A friend and I had a discussion the other day about the dominance of the Windows desktop and if it will stay that way in the future. His contention was that, with the amount of software, the business presence, and the entertainment presence(games, WMA, etc), MS may be weakened, but they would still be the dominant OS on the market.
I, on the other hand, believe that MS has missed it's chance to remain on the cutting edge. Vista will be a vain attemt to repackage Windows, throw in some new features, and shove it all on people making them believe that if they aren't running Vista, they will not be able to take advantage of the latest "innovations" in the industry. I quote the word because Microsoft hasn't really ever been that innovative, they have just done better at repackaging and selling other people's creativity.
Case in point, a recent Slashdot article points out that Office 12 will include the new ability to render documents to PDF. Perhaps I am missing something here, but OO.org has had this for a long time.... and it's not like there aren't a ton of libraries out there to write a PDF renderer, iText and FOP being two that I have used personally in the past. This is not NEW by any stretch of the imagination, just the M$ marketing machine in motion selling people things they can get for free elsewhere. Sales at Microsoft has literally perfected the art of selling snow to Eskimos, wheat and corn to people form Iowa, and sand to Middle Easterners. If M$ has been innovative about anything, it's their sales and marketing departments.
We have to realize that M$ weakness is their current strength, which is proprietary software. Their approach, Windows is our OS and we will write software for our OS, is good for them as long as Windows is dominant. However, there are 3 recent changes in the technology world that, I believe, will change this.
1. At the bottom... Mac going to x86. Everyone knows that Apple/Mac has always done the gui desktop better than Windows. M$ has spent it's entire existence focusing on becoming easier to use than Mac. They still haven't achieved that. In the past, there were two limiting factors to the growth of the MacOS, price and software availablity. As mac becomes available on x86, this price will fall quickly. I believe there will soon be Macs available under $1000. Now that the OS is Unix based, you now have a rock solid OS with the easiest interface in the world to use. This will take over the market for low end users who believe they can't operate a computer if it doesn't say windows as it starts up. The open source and free software movements, for the most part, have filled the gaps as far as software availability. It's only a matter of time before Windows loses this market, which is a mainstay of their users.
2. At the top... Linux. As of now, the user described above would have a very difficult time running a Linux machine. I have used a ton of different distros in my attempt to have a nice Windows alternative for my wife. We settled on Ubuntu. There have been two problems though, one is the hardware is windows specific. I have to do a bunch of fiddling with the drivers to make them work. So linux is not that great of a solution for the low-end user. However, for the developer or advanced user, one who is not afraid of the CLI, most who have tried it have found an excellent replacement for windows. Again, open source and free software have removed the argument of "no available software", so there is a good alternative to Windows for the advanced users.
3. Up the middle... Google to collaborate with Sun on OpenSolaris. While the other parts of that agreement are probably not that big of a deal, I am personally very excited to see what the brainpower at Google does with OpenSolaris. With no real evidence except the usability of the google family of sites, I can see an extremely good gui placed on top of opensolaris to hit the middle-of-the-road user. The one who wants everything to work ot of the box, but is willing to tweak a config file or two. It's all conjecture right now, but I think this is very possible.
As for the other stuff, like games... we have already seen many games for windows ported to Linux(thanks icculus!!!). As Windows loses dominance, we will see games become available on more and more operating systems. Games are fun, but let's face it, IBM won because it inundated the workplace with their machines and people who wanted the same thing at home as they had at work bought them for home use as well. Games don't drive the market. The market has driven games.
Those reasons along with the arrogance of Microsoft a la IBM in the 80s leads me to believe that M$ is headed for the same padded room as IBM.