News

Gartner: Blackcomb Delayed, Longhorn on Tap

• By Stephen Swoyer
• August 21, 2001

Gartner says Microsoft's interim version of the OS – code-named “Longhorn” and planned for release sometime in 2003 - will focus on reliability.

The news generated barely a ripple in the Windows user community, however.

Bill Tillson, a Windows NT systems operations manager with Primus Managed Hosting Solutions, says that his company has only recently started to think about .NET. As a result, he says, the prospect of a delay in Blackcomb doesn’t even begin to register with him as a problem – he’s not even sure whether or not he’ll move to Windows .NET Server.

“If future releases of Microsoft OSes push [.NET] out to consumers with positive results, then the market will force us to respond and implement and support this type of service,” Tillson says. “My personal opinion is that we are much further away from this than the release of Blackcomb.”

Gavin Burris, a visualization systems programmer with the Pennsylvania State University’s visual computing group, offers a more frank assessment.

“[Microsoft] should probably just can [Longhorn] and just work on beefing up the quality and the reliability and security of Windows 2000,” he says. “They’ve never caught on [to the fact] that most of us don’t want to upgrade every two or three years, so they keep on pushing new products and features on us. Or maybe they have caught on, but they don’t care.”

David Smith, vice president of Internet strategies with Gartner Group, says the interim Longhorn release will be intended primarily as a means to shore up the OS’ reliability and stability.

“The major focus of Longhorn is reliability. It’s supposedly the major focus of many releases, but they’ve really spelled it out on this one,” Smith says.

Smith says that Blackcomb is being held up by, among other things, a new unified storage feature that Microsoft plans to bundle with the OS. In this respect, Smith contends, Blackcomb resembles another oft-delayed and ultimately undeliverable Microsoft operating system, code-named “Cairo.” Microsoft began touting Cairo in 1993, and Cairo was also slated to include so-called “unified storage” features.

“Blackcomb smacks awfully similar of Cairo. The big thing that Blackcomb is supposed to deliver is exactly what Cairo was supposed to deliver, which is unified storage,” according to Smith.

“Unified storage” describes a method of storing conventional relational data, unstructured data and individual files in a single repository. According to sources familiar with Microsoft’s plans, the software giant plans to bundle a version of its SQL Server database with Blackcomb as a means to facilitate unified storage services. Some industry watchers have suggested that such a move would open up a whole new can of antitrust worms, but Gartner’s Smith cautions that it’s too early to raise the red flag on a bundling issue of this kind.

“It might be a little premature to get excited about anything like that, because packaging decisions can be very different than technology decisions, and just because an OS might include some technology to implement something, doesn’t mean that that would include the licensing and storage capabilities of a full scale enterprise database on it,” he says.

Microsoft currently offers a spin on unified storage – dubbed “Web store” – in its Exchange 2000 messaging and collaboration platform. Exchange 2000’s Web store repository has apparently been so (un)successful that the software giant has floated the idea of augmenting it – or of replacing it altogether – with a unified storage repository that it plans to incorporate in the next version (code-named “Yukon”) of SQL Server.

But despite Microsoft’s bully-pulpit preaching, or perhaps because of it, analysts such as Gartner’s Smith continue to question the compelling necessity of technologies like unified storage.

“Microsoft thinks it’s important because Bill Gates thinks it’s important. It’s one of Bill’s pet projects,” he says. “There are some theoretical benefits that you can get from it, such as a commonality of how you would address information, but a lot of its [purported] benefits have yet to be fully explained.”

Andrew Baker, director of Internet operations with educational testing service Princeton Review Inc., agrees.

“To me, unified data [equals a] huge security issue,” he says. “The main problem with all of these ‘advances’ is that they require significant infrastructure changes to be realized, and you need to have some decent training to even begin to articulate the benefits to those who will ultimately fund the migrations.”

Like PSU’s Burris, Baker says that Microsoft needs to stop focusing on developing revolutionary new features and start concentrating on delivering consistent, reliable software upgrades.

“We need to get a few upgrades where features we've always wanted to have are made to work properly or more effectively,” he concludes. “ I can think of a lot of things like this which we could do with now, rather than adding yet another buzzword to the feature pot.”