Software is a funny thing. You can’t touch it, but you can certainly pay for it. It weighs nothing, but takes dozens or even hundreds of people to make. And sometimes it just doesn’t work out.
No matter how detailed your specs, how solid your market research, and how optimized your code, there are always winners and losers. Sometimes a competitor comes to market with something better. Other times, unforeseen circumstances make your program useless or outdated. On occasion, companies just screw things up to a point that no patch can fix it.
Let’s crawl through the annals of computing history to showcase software releases that went horribly wrong. Some of the biggest names in the industry show up here—Apple, Google, Facebook, and Microsoft are well-represented—which proves that there’s always a second chance. And a third and a fourth, if you play your cards right. For some of the other development houses on this list, though, their software flops killed their businesses.
In the early days of video gaming, the cost of software was high because it came in cartridge format. Unlike cheap floppy disks, companies like Atari had to shell out for chips and casings for every game they made.
So when the company produced 5 million copies of a game based on Steven Spielberg’s mass hit movie, it represented a serious sunk cost. The game, produced in a staggeringly short four weeks by veteran producer Howard Scott Warshaw, wasn’t very good and less than half of those 5 million copies were sold. It helped usher in the great video game crash of 1983, which would take the industry years to recover from.
In 2014, the (now-defunct) Microsoft Xbox Entertainment Studios chased rumors that Atari dumped 14 truckloads of unsold cartridges and other equipment in a New Mexico landfill in 1983. They were right, and there’s a documentary about it called Atari: Game Over. At least one of those cartridges was sent to the Smithsonian.
Lotus Jazz (1985)
To be fair, Lotus 1-2-3 was one of the elite productivity products of the 1980s. The spreadsheet program was a huge selling point for IBM PCs, letting users collate and compare data faster and more efficiently than ever before. It made Lotus a huge success—making more money than Microsoft at the time. And then it tried to come up with a follow-up and suffered one of the worst sophomore slumps in computer history.
Jazz was a productivity suite folded into a word processor, spreadsheet, and database program for Macintosh systems. Released in 1985, it retailed for a staggering $595, came on four floppy disks that had to be switched and swapped while you ran the program, and it was a tremendous bomb in the marketplace.
As we noted in the November 1987 issue of PC Magazine, this “klutzy multifunction program…fell well short of Lotus’s reputation for fast, innovative highly functional software.” Its black-rubber packaging, with a vest pocket holder for the disks, is a collector’s item today.
GNU Hurd (1990 – Present)
(Image: Carlosar/Wikimedia Commons)
Unix was first developed in the 1970s, and by 1990 the GNU Project decided it was time to replace it with a free offering called GNU Hurd. Thirty years after work on the project started, GNU Hurd has yet to be released as a working operating system for public use. Still, many of the components from GNU were moved over to create the Linux operating system.
Copland (1994 – 1996)
Apple had tried and failed to replace its aging System 7 operating system with a new system that was more stable and supported multitasking. By 1994, work on System 8 had begun under the Copland codename, but the project soon fell into ruin. Ballooned by bad project management and feature creep, Copland became a collection of features rather than a working OS.
The Copland project was officially cancelled in 1996, but it wasn’t all bad for Apple. The company went out and acquired Steve Jobs’ company NeXT and third-party system enhancements to make a new operating system. Features originally developed for Copland were eventually added to future OS releases, and the foundation of the modern macOS was built.
Microsoft Bob (March 1995 – Aug. 1995)
It’s undeniable that Microsoft has been one of the leading forces in how we use computers, but you also can’t deny that much of Windows was swiped wholesale from Apple’s macOS. When Bill Gates & company decided to do their own spin on a user-friendly operating system in 1995, it proved to be an utter disaster.
Microsoft Bob was a bizarrely cheerful newbie-focused OS that retailed for nearly $100 and featured a number of loquacious “guides” who chatted you through your experience. It was heavily advertised, but the hardware requirements were high and the included productivity applications were primitive. A bit of trivia: the widely reviled Comic Sans typeface was originally designed for Bob but never used.
OpenDoc (May 1996 – March 1997)
Modularity is key in many software projects, and Apple wanted to get in on that in the late 1990s. The resulting OpenDoc seemed like a strong idea: create solid, reusable single-task components and let developers combine and intertwine them to build applications.
Unfortunately for everybody involved, those components were incredible memory hogs—just the text editor alone required 2MB of RAM, which was a huge amount in 1997. In addition, documents created in OpenDoc weren’t compatible with other major software platforms, making it difficult to share and edit your work. Apple killed the platform less than a year after launch, unwilling to throw any more money down the hole.
Huge video game hits like Doom, Quake, and Wofenstein 3D all had one thing in common: designer John Romero. So when he left id Software to co-found Ion Storm in 1996, expectations were high. Daikatana—a first-person shooter about a swordmaster who travels through time—was meant to be the new company’s first big release, but nothing went right.
The game was moved to a new engine six months in, leading to considerable delays and development challenges. Understaffing became an issue when programmers jumped ship to get out of the impending failure, many moving to the Austin office to work on the upcoming hit Deus Ex. Making matters worse, the game’s controversial advertising campaign centered around John Romero annoyed gamers amid delays.
In order to fund development, Ion Storm sold a controlling interest in the company to publisher Eidos Interactive. When Daikatana finally released in 2000, it was three years late, had outdated graphics, annoying gameplay features, and could not break 10,000 units sold after its first month. Romero left the company the following year and Ion Storm was shut down by 2005.
Netscape 6 (2000 – 2002)
The browser wars were a hard-fought battle in the late 1990s, with Internet Explorer and Netscape Navigator the two main competitors. In 1997, Microsoft released Internet Explorer 4 and bundled it with Windows installs. It was easily the best version of its browser yet and the world waited for Netscape to respond. And waited. And waited.
Meanwhile, Netscape had released the core of its browser as open source, which would eventually become Firefox. That didn’t help it get a new version out, though, and when it finally shipped in November 2000, Navigator 6.0 (they skipped 5.0 in the interim) was a bloated, buggy mess that didn’t even run on midrange PCs of the time. That disastrous release marked the start of Netscape’s decline into the dustbin of browser history.
Internet Explorer 6 (2001 – 2008)
Remember when Internet Explorer was the best internet browser available? That all fell apart once Microsoft beat out Netscape for supremacy. IE became complacent with no direct competition anymore, and the result was the widely used, but eternally maligned Internet Explorer 6 released in 2001.
This version of the browser did not support modern web standards, which essentially stagnated web development for several years. It also contained numerous security vulnerabilities that Microsoft was slow to patch, with many issues never being fully resolved.
Thanks to these complaints, it made things easier for Mozilla to release Firefox the following year and gain footing in the market. By the time Microsoft released IE7 in 2006, there was real competition again.
Palm OS Cobalt (Jan. 2004 – June 2005)
(Image: Bob Russell/MobileRead)
Long before smartphones hit the market, PDAs dominated the realm of pocket-sized tech gadgets. PalmPilot was the top dog in the ’90s, but a series of corporate mergers and restructuring ultimately sunk the company.
When it came time to update its aging operating system, Palm’s OS Cobalt was devised to lead the company into the future. By this point, the world had moved on from the PDA and Cobalt was unable to secure any licensees in 2004. Palm then decided to focus its attention on developing a Linux OS, but that never materialized and the company no longer exists today.
Joost (2007 – 2012)
Video-streaming services are all the rage these days, but in 2007, two nice European boys were pushing their own platform to let you watch stuff on your computer, and it was called Joost.
The company raised $45 million and racked up a ton of content providers. Using a peer-to-peer system, Joost played video content in a dedicated client. Unfortunately, that P2P network turned out to be a non-starter and the content discovery model was almost unusable.
Downloaders couldn’t get to the shows they wanted to watch, and most closed the app after just a few minutes. The company crashed and burned, first transitioning to a white-label video-hosting service and eventually selling off its assets in 2012.
Windows Vista (2007 – 2016)
Operating system releases are a big deal. Microsoft and Apple put a ton of labor into each iteration of Windows and macOS, and there’s a lot riding on them. So it’s amazing that Windows Vista was such a horrific blunder.
Designed to replace the aging Windows XP in 2007, Vista failed at just about every possible benchmark. It was bloated—50 million lines of code compared with XP’s 40—and buggy; tons of pre-existing apps didn’t even work in it. Long-time users complained that the OS removed features they’d come to love.
Most of all, people just didn’t see why it was necessary. XP had an installed base of some 800 million systems, and those users were still pretty happy with it.
iTunes Ping (2010 – 2012)
The essential Apple conundrum is that the company somehow manages to balance hardware perfection with software incompetence. iPods and iPhones are brilliant. iTunes is an absolute nightmare. So when the company announced that it was getting into the social-networking space in 2010 with Ping, users didn’t have high hopes.
The software was intended to match people with similar tastes, but you could only select three genres of music to be interested in. To make matters worse, just a day after launch, Ping was already inundated with spam accounts and people pretending to be famous musicians to promote themselves. Apple shut it down in 2012.
Yahoo! Screen (2011 – 2016)
Yahoo! Screen began life in 2006 as a video-hosting service called Yahoo! Video before transitioning to streaming content with a new name in 2011. The new venture launched with just eight original shows and a paltry supply of additional content. You probably remember the service for giving Community one more season or for streaming an NFL game.
By 2016, no one was using the service and Yahoo! shut down the service following a $42 million write-down. Yahoo! instead partnered with Hulu to create Yahoo! View, which would stream recent episodes of ABC, NBC, and Fox shows until 2019.
Google+ (2011 – 2019)
The early 2010s were a big year for tech companies starting their own social media platforms, and Google was no stranger to the endeavor. The company had already tried to launch multiple ventures in the space, each less successful than the last. Instead of a “network,” Google+ acted more as a “layer” across all Google Drive, Blogger, and YouTube.
The service launched in 2011 with features that could (hypothetically) challenge Facebook’s dominance. The only problem was no one used it. Though user growth was initially said to be competitive, it was eventually determined that these numbers were bloated by anyone signing up for any Google service.
With user engagement averaging mere minutes or even seconds, Google significantly redesigned the service to cut down on the clutter and shift it toward an “interest-based social experience.” This did nothing to actually make anyone want to use the platform, so it was slated to be shut down.
Google+ is probably most remembered for exposing the personal information of 52 million users, causing the company to turn the lights off four months early in 2019.
Windows 8 (2012 – 2013)
Microsoft followed Vista with the much-loved Windows 7, but later fell short once again with Windows 8. Having missed the boat on the tablet boom, Microsoft tried to make up for it with an OS for desktop and mobile devices.
As innovative as it may have been, Windows 8 introduced a series of design changes that made it tablet-friendly but difficult to navigate on desktop. Eliminating the Start button was a non-starter for many, and adoption was slower than what was projected.
Couple this with the fact consumers were starting to buy more phones and tablets than home computers, and Windows 8 performed terribly. There’s a reason why Microsoft quickly reacted to criticism and released Windows 8.1 as a free update the following year. People just wanted their Start button back. With Windows 10, Windows 8 is all but a memory.
Facebook Home (April 2013 – Jan. 2014)
Not happy with existing on your phone, Facebook decided it wanted to become your phone. The Facebook Home app was devised as a means to add a Facebook skin over your Android phone to make using the company’s services even easier. This app was pre-loaded with the HTC First, the “Facebook Phone” as people called it, but it was only compatible with a small selection of other devices.
Facebook Home made it harder to use your non-Facebook apps, but it’s biggest issue concerned security. One feature allowed you to use Facebook through the lock screen without ever needing to input a password. While this seemed convenient, we noted in our review at the time that it essentially allowed anyone to use your Facebook account.
The HTC First ended up bombing pretty hard, and Facebook Home availability was never expanded. In fact, Facebook backpedaled to make using other apps easier before calling it quits.
go90 (2015 – 2018)
Among the many blunders in Verizon’s foray into media was the go90 streaming service. Launched in 2015, the service was aimed at mobile-first millennials, but that seemed to be the extent of the plan. Content was hard to find, the user experience was bad, there was an unnecessary social component, and it had nothing to make it stand out.
A year later, Verizon acquired a competing service called Vessel and used the company to improve go90, with little effect. By the time it was all said and done, the company had spent $1.2 billion on acquisition, licensing, and production costs. It was officially shut down by Oath (remember that?) in 2018, and its original programing has since found its way to Tumblr.
Artifact (2018 – 2021)
Valve’s PvP digital collectible card game, Artifact, had everything it needed to succeed. It was based on the wildly popular Dota 2 and designed by Magic: The Gathering creator Richard Garfield. Before the game was even released, Valve had high hopes for it as a competitive esport and planned a million-dollar tournament.
Unfortunately, the game was ruined by its monetization model and could never recover. Unlike other games in the genre, Artifact was not free-to-play. You also couldn’t trade cards with other players, so you were forced to buy new card packs to build a better deck in what was considered a pay-to-win scheme.
Within two months of release, player count dropped by 95% and viewership on Twitch fell 97%. Valve tried to fix things by redeveloping the game in 2020, but work on Artifact 2.0 was stopped in 2021 and both versions of the game were made free to play.
Quibi (April 2020 – Dec. 2020)
Don’t you just love ingesting content in short bites? That was the idea behind Quibi, which was intended to serve busy millennials during their commutes or while waiting in line for coffee. Too bad it launched at the beginning of the COVID-19 pandemic when people stopped commuting and leaned into binge watching. Making matters worse, the app initially had no ability to stream content to televisions.
Despite millions of dollars in investments from Disney, Fox, NBCUniversal, Sony, Time Warner, Viacom, and more, and a wide spectrum of original content, Quibi tanked. A week after launch, the app fell out of the top 50 iPhone apps. By June, projections of 7.4 million users were downgraded to 2 million.
Hit with a lawsuit over stolen patented technology, and no buyer in sight, Quibi shut down just eight months after launch. The service’s library has since been purchased by Roku for its lineup of originals. Founder Jeffrey Katzenberg blamed COVID for all this, but in our review, we weren’t convinced it was the only issue.