Microsoft vs the Desktop
Microsoft is stumbling all over itself in the desktop OS arena. Not that overpromising and underdelivering on OS technologies is a new phenomenon for MS (Cairo, anyone?), but with Vista, they've reached a new level of self-parody. The core of the problem comes from wanting to share their great plans for the future with the world as soon as they think of them. So barely a year after XP is out, Microsoft is already talking up the great new features of "Longhorn" that are going to revolutionize desktop computing.
Well, whatever good ideas Microsoft had for Vista which remain in the product (basically fast search and 3D accelerated GUI) have already been implemented by Apple, Google, and the Linux community on all three major desktop platforms. The really big ideas for Vista have long since been spun off into other products that run on XP as well (WinFS and WinFX). When Vista finally makes it to production (early 2007 at the soonest), it's going to seem like a lame ape of OS X, and even its vaunted graphics engine will pale in comparison to Novell's free XGL.
Now comes the news that MS is dropping WinFS altogether. Something tells me the reality of an RDBMS filesystem didn't mesh with what executives were looking for, nor did it fit in with Windows' tight coupling to NTFS. Meanwhile the major supposed benefit, speed of searching, turns out to be doable on a plain old filesystem after all, if you aren't relying on the decade-old MS Index Server to do the cataloging, that is.
So what's the lesson? I'm not even sure. Microsoft has shown that with Office 2007, they are able to deliver a major, even revolutionary, upgrade to an existing large product on time and within expectations. Windows is on another scale entirely, though. Maybe it's time to focus more on incremental improvements. If, after five years of development, all Windows users are going to get is a prettier UI and some friendlier control panels, well it's time to take a new approach.