Talking Seriously About Integration

During their three-hour marathon back-to-back keynotes, Microsoft Chairman Bill Gates and Windows Platform chief Jim Allchin laid out a clear vision for Longhorn and the company’s renewed focus on feature integration. Their speeches kicked off Microsoft’s four-day developers conference. It also was clear that vision is an apt description of Longhorn. While Microsoft has done an excellent job assembling core technologies, Longhorn remains a work very much in the early stages of development. In fact, a number of technologies described my Mr. Gates or Mr. Allchin would appear to be little more than architectural plans at this point.

As Mr. Gates so aptly said, "We’ve got years of work here."

But, as I blogged earlier, that’s not an issue today. Acquainting developers with Longhorn is the goal here; the masses need to wait a little longer.

For faithful developers, Microsoft has a present for them: WinFX. Developers that have been programming to .Net Framework--the core architecture enabling Microsoft Web services to run on Windows platforms--will find WinFX to be fairly familiar. The programming model, which extends from Win32 and Win16 before it, will deliver new application programming interfaces developers will write their software to.

WinFX is part of a multi-pronged effort to integrate Web services directly into Microsoft’s flagship operating system. Because of architectural decisions Microsoft made during the so-called browser wars with Netscape, Internet Explorer is part of the operating system. That means developers have had to write either Windows or browser applications. With Longhorn, Microsoft wants to change that, in part using the new XAML markup language. Talk of XAML sent chills right down my spin.

From one perspective, given Microsoft’s big push around the Web, Web services and Extensible Markup Language (XML), a new markup language makes sense. Conceptually, XAML could make coding Windows applications a more sensible process than it is today. And the approach appears to nicely tap into WinFX.

From another perspective, open standards have ensured Web services had a high degree of portability. That’s something Microsoft had to have early on, particularly considering in the early days non-Windows server platforms dominated enterprises adopting Web services. My concern: That Microsoft might be mucking the openness of Web services standards by introducing a markup language specific to Windows. Indigo, Microsoft’s next-generation Windows communications core, also will integrate Web services. Microsoft also plans to chuck in collaboration and speech technologies.

Avalon, Longhorn’s new graphics subsystem, is built around XML, again which acts as another hook to Web services.

The new Longhorn file system, WinFS, will use database technologies taken straight from the next version of SQL Server, codename Yukon. Through the new file system, Microsoft hopes to make searching for data more meaningful than it is today.

This has been a long-time goal for Mr. Gates, who today said, "Some of you have heard me talk about the unified storage for more than a decade."

I want to save a deeper look at WinFS for another blog, but one thing worth mentioning right now: Longhorn will use a unified address book for supplying contact data and presence to the OS and supporting applications. That’s another, new integrated feature.

Longhorn also will feature a synchronization service, for moving data among multiple devices, such as Pocket PCs. Apple had an early lead on Microsoft with iSync. It will be interesting to see if Microsoft can do better. Synchronization as part of the OS makes a helluva lot of sense.

For that matter, so do many of these other new, integrated features. People forget that Windows is a platform upon which other companies develop software. So, it makes sense for Microsoft to integrate some technologies as they enter the mainstream. The company can then provide better tools to developers and, presumably, more robust software.

Flipside: Integration also can have an anticompetitive effect or straightforwardly be used by a company to gain advantage over competitors. In the case of Microsoft, the company’s Windows monopoly can cause all kinds of seen or unforeseen effects as new features are integrated into the OS. For example, WinFS potentially threatens IBM and Oracle on databases, Google and Yahoo! on search; if Avalon features favor Windows Media technologies that potentially threatens Apple and Real; Web services potentially could harm Sun and IBM, among a long list of companies.

I’m not suggesting any malfeasance on the part of Microsoft; there doesn’t need to be any because of monopoly.

Posted by Joe Wilcox at October 27, 2003 07:39 PM