Microsoft's 13 worst missteps of all time

DOS 4.0, Zune, and Windows 8 are but a few of the landmarks among 25 years of failures Redmond-style

Comments

Over the years, Microsoft's made some incredibly good moves, even if they felt like mistakes at the time: mashing Word and Excel into Office; offering Sabeer Bhatia and cohorts $400 million for a year-old startup; blending Windows 98 and NT to form Windows 2000; sticking a weird Israeli motion sensor on a game box; buying Skype for an unconscionable amount of money. (The jury's still out on the last one.)

Along the way, Microsoft has had more than its fair share of bad mistakes; 2012 alone was among the most tumultuous years in Microsoft history I can recall. This year you can bet that Redmond will do everything in its power to prove 2012 naysayers wrong. To do so, Microsoft must learn from the following dirty baker's dozen of its most dreck-laden decisions, the ones that have had the very worst consequences, from a customer's point of view.

[ Find out how much you know about Windows with our Windows IQ test. | Stay atop key Microsoft technologies in our Technology: Microsoft newsletter. ]

Microsoft misstep No. 13: DOS 4.0

In July 1988, IBM and Microsoft released IBM DOS 4.0, and the wheels fell off with data-eating bugs, corrupted disks, and mismanaged memory. Fingers have been pointing ever since. Microsoft's side says IBM botched the testing; IBM's side says Microsoft shouldn't have expected IBM DOS 4.0 to work on non-IBM hardware.

IBM shipped the partial-fix IBM DOS 4.01 in September, but Microsoft took two more months to ship the clearly identified, and differentiated, MS-DOS 4.01. Many people who bought new computers in late 1988 insisted on DOS 3.3, not 4.0 or 4.01. Customers didn't know what to make of the new versions and largely stuck with the devil they knew, 3.3.

Microsoft misstep No. 12: The evil cuties Bob, Clippy, and Rover

Symbolized by a big yellow blob with nerdy glasses, Microsoft Bob -- code-named "Utopia" -- stands as the quintessential Microsoft failure by which all others must be measured. In a departure from the menu-based interface for Windows 95, which was released seven months after Bob, Microsoft Bob's main screen looked like a cartoon living room, with graphic links to a word processor, finance application, calendar, Rolodex, checkbook, and other programs. Click on the grandfather clock's face and the calendar program appeared. Click on the envelope and the email program sprang to life -- Bob cut a special deal with MCI Mail whereby, for just $5 per month, a Bob owner could send up to 15 emails per month, absolutely free.

Customers could add tiles, er, shortcuts to new applications installed on their Bob-ified computer, and the shortcuts' pictures would appear in the immersive Start menu, uh, cartoon living room, inside picture frames or shipping crates. The living room could be personalized with various colors, decorations, and themes. Stop me if this sounds too familiar.

The eponymous Bob himself looked a lot like Bill Gates -- at least, as much as a smiley face with glasses can resemble the world's greatest philanthropist. Microsoft Bob shipped with cartoon helpers bearing distinctive appearances and personalities: Scuzz the Rat ("Couldn't care less about you. Seldom offers help."); a dog named Rover ("Easy to work with, friendly and helpful. Tries to be your best friend."); Chaos the Cat; Digger the worm; Shelly the turtle; Java the coffee-swigging dinosaur; and a half-dozen more.

Released in 1995 and abandoned in 1996, Bob didn't live very long, but he left quite a legacy -- I mean, in addition to his obvious design influences on Windows 8.

The Rover cartoon helper in Bob appeared, unchanged, as the Search Companion in Windows XP's Windows Explorer. Rover and his similarly designed cohorts -- Merlin the magician, Earl the surfer, and Courtney (the courtesan?) -- offered to help perform local searches on the Windows XP desktop.

More insidious, a perky offshoot known as Clippy "It looks like you're writing a ransom note. Would you like help?" appeared in Office 97. Clippy (and the Dot, Hoverbot, the Einstein-reminiscent Genius, Scribble the cat, Power Pup, Links the cat, Rocky the dog, and Will as in Shakespeare) shipped with Office 97, 2000, and 2003, only to be tossed out on his curled ear with the arrival of Office 2007 and its equally cloying Ribbon interface. I note in passing that Steve Sinofsky was in charge of those Clippy-fied versions of Office.

Another big legacy for Microsoft Bob: Melinda French, product manager for Microsoft Bob, became Mrs. Bill Gates in 1994, while development on Bob was in its final stages. Melinda's a very private person, a driving force for the Bill & Melinda Gates Foundation. Even the people who put together their wedding had to sign NDAs.

Microsoft misstep No. 11: Zune, Kin, Courier, Windows Mobile, and missing the consumer mobile boat

These days, most consumers figure that Apple created the tablet market with its first iPad. Not true: Microsoft has been nipping around the fringes of the tablet market for years. You may not know that Apple had an earlier failure -- a portable of sorts, if not a tablet. In 1989, Apple revealed its first portable computer: a 4-inch-thick, 16-pound behemoth dubbed "Macintosh Portable." The lead-acid batteries had to stay in the machine because it wouldn't run directly on AC power. And it cost only $6,500.

After the luggable MacBrick and the Sony Walkman, but before the iPad, came Apple's iPod, and that's where Microsoft's Zune enters the picture. Apple scored big successes with the iPod in 2002 and 2003. Microsoft started designing the Zune -- portable player, Windows software, music service to compete with the iPod and iTunes -- around the same time. By the time Zune came to market in 2006, Apple already had the iPhone under development. The iPhone ultimately rolled over the iPod, the Zune, and just about every piece of portable consumer electronics ever made. After years of sales that hovered near zero, Microsoft officially killed the Zune in June 2012, with the Zune services side morphing into Xbox Music and Xbox Video -- two products best known to Windows consumers for sporting their own tiles on the Windows 8 Metro Start screen.

Microsoft didn't just miss the boat with Zune. It watched and waved and heckled as the boat roared past. In 2007, Steve Ballmer told USA Today, "Now we'll get a chance to go through this again in phones and music players. There's no chance that the iPhone is going to get any significant market share. No chance. It's a $500 subsidized item. They may make a lot of money. But if you actually take a look at the 1.3 billion phones that get sold, I'd prefer to have our software in 60 percent or 70 percent or 80 percent of them, than I would to have 2 percent or 3 percent, which is what Apple might get."

Of course, Apple's income from the iPhone currently exceeds Microsoft's income -- from all sources.

Ballmer was right about one thing. We got a chance to watch consumer markets unfold with phones and music players. Zune failed miserably. Windows Mobile accounted for 15.7 percent of the U.S. mobile market in January 2010, according to comScore, but in two years Microsoft's flagship mobile product slipped to 3.9 percent. Outside the United States, Windows Mobile never amounted to half a hill of beans.

Then there was the Kin, the short-lived square-ish mobile phone with social networking aspirations. In a move reminiscent of "The Cat in the Hat," in 2010, we were offered Kin One and Kin Two. Like Thing One and Thing Two, the Kin was locked in a software box -- a client-server architecture that didn't allow third-party apps -- although the hardware was based on an ARM design. Verizon started selling Thing One and Thing Two in May 2010; 48 days later, Verizon gave up and sent all of its unsold Kins back to Microsoft.

Last month, Wired ran a series of leaked internal Microsoft videos that show pre-release testing of the Kin, and the results were devastating: Testers couldn't figure out how to perform even the most basic functions. If you bought a phone, you would expect to be able to make a call with it, right? Silly mortals. Heaven only knows why Microsoft continued the project.

Then there's the tablet debacle. Bill Gates talked about the Tablet PC in his Comdex Fall 2000 keynote, and Tablet PC general manager Alexandra Loeb gave the details in a Microsoft press release from November 2000:

No compromises. Immersive. Twelve years ago. Thankfully, she didn't mention "fast" or "fluid."

The first Windows XP Tablet PC Edition tablets appeared in 2002 and drew loud applause from a small group of fans, along with skepticism from many corners. Ultimately, the market spoke. Windows Tablet PC hardly rates a footnote.

Did Microsoft learn its lesson? Consider that in 2006, Microsoft hooked up with Intel and Samsung to work on Project Origami, the Ultra-Mobile PC (UMPC) effort that culminated in a small handful of touch-sensitive machines, some of which ran Windows Tablet PC Edition. Slow, clunky, battery-deprived, and almost universally despised, the Origamis hardly broke the consumer surface.

Then came the Microsoft Courier, arguably Microsoft's most innovative mobile device to date: dual 7-inch touchscreens, hinged in the middle as a book; high-resolution (for the time) camera. Nobody knows for sure, but reports say it would've weighed about a pound, sported Wi-Fi, an inductive charger, and a Home button, driven by an ARM Nvidia Tegra processor. The tile-less Infinite Journal interface would look more like a diary and offer touch or stylus access to a contact list, task organizer, free-form drawing program, email, browser, and maybe even an e-reader.

Reports of the Courier began circulating in 2008, just after the Origami went down in flames and shortly before rumors started leaking about the iPad. As best as anyone can tell, the project was killed in 2010, soon after the first iPad shipped. Ballmer spiked it in a shoot-out between the Courier's biggest proponent J Allard and Steve Sinofsky, who was busy consolidating his vision of Windows 8 on a tablet and didn't want to get sidetracked (or sideswiped) by the Windows CE-based Courier. Ultimately, Sinofsky won, the Courier died, and both J Allard and his boss Robbie Bach left Microsoft shortly after, a brain bust that has significant repercussions to this day.

Does Microsoft "get" consumer mobile? A couple of decades and multiple billions of dollars later, that's still a pertinent question. You can draw your own conclusions.

Microsoft misstep No. 10: Bad Windows: Do they run in cycles?

Windows has had some great runs: Windows 3.1/Windows for Workgroups 3.11, then Windows 95 drove most of the computing world for years. Windows 98 came on strong, and Windows 2000 drew some converts.

But then it all went to Hades in a handbasket. Windows Mistake Edition (officially called "Me" for no discernible reason) was like a Windows 98 service pack, with a couple of features and a lot of bugs thrown in for good measure. Windows Me was, deservedly, the last version of Windows based on the old Windows 9x kernel, which at its heart ran on DOS.

Whistler, better known as Windows XP, came as a breath of fresh air. Built on the Windows NT kernel, XP's reign ran unopposed from its release in 2001, until the appearance of Vista in 2007. Segue to the hisses and boos.

Vista, I'm told, is still listed in the Encarta dictionary as a profane word (or it would be, if Encarta hadn't bit the bucket in 2009). Released to businesses in 2006 and consumers in 2007, Vista tried to improve on XP's security, but only succeeded in alienating an entire generation of Windows users.

Then came Windows 7, and all was right with the world once again. Proponents of the "good Windows, bad Windows" analysis claim that consumer versions of Windows run from great to lousy and back again. That's certainly a discernible trend from Windows 98 onward.

Microsoft misstep No. 9: Windows Ultimate

Windows Vista shipped in several versions, the most extravagant of which was Windows Vista Ultimate. Combining all of the features of Vista Business and Vista Home Premium, it also added the exclusive Windows Vista Ultimate Extras.

Ahem: "Windows Ultimate Extras are programs, services, and premium content for Windows Vista Ultimate. These features are available only to those who own a copy of Windows Vista Ultimate."

That's the promise. The reality was less impressive. Ultimate contained Windows DreamScene, a handful of videos to be used as desktop wallpaper; Hold 'Em Poker; three extra sound packs; and a puzzle game starring a robot -- and was later released free. The, uh, loyal Microsoft customers who spent an extra $100 for Vista Ultimate got burned big time.

Microsoft learned its lesson. Windows 7 Ultimate, under Sinofsky, only promised -- and delivered -- a combination of all the features in Win7 Home Premium and Professional. The loyal customers were never compensated.

Microsoft misstep No. 8: Windows Genuine Advantage

What does the then-richest software company in the world call its copy protection scheme? Why, "genuine advantage," of course.

In the halcyon days of Windows 95, 98, and 2000, anyone installing Windows had to provide a serial number. The serial numbers weren't checked against a master database, so the same serial number could be used to install multiple copies of Windows.

With the consumer version of Windows XP, Microsoft added a validation step, where a specific serial number was matched against a key constructed from serial numbers from PC components. The resulting combination is checked against a central database. Changing your motherboard or network card, or in some cases other combinations of hardware, triggered a revalidation. Fail the validation, and Windows wouldn't work -- you had to call Microsoft to beg for forgiveness.

Hardware manufacturers had a different process for activating machines before they shipped. Volume licensees were given serial numbers that worked in bulk.

Windows XP SP1 cut customers some slack: It added a three-day grace period, so you could continue to use your Windows PC for three days while trying to get it reactivated.

Then the offal hit the fan. Over the course of several years, in many confusing steps, Microsoft rolled out its Genuine Advantage program to all Windows XP and later versions.

The earliest versions of WGA included a program called WGA Notifications that runs whenever you log on and validates your Windows license. They also included an ActiveX control that validated your license every time you tried to download specific Microsoft products, including Internet Explorer 7 and Microsoft Security Essentials. You were also prohibited from using Windows Update to get security patches (although you could download Critical patches manually). WGA reached its nadir in Windows Vista, where failing the WGA check meant your machine was put into "reduced-functionality mode," where you could go onto the Web for an hour at a stretch, before Windows locks up.

Predictably, the validation servers borked; false positives came to light. Microsoft got caught shipping data from unsuspecting PC users to the Microsoft servers once a day. Windows Genuine Spyware became the nom de guerre.

Howls, lawsuits, threats, cracks, and generally furious customers eventually drove Microsoft to backtrack. Reduced-functionality mode evolved into a paper tiger, with harmless balloon notifications about ungeniune wayward ways, and an irritating tendency to turn the desktop wallpaper black.

In Windows 7, WGA became Windows Activation Technology, no doubt to avoid the venom aimed at its predecessor. The newly renamed copy protection scheme retains its toothless nature, even in Windows 8. Microsoft can learn from its mistakes.

Microsoft misstep No. 7. Branding

I don't understand why Microsoft has such a terrible time with branding.

Some of the customer-confusing moves are just silly. In 1991, Microsoft jumped Word for Windows from Version 2 to Version 6, with no intervening version numbers. Word 2.0 was part of Office 3.0, but Word 6.0 was part of Office 4.0. When Microsoft released Office 95, it increased the version numbers of all of its components to 7.0: Word 7.0, Excel 7.0 (also called Excel 95; there was no Excel 6), PowerPoint 7.0 (there were no Versions 5 or 6), and Schedule+ 7.0, which morphed into Outlook.

Some of the branding is worse than silly. Stupid changes made years ago continue to confuse Microsoft users today. Consider how many email clients Microsoft supports at this moment. I can think of seven:

I could rant about most of those branding missteps, but allow me to concentrate on the last one. Can you believe that Microsoft intentionally threw away the name "Hotmail" -- one of the most widely recognized brand names in the world, right up there with Coca-Cola and McDonald's and Toyota and, yes, Microsoft -- turning it into something absolutely nobody understands?

That's worse than crazy. It's self-destructive.

Case in point: Microsoft's attempt to create a database of user IDs and associated data, including individuals' financial information. It all started with Hotmail: sign up for a @hotmail.com account, and Microsoft stored the information you provided.

Then your Hotmail account suddenly became a Microsoft Passport, a fancy name for a single-login authentication service that Microsoft hoped would take over the Internet: Everywhere you go, the marketing material touted, you could log on with your Hotmail account, er, Microsoft Passport.  And if you put your credit card info into Microsoft Wallet (earlier called Passport Express Purchase), buying items anywhere on the Web would only take a couple of clicks.

The privacy groups went ballistic -- like the Joker, only less civilized.

Microsoft bobbed and weaved. To its credit, Microsoft hired one of its most vocal critics, to put reasonable limits on privacy incursion. The name of the service changed -- branding and rebranding as Microsoft Passport Network, MSN Passport, .Net Passport, Windows Live ID. Now we know it as a Microsoft account, and it's baked into Windows 8. The personal financial information came and went, and the terminology/branding seems to have stabilized: If Apple can have one Apple ID or Google a single Google account, you can certainly have a single Microsoft account -- if you can figure out how to get your old Zune, ZunePass, Xbox, and Windows Phone accounts switched over to a Microsoft account.

There's that branding thing again.

Microsoft misstep No. 6. Windows Live

Which brings me to Live. In November 2005, Microsoft announced Windows Live and Office Live, kicking off one of the most confusing branding exercises in the history of international commerce. Decades from now, B-schools will be using Windows Live as an example of branding gone insane.

At first, Windows Live drew on MSN-branded software -- MSN Messenger became Windows Live Messenger, for example; MSN Hotmail became Windows Live Hotmail -- and added an online security scanner, the OneCare antivirus package, and a method for sharing your Internet Explorer favorites across multiple machines. From the beginning, Windows Live was an odd hodgepodge of websites, Web-based applications, and PC-based applications, with a browser add-on tossed in for good measure.

Then it grew. And grew. In August 2006, Microsoft unveiled to testers a website called Windows Live Essentials. Microsoft pulled down the site shortly after, replacing it with a similar site called Windows Live Installer (yes, it was a website) that downloaded Windows Live Messenger, Mail, and Writer. In October 2008, Microsoft announced that Windows Live Installer (the website) would become Windows Live Essentials (the website and software package), and Windows 7 would not ship with Windows Mail, Photo Gallery, or Movie Maker. Instead, Windows 7 would give customers links and encouragement to download and run those applications and several others. At the time, it wasn't clear if Microsoft was yanking those applications out of Windows in order to get Win7 as a whole shipped on time or if the programs were removed over antitrust concerns -- possibly both.

Over the course of many years, Windows Live included about 50 different products, few of which talked to each other, with absolutely no discernible common objective. Basically, whenever the powers that be decided to publish a new program -- online, local, website, whatever -- they branded it "Live" and kicked it out the chute.

Earlier this year, Microsoft started dropping the "Live" moniker, setting off yet another round of branding confusion. Windows Live Essentials became Windows Essentials, and Windows Live SkyDrive became just plain SkyDrive. But Windows Live Mail, Messenger, and Writer still have "Live" in their names.

Microsoft misstep No. 5: Windows 8

Windows 8 is turning out to be a misstep of unprecedented proportions. The fundamental problem, as many of us at InfoWorld have described repeatedly, is the Jekyll-and-Hyde nature of the Frankenstein interface. Clearly, with Windows sales heading deeply down and the entire PC hardware industry going with it, Microsoft had to do something. Relegating the old-fashioned Windows desktop to a tile on a new mobile phone app hasn't done the trick.

Or maybe we're all wrong and Windows 8 will breathe new life into the PC industry. In another six months, we should know for sure.

Microsoft misstep No. 4: Windows 8 branding

There's nothing inherently wrong with Windows 8 -- it's a noble (if flawed) attempt at bringing Windows into a more mobile future. There's plenty inherently wrong with Microsoft branding. When you bring the two together, the combination's absolutely lethal.

Please. I spent a whole day trying to come up with a worse name than "Windows RT." I couldn't do it. Microsoft could call the Metro-side-of-Windows-8-plus-Office-2013-RT just about anything, and it would be better than Windows RT.

So many people right now are so confused about the differences between Windows RT and Windows 8 that we won't hear the end of it for another decade -- or longer. Customers who buy a Windows RT tablet, thinking they were getting Windows, will be sorely upset. I bet the return rate for Windows RT tablets at the retail level goes well into double digits, simply because of the naming confusion.

Then there's Metro: immersive, Modern UI, Windows Store. I don't even know what to call it any more. I can imagine telling my Aunt Mildred that she needs to get a Modern UI Calculator from the Windows Store. I mean, Microsoft itself stopped using the term "UI" nearly a decade ago: It's always UX. You can call Metro apps "Windows Store apps," but not all apps in the Windows Store run on the Metro side (for example, Office 2013 is in the Windows Store). And not all Metro apps originate in the Windows Store -- the Metro Xbox Music app, for example, isn't really a Windows Store Xbox Music app because you don't download or install it from the Windows Store.

There's another Win8 branding inanity as predictable as tomorrow's sunrise. The next version of Windows will probably be called Windows 9 -- cool. Here's what I want to know: What will the next version of Windows RT be called? Windows 9 RT? Windows RT 2.0? Windows SU?

Microsoft misstep No. 3: Missed opportunities in the cloud

In the mid-1990s, Windows senior VP Jim Allchin and Internet Platform and Tools division senior VP Brad Silverberg crossed swords many times, with Silverberg pumping for faster expansion into the cloud and Allchin more intent on building on Windows' success. Allchin won, Silverberg left, and by 1999 the die had been cast.

In the mid-2000s, Windows president Steve Sinofsky and Microsoft Chief Software Architect Ray "of course we're in a post-PC world" Ozzie also had their differences of opinion, with Ozzie pushing hard to expand quickly in the cloud, and Sinofsky more intent on building on Windows' success. Sound familiar? Sinofsky won, Ozzie left, and by 2010, Microsoft had lost its second high-level visionary cloud advocate.

In the intervening 10 years, Microsoft embraced the cloud, but did so with one foot firmly entrenched in Windows and the other in Office. Innovative cloud designs, like Mesh, have been tossed aside, while me-too cloud products like SkyDrive garnered a big budget.

Microsoft's foray into the online advertising market, with the Bing search engine, hasn't gone particularly well in spite of Microsoft's successful attempt to stack the deck in Bing's favor on new Windows computers.

Perhaps the biggest cloud shortcoming for Microsoft, at least from a consumer point of view, is its inability to build an ecosystem that comes anywhere close to the Apple, Android, or even Amazon offerings. Apple's taken an enormous lead in the consumer cloud, with Android scurrying to catch up, and Microsoft not yet in the running.

Microsoft misstep No. 2: Management musical chairs

I've talked about the Microsoft management musical chairs in a series of InfoWorld Tech Watch posts, most recently "Game of thrones: The men who would be Ballmer." Suffice it to say that all of the people capable of providing a steady transition from the reign of Ballmer have left the company.

Jim Allchin. Brad Silverberg. Paul Maritz. Nathan Myhrvold. Greg Maffei. Pete Higgins. Jeff Raikes. J Allard. Robbie Bach. Bill Veghte. Ray Ozzie. Bob Muglia. Steve Sinofsky. They're all legends, in their own way, and Microsoft had many more.

We still have some luminaries. Andy Lees survived the Sinofsky purge. Paul Maritz is still around, having spent years at VMware. Bill Veghte's at HP. There are others still at Microsoft, but most lack the experience to play in that league.

The lack of senior management depth may turn out to be Microsoft's biggest misstep in the early 2010s.

Microsoft misstep No. 1: Internet Explorer 6

Microsoft's greatest misstep of all time? Internet Explorer 6. Consider:

Chances are good that more Windows computers have been infected via Internet Explorer 6 than by any other vector. Flash and Adobe Reader may come close, but IE6 is up there. ActiveX, IE6's evil toady, deserves its own ring in developer hell.

In the process of deploying IE6, Microsoft ran afoul of U.S. antitrust laws. The repercussions of the DoJ action resonated throughout Microsoft's product line for more than a decade, driving all sorts of design decisions that were at least partially influenced by antitrust concerns.

Microsoft lost an enormous amount of public goodwill over IE6. It's as if, suddenly, the average Windows user started to understand that their computer was at risk because of a bad piece of Microsoft software. Web developers did, and do, hate IE6, with its fussy quirks, outright bugs, and absolute disdain for anything reeking of a standard.

Microsoft took more than five years to ship an upgrade -- most likely the biggest misstep of all.

Related articles