Linux Today: Linux News On Internet Time.

Editor's Note: Orphaned Desktop Environments

Jan 26, 2007, 23:30 (9 Talkback[s])
(Other stories by Brian Proffitt)

By Brian Proffitt
Managing Editor

The unification of the Free Standards Group and the Open Source Development Labs into the all-new shiny Linux Foundation has drawn a lot of attention this week. But while a lot of people are interested in the motivation behind the merger, I am wondering what the outcomes of this move might be.

First, and this is the gimmie conclusion, there's going to be a much stronger move to get the two major desktop environments (GNOME and KDE) to work with each other. The OSDL has the Portland Project, the much-needed collaborative project to get the two camps to start coding together. The FSG is all about standardization, since it's in their name. Or, at least it was. Now they're about "Foundation"--whatever that means.

Beyond that, and that fact that they're going to pay the bills for Linus Torvalds, Andrew Morton, and other OSDL Fellows, what's this new Foundation going to be all about?

Here's my suggestion: the LF may end up becoming the glue that holds Linux together. Or saves one of those big desktop environments from obscurity.

Over the course of the last few months, we have seen an escalation (or "surge," if the word suits you) in the competitive rhetoric between Red Hat and Novell and Oracle. Which is a politically correct way of saying that it's been "pile on Red Hat" season lately. Thus far, it's been just thatâ€"rhetoric. But soonâ€"very soon, I think, we may be looking at an event the likes of which we have not seen a major commercial distribution do before.

The dropping of a desktop environment.

The rationale for this is pretty simple: Novell and Red Hat are playing in the enterprise arena, and right now each of them are trying to differentiate themselves from the other.

Business-wise, Novell's big strategy is to partner with vendors (including Microsoft) to get themselves into more customers' shops. Red Hat had that strategy too, with their star partner Oracle, and we can see how well that turned out.

Red Hat's big strategy seems to be "stay the course," which is to say keep doing what they're doing: making the sales, following up on leads, investing in other companies and new technologies.

On the technology side, things are much more fuzzier. After all, both use Linux as their core offering, and there's not too many ways you can technically differentiate Red Hat Enterprise Linux/Fedora from SUSE Linux Enterprise Server/openSUSE. Yes, I know there are some technical differences, some brought about by Red Hat's determination to build special APIs and other hooks for their developer partners, or Novell's desire to do the same. Not to mention toolsets: RHN vs. YaST comes to mind.

But right now, if you are a business that has an application that runs on Linux, you would be hard pressed to find a technical reason to choose between RHEL or SLES. Business reason, yes.

A side show to all of this is the Red Hat/Oracle brouhaha. Oracle announces a distribution based on RHEL, with their own cheaper support. Red Hat responds with the common sense notion that a few dollars of savings per year per server isn't all that great for third-party support. Oracle comes back and announces their Database Management Framework has been re-mixed with new tools added for a Linux management tool that will be free to Unbreakable Linux users (well, middle- and upper-tier UL users, actually). Red Hat zings back an us-too management solution that promises to be better.

And on it goes. This tit-for-tat situation leads me to wonder if we are not about to see a similar battle between Red Hat and Novell. If that's the case, then I could very easily see a desktop environment put on the chopping block by one of these distributors. Not only would focusing on one desktop be a way to differentiate, but it would also enable said vendor to put their support and development efforts into one environment. Theoretically, this could lead to accelerated development and a much better desktop environment.


I want to stop and be clear here. This is not about a commercial distro that would only run one environment. This is Linux after all, and all desktop environments would be able to run... but only one DE would be supported.

The question is, if this theory is viable, who will blink first? There, I have little doubt it will be Novell. They tried to do something very similar to what I am suggesting back in 2004, when they briefly toyed with officially supporting GNOME in their enterprise product line. (Customers could still use KDE, but they'd be on their own.) The hue and cry from the community led Novell to back-pedal, but I think it's still an option for them.

The real question: Which one would Novell choose? You'd have to strongly favor GNOME, since there is still a strong Ximian contingent in the company ranks. But all those ex-SUSE Linux AG folk love KDE (which is the real reason Novell caved on their 2004 decision). This internal duality is, I am sure, the only reason Novell hasn't dumped KDE already.

If Novell made SLES a GNOME environment (maybe letting the KDEers in the company play in the openSUSE sandbox), then what would Red Hat do? Stay with the multi-environment approach? Or cut their own time and effort (read: costs) and choose to support just one DE? And then what would that DE be? GNOME (which Red Hat historically likes) and be beholden to their competitor's development team for improvements? Or a full switch to KDE?

This is the very situation I believe the Linux Foundation is here to prevent. If DEs become more standardized, then the arguments of one DE versus the other become a lot less costly and much more moot. The Big Two could afford to keep things the way they are.

I don't think we need more division between the commercial players. If customers have to choose between environments in their decision to choose a distro, then I fear that this will only dilute the power Linux has overall in the enterprise market. And, if the commercial vendors decide to go with the same environment, what happens to the support and funding for the orphaned one?

The Linux Foundation, if it keeps all the players from moving too far away from the rest, can prevent more technical differentiation than there already is.

Here's to their success.