What GNOME Can Learn from KDE's Recovery

When users complain about GNOME 3, inevitably they compare its release to KDE 4.0's. One KDE developer has told me that he dislikes the comparison, but, in the absence of other parallels, it continues to be made.

However, one part of the analogy that hasn't been explored is KDE's recovery from its user revolt, and whether GNOME is in any position to emulate that as well.

KDE's recovery has not received much notice. It hasn't been covered by the free software media. Often, too, it is overshadowed by those still loyal to the KDE 3 series, who continue to express their dissatisfaction at every opportunity.

All the same, KDE's recovery is as remarkable in its way as the meltdown over the 4.0 release. With each release since then, KDE has continued to add customization features and support for a variety of work flows, until now, with the recent 4.7 release, a considerable portion of the animosity has disappeared as user after user has been won over.

The recovery has not been flawless -- in particular, more effort could be made to engage users in the development process. But, at least to some extent, KDE has succeeded in its attempt to do what seemed impossible when KDE 4.0 was first released.

So is there any chance that GNOME could make a similar recovery? In theory, yes -- which is why I expressed the pious hope that it would when discussing the GNOME 3 meltdown. In practice, though, the possibility seems remote, and I have yet to see any clear signs that the attitudes exist within the GNOME project that would make such a recovery possible.

Where the Analogy Gets Strained

The main reason I am pessimistic that GNOME will follow KDE's example has nothing to do with the alleged animosity between the two desktops. Although some users enjoy the eternal GNOME-KDE flame war, developers largely ignore it.

Not only have GNOME and KDE recently completed their second desktop summit, but there is also a constant, informal exchange of ideas between the two projects. This exchange is largely invisible to users, although hints of it occasionally become obvious in such features as the use of "activities" to refer to virtual workspaces.

Instead, my pessimism is due to the limits of the analogy, and how GNOME has reacted so far.

To start with, the reception of KDE 4.0 could not have been anticipated, because the resulting user revolt had few parallels. The transition between each release series had stirred some complaints on KDE mailing lists, but not such a tsunami of complaints.

In fact, Aaron Seigo told me five months after the release that, "Personally, I naively thought that we had really proven ourselves with KDE 3. We had taken the promise of KDE 2 and matured it to KDE 3.5.9. And then we were going to attempt to replicate the results of that previous effort and take it to a whole new level."

By contrast, GNOME had the example of KDE 4.0 to illustrate what can happen when innovation happens faster than users can adjust. So far as I can tell, GNOME's main reaction was not to change development plans, but to develop a marketing campaign to prepare users.

Since this campaign did not circulate widely, and was centered on benefits that nobody was demanding, such as an easier interface and "distraction-free" computing, it failed. More importantly, it suggests a problem learning from history, and an inability to change directions that makes the idea of future efforts to follow KDE's example seem unlikely.

Just as importantly, although KDE 4.0 seemed to contain a series of radical changes, most of the actual changes were in the creation of back end sub-systems and largely invisible to users. What was placed in front of users were extended traditional desktop concepts such as icons and virtual workspaces, but in such a way that users could ignore the innovations if they chose, and -- aside from a smaller range of configuration options -- continue to work in much the same way as they had in the KDE 3 series.