From John Gruber’s article about Windows 7 adoption:
The truth is that no one really knows why Vista fared so poorly in the market. It defies a simple explanation.
I’ll try to provide one: Most people had no reason to upgrade, and plenty of reasons not to. I’d guess that Gruber’s theory stated earlier in the article is spot on:
What if the reason why most PCs are still running XP has nothing to do with whether Vista is “good” or “bad”, but rather is the result of indifference on the part of whoever owns these untold millions of XP machines, be they at home or in a corporate IT environment. I.e., that switching to Vista, regardless of Vista’s merits, seemed like too much work and too much new stuff to learn; that the nature of the PC as a universal commodity is such that most of them belong to people who value “old and familiar” more than “new and improved but therefore different”.
I think this explains Vista’s poor adoption, but as part of a much bigger problem: the lack of a reason for regular people to ever upgrade to any new OS release.
Our industry has collectively taught average people over the last few decades that computers should be feared and are always a single misstep from breaking. We’ve trained them to expect the working state to be fragile and temporary, and experience from previous upgrades has convinced them that they shouldn’t mess with anything if it works. They’ve learned to ignore our pressures to always get the latest versions of everything because our upgrades frequently break their software and workflow. They expect unreliable functionality, shoddy software workmanship, unnecessary complexity, broken promises from software marketers, and degrading hostility from their office’s IT staff.
When we tell them that the new OS is faster and better, only to have the upgrade break a piece of software that we don’t care about but they really do, we burn our likelihood that they’ll ever willingly upgrade again. Every time we tell them that they can now easily edit video or make DVDs, only to have them abandon their first effort in frustration and never attempt it again because our software sucks, we drive them closer to indifference or resentment toward future technology.
So when our nontechnical aunts refuse to upgrade from their Pentium III PCs with Windows 98 that work well enough for them and are set up exactly how they prefer, we have nobody to blame but ourselves.
The upgrade market for average PC owners is dead. We killed it.
It was never very strong to begin with, because most people don’t know or care about OS updates, as Gruber pointed out. They get the new OS whenever they buy their next computer, because we don’t give them a choice.
What’s driving computer sales? Well, in the ’80s and ’90s, we were making huge hardware advances that affected a large portion of normal people. Every few years, people needed to upgrade their computers for a new killer app or a massively improved generation of their existing apps.
But the rate of such changes that are relevant to average people has plummeted in the last decade. Graphical interfaces, multitasking, SimCity, porn, email, shopping, and dating sold a lot more new computers than nearly anything we’ve come up with since 2000 except malware. (I honestly believe that malware carried computer sales for most of the last decade. That only worked because we’ve taught people, with a combination of misinformation and omission, two great lies: that computers slow down over time, and that the only way to fix a malware infestation is to buy a new computer.)
Hardware, software, and the internet have all reached mature plateaus of dramatically slowed innovation. In 1998, when everyone was happily using long filenames and browsing the internet and playing their first MP3s and editing their first scanned photos to email to their relatives, a five-year-old computer couldn’t easily do any of these things.
But what common tasks in 2009 can’t be accomplished by a 2.8 GHz Pentium 4 PC with Windows XP SP2 and a cable internet connection — the average technology of 2004? Not much that regular people actually do.
We’re still burning their trust, time, and money, but we’re offering much less in return. It shouldn’t surprise any of us that they stopped caring.