Longhorn in 2006: Can Microsoft Really Pull This Off?
|
Let the (reopened) Longhorn date pools begin!
Will Microsoft really ship a major new version — something that's more than just a service pack upgrade — Microsoft watchers have been guesstimating since well before the first Longhorn Professional Developer Conference (October 2003) when Microsoft's XP successor would ship. Starting in earnest last August, when Microsoft gutted Longhorn by exorcising the WinFS file system in order to get it out the door, speculation intensified, regarding whether or not Redmond could make good on its reset 2006 delivery target. This week, Microsoft's biggest Windows pooh-bah, group vice president Jim Allchin, stated for the record that Microsoft can and will deliver the final version of Longhorn to PC and software makers in time for them to package it up for "holiday 2006." (As usual, all of Microsoft's usual quality disclaimers apply: If the product is not solid, all ship promises are off.) Allchin, who stumped for Longhorn this week on a cross-country press tour, told reporters that Microsoft is on track to ship the oft-delayed Longhorn next year. The timetable: We told Allchin we were skeptical. Since when has Microsoft been able to deliver in little more than a year a new release of Windows (or even a Windows service pack, for that matter)? Was Longhorn in danger of becoming another Cairo, the object-oriented-file-system-rich version of Windows that disappeared with nary a trace? Allchin justifiably reminded us that Longhorn's been in development for quite a while now. How quickly we forgot about those leaked Longhorn alpha builds back in 2003 and long-ago promises of a 2004 final Longhorn delivery date! Sure, almost the entire Windows development team got pulled off Longhorn to focus on getting Windows XP Service Pack 2, Allchin said. But they're back on Longhorn with a vengeance, Allchin reassured us.
"We did (Windows) XP in 18 months," give or take, Allchin said. So why not Longhorn? After all, Longhorn Build 5058 of which Allchin has been running on his main work desktop machine for two and a half months, has yet to crash, he told us. It sounded plausible…. But Beta 1 is only going to be about one-third feature-complete when it hits this summer, Allchin acknowledged. And while Microsoft has pared back on Longhorn features, it is still promising some big major ones, including the first elements of its Next Generation Secure Computing Base; a new "visualize and organize" search system; full IPv6 networking support and more. And there are all those security hoops through which Microsoft products, especially ones as major as Windows, must jump before seeing the light of day…. We want to believe. But we've been lulled by the Longhorn vision one too many times. We admit it: We're leery that Microsoft can deliver the features and functionality Allchin promised this week by Q3/Q4 2006. What do you think? Can Microsoft pull this off? Will Longhorn ship next year? And if so, what will make it a must-have Windows release, rather than a more minor, interim update, in your opinion? Talk back below. Or write me at mswatch@ziffdavis.com and |
Comments (4)
Yep, and I will see my stocks at 2000 NASDAQ top levels at the VERY same time!!!
RU kidding me!?
MSFT keeping its promises and delivering a decent product on time...
Posted by felixml | April 15, 2005 2:58 PM
... there is another significant reduction in feature set they'll make it. Of course, then we blur the lines between XP SP3 and Longhorn.
The problems are always the same: security, code bloat and backwards compatibility. The more I think about it, the more I come to believe that Microsoft is its own worst enemy. Their obsession with the proprietary nature of their API's, plus the intentional proliferation and obfuscation of the "internal" ones for Microsoft developers only, yield an environment where it is impossible to test every branch of the code. Apple went through the API hell just once, with the transition to OS X. Even then, the result was a relatively slim OS based on open source, with all API's fully exposed. No wonder Apple is pulling ahead!
Posted by Eric Fisher | April 15, 2005 4:33 PM
I think Microsoft is going through what Apple did with its Copland OS. If you look at what Apple did with Copland it sounds exactly like what is happening with Longhorn. Ultralong development cycle, features coming and going, release dates sliding and sliding, it sounds all too familiar. Apple finally trashed Copland and purchased the NextStepOS. Apple is releasing the next OS X on schedule in a few days with every feature promised well over a year ago and more as well.
Since I use both I can say without a doubt the OS X is the better OS. I hope to see Longhorn be as good but at this point if Microsoft releases it on schedule I think it will be little more than XP3. As stated hidden API's makes for bad programs and it is beginning to show.
Posted by Randy Smith | April 20, 2005 11:23 PM
MS will do whatever it takes to meet the August "new PC for college" and Winter "new PC for Christmas" shopping seasons in 2006. Allowing program files to be re-written after having been read and marked "in use" (like a Linux RPM upgrade, avoiding most "restart your computer" messages) is a great feature by itself. And real user context, where a User DOESN'T have destructive access to system files while browsing the Internet under User credentials, is a huge improvement. These things are already built. Lots of drivers which insist on putting "work data" into system directories, (such as my HP Printer drivers ;-(( will need to be modified. But the work involved in doing this is not huge. The new Display/Print system sounds really neat, but people's old stuff will continue to work in a "compatibility mode" while MS fleshes out this feature, and others, as Updates.
I think that IPV6, having been badly mis-designed from the start (i.e., it has terrible backward compatibility with IPV4) will be really slow on the uptake. All the web sites which MS customers want to get to are IPV4 sites, and that won't change in the next 15 months. So MS doesn't need to squeeze out every last obscure IPV6 bug before Longhorn goes Gold.
Thus, Longhorn should be pretty compelling upon its Release: Even if the only feature-complete stuff turned out to be the "true User Mode" security improvements (and there will be plenty of other ready-to-run enhancements), I'll take the leap quickly.
(Although, you might want to consider that I'm posting this from Mozilla on a Mandriva-Linux machine... I only use Windoze-XP when I have to duplicate my customers' problems.)
Posted by Rick Stockton | May 19, 2005 8:09 PM