Recent discussion regarding Windows 8 sales sats shows a profound misconception, as to the way that the computer market has functioned over the past 30-odd years. And a total ignorance of how companies succeed, or fail.
Everyone, including Microsoft, seems to have concluded that Apple’s success has been based on brilliant, game-changing innovation. Nothing could be further from the truth. If we look at the complete record — not just the usual high points — we’ll see clearly that steady evolution and a favorable cost-benefit ‘value proposition’ — not revolutionary new products — are what actually define the process. As in most other markets.
In hindsight, Apple’s success with the Macintosh seems revolutionary. “Wow: a total game-changer. The Mac allowed Apple to rule the computing industry for thirty years!” But of course, that’s not how it happened.
Apple did rule the computing industry with the Apple ][, which in turn was a nice evolutionary step forward compared to things like the Altair, or teletype-console terminals. The Apple ][ had a vastly better cost-benefit balance than what went before, and this helped it spread beyond business and education into the home. Apple tried to capitalize on the success of Apple ][ with the Apple III, but that computer was such an utter, dismal failure that no one today even remembers it. It was too different, too incompatible, too expensive. It failed to take what worked and move it smoothly forward.
Next, Apple really went for broke, with the Lisa. That thermonuclear failure almost put the company out of business for good. Lots of people admired the Lisa, but at $9,000 or so (in 1980s dollars!), nobody bought it. It was only with the Macintosh that Apple just barely managed to hit the ‘window of opportunity.’ After the Lisa, the Mac seemed somehow less unfamiliar. And it offered a cost-benefit proposition that, while still on the steep side, appealed to a worthwhile segment of users.
What’s tends to be forgotten is how thin that slice of the user pie actually was. The Mac was never mainstream, throughout the first two decades of its life. It was seen as a luxury product, and won devoted loyalty from only about 5% of individual users. It failed utterly, in the all-important business market.
Meanwhile, businesses large and small were following their own steady, intensely pragmatic evolutionary path. CP/M had won near-universal acceptance, as a comfortable step down from the previously dominant big-iron systems. The Apple ][ started to make some headway, but lacked support for things like lower-case letters and 80-column display, while adding features like really bad color, that business had no interest in. With the Apple III and then the Mac, Steve Jobs threw away any hope of success in the business market, in favor of a search for magical user experiences.
It came as no shock that a company with powerful expertise in business was the one to find the next logical stepping-stone for that market. IBM designed the PC to be similar to CP/M. But instead of using CP/M-86, IBM standardized almost on the spur of the moment on a very similar, but more proprietary operating system, PC-DOS. This software was built quickly and cheaply by a small company, Microsoft, and acquired on a non-exclusive basis. Microsoft retained the — seemingly worthless — right to market the software on its own. (CP/M-86 was actually available on the PC, which shows just how strong the line of continuity really was.)
The PC succeeded because it was both familiar and cost-effective. (And partly because it was open. Open not as in public domain, but as in setting an ad hoc standard that many companies could exploit.) It was cost-effective, partly because it eschewed frills that business had no use for. It allowed businesses to run Lotus 1-2-3, WordPerfect and maybe some accounting package, with the minimum overhead (both processor cycles and dollars). With PC/MS-DOS, Microsoft shrewdly defined the thinnest software layer needed to give users ready access to the hardware technology of the day. Today, Microsoft has forgotten that this technical efficiency — not market trickery, or brilliant design, or jingoistic marketing — was the true secret of its success.
The Mac continued to be seen as a luxury-priced toy. No business was going to pay the vast premium required to give its users a ‘nicer computing experience.’ The graphical user interface (GUI) created a huge overhead, exposed features business didn’t need, while offering nothing they did need. Lotus 1-2-3 wasn’t even available on the Mac. (Microsoft eventually did Apple a huge favor by shipping Excel first on the Mac… largely in exchange for not being sued over Windows.)
Ironically, despite being ‘the graphical computer,’ the Mac also failed in the burgeoning electronic games market. Not because it was poorly designed or unsophisticated, but because it was inflexible and overly ambitious. The Mac’s hardwired minimum screen resolution of 640×480 was too high to allow games like the 3D hit Doom to run well. There were simply too many pixels to push, for the available CPU horsepower. The PC, with its VGA ‘Mode X’ 320×240 resolution, sailed past the nominally-superior Mac for real-time animation. When id’s sequel, Quake, ushered in the first wave of 3D accelerators (first from Rendition, then, more successfully, from 3Dfx), the Mac was left behind in the dust. Alas, it had failed to carry forward the geeky and seemingly unnecessary ‘expansion slots’ that had been standard in all previous computers (including Apple’s own).
We are told that Apple next “revolutionized” the world of consumer electronics by inventing the portable digital-audio player. Of course, this never happened. Portable MP3 players evolved from Sony’s WalkMan, which in turn was a portable version of the once-ubiquitous mini-tapedeck. And MP3 players were pioneered not by Apple, but by Rio, Creative Labs, and a number of other now-forgotten companies. Apple built smoothly and cleverly upon that work, and offered a nicer player to customers who already knew what they wanted. Apple also saw a missing link on the content side. It didn’t create music downloads — but did figure out how to monetize the previous non-commercial free-for-all.
A few years later, Apple similarly did not ‘invent’ the iPhone. It merely added well-accepted cell-phone functionality to the already successful iPod. Neither did it invent the handheld computer — at least, not successfully. The Newton crashed and burned just as badly as the Lisa, and for roughly the same reasons. Microsoft, meanwhile, did some conservative evolving of its ubiquitous Windows platform, created a version that could run well on handheld hardware, and had a huge success with Windows CE (later Windows Mobile). Of course, Microsoft then failed to predict the importance of cellular connectivity, and Apple was again able to pull ahead. Not by revolutionizing anything, but by feeling the pulse of consumer demand.
Microsoft also created the tablet, with Apple nowhere to be seen at the time. Again, the change was evolutionary — essentially just a few tweaks to the basic Windows software. Tablets gained some fans, and some vertical applications, but failed to set the world on fire. The niche was simply not wide enough. Again, Microsoft failed to correctly sense the mood of the market. It virtually abandoned all its very successful work in handhelds and tablets, while Apple listened, learned and adapted. Apple realized that users would be happy to accept lower functionality and proprietary limitations (remember MS-DOS?), in exchange for greater efficiency (in this case, portability) and a better cost-benefit ratio (much lower prices). But again, Apple didn’t whip up the iPad out of whole cloth. It simply expanded the already successful iPhone to target the form factor narrowly missed by Microsoft’s more ambitious ‘tablet PC.’
And so, at last, we come to the present. Windows 8. If you’ve followed my little history lesson this far, you should already know why Windows 8 is not just a failure, but a disastrous mistake.
People are comparing Windows 8 to Vista. In fact, it’s more like Microsoft’s Lisa, or Newton. Or better yet, its OS/2… another misconceived operating system I didn’t even mention. Like OS/2, Windows 8 is the answer to a question nobody asked. Like OS/2, it’s needlessly unfamiliar — yet offers a less-favorable cost-benefit ratio than what went before. While the iPad fits its market niche like a key to a lock, Windows 8 tries to force open a new niche where none exists.
Note that this criticism remains valid even if you insist that Windows 8 is well-designed. The Lisa was a thing of wonder and beauty. OS/2 was a clever, powerful piece of software with many excellent features still not available today in any mainstream OS. The Mac was vastly ‘superior’ to MS-DOS. But each was mis-targeted, asking customer to give up too much, learn too much, and pay too much for the benefits achieved.
On top of this, of course, Windows 8 is not well-designed. It’s a two-headed bastard hybrid, birthed into an increasingly sophisticated market that demands specialization and excellence in every niche. But that’s not the main reason it is failing. Windows 8 is a grandiose idea in search of a market that doesn’t exist. It suits hybrid devices that no one asked for, but fails to carry forward the immense momentum of Windows. It forces users to abandon the familiar, and in return offers no genuine advancement in cost-benefit ratio. Windows 8 is not a repetition of the Vista disaster; it’s a repetition of the OS/2 disaster. Only this time, Microsoft is playing the part of IBM — an aging company, attempting a last ill-considered masterstroke, hoping to twist the market to its own needs… while abandoning its previous awareness of users’ needs.
OS/2 failed not because it was a bad product, but because it was the wrong product. Windows 8 isn’t as good as OS/2 was in its time, and is likely to fail even more miserably. If Microsoft doesn’t realize this soon, it could easily end up in a similar state to IBM: a burnt-out husk of a company, with no option but to abandon its leadership position in the rapidly-advancing personal computer space, and fall back on its remaining ability to quietly milk affluent enterprise customers.