You're painting a fairly incomplete picture here. First off, I'd venture to say that there are probably a few (millions upon millions upon millions) more users/customers in the 2000's than in the 1980's. Current customers are mainly focussed on high level features and usability as functions of the OS whereas anyone running DOS in the 1980's (necessarily computer savvy) not so much.

It takes a lot more developers, resources, and time to support automatic virtual memory allocation behind the scenes, plug-and-play, file indexing across memory and disk, and other modern OS trappings than it does to support printing filenames to the screen when the user types "dir".

Another consideration is that these previous many of these previous OS build on one another or patch one another. The transition from one version of MS DOS to another is better compared to the transition from one service pack of XP to another, in terms of aim and complexity.

A more appropriate timeline would list DOS from 80-95, Windows 3.1 from 92-95, Windows 95 from 95-98, Windows 98 from 98-2000, Windows XP 01-09, Windows Vista 06-09, Windows 7, 09-??

So, Microsoft has DOS out for 15 years, sustaining the hobbyists, elecrtical engineers, wealthy tinkerers, and the smattering of other people with personal computers in an age when that was relatively uncommon. As the internet sensation got in full swing along with the first GUI Windows, Microsoft rushed into development with 95 and then 98, working as quickly as possible to come out with new OS and new features to support the explosion of personal computers and the rapidly increasing hardware demands. In their rush, Microsoft made these OS rather poorly constructed from a design standpoint, so they fused that line of development with the more stable and better designed "professional" lines and made Windows XP. Windows XP was stable (comparatively) and very extensible - quite probably the best OS they have ever released.

As hardware grew far past what anyone but hardcore application developers and hardcore gamers actually needed, the push to release a new OS or be left in the dust faded (also, Microsoft had a strong monopoly on the market, by this time, so it didn't really matter). So, Microsoft took its time and spent years and billions on developing Longhorn (Vista), to be released in 2004. The product ran far behind schedule, with a shifting series of goals and problems and was not released until 2006.

At this time, Microsoft experienced something it had never previously experienced. Commercial users had had time to adjust to their OS product and get it working just the way they liked it. For the first time since it was releasing GUI OS, the majority of users didn't just automatically upgrade. And not only that, but OEM hardware manufacturers staged a mutiny when their sales of new machines plummeted and, for the first time again, OEM's defied Microsoft and continued, for years to offer older OS software on new machines.

Compared with the excitement of previous Microsoft releases, this was a complete flop. I don't know if you recall the behavior of the public for the release of Windows 98 and XP, but there were people lined up at stores to buy those OS. It generated buzz and excitement. Half a decade later, the public's reaction was a collective yawn and, when they discovered that the OS resource requirements, for all intents and purposes required them to scrap whatever existing hardware they had in favor of more expensive hardware, annoyance.

The business world, for its part, didn't even blink at the release of Vista (and that's what really matters to them). This was much different behavior than its reactions to NT, Windows 2000, and Windows XP Pro, all of which were adopted fairly quickly and readily by companies ordering new hardware for the rank and file.

If that's your view, I wouldn't apply for a job in management at Microsoft. Blowing sunshine up their own asses probably isn't going to help sales and profitability.

It's not quite as impressive when you consider that it took Windows XP about a year from its release to have the highest market share of any OS. Vista is working on year 3 and still hasn't achieved half of the popularity of its predecessor.

Now, let's extrapolate. If Microsoft agreed with your rosy assessment of Vista and wasn't rushing a new OS out the door, opting instead to let Vista hang around as long as XP, we could probably expect to see Vista overtake XP by 2013, or 7 years after its release for production. If we compare it to XP, we'd have to look at XP marketshare in 2008. If XP were as much of a "success" as Vista is, the majority of people would have still been using Windows 98 last year, and Dell would still have been offering new computers running Windows 98.

__________________
"Government big enough to supply everything you need is big enough to take everything you have... The course of history shows that as a government grows, liberty decreases."

-Thomas Jefferson