Microsoft continues to confound. They’ve officially rolled out their x86_64 lineup (XP and Server 2003 x64) with very little fanfare and with a tepid response from all but the most jubilant enthusiasts. Lack of driver support is still an issue, a year and a half after the first beta. They informed us that they changed their hundreds of servers over to x64 variants, which some /.ers thought was ballsy, but really wasn’t, since the betas themselves were pretty stable. Microsoft doesn’t really have to worry about unsupported hardware, gaming issues, and the sort of application support that home users worry about. For server applications, x86_64 has long been reliable, ever since 64-bit ports of Linux appeared in late 2003. For database applications, &c., the extra computing power is already paying performance dividends.

These Windows releases are essentially just a stepping stone to Longhorn, which, so I read, will be better supported with regard to 64-bit software and drivers. No one, especially in the server market, is jumping up to invest in a bridge release.

On the subject of Longhorn, however, the situation becomes even bleaker. Microsoft has been excising features from it in a desperate attempt to keep their timeline (2005 beta, Q1 2006 gold). The much-touted WinFS (which I didn’t really give a damn about: it seemed the sort of thing that would hog system resources for the benefit of casual users) was dropped. Palladium, the secure computing platform, was partially dropped just recently. The OS itself looks less and less impressive as time goes on: build 5048 (the 2005 WinHEC build) has a visual style similar in structure to the godawful Luna theme but less wretched colours. More widgets, maybe, though they thankfully dumped the damn sidebar. Balloonish curves. To be honest, I can’t figure out what exactly it taking so long, or even what Longhorn has to offer at this point. It’s looking to be less like a nextgen OS and more like XP Service Pack 3 (which might actually come out before Longhorn does).

Microsoft can’t really afford to stagnate like it is. 5 years between releases, and all they have to offer is a truncated OS with maybe an expanded driver base and a new platform for graphics (DirectX 10, or Avalon, mostly for widgets at this point: programmeres and games won’t take advantage of it for a long time). Apple’s new OS X release (10.4 ‘Tiger’) is being released, not that anyone particularly cares about it except current Mac users (what, 1% market share?). Linux is still chugging along, expanding its market, slowly but surely, even on the desktop. I’m not sure that it will ever eclipse Microsoft in terms of its development, since it tends to lack the clout to push new things through the market, though.

§590 · April 29, 2005 · Tags: , , ·

3 Comments to “More on Microsoft”

  1. rob says:

    You really can’t see the benefit of WinFS? If you haven’t already, look into BeFS (the filesystem used on BeOS) and see how cool it was.

    Et aussi, they didn’t dump the sidebar in the WinHEC build btw, they just hid it for some crazy reason.

    But yeah, I’ve no idea what on earth they’ve spend the last four years doing. Axing features left right and centre and pushing back deadlines – it’s pretty stupid of them to throw away their clear lead in the desktop market like this. Hopefully Linux can plug the gap; don’t see it happening myself, though.

  2. rob says:

    Some links on BeFS for you: (nice interview about the creation of BeFS, and how it compares to WinFS [despite them doing it 15 years before]) (a bit deep, though) (an open source implementation)

  3. Ben says:

    I understand on some level that WinFS will be an improvement, but most of the information I’ve seen so far has been of a “It will help computer illiterates find stuff” variety. I wasn’t aware that database-driven FSs were this old.

    I really wish I could make the switch to Linux. I keep trying, but I’ve too much knowledge and configuration time invested in Windows. Maybe some day, when I stop playing so many games.

Leave a Reply