Last October, Microsoft chairman and chief software architect Bill Gates unveiled an early technical preview of Longhorn — the planned next-generation Windows operating system — to attendees at the Redmond, Washington-based company’s Professional Developers Conference.
Jim Allchin, group vice president of the Platforms Group at Microsoft, previewed Avalon, the graphics subsystem behind Longhorn’s new presentation technologies; WinFS, Longhorn’s storage subsystem; and Indigo, the basis for the pending operating system’s communications technologies. Companies already developing in Microsoft’s .NET environment are well positioned to take advantage of Longhorn’s advances, Allchin told the thousands of developers gathered at the event.
One independent developer put it more bluntly. “We have to get people on the .NET platform because that’s where Longhorn is going. Longhorn is so different [that] if you’re not developing on .NET, you may as well forget it,” said Robert McLaws, president and chief software architect of Interscape Technologies, a .NET development firm that also runs LonghornBlogs.com (*correction), in an interview with the E-Commerce Times.
However, although Longhorn is tentatively scheduled for release in 2005 or 2006, some industry executives predict the operating system will not ship until 2008 or 2009. Will developer ire run high if such a delay occurs, considering the prerelease hype?
According to many IT professionals, even if Longhorn takes a long time to cometo market, Microsoft’s communications with developers, coupled with the software’s anticipated benefits, should reduce or eliminate potential backlash.
Delays Are Status Quo
Oddly enough, Microsoft’s reputation for delayed ship dates might nothurt, either.
“To some extent, Microsoft enterprise customers and partners — ISVs [and] system integrators — have come to view Microsoft’s road maps with a degree of skepticism and, therefore, will build an element of contingency into their planning,” Neil Macehiter, research director at London-based Ovum, told the E-Commerce Times. “If they don’t, then they should. In summary, do I think there will be a backlash? No, because Microsoft has given itself enough time to manage expectations.”
Even if Longhorn does ship within two years, it is still early for many enterprise IT departments and independent software vendors (ISVs) to invest heavily in development, added David Temkin, CTO of Laszlo Systems, a San Francisco-based developer of server software for rich Internet applications. As a result, they are unlikely to be thrown off course if the OS is delayed.
“I think you’re getting a lot of developers looking at [Longhorn],” he told the E-Commerce Times. “We expect that [corporate applications] is where it’ll get its initial success. But even when you’re talking about a corporate environment, I think it’s a little early for developers to be hunkering down and working on it.”
Later in Longhorn’s evolution, though, companies like Laszlo plan to enter the fray. “Our interest, when Longhorn comes out, is to be compatible with Longhorn on the client side,” Temkin said. “We attended Microsoft PDC. We’ve been researching it, learning about it.”
Laszlo’s mildly involved approach may be a good one. After all, whether Microsoft releases Longhorn in 2005 or 2009, developers need to begin scrutinizing early code, the company’s developer Web site and the Internet so that they are prepared to leverage the OS when it does ship, McLaws of Longhornblogs.com said.
“People need to know about this stuff now because it’s at least two years out,” he told the E-Commerce Times. “If it comes out and you’re not ready for it, it’s your fault. It’s the earliest Microsoft has ever released an operating system for people to use.”
Ovum’s Macehiter agreed, saying that while developers may not be involved heavily in writing actual code today, they should take a bird’s-eye view of Longhorn and its capabilities.
“I am sure that Microsoft ISVs, just as they invested in .NET early on, are investigating the new capabilities of Longhorn — WinFS and Avalon — to assess the implications for their applications and solutions,” he said. “They will — or at least, should — be doing this from a high-level architecture [and] design perspective, rather than low-level implementation.
“The code that is available for Longhorn is early access and thus subject to change, but the high-level principles are less likely to be so,” he added. “ISVs that have not yet invested in .NET — managed code, Web services — will certainly be doing so as this provides an evolutionary path toward Longhorn, similarly with Yukon, the next release of SQL Server due next year.”
Par for the Course
Many software developers miss ship dates, added Temkin, noting that “it’s just standard operating procedure in the industry.”
However, given its size and market reach, Microsoft makes an easy target, McLaws said. “At this point … they’re damned if they do and damned if they don’t [ship by 2006],” he noted, adding that a longer lead time will give developers more time to explore and test the OS before it goes gold.
“Microsoft is doing a really good job of listening to their customers and developers,” McLaws said. “Any date slips are understandable. Just the fact that they’re able to nail down that [we’re] going to have it this decade is amazing.”
*Editor’s Correction Note: In the original version of this article, we incorrectly stated Robert McLaws was the president and chief software architect of LonghornBlogs.com. In fact, McLaws is president and chief software architect of Interscape Technologies, a .NET development firm that also runsLonghornBlogs.com.