f1.htm - Open OSCAR Server: AIM and ICQ compatible server project (GitHub): https://github.com/mk6i/open-oscar-server --- # Why Netscape Navigator Lost to Internet Explorer URL: https://404memoryfound.com/posts/how-netscape-lost-the-browser-war.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-22 Updated: 2026-09-12 Topics: Software & Apps, Business Blunders **Summary:** Netscape Navigator lost the browser war because Microsoft made Internet Explorer free to everyone and shipped it inside Windows, which ran on more than 90 percent of new PCs. Navigator still led by roughly 51 percent to 40 percent at the end of 1997, was overtaken during 1998, and support for Netscape browsers ended on March 1, 2008. Internet Explorer outlived it but was itself retired on June 15, 2022. **Key facts:** - Launched: Netscape Navigator, December 1994, Netscape Communications - Peak usage: 70 percent share, cited in Bill Gates' memo of May 26, 1995 - Status today: Discontinued; AOL ended Netscape browser support March 1, 2008 - Owner today: The NETSCAPE trademark (US Reg. 2027552) is held by Yahoo Inc. - What replaced it: Firefox 1.0, built from Netscape's code, shipped November 9, 2004 ## What made Netscape Navigator the default browser of 1995? Netscape Communications shipped Netscape Navigator in December 1994, months after Marc Andreessen left the University of Illinois team behind NCSA Mosaic (https://404memoryfound.com/posts/what-happened-to-ncsa-mosaic-browser.html). Navigator drew images inline while a page was still loading, which made the web feel like something to use rather than something to study. Netscape gave the browser away to individuals and educators and charged businesses for it, so the installed base grew far faster than the revenue. That trade looked brilliant on August 9, 1995. The underwriters priced the IPO at $28 a share, buy orders were heavy enough that trading could not open for nearly two hours, and the first print came at $71. The stock touched $74.75 and closed at $58.25, which gave a 16-month-old company with no profits a market value of about $2.9 billion. Andreessen was 24 years old. Netscape did not want to sell a browser so much as become a platform, a layer sitting above Windows that would reduce the operating system to a detail. Saying so out loud is what turned a software company into a target. ## How did Bill Gates decide to go after Netscape? On May 26, 1995, Gates sent Microsoft's executive staff a memo titled "The Internet Tidal Wave," later entered as a government exhibit in United States v. Microsoft. It named the problem in one line: "a new competitor 'born' on the Internet is Netscape. Their browser is dominant, with 70% usage share, allowing them to determine which network extensions will catch on." Microsoft licensed Mosaic code from Spyglass and shipped Internet Explorer 1.0 in August 1995 inside the Windows 95 Plus! pack. It was slow and thin, and almost nobody used it. Internet Explorer 2.0 was better and still lost. Internet Explorer 3.0, released in August 1996, was the first version that competed on features, with style sheets, ActiveX controls, Java applets and JScript. The decision that mattered was not technical. Netscape charged companies for Navigator. Microsoft made Internet Explorer free to every user, business licences included, and never charged again. One product carried a price and the other did not, and the price was the line that paid Navigator's engineers. ## Why did bundling Internet Explorer with Windows work? Windows ran on more than 90 percent of new personal computers through the second half of the 1990s. A browser that shipped inside Windows therefore reached the buyer before any download could. Microsoft folded Internet Explorer into Windows 95 service releases and then into Windows 98, which arrived in June 1998 with the browser wired into the shell itself. Contracts finished the job. Microsoft signed computer manufacturers, internet service providers and content companies to carry or promote Internet Explorer, and the Justice Department later established that manufacturers who pushed Navigator risked their Windows licensing terms. AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html), at the time the largest online service in America, agreed in 1996 to build its client on Internet Explorer instead of Navigator in exchange for a position on the Windows desktop. The share numbers turned inside two years. At the end of 1997 Navigator still held roughly 51 percent of the market against roughly 40 percent for Internet Explorer. On October 1, 1998, The Washington Post printed the result as a headline: Microsoft's web browser had overtaken Netscape's. Navigator never led again. ## What did the court say Microsoft actually did? The United States sued Microsoft in May 1998. On November 5, 1999, Judge Thomas Penfield Jackson issued 207 pages of findings of fact and wrote that "viewed together, three main facts indicate that Microsoft enjoys monopoly power," citing a share of Intel-compatible PC operating systems that was extremely large and stable and protected by a high barrier to entry. The findings held that Microsoft had used that power against a list of threats that included Netscape, Java, Apple and RealNetworks, and that internal messages showed manufacturers being pressured over which browser they promoted. On April 3, 2000, Jackson ruled that Microsoft had violated Sections 1 and 2 of the Sherman Act. On June 7, 2000, he ordered the company split in two, an operating-system business and everything else. None of it reached Navigator in time. The D.C. Circuit upheld the monopoly-maintenance finding in June 2001, threw out the breakup and removed Jackson from the case, and the settlement that followed in November 2001 imposed rules on conduct rather than structure. The browser the case was about had already lost. ## Who owns Netscape Navigator today? Nobody sells it or updates it. AOL announced the purchase of Netscape Communications on November 24, 1998 in a stock deal valued at $4.2 billion on the day, and closed it on March 17, 1999, by which point AOL's higher share price put the number nearer $10 billion. AOL was buying the Netcenter portal and the server business more than the browser. Netscape-branded releases continued as far as version 9, and AOL ended all support on March 1, 2008. The trademark outlived the software. US registration 2027552 for NETSCAPE was filed by Netscape Communications Corporation and has since passed to AOL Inc., then Oath Inc., then Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) Inc., which holds it now. For what survives of the name and the portal, see whether Netscape still exists (https://404memoryfound.com/posts/is-netscape-still-around.html). The code did better than the company. On March 31, 1998, months before the AOL deal, Netscape published the Communicator source under an open licence as the Mozilla project. The team eventually threw out that codebase and rebuilt from scratch, which produced Firefox 1.0 on November 9, 2004. ## What is left of the first browser war in 2026? Internet Explorer is gone as well. Microsoft retired Internet Explorer 11 on June 15, 2022, writing that "after 25+ years of helping people use and experience the web, Internet Explorer (IE) is officially retired and out of support," then disabled it through a Microsoft Edge update on supported builds of Windows 10. Its replacement tells the story. Microsoft Edge is built on Chromium, the open-source engine behind Google Chrome, and carries an Internet Explorer mode for old corporate web applications that Microsoft has said it will support through at least 2029. The company that won by controlling the engine now ships someone else's. Firefox is the only widely used browser still shipping an engine descended from the code Netscape released in 1998. The rest of Netscape's work is everywhere and nobody notices it: JavaScript, SSL and the HTTP cookie all came out of the company that lost, and every browser opening this page implements all three. ## Frequently Asked Questions ### Who won the first browser war? Microsoft won the first browser war. Internet Explorer overtook Netscape Navigator in usage during 1998, having trailed it by roughly 40 percent to 51 percent at the end of 1997, and Navigator never led again. AOL bought Netscape Communications in 1998 and ended support for Netscape browsers on March 1, 2008. ### Did Microsoft break the law to beat Netscape Navigator? A federal court said it did. On November 5, 1999, Judge Thomas Penfield Jackson found that Microsoft held monopoly power in PC operating systems, and on April 3, 2000, he ruled that it had violated the Sherman Act, in part by tying Internet Explorer to Windows to cut Netscape Navigator's share. The D.C. Circuit upheld the monopoly-maintenance finding in June 2001 but cancelled the ordered breakup, and Microsoft settled in November 2001. ### Is Internet Explorer still available in 2026? No. Microsoft retired Internet Explorer 11 on June 15, 2022 and then disabled the desktop application through a Microsoft Edge update on supported versions of Windows 10. The only way to open an Internet Explorer page in 2026 is Internet Explorer mode inside Microsoft Edge, which Microsoft has said it will support through at least 2029. **Sources:** - Court's Findings of Fact, United States v. Microsoft Corporation, November 5, 1999 (US Department of Justice): https://www.justice.gov/atr/us-v-microsoft-courts-findings-fact - Buyers Drive Up Netscape Stock Price, The Washington Post, August 10, 1995: https://www.washingtonpost.com/archive/business/1995/08/10/buyers-drive-up-netscape-stock-price/1c05dc83-3328-4a4a-97b5-e6ecf737f83f/ - Microsoft's Web Browser Overtakes Netscape's, The Washington Post, October 1, 1998: https://www.washingtonpost.com/archive/business/1998/10/01/microsofts-web-browser-overtakes-netscapes/385a60e8-3e7e-43b6-9700-8710f178278c/ - Internet Explorer 11 has retired and is officially out of support, Microsoft Windows Experience Blog, June 15, 2022: https://blogs.windows.com/windowsexperience/2022/06/15/internet-explorer-11-has-retired-and-is-officially-out-of-support-what-you-need-to-know/ - NETSCAPE, US Trademark Registration No. 2027552, USPTO record: https://trademarks.justia.com/745/74/netscape-74574057.html --- # Windows XP Bliss Wallpaper: Where the Hill Is Now URL: https://404memoryfound.com/posts/windows-xp-wallpaper-bliss-story.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-22 Updated: 2026-09-12 Topics: Software & Apps, Internet Culture **Summary:** The Windows XP Bliss wallpaper is an unretouched photograph of a real hillside, shot by Charles O'Rear in January 1996 off Fremont Drive between Napa and Sonoma, California. The hill still exists on private land, but it has been planted with grapevines since at least 2006. It went bare and green again for a few days in February 2026, and a passing driver photographed it. **Key facts:** - Photographer: Charles O'Rear, National Geographic 1971 to 1995 - Shot: January 1996, Mamiya RZ67 medium format, Fujifilm Velvia - Location: Private hillside off Fremont Drive, between Napa and Sonoma, California - Status today: Vineyard land; briefly bare and green again on 18 February 2026 - Windows XP: Launched 25 October 2001, support ended 8 April 2014 ## Who took the Windows XP Bliss wallpaper photo? Charles O'Rear shot Bliss in January 1996. He had spent 1971 to 1995 on assignment for National Geographic, and by the middle of the decade he was living in wine country and photographing it. The picture was not an assignment. O'Rear was driving Fremont Drive, the stretch of California State Route 12 and 121 that links Napa and Sonoma, when the hillside came up green under a broken sky. He pulled over and shot it on a Mamiya RZ67, a medium format film camera, loaded with Fujifilm Velvia. He has never claimed more credit than that. "I didn't create this. I just happened to be there at the right moment and documented it," O'Rear said of the frame in Amateur Photographer's account of the shot. He told the Napa Valley Register in 2010 that he had watched that field for years and knew what January could do to it: "I always carry a camera and, one day, it happened." Microsoft named the image Bliss. O'Rear's own title for it was Bucolic Green Hills. ## Why was the grass in Bliss so green? Two accidents lined up. The hillside had carried grapevines, and those vines had been torn out after a phylloxera infestation, the root louse that forced growers across Napa and Sonoma to replant through the late 1980s and 1990s. Bare ground plus a wet Northern California January produced a slope covered in grass and nothing else. The second accident was the film. Fujifilm Velvia is a transparency stock built to saturate greens and blues, and a medium format frame holds detail that 35mm film would have thrown away. Whenever readers insist the picture must be computer generated, O'Rear points at that combination of camera and stock rather than at any software. That is not the same as saying the file on your old desktop is untouched. O'Rear did not retouch the original, but Microsoft cropped the frame and pushed the saturation further for the version that shipped with Windows XP. The wallpaper is a tighter, brighter edit of what he handed over. Nothing in the scene is fake, and nothing in it is rare either. It is a hill in a farming county on a good morning, and the strangeness is only that a billion office workers ended up staring at it. ## How much did Microsoft pay Charles O'Rear? The exact number has never been published. O'Rear signed a confidentiality agreement with Microsoft and has kept to it. What he has said is that the payment was the second largest sum ever paid to a living photographer for a single image, and reporting on the deal puts it in the low six figures. PetaPixel, writing in 2017, gave the figure as more than $100,000. The contrast with the rest of the Windows XP set is the part photographers remember. Peter Burian shot Autumn, another XP default background, in October 1999 while testing lenses, filed it with Corbis as a royalty free image, and took home $45. Getting the picture to Microsoft was its own problem. O'Rear had to hand over the original transparency, the physical piece of film, and couriers turned the job down once they understood the slide was worth more than their insurance would cover. Microsoft bought him a plane ticket and he carried it on board himself. The image had reached Microsoft through Corbis, the stock library owned by Bill Gates, which had absorbed O'Rear's agency. Microsoft acquired full rights in 2000, the year before Windows XP shipped. ## What happened to the hill after 1996? It went back to being farmland. The slope sits on private property off Fremont Drive in Sonoma County, and the vines that phylloxera had killed were replanted. There is no plaque, no pull-in and no sign telling drivers what they are passing. The artist duo Goldin+Senneby visited the site with O'Rear in November 2006 and re-photographed the same view for a work called After Microsoft, first shown in Paris in 2007. Their version is rows of grapevines under a flat grey sky. Set the two frames side by side and the point makes itself. Then came February 2026. A driver passing on 18 February found the hill bare and green again, the vines gone and the winter rain doing what it had done thirty years earlier, and posted a phone snapshot to Reddit. Tom's Hardware described the alignment as a "super rare" event, which is fair. For most of the past twenty years the hill has been either under vines or browned off by drought, and the window in which it resembles the wallpaper runs for a couple of weeks at most. So the honest answer to whether you can go and look at Bliss is that the hill is there, it belongs to somebody, it is usually a vineyard, and about once a generation it agrees to pose again. ## Is the Bliss wallpaper still around in 2026? Windows XP shipped on 25 October 2001 and Microsoft ended support on 8 April 2014. StatCounter's 2026 figures put Windows XP under 0.2 percent of Windows desktops, which is a rounding error as a share and still a real number of tills, lab benches and factory terminals. The wallpaper has outlived the operating system by a wide margin. Bliss works as a meme template, a Minecraft build, a gallery wall text and a T-shirt, and its thirtieth anniversary in January 2026 was covered as news, including by ABC News in Australia in March 2026. Microsoft has never repeated the trick, and the reason is structural rather than artistic. Windows 11 rotates a fresh Bing photograph onto the lock screen every day, macOS ships a new wallpaper with each release, and every phone owner picks their own. No single default image gets a decade of undivided attention now. The last Microsoft artifacts that reached that kind of saturation were Brian Eno's Windows 95 startup sound (https://404memoryfound.com/posts/windows-95-startup-sound-brian-eno.html), the screensavers that ran on every idle office monitor (https://404memoryfound.com/posts/what-happened-to-screensavers-flying-toasters.html) and Clippy (https://404memoryfound.com/posts/what-happened-to-clippy-microsoft.html), and all three belong to the same closed era. O'Rear has stopped working. He and his wife moved to Brevard, North Carolina in 2017, and he has been retired from photography since. ## Frequently Asked Questions ### Is the Windows XP wallpaper a real photo? Yes. The Windows XP wallpaper, named Bliss by Microsoft, is a film photograph taken by Charles O'Rear in January 1996 on a Mamiya RZ67 medium format camera loaded with Fujifilm Velvia. O'Rear did not retouch the original frame; the saturated green came from the film stock and from a hillside stripped of grapevines after a phylloxera infestation. Microsoft did crop the image and raise its saturation before shipping it in 2001. ### Where is the Windows XP Bliss hill located? The Bliss hill stands on private land off Fremont Drive, the run of California State Route 12 and 121 between Napa and Sonoma in Sonoma County, California. Charles O'Rear photographed it there in January 1996. The site is a working vineyard with no marker or viewing area, so there is nothing to visit beyond a roadside view of somebody else's grapes. ### How much was Charles O'Rear paid for the Windows XP wallpaper? Charles O'Rear has never disclosed what Microsoft paid him for Bliss, because he signed a confidentiality agreement when the company bought full rights in 2000. He has said it was the second largest sum ever paid to a living photographer for one image, and PetaPixel reported in 2017 that the figure was above $100,000. By comparison, Peter Burian earned $45 for Autumn, another Windows XP default wallpaper. **Sources:** - Amateur Photographer: 25 years later, Bliss by Charles O'Rear: https://amateurphotographer.com/iconic-images/bliss-by-charles-orear-iconic-photograph/ - Napa Valley Register: Say goodbye to 'Bliss': https://napavalleyregister.com/star/lifestyles/say-goodbye-to-bliss/article_2c485132-b504-11e3-85ef-0019bb2963f4.html - PetaPixel: Microsoft Paid 'Bliss' Photog $100K+ and 'Autumn' Photog $45: https://petapixel.com/2017/12/23/microsoft-xp-bliss-photog-paid-100k-autumn-photog-paid-45/ - ABC News: Microsoft's default wallpaper Bliss turns 30 this year: https://www.abc.net.au/news/2026-03-23/microsoft-default-wallpaper-bliss-most-viewed-photo-in-history/106472702 - Tom's Hardware: 30 years later, the Bliss green hill from Windows XP: https://www.tomshardware.com/software/windows/30-years-later-the-iconic-green-hill-from-windows-xps-bliss-wallpaper-is-still-thriving-new-picture-captures-the-hill-looking-almost-identical-to-the-original --- # Y2K Explained: What Really Broke on January 1, 2000 URL: https://404memoryfound.com/posts/y2k-bug-what-actually-happened.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-19 Updated: 2026-09-09 Topics: Software & Apps, Internet Culture, Money & Tech **Summary:** The Y2K bug was a genuine defect in decades of two-digit date code, and most of it was found and repaired before January 1, 2000, at a cost the US Commerce Department put at roughly $100 billion in the United States alone. Enough of it survived to prove the risk was real, including a spy satellite ground station that went dark for hours, 800 slot machines in Delaware that read the date as 1900, and $50 million in Medicare payments delayed by a day. **Key facts:** - The bug: Years stored as two digits, so 00 read as 1900 instead of 2000 - US cost: About $100 billion for repairs from 1995 to 2001 (Commerce Department, 1999) - Federal share: $8.34 billion, plus $3.35 billion in emergency supplemental funds - Worst US failure: A spy satellite ground station at Fort Belvoir, offline for hours on December 31, 1999 - Status today: Fixed, though windowing patches and the 2038 Unix rollover are still out there ## Why did two digits nearly break the world's computers? Y2K was a storage habit that outlived its excuse. When memory and disk space were expensive, programmers wrote the year as two digits, so 1971 became 71 and 1995 became 95. Every record in every file saved two bytes, and almost nobody writing that code in the 1960s and 1970s expected it to still be running at the end of the century. The failure begins the moment a machine reads 00 and decides it means 1900. Ages come back as 100 instead of 0. Interest accrues in the wrong direction. Lease dates, prescription dates and expiration dates land a century early. Sorting collapses, because 00 files ahead of 99. By the late 1990s that habit sat inside payroll systems, billing systems, medical devices, factory controllers, and the consumer software boom that followed the Windows 95 launch (https://404memoryfound.com/posts/the-night-windows-95-launched.html), all the way down to the desktop spreadsheet era that Lotus 1-2-3 (https://404memoryfound.com/posts/what-happened-to-lotus-1-2-3-spreadsheet.html) helped create. There was no master list of affected systems. Finding the bug meant reading the code, line by line. Embedded chips were the second fear, and that one shrank on inspection. The Senate Special Committee on the Year 2000 Technology Problem reported that testing in the last quarter of 1999 pointed to an embedded chip failure rate of 0.001 percent, against the 2 to 3 percent projected in late 1998 and early 1999. ## Did anything actually break on January 1, 2000? Yes, and the most serious American failure was military. A ground station at Fort Belvoir, Virginia, that processes imagery from National Reconnaissance Office satellites failed at 7 p.m. Eastern on December 31, 1999, which was midnight GMT. Deputy Defense Secretary John Hamre said the outage ran a matter of hours before backup procedures restored partial processing, and the Pentagon, which had spent $3.6 billion on Y2K work, called it its most significant glitch. The rest of the list is small, specific and oddly domestic. The Senate committee's final report catalogs 800 slot machines in Delaware that shut down after reading the date as January 1, 1900. It records $50 million in Medicare payments delayed a day by a date problem at a bank handling the electronic transfer, and one Medicare contractor that received about 11,000 claims stamped 1900 or 2099. Seven nuclear power plants reported minor glitches in non-safety systems, all fixed quickly. Two Federal Aviation Administration systems tripped, the Low Level Wind Shear Alert System at eight sites and the Kavouras Graphic Weather Display System. A water utility in Akron, Ohio sent shutoff notices to 3,000 customers who did not deserve them. A Godiva store in New York lost its systems, registers included, and had them back inside three hours. In Champaign, Illinois, 8,500 phone subscribers opened January statements that appeared to be a century old. The committee's summary was blunt about scale. Hundreds of computer problems were reported after January 1, most were corrected quickly, and none caused serious disruptions. ## What did fixing the Y2K bug cost? The Commerce Department put US spending at about $100 billion in November 1999, covering repairs from 1995 through 2001, which worked out to roughly $365 for every American. Commerce Secretary William Daley defended the bill by saying "the potential cost of not doing anything was far greater." Set against what a 1999 dollar was worth (https://404memoryfound.com/posts/thousand-dollars-then-now.html), it remains one of the largest maintenance projects ever run on civilian software. The federal slice was comparatively modest. The Office of Management and Budget counted $8.34 billion in federal Y2K spending, and Congress added $3.35 billion in emergency supplemental funding, $1.1 billion of it for the Defense Department and $2.25 billion for civilian agencies. Bigger numbers were floating around, and they never converged. Gartner Group estimated US spending at $150 billion to $225 billion. International Data Corporation's Project Magellan put the worldwide figure at $320 billion, with $134 billion of it spent in the United States. A Newsweek journalist estimated global spending at $500 billion. One common rule of thumb priced remediation at a dollar per line of code, and the Senate committee concluded plainly that the worldwide cost may never be known. ## Was Y2K a hoax? No, and the people best placed to call it one did the opposite. The Senate committee that spent two years investigating the problem judged that the level of effort was justified and that public and private spending was necessary, because testimony and research through 1998 and early 1999 convinced it that the threat was real and the consequences of inaction too severe to gamble on. The hoax reading also leans on a number that does not exist. There is no reliable global tally of what broke, because, as the committee put it, no company or country has any incentive to publicize its computer failures. Thirty-two countries, including Australia, Brazil, Great Britain, Canada, Germany and Norway, reported zero incidents to the International Y2K Cooperation Center while news services were reporting incidents inside those same countries. What that leaves is a quiet January and an unprovable counterfactual. Every documented failure of January 2000 happened in a system that had not been remediated, which is evidence for the fixes rather than against them, but nobody can rerun 2000 with the money unspent. ## Why did countries that spent almost nothing come out fine? This is the strongest argument the skeptics have, and the answer is mostly arithmetic. Writing in January 2000, Slate noted that Italy and Paraguay spent far less than the United States and fared about as well, then pointed out that Paraguay's capital had roughly one telephone line for every 11 people. Countries spent in proportion to how much computing they had to lose. The Senate committee reached the same conclusion from the other direction. It acknowledged that US per capita Y2K spending ran higher than Italy's, Spain's or Russia's, and attributed that to a country with about a quarter of the world's computers running systems more interconnected and more complex than most. Nations that started late also inherited American diagnostic tools, testing experience and published compliance information for free. There is a catch buried in the international results. The committee noted that most overseas fixes were temporary clock rollbacks that still had to be made permanent afterward, which means some of that low spending was deferred, not avoided. ## Is another Y2K coming? Some of Y2K was patched rather than solved. A common shortcut called windowing left two-digit years in place and simply told the software which century to assume, with a pivot year deciding the split. The committee flagged the obvious risks, that pivot years eventually arrive, that two systems trading dates can assume different pivots, and that repairs introduce fresh defects of their own. New York City got the demonstration exactly 20 years later. On January 1, 2020, all 14,000 of the city's parking meters, covering about 85,000 spaces, stopped accepting credit and prepaid cards because the payment software hit a hardcoded expiration date. Coins and the ParkNYC app still worked, and crews had to reconfigure the meters one at a time. The date most often named as the sequel is 2038, when timers in some Unix systems roll over the way GPS clocks did in August 1999. The Senate committee expected it to be narrower than Y2K, on the reasoning that two-digit years were a universal programming habit while 2038 is a limitation of particular systems, and that Y2K left the industry far harder to surprise with a date. The committee itself disbanded on February 29, 2000, once the leap day everyone had also worried about passed without incident. ## Frequently Asked Questions ### Did anyone die because of the Y2K bug? No deaths were attributed to Y2K. The Senate committee's final report counted hundreds of reported computer problems after January 1, 2000, none of which caused serious disruptions. The medical entries are minor by design, such as a tissue processor at a VA hospital that would not run on automatic with a date of 2000 but worked manually. ### Why did programmers use two-digit years in the first place? Storage and memory were the constraint. Dropping the century from every date in every record was a real saving when the code was written in the 1960s and 1970s, and the software was not expected to survive to 2000. Undoing it was expensive precisely because it was everywhere, with one widely used estimate pricing remediation at a dollar per line of code. ### What is the Year 2038 problem? Timers in some Unix systems count seconds and will roll over in 2038, in the same way GPS system clocks rolled over in August 1999. The Senate committee predicted in February 2000 that it would not be pervasive, since it comes from limits in specific software rather than from an industry-wide habit like two-digit years. **Sources:** - US Senate Special Committee on the Year 2000 Technology Problem, Y2K Aftermath: Crisis Averted, final committee report, February 29, 2000: https://www.govinfo.gov/content/pkg/GOVPUB-Y1_3-PURL-LPS90964/pdf/GOVPUB-Y1_3-PURL-LPS90964.pdf - The New York Times, Y2K: In One of Few Problems, Link to Spy Satellite Fails, January 2, 2000: https://sgp.fas.org/news/2000/01/nyt010200.html - Government Executive, Y2K costs government, businesses $100 billion, November 1999: https://www.govexec.com/federal-news/1999/11/y2k-costs-government-businesses-100-billion/5119/ - Slate, Were the Y2K Preparations in Vain?, January 2000: https://slate.com/news-and-politics/2000/01/were-the-y2k-preparations-in-vain.html - Engadget, Y2K-type glitch is causing NYC parking meters to reject credit cards, January 3, 2020: https://www.engadget.com/2020/01/03/nyc-parking-meter-software-glitch/ --- # Did Napster Kill the Music Industry? The 2026 Math URL: https://404memoryfound.com/posts/napster-destroyed-music.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-17 Updated: 2026-09-11 Topics: Music & Entertainment, Software & Apps **Summary:** Napster did not kill the music industry, but it did end the CD business the industry was built on. US recorded music revenue fell from an all-time high of $14.6 billion in 1999, the year Napster launched, to about $8.5 billion in 2009. It reached a record $11.5 billion in 2025, with paid streaming subscriptions supplying 55.3% of the total. **Key facts:** - Launched: 1999, by Shawn Fanning - Shut down: July 2001, to comply with the injunction in A&M Records v. Napster - Industry peak before Napster: $14.6 billion in US recorded music sales in 1999 - Low point after: About $8.5 billion in 2009, roughly $6 billion a year below the peak - Status today: US recorded music revenue hit a record $11.5 billion in 2025, across 106.5 million paid subscriptions ## How far did US music revenue actually fall? US recorded music sales hit an all-time high of $14.6 billion in 1999, the year Napster launched. Ten years later, in 2009, the same industry took in about $8.5 billion. Roughly $6 billion a year disappeared from a business that had grown almost every year through the 1990s. It did not arrive as a single crash. It arrived as ten consecutive losing years, which is harder to fight and much harder to explain to shareholders. Napster was not the only cause. The CD boom of the 1990s was driven partly by listeners rebuying albums they already owned on vinyl and cassette, and that replacement cycle was running out on its own. Broadband, CD burners and cheap hard drives all landed in the same window. What made the fall structural rather than cyclical was unbundling. A CD sold a dozen songs for one price whether you wanted a dozen or one. Once single tracks could be pulled out of an album, the album stopped being the unit of sale, and the pricing model built on it went with it. No court ruling could put that back, because the change was in what listeners expected to buy, not in what any one company was offering. ## Why did the record labels sue instead of opening a store? The Recording Industry Association of America sued Napster in December 1999, months after it appeared. Metallica sued separately, and on July 11, 2000 drummer Lars Ulrich testified before the Senate Judiciary Committee, where he argued there was no honest way to assemble a collection the size of a Napster user's: "Walk into a record store, grab what you want and walk out." Ulrich became the public face of the fight, which suited the labels. The quieter argument inside the companies was about margin. A CD at full price funded everything: advances, studios, radio promotion, the retail chains. Any authorized download store would have sold the same music for less, and nobody wanted to be the executive who halved the industry's revenue per customer. So they defended the disc. For about five years the most convenient music software in the world was also the illegal kind, and every shutdown produced a replacement: LimeWire (https://404memoryfound.com/posts/what-happened-to-limewire-file-sharing.html), Kazaa (https://404memoryfound.com/posts/what-happened-to-kazaa-p2p-wars.html), BitTorrent trackers. Suing one network never removed the demand that built it. ## What actually shut Napster down in 2001? A federal court, in stages. In July 2000, US District Judge Marilyn Hall Patel granted the record companies a preliminary injunction ordering Napster to stop the trading of copyrighted recordings on its network. The Ninth Circuit stayed that order, heard the appeal, and on February 12, 2001 largely upheld it in A&M Records, Inc. v. Napster, Inc. The appeals court held that Napster could be liable for both contributory and vicarious copyright infringement. It knew specific infringing files sat on the system, it had the ability to police them, and it benefited from the traffic they brought. The court also rejected the argument that users swapping complete songs were making fair use. A revised injunction followed in March 2001, requiring Napster to keep infringing titles out of its index. Filtering a catalogue named by its own users was never going to work, and in July 2001 Napster switched the network off. The company filed for bankruptcy protection soon after. What survived was a name, bought and sold ever since (https://404memoryfound.com/posts/who-owns-napster-now.html) by companies with no connection to the original service. ## Is the music industry bigger now than before Napster? In plain dollars, yes, and only recently. US recorded music revenue reached a record $11.5 billion in 2025, up 3.1% on 2024, according to the RIAA year-end report published in March 2026. Streaming supplied $9.75 billion of that total. Paid subscriptions are the business now. They brought in $6.38 billion in 2025, or 55.3% of every dollar the industry earned, across 106.5 million paid accounts in the United States. "The last 20 years have been marked by unprecedented transformation for recorded music," RIAA chairman and chief executive Mitch Glazier said in the 2025 year-end report, pointing to the rise of streaming and the return of vinyl. The caveat is inflation. $11.5 billion in 2025 is not the same money as $14.6 billion in 1999, and 26 years sit between those two figures, so the real peak is still the one Napster interrupted. If you want the conversion, we did the math on what a 1999 dollar is worth today (https://404memoryfound.com/posts/thousand-dollars-then-now.html). ## What did listeners actually take from Napster? The expectation, not the software. Before 1999, access to music was rationed by what a store stocked and what you could afford that week. Napster replaced that with a search box, and once people had used one, no service could go back to selling scarcity. Everything that has made money since has had to match that standard first and charge second. Apple's iTunes Store did it by selling songs one at a time. Subscription streaming did it by dropping ownership altogether: pay monthly, get the catalogue, keep nothing. The trade was real, and artists carry most of it. Album sales paid out in lumps per unit. Streaming pays fractions per play, which rewards deep catalogues and constant releases over one strong record. Touring, merchandise and sync licensing now carry weight they never had to carry in 1999. Vinyl came back as a collector's format in the same stretch, which says something about what people will still pay to own. The business survived the thing it said would end it, but not in the form it was defending. The labels won in court in 2001 and still lost the product argument, which is the part worth remembering. ## Frequently Asked Questions ### Did Napster kill the music industry? No. Napster, the file-sharing service that ran from 1999 until July 2001, ended the CD business the music industry had been built on: US recorded music sales fell from $14.6 billion in 1999 to about $8.5 billion in 2009. The industry itself recovered on subscription streaming and set a record of $11.5 billion in 2025. ### How much money did the music industry lose after Napster? US recorded music revenue fell from an all-time high of $14.6 billion in 1999, the year Napster launched, to about $8.5 billion in 2009, roughly $6 billion a year. Revenue did not reach record territory again until the streaming era, hitting $11.5 billion in 2025. ### Is Napster still around in 2026? The original Napster file-sharing network has been gone since July 2001, when the company switched it off to comply with the injunction in A&M Records v. Napster. The Napster name survived as a brand, has changed hands repeatedly since, and belongs to owners with no connection to the service that launched in 1999. **Sources:** - RIAA, 2025 Year-End Music Industry Revenue Report: https://www.riaa.com/reports/2025-year-end-music-industry-revenue-report-riaa/ - Billboard, 'RIAA 2025 Music Report: Revenue Tops $11.5B as Streaming, Vinyl Grow': https://www.billboard.com/pro/riaa-2025-music-report-revenue-streaming-vinyl/ - The Music Business Journal, 'A Decade to Remember: Changes in the Music Industry, 1999 to 2009': https://www.thembj.org/2009/10/a-decade-to-remember-changes-in-the-music-industry-1999-2009/ - CNN transcript, Senate Judiciary Committee hearing on downloading music, July 11, 2000: https://www.cnn.com/TRANSCRIPTS/0007/11/se.01.html - A&M Records, Inc. v. Napster, Inc., 239 F.3d 1004 (9th Cir. 2001): https://law.justia.com/cases/federal/appellate-courts/F3/239/1004/636120/ --- # What Is $1,000 From 1999 Worth in 2026 Dollars? URL: https://404memoryfound.com/posts/thousand-dollars-then-now.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-15 Updated: 2026-09-11 Topics: Money & Tech, Then vs Now **Summary:** $1,000 from 1999 has the same buying power as roughly $2,000 in 2026, according to the Bureau of Labor Statistics Consumer Price Index. That average hides a split: a thousand dollars buys far more computer than it did in 1999, and far less house or health insurance. The average employer family health premium alone went from $5,791 in 1999 to $26,993 in 2025. **Key facts:** - Status today: A 1999 dollar is worth about 50 cents of 2026 buying power - $1,000 in 1999: About $2,000 in 2026 dollars (BLS CPI) - Median household income, 1999: $41,994 (U.S. Census Bureau) - Family health premium: $5,791 in 1999, $26,993 in 2025 (KFF) - Median home price today: $434,100 in July 2026 (National Association of Realtors) ## Why a thousand dollars felt like real money in 1999 In 1999, a thousand dollars was a meaningful slice of a household's year. The U.S. Census Bureau put median household income for 1999 at $41,994, so $1,000 was about 2.9 percent of what the household in the middle of the country earned across twelve months. It was also the going price of the most expensive thing in most living rooms that was not furniture. CNN Money ran the headline "Average PC below $1000" on December 21, 1998, reporting that the average price of a personal computer sold at U.S. retail had dropped under a thousand dollars. That is the anchor people remember. A thousand dollars was a computer: a tower, a monitor, a keyboard and a 56k modem (https://404memoryfound.com/posts/dialup-speed-then-vs-now.html) that tied up the phone line. Gateway (https://404memoryfound.com/posts/what-happened-to-gateway-2000-cow-box-computer.html) shipped it in a cow-spotted box, and a family expected the machine to last three or four years before it felt slow. Everything below measures the same $1,000 against the things it used to cover, using price series that run continuously from 1999 to now. ## Is $1,000 still a lot of money in 2026? Measured against the Consumer Price Index, it is worth half of what it was. $1,000 from 1999 has the same buying power as roughly $2,000 in 2026, according to the Bureau of Labor Statistics CPI inflation calculator. Prices across the basket the BLS tracks have about doubled in 27 years. Turn that around and a 2026 dollar buys about half of what a 1999 dollar bought. A thousand dollars sitting in a 2026 checking account is, in 1999 terms, about $500. That is the headline number, and it is also the number that misleads people. The CPI is an average of a basket, and almost nothing in the basket actually doubled. Some categories tripled or quadrupled. Others fell by more than half. Which ones a household buys decides whether 27 years feels like a doubling or something a lot worse. Three categories carry most of the gap: housing, health insurance and computers. ## Why housing pulled away from the inflation average Housing is the category that broke the average. The National Association of Realtors put the median price of an existing home at $434,100 in July 2026, a record, and the 37th straight month of year-over-year price increases. General prices roughly doubled since 1999. Home prices did considerably more than that, and the monthly payment attached to them moved with mortgage rates on top of the price itself. "Without a doubt, the affordability is a major challenge for people who want to become homeowners, which is the reason why we need more supply," NAR chief economist Lawrence Yun said in June 2026. What $1,000 does here has changed in kind rather than degree. In 1999 a thousand dollars was a real contribution to a down payment on a median-priced house. In 2026 it is a fraction of one month's carrying cost on one, which is why the same sum feels so much lighter even though the CPI says it only lost half its value. ## Is health insurance where the purchasing power went? For a lot of households, yes. KFF has run the same employer health benefits survey every year since 1999, which makes it one of the cleanest 27-year price series in American life. In 1999, the average annual premium for employer-sponsored family coverage was $5,791. In 2025 it was $26,993. That is about 4.7 times, against roughly 2 times for prices in general. Workers paid $6,850 of the 2025 figure out of their own paychecks, which by itself is more than the entire family premium cost in 1999. KFF president and CEO Drew Altman is not forecasting relief. "With GLP-1s, increases in hospital prices, tariffs and other factors, we expect employer premiums to rise more sharply next year," he said in October 2025. Set that against the thousand dollars. In 1999, $1,000 covered about two months of the average family premium. In 2025 it covered about two weeks. Nothing else in a normal budget moved that far that fast. ## Why computers went the other way Technology is the one large category that moved down instead of up. The CNN Money milestone in December 1998 was news precisely because a sub-$1,000 average was new. Within a year the average had fallen further, and the cheap machine kept getting less cheap-feeling. In 2026 the same thousand dollars sits near the top of the mainstream laptop market rather than in the middle of it. The middle moved down underneath it, and the capability of an entry-level machine moved past anything sold in 1999. Two forces did that, and neither shows up cleanly in a price index. Components got cheaper per unit of performance every year, and the software most people use moved into the browser, so the hardware stopped needing to be impressive to be enough. The same arithmetic run across forty pieces of 90s hardware converted into 2026 dollars (https://404memoryfound.com/posts/90s-tech-inflation-calculator.html) shows the pattern holding almost everywhere. The closer a product sits to a silicon chip, the worse it does at holding its price. ## What the comparison gets wrong if you stop at the CPI The doubling figure is correct and incomplete. It tells you what an average basket costs. It does not tell you that the basket itself changed shape. A 1999 household paid separately for long-distance calls, a film camera, a camcorder, a CD collection, a road atlas, an encyclopedia and a Friday night video rental. One 2026 phone absorbed all of them. Those substitutions never register as savings, because a price index tracks the price of things rather than the disappearance of things. The reverse is the part people feel. Housing and health insurance are not subject to any of that substitution. Neither can be downloaded, made smaller, or folded into a device already in a pocket, and both are exactly the categories that ran past the average. So the short answer holds: $1,000 from 1999 is worth about $2,000 in 2026. The longer answer is that the same money now buys far more computer, far less house, and about a fifth as much family health coverage. ## Frequently Asked Questions ### How much is $1,000 from 1999 worth in 2026? $1,000 from 1999 has about the same buying power as $2,000 in 2026, based on the Bureau of Labor Statistics Consumer Price Index. Prices roughly doubled over the 27 years from 1999 to 2026, so a 2026 dollar buys about half of what a 1999 dollar bought. ### Did $1,000 really buy a computer in 1999? Yes. CNN Money reported in December 1998 that the average price of a PC sold at U.S. retail had fallen below $1,000, and prices kept falling through 1999. A thousand dollars in 1999 bought a mainstream desktop computer, and that machine was usually the most expensive electronic item in the house. ### What was the median household income in 1999? Median household income in the United States was $41,994 in 1999, according to the U.S. Census Bureau's Census 2000 brief on household income. Converted with the Consumer Price Index, that is roughly $84,000 in 2026 dollars. **Sources:** - U.S. Bureau of Labor Statistics, CPI Inflation Calculator: https://www.bls.gov/data/inflation_calculator.htm - U.S. Census Bureau, Census 2000 Brief: Household Income: 1999: https://www.census.gov/library/publications/2005/dec/c2kbr-36.html - KFF, 2025 Employer Health Benefits Survey: https://www.kff.org/health-costs/2025-employer-health-benefits-survey/ - National Association of Realtors, Existing-Home Sales: https://www.nar.realtor/research-and-statistics/housing-statistics/existing-home-sales - CNN Money, 'Average PC below $1000', December 21, 1998: https://money.cnn.com/1998/12/21/technology/pcdata/ --- # 90s Tech Prices in 2026 Dollars: Full 40-Item Table URL: https://404memoryfound.com/posts/90s-tech-inflation-calculator.html Author: Theo Lindqvist (404 Memory Found) Published: 2026-03-13 Updated: 2026-09-09 Topics: Money & Tech, Then vs Now **Summary:** A Sony PlayStation cost $299 at its US launch in 1995, which is about $655 in 2026 dollars, and a Nikon D1 body cost $5,500 in 1999, or roughly $11,025 today. This page lists 40 gadgets, computers, cameras, music players, software titles and online services from 1989 to 2007 with their US launch prices and the 2026 equivalent. Every launch price links to the source it came from, and the conversion uses the BLS CPI-U index. **Key facts:** - Items priced: 40, from 1989 to 2007 - Inflation method: CPI-U, launch-year annual average to July 2026 (index 333.918) - Priciest here: Nikon D1, $5,500 in 1999, about $11,025 in 2026 dollars - Cheapest here: Furby, $35 in 1998, about $72 in 2026 dollars - Last updated: September 9, 2026 ## How these 2026 numbers were computed Every launch price below is a US suggested retail price for the base configuration at its American release, taken from the page linked in the Source column. Later price cuts, holiday bundles and street prices are excluded, which is why some numbers look higher than the figure you remember paying. The conversion uses the Consumer Price Index for All Urban Consumers (CPI-U), US city average, all items, not seasonally adjusted. That is the same series behind the Bureau of Labor Statistics inflation calculator. Each launch price is multiplied by 333.918, the CPI-U reading for July 2026, then divided by the CPI-U annual average for the year the product shipped in the United States. The annual averages used here are 123.967 for 1989, 130.658 for 1990, 136.192 for 1991, 140.317 for 1992, 144.458 for 1993, 148.225 for 1994, 152.383 for 1995, 156.850 for 1996, 163.008 for 1998, 166.575 for 1999, 172.200 for 2000, 177.067 for 2001, 188.883 for 2004, 195.292 for 2005, 201.592 for 2006 and 207.342 for 2007. In plain multipliers, a 1989 dollar is worth 2.694 dollars in 2026, a 1995 dollar 2.191, a 1998 dollar 2.048 and a 2001 dollar 1.886. Two limits are worth stating up front. CPI-U tracks a whole household basket, so it says nothing about how much better the hardware got, and a 1994 digital camera at 640 by 480 pixels is not the same product as a phone camera today. CPI-U is also a national average that excludes state sales tax, so what a shopper actually handed over at the register was higher than the MSRP in most states. ## Which gadgets got cheaper in real terms, and which got pricier Software and imaging fell hardest. Adobe Photoshop 1.0 listed at $895 in 1990, which is $2,287 in 2026 dollars, for a single-machine license with no updates. Microsoft Encarta (https://404memoryfound.com/posts/what-happened-to-microsoft-encarta.html) shipped at $395 in 1993, or $913 today, to put an encyclopedia on one CD-ROM. A full retail copy of Windows 95 was $209.95, or $460 now. Cameras collapsed even faster. The Nikon D1 body cost $5,500 in 1999, which is $11,025 in 2026 dollars, and the Apple QuickTake 100 asked $749 in 1994, or $1,687, for eight photos at VGA resolution. Phones followed the same curve. The Motorola StarTAC was $1,000 in 1996, worth $2,129 today, and the first iPhone at $599 in 2007 works out to $965. Game consoles are the surprise in the other direction. They did not get dramatically cheaper, and at the top of the range they got dearer. The original PlayStation at $299 in 1995 equals $655 in 2026 dollars, but the GameCube (https://404memoryfound.com/posts/why-nintendo-gamecube-lost-console-war.html) at $199 in 2001 is only $375 and the Wii at $249.99 in 2006 is $414. The Nintendo Switch 2 launched in the United States on June 5, 2025 at $449.99, above both of them in real terms. Handhelds crept up too: the 1989 Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html) equals $242 today, while the 2004 Nintendo DS equals $265. ## What a full 1998 setup cost versus a 2026 one Take a household kitting itself out in 1998 with the machines this site keeps writing about. A Bondi Blue iMac G3 (https://404memoryfound.com/posts/how-imac-g3-saved-apple-killed-beige-box.html) at $1,299, an Iomega Zip 100 drive at $199 to move files around, a Diamond Rio PMP300 at $200 to carry roughly 30 minutes of MP3s, and AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html)'s unlimited dial-up plan at $19.95 a month, the price it set in December 1996. That is $1,698 of hardware plus $239.40 for the first year online. Converted item by item, it comes to about $4,011 in 2026 dollars. Add a 14-hour TiVo (https://404memoryfound.com/posts/what-happened-to-tivo-dvr-pioneer.html) at $499 in 1999, worth $1,000 today, and the same living room clears $5,000. The 2026 comparison is not one device against another, it is one device against a shelf. The 2007 iPhone, $965 in today's money, absorbed the $1,687 digital camera, the $410 MP3 player and the $1,616 Apple Newton MessagePad (https://404memoryfound.com/posts/what-happened-to-apple-newton.html) in a single $599 purchase. The gadget that got cheaper was rarely any one gadget. It was the whole pile of them collapsing into one. Some of the pile never earned its price at any exchange rate. The Virtual Boy (https://404memoryfound.com/posts/what-happened-to-virtual-boy-nintendo.html) asked $179.95 in 1995, or $394 in 2026 dollars, and Nintendo pulled it inside a year. The 3DO (https://404memoryfound.com/posts/what-happened-to-3do-700-dollar-console.html) at $699.99 in 1993 is $1,618 today, and it sold accordingly. ## The full table: 40 gadgets at launch and in 2026 dollars Forty items, from the 1989 Sega Genesis to the 2007 iPhone, sorted by US launch year. The RAZR V3 figure is its Cingular launch price with a two-year contract rather than an unlocked MSRP, and the AOL row is a monthly subscription, not a purchase. Journalists and researchers are welcome to reuse the table with a link back. ProductYearLaunch price (US MSRP)In 2026 dollarsSourceSega Genesis1989$189$509Wikipedia (https://en.wikipedia.org/wiki/Sega_Genesis)Nintendo Game Boy1989$89.95$242Wikipedia (https://en.wikipedia.org/wiki/Game_Boy)Atari Lynx1989$179.95$485Wikipedia (https://en.wikipedia.org/wiki/Atari_Lynx)Adobe Photoshop 1.0 (Mac)1990$895$2,287Wikipedia (https://en.wikipedia.org/wiki/Adobe_Photoshop_version_history)Super Nintendo (SNES)1991$199$488Wikipedia (https://en.wikipedia.org/wiki/Super_Nintendo_Entertainment_System)Sega Game Gear1991$149.99$368Wikipedia (https://en.wikipedia.org/wiki/Game_Gear)Sony MZ-1 MiniDisc recorder1992$750$1,785Wall Street Journal (https://www.minidisc.org/wsj_article.html)Apple Newton MessagePad1993$699$1,616Wikipedia (https://en.wikipedia.org/wiki/MessagePad)3DO Interactive Multiplayer1993$699.99$1,618Wikipedia (https://en.wikipedia.org/wiki/3DO_Interactive_Multiplayer)Atari Jaguar1993$249.95$578Wikipedia (https://en.wikipedia.org/wiki/Atari_Jaguar)Microsoft Encarta1993$395$913Wikipedia (https://en.wikipedia.org/wiki/Encarta)Apple QuickTake 100 camera1994$749$1,687Wikipedia (https://en.wikipedia.org/wiki/Apple_QuickTake)IBM ThinkPad 701C1995$1,999$4,380Wikipedia (https://en.wikipedia.org/wiki/IBM_ThinkPad_701)Windows 95 (full retail)1995$209.95$460Seattle Times (https://archive.seattletimes.com/archive/?date=19950607&slug=2125167)Sony PlayStation1995$299$655Wikipedia (https://en.wikipedia.org/wiki/PlayStation_(console))Sega Saturn1995$399$874Wikipedia (https://en.wikipedia.org/wiki/Sega_Saturn)Nintendo Virtual Boy1995$179.95$394Wikipedia (https://en.wikipedia.org/wiki/Virtual_Boy)Iomega Zip 100 drive1995$199$436How-To Geek (https://www.howtogeek.com/658287/even-25-years-later-the-iomega-zip-is-unforgettable/)Palm Pilot 10001996$299$637PCWorld (https://www.pcworld.com/article/523791/history_of_palm.html)Motorola StarTAC1996$1,000$2,129Wikipedia (https://en.wikipedia.org/wiki/Motorola_StarTAC)Nintendo 641996$199.99$426Wikipedia (https://en.wikipedia.org/wiki/Nintendo_64)AOL unlimited dial-up (monthly)1996$19.95$42 per monthDeseret News (https://www.deseret.com/1996/12/2/19280328/aol-s-flat-rate-is-a-good-deal-offered-in-an-insensitive-way/)Apple iMac G31998$1,299$2,661Macworld (https://www.macworld.com/article/191974/imacanniversary.html)Diamond Rio PMP3001998$200$410Wikipedia (https://en.wikipedia.org/wiki/Rio_PMP300)Furby1998$35$72Wikipedia (https://en.wikipedia.org/wiki/Furby)Sega Dreamcast1999$199$399Wikipedia (https://en.wikipedia.org/wiki/Dreamcast)Nikon D1 (body only)1999$5,500$11,025Wikipedia (https://en.wikipedia.org/wiki/Nikon_D1)TiVo (14-hour)1999$499$1,000Encyclopedia.com (https://www.encyclopedia.com/books/politics-and-business-magazines/tivo-inc)Sony PlayStation 22000$299$580Wikipedia (https://en.wikipedia.org/wiki/PlayStation_2)Microsoft Xbox2001$299$564Microsoft (https://news.microsoft.com/source/2001/05/16/microsoft-announces-launch-details-for-xbox-in-north-america/)Nintendo GameCube2001$199$375Wikipedia (https://en.wikipedia.org/wiki/GameCube)Game Boy Advance2001$99.99$189Wikipedia (https://en.wikipedia.org/wiki/Game_Boy_Advance)Apple iPod (5 GB)2001$399$752Apple (https://www.apple.com/newsroom/2001/10/23Apple-Presents-iPod/)Nintendo DS2004$149.99$265Wikipedia (https://en.wikipedia.org/wiki/Nintendo_DS)Motorola RAZR V3 (with 2-year contract)2004$500$884NBC News (https://www.nbcnews.com/id/wbna6504736)Sony PSP2005$249$426Wikipedia (https://en.wikipedia.org/wiki/PlayStation_Portable)Xbox 360 Premium2005$399$682Wikipedia (https://en.wikipedia.org/wiki/Xbox_360)Nintendo Wii2006$249.99$414Nintendo (http://www.nintendoworldreport.com/pr/12050/official-wii-launch-details)Microsoft Zune 302006$249.95$414Wikipedia (https://en.wikipedia.org/wiki/Zune)Apple iPhone (8 GB)2007$599$965Apple (https://www.apple.com/newsroom/2007/01/09Apple-Reinvents-the-Phone-with-iPhone/) ## Frequently Asked Questions ### How much did a PlayStation cost in 1995 in today's money? The original Sony PlayStation went on sale in North America on September 9, 1995 at $299. Using the CPI-U annual average for 1995 and the July 2026 index, that is about $655 in 2026 dollars. The Sega Saturn, which beat it to market at $399, works out to roughly $874. ### What was the most expensive gadget of the 90s? On this list it is the Nikon D1, which cost $5,500 for the body alone in 1999 and equals about $11,025 in 2026 dollars. Among things ordinary households actually bought, the IBM ThinkPad 701C at $1,999 in 1995 is the peak at roughly $4,380, followed by the $1,000 Motorola StarTAC at $2,129. ### Why were 90s computers so expensive? Chips, memory and displays were made in far smaller volumes, and a computer had to include parts that later became free or built in, such as a modem, a CD-ROM drive, a sound card and an operating system license worth $209.95 on its own. Retail software carried the same load: a single Photoshop license listed at $895 in 1990. Volume, integration and price competition did most of the work of bringing those numbers down, not inflation. **Sources:** - U.S. Bureau of Labor Statistics, CPI Inflation Calculator: https://www.bls.gov/data/inflation_calculator.htm - Seattle Times: Microsoft Suggests $209.95 Retail Price For Windows 95 (June 7, 1995): https://archive.seattletimes.com/archive/?date=19950607&slug=2125167 - Microsoft: Microsoft Announces Launch Details for Xbox in North America (May 16, 2001): https://news.microsoft.com/source/2001/05/16/microsoft-announces-launch-details-for-xbox-in-north-america/ - Apple: Apple Presents iPod (October 23, 2001): https://www.apple.com/newsroom/2001/10/23Apple-Presents-iPod/ - Apple: Apple Reinvents the Phone with iPhone (January 9, 2007): https://www.apple.com/newsroom/2007/01/09Apple-Reinvents-the-Phone-with-iPhone/ - Nintendo: Official Wii Launch Details (September 14, 2006): http://www.nintendoworldreport.com/pr/12050/official-wii-launch-details - PCWorld: From Palm Pilot to Palm Pre, A Brief History of Palm's Handhelds: https://www.pcworld.com/article/523791/history_of_palm.html - Deseret News: AOL's flat rate is a good deal offered in an insensitive way (December 2, 1996): https://www.deseret.com/1996/12/2/19280328/aol-s-flat-rate-is-a-good-deal-offered-in-an-insensitive-way/ --- # 90s Websites Still Online in 2026: 5 That Survived URL: https://404memoryfound.com/posts/90s-websites-still-online.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-11 Updated: 2026-09-11 Topics: Internet Culture, Then vs Now **Summary:** A handful of 1990s websites are still online at their original addresses in 2026, and the rest survive only as archive copies. The original 1996 Space Jam site is live at spacejam.com/1996, where Warner Bros. moved it in 2021, and info.cern.ch, published in August 1991, was restored by CERN in 2013 and still answers today. Everything else runs through the Wayback Machine, which passed one trillion archived pages in October 2025. **Key facts:** - Status today: Live in 2026: info.cern.ch, spacejam.com/1996, berkshirehathaway.com, fogcam.org - Oldest still online: info.cern.ch, published August 1991, restored by CERN in 2013 - Space Jam site: Moved from spacejam.com to spacejam.com/1996 in 2021 - Pages archived: More than 1 trillion, Internet Archive, October 2025 - Oldest running webcam: FogCam, San Francisco State University, running since 1994 ## Is the Space Jam website still online in 2026? Yes, but not at the address people remember. The 1996 promotional site for Space Jam sat at spacejam.com untouched for 25 years: a black starfield with a ring of planets labeled Jam Central, Planet B-Ball, Lunar Tunes, Junior Jam and Stellar Souvenirs. In 2021, ahead of the 16 July release of Space Jam: A New Legacy, Warner Bros. handed the main domain to the sequel and moved the original to spacejam.com/1996. That move is the reason the page still exists. Warner Bros. could have deleted the directory. Instead the studio gave it a permanent shelf, and the 1996 files load in 2026 the way they loaded on a 28.8k modem: tables for layout, animated GIFs, no CSS, no JavaScript worth the name, a fixed width built for a 640 by 480 screen. What survives is not a reconstruction or a fan copy. It is the same HTML that Warner Bros. Online published while the movie was in theaters, which makes it the most visited piece of 1990s web design on the internet. The links still work. The trailer still downloads. The copyright line still reads 1996. ## What was the first website, and can you still open it? Tim Berners-Lee published the first website at CERN in August 1991, on a NeXT machine, at info.cern.ch. It was a page about the WorldWideWeb project itself: what hypertext was, how to get a browser, how to put your own documents online. There were no images on it, because the HTML of 1991 had no way to show one. The original page went missing as the web expanded through the 1990s. CERN rebuilt it in 2013, for the twentieth anniversary of the day in April 1993 when the laboratory released the web into the public domain. The restored file sits at its first URL and has stayed there ever since. "We are in a unique moment where we can still switch on the first web server and experience it," CERN web manager Dan Noyes told BBC News in 2013. That is the whole case for keeping these pages alive, made by the institution with the strongest claim to the first one. Opening it in 2026 takes about a second, because it is a few kilobytes of plain text and nothing else. It is the fastest page you will load all week. ## Why does Berkshire Hathaway still run a 1997 website? berkshirehathaway.com is the strangest survivor on this list, because it was never abandoned. It is the live, official site of one of the largest companies in the world, and it looks like 1997: white background, a column of blue hyperlinks, a text advertisement for Geico and a postal address in Omaha for sending feedback. The source code gives the game away. The pages were built in Microsoft FrontPage 3, Adobe PageMill 2 and Microsoft Word, and have been edited by hand since. Warren Buffett stepped down as chief executive at the end of 2025 after six decades running the company, and the site did not change. The Wall Street Journal went looking at it in 2026 and found shareholders who want it left exactly as it is. The site, the paper wrote, "is still very much a reflection of the guy who says he's only sent an email once." One investor told the Journal she would sell her shares if the site were redesigned. It is the clearest proof that old pages do not survive by neglect alone. Somebody has to decide, repeatedly, that changing them costs more than keeping them. ## Who keeps a 1994 webcam pointed at a San Francisco campus? FogCam started in the fall of 1994 as a student project in the Department of Instructional Technologies at San Francisco State University, built by Jeff Schwartz and Dan Wong. It posts a still image of a campus view and has done it for more than 30 years. Once the Trojan Room Coffee Pot camera at the University of Cambridge was switched off in 2001, FogCam became the oldest webcam still running anywhere. It nearly ended. In August 2019 its creators announced they would shut it down at the end of that month, citing the loss of a decent vantage point and no support from the university. The announcement traveled, San Francisco State reversed course and took over the camera, and FogCam stayed on. It still updates in 2026, and a separate volunteer project now captures every frame it publishes. Craigslist (https://404memoryfound.com/posts/what-happened-to-craigslist-ugliest-website-worth-billions.html) belongs in the same family for a different reason. Nobody preserved it. It simply never stopped being a page of blue links on a white background, because the layout worked and the traffic never punished it. ## Which 90s sites did not survive? Most of them. The Dole-Kemp site from the 1996 presidential race is the one people miss: a complete campaign website that stayed live for a quarter century after the election, then went offline in 2021, as Boing Boing reported at the time. A Wayback Machine copy and a preservation page kept by 4President.org are what remain of it. GeoCities (https://404memoryfound.com/posts/history-of-geocities-websites.html), the free homepage service that hosted a generation of personal pages, was shut down by Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html), and volunteers at Archive Team crawled everything they could reach before the deadline and handed the copy to the Internet Archive. NCSA Mosaic (https://404memoryfound.com/posts/what-happened-to-ncsa-mosaic-browser.html), the browser most of these pages were built for, stopped shipping long before the pages did. The pattern holds across all of them. A 1990s page survives when one institution decides that keeping it costs less than the meeting about deleting it, or when one named person takes responsibility for the server. It dies when a company reorganizes and nobody in the new structure owns the domain. ## How do you visit a 90s site that is already gone? The Wayback Machine. The Internet Archive has been copying web pages since 1996, and in October 2025 it announced that it had passed one trillion archived pages. Type any address into web.archive.org and you get a calendar of snapshots reaching back as far as the crawler got. Two limits matter. The archive keeps what the crawler could fetch, so pages behind logins, Flash movies and anything generated by a server script often come back broken or blank. And a snapshot is a single day in the life of a site, not the site: the links inside it may jump to a copy from a different year, or to nothing at all. For the survivors in this article you do not need any of that. info.cern.ch, spacejam.com/1996, berkshirehathaway.com and fogcam.org all answer on their own domains in 2026. They are the last places on the web where the 1990s is still being served live rather than reenacted. ## Frequently Asked Questions ### Is the Space Jam website still up? Yes. The original 1996 Space Jam website is online at spacejam.com/1996, where Warner Bros. moved it in 2021 when the main spacejam.com address was given over to Space Jam: A New Legacy. The pages are the same HTML the studio published in 1996, Jam Central and Lunar Tunes included. ### What is the oldest website still online? info.cern.ch, the page Tim Berners-Lee published at CERN in August 1991 to explain the WorldWideWeb project. The original was lost as the web grew through the 1990s, and CERN restored it in 2013 at the same address, where it is still readable in 2026. ### Where can I see old websites from the 90s? The Wayback Machine at web.archive.org holds snapshots of most of the web since 1996 and passed one trillion archived pages in October 2025. Enter any address and pick a date. For the 1990s sites that never went away, including info.cern.ch, spacejam.com/1996, berkshirehathaway.com and fogcam.org, you can simply visit the live domain. **Sources:** - Space Jam (1996), the original Warner Bros. site: https://www.spacejam.com/1996/ - Restoring the first website, CERN Archive: https://first-website.web.cern.ch/ - Celebrating 1 Trillion Web Pages Archived, Internet Archive: https://blog.archive.org/trillion/ - Berkshire Has a Website From the '90s and Buffett Fans Say Don't Mess With It, The Wall Street Journal (2026): https://www.securitiesdocket.com/2026/05/01/berkshire-has-a-website-from-the-90s-and-buffett-fans-say-dont-mess-with-it-wsj/ - SF State Saves FogCam, Internet's Longest-Running Webcam, KQED: https://www.kqed.org/news/11771300/sf-state-saves-fogcam-internets-longest-running-webcam-after-it-almost-shuts-down --- # Dial-Up Speed vs 2026 Internet: How Slow Was 56k? URL: https://404memoryfound.com/posts/dialup-speed-then-vs-now.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-09 Updated: 2026-09-11 Topics: Internet Culture, Then vs Now **Summary:** Dial-up internet still exists in 2026, but only just. AOL discontinued its dial-up service on September 30, 2025, and Census Bureau figures put the number of Americans relying on dial-up alone at about 163,400 in 2023. A 56k modem in the United States peaked at 53.3 kilobits per second, roughly 1,876 times slower than the 100 Mbps floor the FCC has used to define broadband since March 2024. **Key facts:** - Status today: Still sold by NetZero and Juno; AOL ended dial-up on September 30, 2025 - Peak speed: 53.3 kbit/s downstream, 33.6 kbit/s upstream (V.90, 1998) - Broadband floor today: 100 Mbps down, 20 Mbps up (FCC benchmark, March 2024) - Users left: About 163,400 Americans on dial-up alone in 2023 (Census Bureau) - Price today: Juno still offers a free dial-up plan capped at 10 hours a month ## How fast was 56k dial-up in real numbers? V.90, the standard that ended the 56k modem wars, arrived in 1998. It replaced two incompatible schemes, x2 from US Robotics and K56flex from Rockwell and Lucent, and for the first time the speed you got did not depend on whose equipment your provider had bought. On paper V.90 ran 56 kilobits per second downstream and 33.6 kilobits per second upstream. It never reached 56 in the United States. FCC rules cap how much signal power can be pushed down a copper pair, and that cap set the real ceiling at 53.3 kilobits per second. EE Times, in its introduction to the standard, put typical downstream throughput at 40 to 50 kilobits per second once line noise was accounted for. Turn 53.3 kilobits into bytes and you get about 6.5 kilobytes a second. A 500 KB photograph takes about 75 seconds. A four-megabyte MP3 takes about ten minutes. Those are best-case numbers, on a clean line, with nothing else running. V.90 was lopsided on purpose. The fast path worked only downstream, because the phone company's end of the link was already digital, while the trip back from the house stayed analog and stayed at 33.6 kilobits per second. V.92 raised uploads to 48 kilobits per second in 2000, and almost nobody noticed, because cable and DSL had already arrived. ## What a 53.3 kbit/s connection felt like to use Images arrived in visible bands. The browser painted each row of pixels as it came off the wire, so a single 500 KB photograph unrolled over roughly 75 seconds while you watched. The connection owned the phone line. A modem session was a phone call, so an incoming call could drop it, and anyone lifting a handset elsewhere in the house did the same. Nothing resumed. A download that died at 90 percent started again at zero. That is why people treated the internet as somewhere you visited rather than somewhere you stayed. You dialed in, grabbed what you wanted, hung up, and read it later. Offline browsers existed for exactly this job: they pulled entire sites down in one session so you could click around afterward with the modem silent. Mail worked the same way. Connect, download the inbox, disconnect, write the replies at your own pace, then dial back in to send them. GeoCities pages (https://404memoryfound.com/posts/history-of-geocities-websites.html) were built for that rhythm, which is part of why the surviving ones still load instantly on hardware three decades newer. ## Why the 90s web looked the way it did Every design decision on the 1990s web was a bandwidth decision. At 6.5 kilobytes a second, a 100 KB banner image is a 15-second tax on every single visitor, so designers did not use one. Pages were text first, with small flat-color graphics, layout held together by tables, and headline type rendered as an image only when there was no alternative. Animation meant a short looping GIF. Video meant a link to a file you would leave downloading overnight and probably never watch. The constraint produced a web that looks crude now and was fast in a way the modern web is not. No analytics bundles, no ad auctions running in the background, no fonts fetched from three separate domains. The page was the page. It also set a hard ceiling on who could publish. Uploading at 33.6 kilobits per second meant a few hundred kilobytes took minutes, so posting photographs was a chore and posting video was not something ordinary people did at all. The cable modem rollouts of the late 1990s, Excite@Home (https://404memoryfound.com/posts/what-happened-to-excite-at-home-broadband.html) among them, are what removed that ceiling. ## Is dial-up internet still available in 2026? Yes, though the biggest name in it is gone. AOL discontinued its dial-up service on September 30, 2025, along with the AOL Dialer and the AOL Shield browser. The company's own notice offered no reason beyond housekeeping: "AOL routinely evaluates its products and services and has decided to discontinue Dial-up Internet." Smaller providers kept going. PCWorld's roundup of who still sells dial-up after AOL's exit names NetZero (https://404memoryfound.com/posts/dial-up-internet-still-available-2026.html) and Juno among the operators still taking signups in the United States, with Juno still offering a free plan capped at 10 hours a month. The remaining customer base is small and mostly rural. Census Bureau figures quoted throughout the 2025 coverage of the AOL shutdown put the number of Americans relying on dial-up alone at about 163,400 in 2023. That is a rounding error against national internet use, and it is also a real set of households where nothing faster reaches the property line. For most people the service died long before the technology did. AOL still sells a paid membership (https://404memoryfound.com/posts/does-aol-still-exist-today.html), just not one that dials anything. ## How much faster is internet in 2026? The cleanest comparison is the regulatory one. In March 2024 the FCC raised the benchmark for fixed broadband in the United States to 100 megabits per second down and 20 megabits per second up, a four-fold increase on the 25/3 standard it had set in 2015. Chairwoman Jessica Rosenworcel was blunt about the delay in her statement on the decision: "Honestly, this fix is overdue." Run the ratio. 100 megabits per second is 100,000 kilobits per second, which divided by dial-up's 53.3 comes to about 1,876 times faster. That is the floor, the slowest connection the federal government is willing to call broadband. The FCC also set a long-term goal of 1 gigabit down and 500 megabits up, which would be roughly 18,760 times a 56k modem's best day. Stated in time rather than multiples: the four-megabyte MP3 that cost ten minutes on a clean 53.3 kbit/s line takes about a third of a second at 100 megabits per second. The wait did not get shorter. It stopped existing. ## Frequently Asked Questions ### How slow was dial-up internet? Dial-up internet on a 56k modem topped out at 53.3 kilobits per second in the United States, about 6.5 kilobytes a second, with typical real-world speeds of 40 to 50 kilobits per second. At that rate a 500 KB photograph took roughly 75 seconds and a four-megabyte MP3 took about ten minutes. The FCC broadband benchmark set in March 2024, 100 megabits per second, is about 1,876 times faster. ### Does dial-up internet still exist in 2026? Dial-up internet still exists in 2026, but AOL, the service most Americans associate with it, discontinued its dial-up product on September 30, 2025. Smaller providers including NetZero and Juno still sell dial-up access, and Juno still offers a free plan limited to 10 hours a month. Census Bureau figures put the number of Americans relying on dial-up alone at about 163,400 in 2023. ### Why did dial-up modems make that screeching sound? A dial-up modem session was an ordinary voice phone call, so the negotiation between the two modems was audible on the line. The tones and hiss were the modems testing the circuit and settling on a rate, which under the V.90 standard used from 1998 could land anywhere between 28 kilobits per second and the 53.3 kilobit ceiling. Once the handshake finished the modem muted its speaker, which is why the noise always stopped. **Sources:** - NPR: Say bye-bye to the beeps and boops of AOL's dial-up internet service (2025): https://www.npr.org/2025/08/12/nx-s1-5499539/aol-dial-up-ending - Forbes: AOL Ditches Dial-Up, Sparking Nostalgia (2025): https://www.forbes.com/sites/lesliekatz/2025/08/11/aol-ditches-dial-up-sparking-nostalgia-and-wait-thats-still-a-thing/ - FCC: Rosenworcel Statement, FCC Increases Broadband Speed Benchmark (March 2024): https://www.fcc.gov/document/fcc-increases-broadband-speed-benchmark/rosenworcel-statement-0 - EE Times: An Introduction to the V.90 (56K) Modem: https://www.eetimes.com/an-introduction-to-the-v-90-56k-modem/ - PCWorld: Despite AOL's exit, dial-up internet isn't going away (2025): https://www.pcworld.com/article/2874440/list-of-dialup-internet-providers-in-2025.html --- # What Was the First Thing Sold Online? PizzaNet vs a CD URL: https://404memoryfound.com/posts/first-online-purchase-pizza.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-07 Updated: 2026-09-10 Topics: Internet Culture, Then vs Now **Summary:** Pizza Hut's PizzaNet went live in Santa Cruz, California in August 1994 and took a large pepperoni, mushroom and extra cheese pizza as its first order, which is why Pizza Hut still calls it the world's first online purchase. The stronger claim belongs to NetMarket, which sold a Sting CD for $12.48 plus shipping on August 11, 1994, in the first secure credit card sale of a physical good. PizzaNet itself is long gone, and Pizza Hut changed hands in 2026. **Key facts:** - First PizzaNet order: A large pepperoni, mushroom and extra cheese pizza, Santa Cruz, California, 1994 - Announced: August 22, 1994, by Pizza Hut and The Santa Cruz Operation - Earlier claim: A Sting CD sold on NetMarket for $12.48 plus shipping, August 11, 1994 - Status today: PizzaNet is gone; Pizza Hut takes orders through its own site and app - Owner today: LongRange Capital, which completed its purchase from Yum! Brands on September 1, 2026 ## What did PizzaNet actually sell in August 1994? PizzaNet was a pilot run out of a single Pizza Hut franchise in Santa Cruz, California. The software came from The Santa Cruz Operation, the Unix company based in the same town, and the two announced the program at SCO Forum on August 22, 1994 as a pilot that let computer users order pizza delivery from their local Pizza Hut over the internet. The form lived at pizza.net. A customer chose a size, checked off toppings, typed an address and a phone number, and submitted it. The order travelled to a Pizza Hut server in Wichita, Kansas running SCO software, and was relayed back to the Santa Cruz restaurant. Pizza Hut's own account of the launch says the first order to arrive that way was a large pepperoni, mushroom and extra cheese pizza. Two details are worth holding on to, because both get garbled in the retellings. PizzaNet was a web page, not a dial-up bulletin board, and using it meant having a graphical browser at a time when almost nobody did. NCSA Mosaic (https://404memoryfound.com/posts/what-happened-to-ncsa-mosaic-browser.html), released in 1993, was barely a year old. ## How PizzaNet worked without taking a single payment PizzaNet did not process money. There was no card field, no encryption and no merchant account behind it. The web form collected an order and a phone number, and a person at the Santa Cruz restaurant called the customer back to confirm before anyone touched dough. Payment happened at the door, in cash or on a card, the way it always had. The callback was a fraud control as much as a courtesy. An open web form with no identity check is an invitation to send forty pizzas to someone you dislike, and the staff worked that out fast. This matters for the record, because the fight over what came first is really a fight about definitions. If placing an order counts, PizzaNet is in the running. If the sale has to complete online, with the payment crossing the network, PizzaNet is not a sale at all. It is an early order form bolted onto an ordinary phone-and-cash transaction. Pizza Hut has never been shy about the claim regardless. Chief marketing officer Carrie Walsh said in 2014 that the company wanted to celebrate that "before consumers could buy books, clothes, music or vacation packages via the internet, they could place an online order for a Pizza Hut pizza." ## Why a Sting CD beat the pizza by eleven days On August 11, 1994, a 21-year-old named Dan Kohn ran a transaction through NetMarket, the online store he had helped start. A friend in Philadelphia bought a compact disc of Sting's album Ten Summoner's Tales for $12.48 plus shipping, and the card number crossed the network encrypted. The New York Times covered it the next day under the headline Attention Shoppers: Internet Is Open. Kohn's line to the paper is the one people still repeat: "Even if the N.S.A. was listening in, they couldn't get his credit card number." The point was never the CD. The point was that a stranger's card details could travel over a public network without being readable in transit, which is the foundation every online store since has been built on. Eleven days later, Pizza Hut and SCO announced PizzaNet. On the ordinary reading, where a first sale means the first time money and goods changed hands through the network, the CD wins on the date and on the mechanism. The pizza is the better story, which is most of why it is the one that stuck. Retail on the web needed the encryption more than it needed the pepperoni. ## Was anything bought or sold online before 1994? Yes, if the definition is loosened, and it has been loosened in two directions. The oldest claim is a drug deal. In his 2005 book What the Dormouse Said, John Markoff reported that students at Stanford's artificial intelligence lab used ARPANET accounts in the early 1970s to arrange a sale of a small quantity of marijuana to counterparts at MIT. No money moved over the network. They used it to fix a meeting place, which makes it the first online negotiation rather than the first online sale. The second direction is commercial and much less romantic. Shopping over closed, subscription-only networks predates the web by a decade. CompuServe (https://404memoryfound.com/posts/what-happened-to-compuserve-online-service.html) ran a shopping mall where subscribers ordered real merchandise from real retailers through a dial-up service that was not the internet. By that standard the web in 1994 did not invent online shopping. It took online shopping out of a walled garden and put it somewhere anyone with a modem and a browser could reach, which is also why 1994 produced the first banner ad (https://404memoryfound.com/posts/first-banner-ad-hotwired-1994-history.html) ten weeks after the pizza. ## Who owns Pizza Hut today, and does PizzaNet still exist? PizzaNet does not exist. It ran in one city, for a pilot period, and pizza.net is not a Pizza Hut ordering site now. What replaced it is the thing PizzaNet was built to test: a national ordering site, then an app, then delivery marketplaces that route orders the restaurant never sees a phone call for. Pizza Hut itself changed hands in 2026. Yum! Brands, its parent for more than two decades, sold the chain in two pieces. Yum China Holdings bought the mainland China business for $1.2 billion in a deal that closed on August 7, 2026. The private equity firm LongRange Capital bought everything else for about $1.5 billion and announced the acquisition complete on September 1, 2026. So the answer people are searching for is that Pizza Hut is owned by LongRange Capital outside mainland China, and by Yum China Holdings inside it. Yum! Brands, the parent through almost the whole of the chain's online life, no longer owns any of it. The company that ran the first famous online pizza order spent 2026 being taken apart and sold. ## Frequently Asked Questions ### What was the first thing sold online? The most commonly cited answer is a compact disc. On August 11, 1994, NetMarket sold a copy of Sting's Ten Summoner's Tales for $12.48 plus shipping, which The New York Times reported the next day as the first secure retail transaction on the internet. Pizza Hut's PizzaNet, whose first order was a large pepperoni, mushroom and extra cheese pizza, was announced eleven days later on August 22, 1994. ### Was the first online purchase really a Pizza Hut pizza? Not by the usual test. Pizza Hut's PizzaNet, launched in Santa Cruz, California in August 1994, collected orders through a web form but took no payment online, and staff phoned each customer to confirm before money changed hands at the door. The NetMarket sale of a Sting CD on August 11, 1994 moved the order and the encrypted card number over the network together, which is why it is usually credited first. ### Who owns Pizza Hut now? Pizza Hut is owned by LongRange Capital, a private equity firm that completed its roughly $1.5 billion purchase of the chain from Yum! Brands on September 1, 2026. The mainland China business went separately to Yum China Holdings for $1.2 billion in a transaction that closed on August 7, 2026. **Sources:** - SCO and Pizza Hut Announce Pilot Program (biz.sco.announce, August 1994): https://groups.google.com/g/biz.sco.announce/c/YSc-pXPPOig - Pizza Hut Celebrates 20th Anniversary of World's First Online Purchase (PR Newswire, 2014): https://www.prnewswire.com/news-releases/pizza-hut-celebrates-20th-anniversary-of-worlds-first-online-purchase-with-50-percent-off-online-deal-for-hut-lovers-members-238428021.html - Attention Shoppers: Internet Is Open (The New York Times, August 12, 1994): https://www.nytimes.com/1994/08/12/business/attention-shoppers-internet-is-open.html - What Was the First Thing Sold on the Internet? (Smithsonian Magazine): https://www.smithsonianmag.com/smart-news/what-was-first-thing-sold-internet-180957414/ - LongRange Capital Completes Acquisition of Pizza Hut from Yum! Brands (September 1, 2026): https://www.longrangecapital.com/news-article/longrange-capital-completes-acquisition-of-pizza-hut-from-yum-brands --- # Did Blockbuster Turn Down Netflix for $50 Million? URL: https://404memoryfound.com/posts/blockbuster-netflix-acquisition.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-05 Updated: 2026-09-10 Topics: Business Blunders, Music & Entertainment **Summary:** Blockbuster did turn down Netflix. In 2000 Netflix co-founders Reed Hastings and Marc Randolph offered to sell their DVD-by-mail company for about $50 million, and Blockbuster chief executive John Antioco declined. Blockbuster filed for Chapter 11 in September 2010, its assets went to DISH Network in 2011, and the brand sits inside EchoStar today, while Netflix reported $45.2 billion of revenue for 2025. **Key facts:** - The offer: About $50 million for Netflix, pitched to Blockbuster in 2000 - Who said no: John Antioco, Blockbuster chief executive from 1997 to 2007 - Status today: Gone as a chain. One franchised store still trades in Bend, Oregon - Owner today: EchoStar Corporation, through DISH Network, which bought the assets in 2011 - Netflix now: $45.2 billion of revenue in 2025, roughly 900 times the 2000 asking price ## What happened in the Dallas meeting in 2000? Early in 2000, Netflix co-founders Reed Hastings and Marc Randolph flew to Dallas with their chief financial officer to see Blockbuster chief executive John Antioco. Netflix was a DVD-by-mail service with no stores, no late fees and a cash problem that was getting worse as the dot-com market fell apart. The pitch was that Blockbuster should buy Netflix and let it run the online half of the business. The price Randolph and Hastings had in mind was about $50 million. Blockbuster at that point was the video rental business in the United States, with thousands of stores and a rental model that had worked for 15 years. Randolph told the story in his 2019 book That Will Never Work and has repeated it in interviews since. His account is that Antioco was struggling not to laugh, and that the verdict from the Blockbuster side was that "the dot-com hysteria is completely overblown." Netflix, in that reading, was a niche mail-order outfit riding a bubble. Randolph put it more bluntly to Fortune in 2023: Blockbuster "laughed us out of the room." No counteroffer followed, and the meeting ended without a deal. Netflix went public in May 2002 instead. ## Why did Blockbuster say no to a $50 million price? The refusal was not really about the price. It was about what Netflix would have cost Blockbuster to own, because the two companies made money in opposite ways. Late fees were the difference. Blockbuster collected roughly $800 million in late fees in 2000, about 16 percent of its revenue that year. That money arrived with no extra inventory, no extra staff and no extra shelf space attached to it. It was the most profitable line the company had. Netflix charged a flat monthly fee and had no due dates at all, which is the same as saying its whole product was built out of the thing Blockbuster charged for. Buying Netflix in 2000 would have meant telling investors that the highest-margin revenue in the business was a liability, and doing it while the stores, the leases and the staff to run them stayed on the books. Blockbuster's leadership took the other route and defended the model that was working. That is a familiar shape in this era. Yahoo's decision to pass on Google (https://404memoryfound.com/posts/yahoo-rejected-google.html) and Microsoft's attempt to buy Nintendo (https://404memoryfound.com/posts/microsoft-tried-buy-nintendo.html) both look odd in hindsight for the same reason: the asking price was small, and the thing being sold did not fit the buyer's existing business. ## Did Blockbuster ever try to fight back? It did, and the attempt got closer than the punchline version of this story allows. Blockbuster launched an online rental service of its own and then, in early 2005, scrapped late fees across its stores, which meant giving up the revenue line that had made Netflix unbuyable five years earlier. Late in 2006 the company added Total Access, which let online subscribers swap a mailed DVD for a fresh rental at any store. Netflix could not match that, because Netflix had nowhere to walk into. For several quarters Blockbuster was taking subscribers rather than losing them. It also cost money, and the spending brought in Carl Icahn, who bought into the stock in 2005 and fought the board over strategy and pay. Antioco left in 2007, and the online push was cut back under his successor. Antioco's own verdict came in Harvard Business Review in 2011: "I firmly believe that if our online strategy had not been essentially abandoned, Blockbuster Online would have 10 million subscribers today." He was writing four years after his exit, and one year after the company he had run filed for bankruptcy. ## Who owns Blockbuster today? EchoStar Corporation does, indirectly. Blockbuster filed for Chapter 11 bankruptcy protection in September 2010. In April 2011, DISH Network won the bankruptcy auction with a bid valued at about $320 million and completed the purchase on April 26, 2011, taking the stores, the trademark and the DVD-by-mail operation. DISH kept the brand alive for a few years, ran the mail service, and then shut the remaining company-owned stores at the start of 2014. DISH Network has been a subsidiary of EchoStar Corporation since December 2023, so the Blockbuster trademark now sits inside EchoStar, a satellite and wireless company with no video rental business. The stores are not entirely gone. One franchised location in Bend, Oregon has kept trading since the corporate chain closed, and it now works partly as a shop and partly as a destination for people who want to see one. The last Blockbuster store and how it survives (https://404memoryfound.com/posts/is-blockbuster-still-open-last-store.html) is a separate story from the deal that went wrong in 2000. So the short answer to the question people type: Blockbuster is not in business as a rental chain in 2026. The name is an asset on a telecom balance sheet, and the last store is a franchise that outlived its franchisor. ## What is Netflix worth next to the $50 million it asked for? Netflix reported revenue of $45.2 billion for 2025 in its annual report to the Securities and Exchange Commission, up 16 percent on 2024, with operating income of $13.3 billion and net income of $11.0 billion. The company guided to between $50.7 billion and $51.7 billion of revenue for 2026. Set that against the number in the Dallas meeting. One year of Netflix revenue is roughly 900 times the $50 million Blockbuster declined to pay for the entire company in 2000. The gap did not open because of DVDs by mail. Netflix added streaming in 2007, which removed postage, warehouses and shipping time from the product in one move, and made the store network Blockbuster had spent decades building irrelevant rather than merely inconvenient. Blockbuster's answer to mail-order rental, Total Access, was a good answer to the wrong question. That is the part worth keeping from the story. Blockbuster did not lose because it failed to spot Netflix. It spotted Netflix, was offered it cheaply, and then competed hard against the version of Netflix that existed in 2000 rather than the one that existed in 2010. ## Frequently Asked Questions ### Did Blockbuster really turn down Netflix for $50 million? Yes. Netflix co-founders Reed Hastings and Marc Randolph offered to sell Netflix to Blockbuster for about $50 million at a meeting in Dallas in 2000, and Blockbuster chief executive John Antioco turned the offer down. Randolph published his account of the meeting in his 2019 book That Will Never Work and has repeated it in interviews since. ### Why did Blockbuster fail? Blockbuster failed because its most profitable revenue depended on the model Netflix removed. Late fees brought Blockbuster about $800 million in 2000, roughly 16 percent of revenue, and the company defended that model rather than buying the DVD-by-mail service that had no due dates. Blockbuster scrapped late fees in 2005, cut back its online push after 2007, and filed for Chapter 11 in September 2010. ### Who owns Blockbuster now? EchoStar Corporation owns the Blockbuster brand in 2026, through DISH Network, which bought Blockbuster's assets out of bankruptcy in April 2011 for a bid valued at about $320 million. DISH became a subsidiary of EchoStar in December 2023. The last company-owned stores closed at the start of 2014, leaving one franchised store in Bend, Oregon. **Sources:** - Marc Randolph on Blockbuster rejecting the chance to buy Netflix, Fortune, April 2023: https://fortune.com/2023/04/14/netflix-cofounder-marc-randolph-recalls-blockbuster-rejecting-chance-to-buy-it/ - John Antioco, 'How I Did It: Blockbuster's Former CEO on Sparring with an Activist Shareholder', Harvard Business Review, April 2011: https://hbr.org/2011/04/how-i-did-it-blockbusters-former-ceo-on-sparring-with-an-activist-shareholder - 'Lessons from the Rise of Netflix and the Fall of Blockbuster', Cato Institute: https://www.cato.org/commentary/lessons-rise-netflix-fall-blockbuster - 'DISH Network Completes Acquisition of Blockbuster Assets', DISH Network press release, April 26, 2011: https://www.prnewswire.com/news-releases/dish-network-completes-acquisition-of-blockbuster-assets-120735549.html - Netflix, Inc. Form 10-K for fiscal year 2025, SEC EDGAR: https://www.sec.gov/Archives/edgar/data/1065280/000106528026000034/nflx-20251231.htm --- # Did Yahoo Turn Down Google for $1 Million in 1998? URL: https://404memoryfound.com/posts/yahoo-rejected-google.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-03 Updated: 2026-09-09 Topics: Business Blunders, Internet Culture **Summary:** Yahoo never bought Google, and Google has never bought Yahoo. The famous claim that Yahoo passed on Google for $1 million in 1998 traces to later retellings rather than any record, while the documented near-miss came in 2002, when Yahoo offered a reported $3 billion and Google's founders wanted $5 billion. Yahoo is now owned by Apollo Global Management, and Alphabet was worth about $4.04 trillion in September 2026. **Key facts:** - The 1998 story: Yahoo reportedly passed at $1 million (disputed, no primary record) - The documented offer: Excite, $750,000 cash plus about 1% of Excite, declined by CEO George Bell - The 2002 talks: Yahoo offered a reported $3 billion; Google asked $5 billion; no deal - Owner today: Apollo Global Management (90%), Verizon (10%), since September 1, 2021 - Status: Yahoo runs as a private company; Alphabet worth about $4.04 trillion (September 2026) ## Did Yahoo ever buy Google, or Google ever buy Yahoo? No, in both directions. Yahoo (https://404memoryfound.com/posts/who-owns-yahoo-now.html) never owned Google, and Google has never owned Yahoo. The two were rivals from 1998 onward, business partners between 2000 and 2004, and takeover targets for other buyers after that. The confusion comes from two real events that get squashed into one story. Yahoo paid Google to supply its search results starting in June 2000, so for about four years a search on Yahoo really was a Google search. And Yahoo tried to buy Google outright in 2002, which is the closest either company came to owning the other. Today the two sit on opposite ends of the internet economy. Yahoo is a private company controlled by the buyout firm Apollo Global Management. Google's parent, Alphabet, was worth about $4.04 trillion as of September 9, 2026. ## Is the 1998 $1 million story true? The version that circulates every few months goes like this: in 1998, Larry Page and Sergey Brin offered to sell their new search engine to Yahoo for $1 million, and Yahoo turned them down. Wikipedia's Yahoo entry does carry that claim, dated 1998. Here is the part the viral posts leave out. That claim is sourced to later news retellings, not to a contemporary document, a securities filing, a named Yahoo executive, or an interview with anyone who sat in the room. No public record fixes the meeting, the date, or the price, so the $1 million number is best treated as disputed. The retellings do not even agree on the buyer. Some name Yahoo, some name Excite, and some name the search engine that ruled the web before Google (https://404memoryfound.com/posts/what-happened-to-altavista-search-engine.html). One version does have a first-hand account, and it is not Yahoo's. Excite's chief executive, George Bell, has described his own terms: $750,000 in cash plus roughly 1 percent of Excite, in talks he places around 1998. Bell says the deal died on a condition rather than a price, because Page wanted Excite to rip out its own search technology and run Google's instead. Excite ran the two engines side by side and, by Bell's telling, "couldn't see much difference" in the results. Bell also rejects the tidier explanation that circulates online, that Excite refused because better answers would push visitors off the portal faster. He called that version baloney in a 2014 interview. The fuller account of how Excite handled the offer (https://404memoryfound.com/posts/what-happened-to-excite-search-engine-google.html) is a story of its own. ## Why did Yahoo pay Google to run its search in 2000? This is the piece that gets forgotten, and it is the best documented event in the whole story. On June 26, 2000, Yahoo and Google announced an agreement making Google the default search results provider for Yahoo.com, sitting behind Yahoo's human-edited directory. Google said at the time that it already powered 76 portal and destination sites worldwide, and the two companies expected the integration to be live within 30 days. So Yahoo did end up buying Google, in a sense, one query at a time and on Google's terms. The deal put Google's name in front of Yahoo's enormous audience during exactly the years Google needed the exposure most. Yahoo spent the following four years building search technology of its own and swapped Google out in 2004. By then Google had a brand, a business model, and its own front page habit among users. ## How close did Yahoo come to buying Google in 2002? Closer than the 1998 story, and with more plausible numbers. Under chief executive Terry Semel, Yahoo opened talks in 2002 to buy Google outright. The widely reported outcome: Yahoo put roughly $3 billion on the table, Google's founders wanted $5 billion, and Semel walked. Those figures also deserve a disputed label. They come from books and press accounts written years afterward rather than from filings, and different retellings shuffle which side named which number. What is clear is that the two companies talked, no deal happened, and the gap was in the billions rather than the millions. Google went public on August 19, 2004, closing its first day of trading worth about $27.2 billion. Any window Yahoo had to buy it closed that morning. Four years later the roles reversed completely: on February 1, 2008, Microsoft bid $44.6 billion for Yahoo, and Yahoo's board said no to that too. ## Who owns Yahoo now? Yahoo is owned by funds managed by Apollo Global Management, which holds 90 percent. Verizon kept 10 percent. The deal closed on September 1, 2021 and was valued at about $5 billion, made up of $4.25 billion in cash plus $750 million in preferred interests. At close, the group of properties reported roughly 900 million monthly active users and included Yahoo Mail, Yahoo Finance, Yahoo Sports, AOL (https://404memoryfound.com/posts/does-aol-still-exist-today.html), TechCrunch, and Engadget. The path there was a long markdown. Verizon paid $4.4 billion for AOL in 2015, then agreed to buy Yahoo's operating business for $4.83 billion in 2016. That price was cut to $4.48 billion after Yahoo disclosed its data breaches, and the sale closed on June 13, 2017. The leftover shell, which mostly held stakes in Alibaba and Yahoo Japan, was renamed Altaba and wound down, making its final distribution to shareholders in October 2020. Yahoo the brand kept running through all of it. The acquisitions Yahoo made on the way down are their own story, including the billion dollars it spent on Tumblr (https://404memoryfound.com/posts/what-happened-to-tumblr-yahoo-billion-dollar-mistake.html). ## What did saying no actually cost Yahoo? Run the numbers on the same two companies over 22 years. Google's parent went from about $27.2 billion on its first day of trading in August 2004 to about $4.04 trillion in September 2026. Yahoo's stock peaked at $118.75 on January 3, 2000, near the top of the dot-com bubble, and the whole operating business sold for $4.48 billion in 2017. Put another way, Yahoo's entire operating business fetched roughly a tenth of what Microsoft had offered for it nine years earlier, and a rounding error against the company it once tried to buy. Apollo's $5 billion in 2021 bought Yahoo, AOL, and a stack of media brands for less than the $5 billion Google's founders reportedly wanted for Google alone in 2002. The cheap lesson is that Yahoo missed one meeting in 1998. The more accurate one is that Yahoo had at least three chances, in 1998, in 2000 and in 2002, and each time it treated search as a feature to rent rather than the business itself. ## Frequently Asked Questions ### Has Google ever bought Yahoo? No. Google has never acquired Yahoo or any stake in it. The two were search partners between June 2000 and 2004, when Google supplied Yahoo's results, and competitors before and after that. ### How much did Yahoo offer for Google in 2002? Reported accounts put Yahoo's offer at about $3 billion, against a $5 billion asking price from Google's founders, and the talks collapsed over that gap. Those numbers come from later books and press retellings rather than filings, so they should be read as reported rather than confirmed. ### Is Yahoo still around in 2026? Yes. Yahoo operates as a private company under Apollo Global Management, which bought 90 percent of it from Verizon in a deal valued at about $5 billion that closed on September 1, 2021. Yahoo Mail, Yahoo Finance and Yahoo Sports are all still running, alongside AOL and TechCrunch under the same owner. **Sources:** - Google press release: Yahoo! Selects Google as its Default Search Engine Provider (June 26, 2000): http://googlepress.blogspot.com/2000/06/yahoo-selects-google-as-its-default.html - Internet History Podcast: The Real Reason Excite Turned Down Buying Google (George Bell interview): https://www.internethistorypodcast.com/2014/11/the-real-reason-excite-turned-down-buying-google-for-750000-in-1999/ - TechCrunch: Apollo completes its $5B acquisition of Verizon Media, now known as Yahoo: https://techcrunch.com/2021/09/01/apollo-completes-its-5b-acquisition-of-verizon-media-now-known-as-yahoo - StockAnalysis: Alphabet (GOOGL) market cap and IPO history: https://stockanalysis.com/stocks/googl/market-cap/ - Wikipedia: Yahoo!: https://en.wikipedia.org/wiki/Yahoo! --- # Microsoft Tried to Buy Nintendo and Got Laughed At URL: https://404memoryfound.com/posts/microsoft-tried-buy-nintendo.html Author: Dana Reyes (404 Memory Found) Published: 2026-03-01 Updated: 2026-09-09 Topics: Business Blunders, Gaming **Summary:** Microsoft tried to buy Nintendo in 1999, while the Xbox was still an unreleased project, and Nintendo refused. Former Xbox executive Kevin Bachus told Bloomberg that Nintendo "just laughed their asses off" during the meeting. Microsoft published the approach letter itself in 2021, and shipped the Xbox on 15 November 2001 instead. **Key facts:** - When: October 1999 letter, follow-up meeting January 2000 - Who sent it: Rick Thompson, Microsoft VP of hardware, to Jacqualee Story at Nintendo of America - Price offered: Never disclosed by either company - Also approached: Electronic Arts, Midway Games and Square - What happened instead: Xbox launched 15 November 2001 at 299 dollars ## Did Microsoft really try to buy Nintendo? Yes, and two former Xbox executives have described it on the record. In Bloomberg's oral history of the Xbox, published for the console's twentieth anniversary, Kevin Bachus, who ran third-party relations on the project, said Steve Ballmer sent the team to ask whether Nintendo would consider being acquired. Nintendo's answer, in Bachus's words, was that they "just laughed their asses off." Bachus described it as an hour of somebody laughing at you. The reaction makes sense from where Nintendo sat. In 1999 Microsoft sold operating systems and office software. It had no console on the market, no console retail channel and no first-party game studios worth the name. Nintendo had the Game Boy (https://404memoryfound.com/posts/what-happened-to-game-boy-nintendo-handheld.html), the Nintendo 64, Mario and Pokemon. Nintendo was not the only company Microsoft called. The same reporting says Microsoft approached Electronic Arts, Midway Games and Square while it worried that its unnamed console would launch without enough games. None of those approaches produced an acquisition. Microsoft ended up assembling a launch lineup the slow way. ## What was in the October 1999 letter to Nintendo? Microsoft published the letter itself in November 2021, inside a web museum it built for the Xbox's twentieth anniversary. It was written by Rick Thompson, then Microsoft's vice president of hardware for the Xbox project, and addressed to Jacqualee Story, executive vice president of business affairs at Nintendo of America. The wording is diplomatic rather than predatory. Thompson thanks Story for trying to arrange a meeting with Genyo Takeda and Hiroshi Yamauchi to discuss "a possible strategic partnership between Nintendo and Microsoft on future video game platforms." No price appears anywhere on the page. Microsoft has never disclosed what it was willing to pay, and no filing or court record has produced a figure. The letter also mentions Dolphin, the code name for the console that shipped as the GameCube, and offers Microsoft's help in making Dolphin the best it could be. That is the real tell. Microsoft wanted inside Nintendo's next machine as an owner, a partner or a supplier, in that order of preference. The GameCube (https://404memoryfound.com/posts/why-nintendo-gamecube-lost-console-war.html) shipped in 2001 on an IBM processor and a Panasonic optical drive instead. ## Why did Nintendo turn Microsoft down? Because the pitch asked Nintendo to give up the half of the business it had spent forty years defending. Bob McBreen, head of business development for the Xbox, told Bloomberg that Nintendo executives came to Microsoft's offices in January 2000 to work through a joint venture, and that Microsoft handed over the full technical specifications of the Xbox. McBreen summarised the argument without much varnish: Nintendo's hardware was weak next to the PlayStation, so Nintendo should stick to Mario and let Microsoft build the boxes. Nintendo had spent the previous decade doing the opposite of that. It co-designed its own silicon, chose its own storage formats and policed which games shipped on its systems. That control is the oldest habit the company has. Nintendo has been in business since 1889, when Fusajiro Yamauchi began making hanafuda playing cards in Kyoto, and his great-grandson Hiroshi Yamauchi was still president when the letter arrived. The NES rebuilt the American console market (https://404memoryfound.com/posts/how-nintendo-nes-saved-video-games-1983-crash.html) after the 1983 crash on that exact principle. Nintendo decided what shipped. ## What did Microsoft build after the answer was no? The original Xbox (https://404memoryfound.com/posts/what-happened-to-original-xbox-microsoft-gaming.html) launched in North America on 15 November 2001 at 299 dollars, with a built-in hard drive, an Ethernet port and Halo as the reason to buy one. Microsoft had gone from being laughed out of a meeting to shipping a console in roughly two years. The scoreboard was split. The Xbox sold about 24 million units worldwide before Microsoft retired it, ahead of the GameCube at 21.74 million and far behind the PlayStation 2 (https://404memoryfound.com/posts/playstation-2-best-selling-console-ever.html) at roughly 160 million. Microsoft outsold Nintendo in that generation and still lost the console war, and it lost money on the hardware while doing it. The consolation prize turned out to be the thing Microsoft could not have bought. Building the box forced it to build the studios, the online service and the publishing relationships underneath it, and those became the durable part of the business. An acquired Nintendo would have handed Microsoft a games catalogue in 2000 and taught it nothing about running a platform. Nintendo, for its part, did not need saving. The Nintendo 64 finished its life at 32.92 million units. The DS and the Wii followed, and the Switch had passed 154 million units by Nintendo's November 2025 results, which puts it level with the DS and within reach of the PlayStation 2 at the top of the all-time hardware list. ## Is Microsoft still trying to buy Nintendo? The idea outlived everyone who pitched it in 1999. In September 2023, unredacted documents from the FTC's case against Microsoft over the Activision Blizzard deal appeared on a court portal and were reported by Forbes and others. One of them was a 2020 email from Phil Spencer, the head of Xbox. Spencer wrote that "Nintendo is THE prime asset for us in Gaming" and that "getting Nintendo would be a career moment." He added that Microsoft had held numerous conversations with Nintendo about closer collaboration and believed it was the American company best positioned to try. Twenty-one years after the first letter, the answer was still no, and this time nobody had to say it in a meeting. What the two companies do have is a supplier relationship on Nintendo's terms. Minecraft, which Microsoft owns, sells on Nintendo hardware. In February 2023, while fighting for the Activision deal, Microsoft signed a binding ten-year agreement to bring Call of Duty to Nintendo platforms. Microsoft got to publish on Nintendo's store. It never got the boardroom. ## Frequently Asked Questions ### When did Microsoft try to buy Nintendo? The documented approach is a letter dated October 1999 from Microsoft's Rick Thompson to Jacqualee Story at Nintendo of America, followed by a meeting at Microsoft's offices in January 2000. Both happened before the Xbox was on sale. ### How much did Microsoft offer for Nintendo? No public figure exists. Microsoft published the approach letter in 2021 and it names no price, and neither company has ever confirmed one. Any specific dollar amount you see quoted for the deal is not sourced to Microsoft or Nintendo. ### Did Microsoft try to buy Nintendo again? It considered it. A 2020 email from Xbox head Phil Spencer, made public in the September 2023 FTC document leak, called Nintendo "THE prime asset for us in Gaming." No offer followed, and Nintendo remains independent. **Sources:** - Kotaku: Former Xbox Executives Say Nintendo 'Laughed Their Asses Off' When Microsoft Offered To Buy It: https://kotaku.com/former-xbox-executives-say-nintendo-laughed-their-asses-1846000965 - GameSpot: Microsoft Reveals The Letter It Sent When It Tried To Buy Nintendo: https://www.gamespot.com/articles/microsoft-reveals-the-letter-it-sent-when-it-tried-to-buy-nintendo/1100-6498351/ - Forbes: Unredacted FTC Filing Shows Talks About Potential Nintendo Acquisition: https://www.forbes.com/sites/siladityaray/2023/09/19/microsofts-xbox-plans-leaked-unredacted-ftc-filing-show-talks-about-potential-nintendo-acquisition/ - Nintendo Investor Relations: Dedicated Video Game Sales Units: https://www.nintendo.co.jp/ir/en/finance/hard_soft/index.html - Wikipedia: Xbox (console): https://en.wikipedia.org/wiki/Xbox_(console) ---