I'm using the expression "Vista Forever" to refer to the long, slow decline of Windows after Microsoft realizes the only way to continue to make money from it is to cease further development beyond bug fixes, security updates, and drivers for new hardware. It presumes that Microsoft will come to their senses before having invested heavily in yet another major iteration of the platform, and that Vista will be the final one. It also presumes that Microsoft will continue to take your money for a copy of Vista for years to come.
The Windows market hates change. That's how Microsoft got so hemmed in by legacy code in the first place. Once it's developed sufficiently to serve their needs, many Windows users would prefer that it stopped changing, as in 'XP is good enough, why do we need Vista?' and 'Does this mean I'm going to have to upgrade all my software again?' (The answer to the first of these is "because Microsoft spent $6 Billion to develop Vista," and the answer to the second is "not necessarily, although it's quite possible.")
A stable system software environment would mean gradual improvement in the performance of other software, as minor bugs are located and fixed. A stable API (basically the same thing) would mean that alternatives, like Wine, have a chance to mature, and you'd eventually be able to run most Windows software on a variety of other platforms, even without a copy of Windows installed. A stable API would also mean that some software which currently exists only on other platforms would finally be ported to Windows, since doing so would become trivial.
Better yet, Windows Forever inevitably means that a license for a copy of Windows would become cheaper and cheaper, as other platforms in a better position to do so continue to evolve rapidly (for a small fraction of what equivalent advances in Windows might cost), further eroding Windows' market share and forcing MS to compete by lowering prices.
As for Microsoft, they'll continue to suck in some revenue from Windows and - once they've written off the $6 Billion development cost of Vista - post profits. They'll also be in a far better position to concentrate on application software and game machines. They'll be fine, really.
It's a win-win scenario. Cheer up! ;-)