When Microsoft Can't Do Windows

The delay in shipping Longhorn betrays major R&D troubles

When Microsoft Corp. (MSFT ) announced on Aug. 27 that its next version of Windows, code-named Longhorn, would ship as much as a year later than expected, in late 2006, the company downplayed the news. Don't be fooled, because the setback is, in fact, a big deal. For one thing, it could damage the reputation of Chairman William H. Gates III, who put more than a third of his time into the project since he became the company's chief software architect in 2000. It also means there won't be advances in the operating system for at least the next two years. That means customers will have one less reason to buy new PCs. And it gives an opening to the software giant's rivals, who are going after Microsoft as if it were an aging prizefighter. With every Microsoft misstep, their offerings become more attractive to both consumers and corporations.

The Longhorn mess points to a recurring problem at Microsoft: Hype often gets way ahead of the company's computer science. "The priority is more on marketing than development," says Alan Paller, research director at the SANS Institute, a computer security training organization. That doesn't bode well for a company that hopes to lead the computer industry in the 21st century.

With Longhorn, as with other past products, Microsoft talked up a technology it couldn't quite deliver. It set out to overhaul a central piece of desktop computing -- the way users search and store information. But it proved too difficult a challenge -- especially at a time when it is under tremendous pressure to make its current version of Windows safe from a plague of viruses, worms, spam, and pop-up ads.


Often criticized for its lack of innovation, Microsoft might have hushed naysayers had it pulled off the much anticipated Windows File System feature of Longhorn. Shelved at least for a while, WinFS would have made it much simpler for computer users to store and search for everything from e-mail to spreadsheets to Web pages on their computer, a network, or the Internet. It's still a long-term goal. "We're not going to temper our ambition," says Will Poole, senior vice-president of the Windows client business.

The news that Longhorn is late has Microsoft rivals licking their chops. They have a two-year window of opportunity to improve their products and win over customers. "Microsoft is continuously promising these advancements but isn't able to deliver on the promises," says Mike Ferris, product marketing manager at Red Hat Inc., which sells the rival Linux operating system. "Now customers can look at how non-innovative this is going to be, and they can look at open source as an alternative." Red Hat, one of the leaders in Linux software for server computers, in August started selling a desktop version for a subscription cost of less than $6 per user per year.

Some in the industry believe that Microsoft could learn from one of its rivals -- Apple Computer Co. The Mac maker does a great job of keeping both its new hardware and software under wraps, then thrilling its fans with whizzy breakthroughs -- something it did once again in late August with the unveiling of its gorgeous new iMac computer. In contrast, when Microsoft doesn't deliver on its promises, it loses credibility with PC makers, software publishers, and customers.

Last spring, as the first signs of Longhorn delays emerged, techies began referring to the operating system as Longwait. Now, with WinFS cut, critics have renamed the software Shorthorn. The name-calling poses no threat to Microsoft's hegemony. But its stumble with Longhorn gives rivals an opening at a time when Microsoft needs to get its business revving again.

By Jay Greene in Seattle

    Before it's here, it's on the Bloomberg Terminal.