Novell's Linux power play

With its market share in the networking services market shrinking from over 70 percent to under 20 percent in the last 10 years, Novell has been searching for ways to increase its revenue and, more importantly, its strategic relevance to corporate technol
Written by Tim Landgrave, Contributor

With its market share in the networking services market shrinking from over 70 percent to under 20 percent in the last 10 years, Novell has been searching for ways to increase its revenue and, more importantly, its strategic relevance to corporate technology buyers. Microsoft has not only decimated Novell’s networking business, but the rise of Office as the dominant corporate productivity application also gutted Novell’s vision of becoming a major player in the office suites market for Windows desktops.

What was left for Novell? It had a couple of key assets, including a robust directory service, a rich software distribution product, and an installed base looking for leadership from a company that itself appeared to have a vacuum at the helm.

But Novell has seen the future — and it’s Linux. In the last year, Novell has transformed itself from a Linux also-ran to a company that’s poised to become a dominant provider in the one space where Linux has no real champion—the rich desktop. Novell’s recent purchase of a little-known Linux company, Ximian, demonstrates its interest in moving the Linux desktop forward. Let's look at the effect of Novell’s Ximian acquisition and how it may allow Novell to become the Linux rich-client champion.

Who cares about the rich desktop?
With all of Sun’s barking about the importance of the Java Application Server, it's been unable to convince corporations to give up their Windows PCs for UNIX/Linux terminals and networked productivity applications spawned from Java servers. And for all its success on the server, Java has been a failure as a platform for desktop client development. As long as the Linux camp ignores the prevalence of Windows on the desktop, it has little hope of capturing the enterprise market. But without a champion for the desktop, the status quo isn’t likely to change.

Ximian's desktop history
A little company called Ximian has been working for years to make the rich Linux desktop a reality. Ximian’s developers created Gnome, one of the two dominant rich Linux clients or shells, as they’re referred to in the *NIX community (the other being the KDE shell). But beyond creating a simple GUI for Linux, Ximian created an entire framework (called Bonobo) that makes it easier for developers to create rich client applications. The Bonobo extensions include support for creating compound documents, object linking and embedding, service activation, storage, menus, toolbars, and printing.

Using these technologies, Ximian created Evolution, a rich groupware and collaboration client that can work with servers that implement the POP3, IMAP, vCal, and iCalendar standards or with Microsoft Exchange using its Ximian Connector for Microsoft Exchange 2000. Ximian has combined its Evolution client, the Gnome 2.2 shell, the Ximian Edition of OpenOffice.org, and other key client technologies to create Ximian’s Desktop 2 product (XD2). XD2 gives enterprises the ability to deploy a standard Linux desktop that includes productivity applications, software distribution management, and a rich printing subsystem. But it doesn’t solve one of the fundamental problems with the rich Linux desktop—the dearth of applications.

The Ximian (now Novell) Mono play
Although having a standard desktop distribution was important for Novell to be considered a serious Linux player in the enterprise, I believe that Mono will be the more important technology for Novell. This puts Novell in a unique position among all of the other Linux backers. Novell will be the only provider able to release a version of its Java Application Server that can host either J2EE or .NET applications. And if done correctly, it will be the only company that can provide out of the box interoperability between the two platforms at the lowest layer, that is, the ability to share security credentials and application state anywhere in the application.

But the more interesting possibility lies on the desktop. By creating a Windows Forms-compatible .NET library using the Gnome 2.2 shell and the Bonobo framework, Novell/Ximian has the opportunity to create a Windows Forms-compatible environment for which corporate developers, ISVs, and hobbyist programmers can create applications that run on either Windows or Linux (more likely XD2) without modification. This would give the Linux world a true competitor to the Windows rich client environment and would do so in a way that software developers could reuse a huge amount of their existing intellectual property.

By implementing Windows Forms classes on Linux that map directly to the Gnome interface (replacing the underlying Windows GUI calls), Novell/Ximian will make rich client application development commonplace rather than a rarity. And after Novell integrates its directory services, GroupWise, NetWare services, and ZenWorks software distribution products with the Ximian product line, it will have a nearly complete rich desktop strategy in place for Linux.

The missing link
There’s one thing still missing from the strategy. The Linux camp still needs a development tool that will do what Microsoft’s Visual Basic did for Windows 3.0. They need a tool that makes it simple for a small ISV or hobbyist developer to create robust applications quickly without having to be a Java or J2EE expert. The closest tool out there today is Borland’s Delphi. Perhaps it’s time for Novell to make another attempt to purchase Borland outright. That acquisition would make Novell a serious player in the Linux market and put it in the same league with the big boys—Sun, IBM, BEA, and Oracle.

TechRepublic originally published this article on 2 September 2003.

Editorial standards