Mozilla, Gnome mull united front against Longhorn

Representatives from two open-source foundations, Mozilla and Gnome, met last week to consider a joint course of action aimed at keeping their respective Web and desktop software products relevant once Microsoft releases the next major overhaul of its Windows operating system, known as Longhorn.

Microsoft now has "a single team for Web and native desktop rendering," noted one participant, according to meeting minutes posted on the Gnome Web site. "Gnome and Mozilla need to align to counter this."

Mozilla is an open-source browser development project. Gnome, which stands for GNU Network Object Model Environment, is an open-source user interface for use with Linux and other Unix systems.

The April 21 meeting, attended by veteran Mozilla and Gnome organizers including JavaScript inventor Brendan Eich and Ximian co-founder Nat Friedman, is but one manifestation of the open-source community's Longhorn jitters. Microsoft has promised that Longhorn will fuse Web browsing and desktop computing to an unprecedented degree.

Microsoft said last year that it would discontinue standalone versions of its Internet Explorer browser to focus development energies on Longhorn.

Competitors fret that when Longhorn launches, standalone browser and desktop applications may find themselves consigned to the computing paradigm scrapheap.

The open-source developers may have time on their side. Microsoft earlier this month said it won't release Longhorn until at least the first half of 2006, having decided instead to focus this year on getting out a major security upgrade, known as WindowsXP Service Pack 2, for its current operating system.

Microsoft also faces unknown fallout from a decision last month by the European Union to force the software maker to supply a version of its Windows operating system without its Media Player software. Microsoft has appealed the ruling, and a final decision could be years away. But it could set a precedent on how the company builds its software that could affect Longhorn, which will introduce many new features.

While Microsoft has delayed Longhorn's release repeatedly, the company has advanced vital components and related technologies, including the Extensible Application Markup Language (XAML), the Avalon graphics and user interface technology, and the .Net Web services framework.

A dangerous combination
Taken together, that arsenal is costing open-source competitors sleep.

"What makes Longhorn dangerous for the viability of Linux on the desktop is that the combination of Microsoft deployment power, XAML, Avalon and .Net is killer," Ximian co-founder Miguel de Icaza wrote in a recent blog posting. "It is what Java wanted to do with the Web--but with the channel to deploy it and the lessons learned from Java's mistakes. The combination means that Longhorn apps get the Web-like deployment benefits: (You can) develop centrally, deploy centrally and safely access any content with your browser."

		Get Up to Speed on... Open source Get the latest headlines and company-specific news in our expanded GUTS section.

A key weapon in any planned counterattack could be Mozilla's Extensible User Interface Language (XUL), a 5-year-old scheme for building desktop applications' user interfaces out of lightweight Web markup languages like XML (Extensible Markup Language) and CSS (Cascading Style Sheets).

The original impetus for XUL was to make the Mozilla browser itself lighter and faster by creating its interface with Web standards. But out of the resulting technology Mozilla developers speculated they could spark a "programming revolution."

So far, XUL has failed to catch on, and Microsoft questioned whether Mozilla's technology would do much to help Gnome ward off Longhorn's promised threat.

XAML, Microsoft warned, is more potent than XUL in its ability to reflect exactly what's in the operating system.

"XUL is not the multipurpose declarative language that Gnome probably wants," said Ed Kaim, product manager for the Windows developer platform. "People say that when all you've got is a hammer, everything looks like a nail. In the same way, people are trying to figure out how to crush XUL into an OS it really wasn't designed for. The browser is great for a lot of things, but when it comes to robust client side applications, it's not the best."

Another trick will be in reconciling XUL with Gnome's existing user interface technology.

"There are ways to marry them," said Bruce Perens, an open-source consultant who serves as executive director of the Desktop Linux Consortium, a marketing organization. "But it's very difficult to get the two teams working in the same direction. They both went on a several-year tour of technical creation where they sat down and created everything they needed to do GUI (graphical user interface) applications--and they didn't create the same thing. Now to get them together it would take some number of years to resolve the technical diversions."

Gnome already relies on some Mozilla software and produces a Mozilla-based browser called Epiphany.

Mozilla also produces a version of its Firefox browser for Linux and Gnome, and one of the points of discussion between the two groups is to produce a browser that combines the native Gnome interface elements of Epiphany and the cross-platform capabilities and 200 extensions or plug-ins that come with Firefox.

But it is the development framework that poses the greater challenge and holds the higher stakes.

"As we look at the challenges coming our way, we must remain competitive and retain an aggressive agenda to provide a rich user experience on all platforms," said Mozilla spokesman Bart Decrem. "XUL has come a long way since it first came out, and the combination of Gecko and XUL is a great starting point for delivering rich applications to the desktop."