In January 2008, the long awaited KDE 4.0 was released. Immediately, a user revolt erupted.
KDE 4.0 was too radical a change, too lacking in features or stability, too much a triumph of developer's interests over user's -- the accusations seemed endless, and only began to quiet six months later when KDE 4.1 began addressing the shortcomings. Partly, the hostility continues to this day, although for many the KDE 4 series has long ago proved itself.
This April, GNOME 3.0 is scheduled for release. Just as KDE 4.0 was a radical departure from KDE 3.5, so GNOME 3.0 is a radical departure from GNOME 2.32. But will its release trigger another user revolt? Or has the GNOME project -- perhaps learning from KDE's experience -- managed expectations well enough to prevent history from repeating itself?
Certainly, GNOME has tried much harder to handle its own break with the past differently than KDE managed KDE 4.0. But the KDE revolt resulted from multiple causes, and, although GNOME has addressed some of those causes, the underlying problems of the project's relationship to its users remains in some ways disturbingly similar to those faced by KDE three years ago.
Like KDE 4.0, regular snapshots of GNOME 3.0 have been released as it developed. For the last couple of releases in the GNOME 2.0 series, a GNOME 3.0 preview has been available, although it has not received widespread attention until the beta release earlier this month. In fact, many users seemed unaware of the preview.
However, in general, GNOME is showing much more caution than KDE did. KDE 4.0 was a developer's release that slipped into general circulation partly because of miscommunication and partly because of each distribution's wish to ship the latest software. Consequently, it was not stable or fully-featured, and, according to KDE developers today, was never meant to be. By contrast , GNOME 3.0 has been delayed twice.
Both times, GNOME has explained the delay as a wish to perfect the software. When the release was rescheduled for September 2010 instead of the original April 2010 target, the explanation was that "our community wants GNOME 3.0 to be fully working for users."
Similarly, when the release was postponed a second time, from September 2010 to March 2011, the official reason was to allow "adequate time not only for feature development, but user feedback and testing." Unlike KDE, GNOME has no intentions of producing a developer's release, or of justifying any complaints about lack of stability or features.
Since then, there has been an additional delay of a few weeks. Although no reason has been given for this new delay, slippage of this length is common in free software. It also gives time for the GNOME documentation and marketing teams to have a hackfest to prepare for the release.
In fact, GNOME shows every sign of attempting to manage expectations and rumors about the release. For at least half a year, there has been a page dedicated to debunking rumors about GNOME 3.0. More recently, a website dedicated to the release has maintained a Common Questions and Answers page, as well as a home page that is as much a product sheet as anything you would see in a PDF file released by a commercial company.
All in all, GNOME has made a considerable effort to keep users informed -- far more than KDE ever did in its similar circumstances. However, these efforts are not widely publicized. They were not brought to the attention of the journalists who write about free software -- instead, all users were left to discover these efforts for themselves.
GNOME is apparently preparing publicity for the general release, but, considering that GNOME 3.0 has been two years in the making -- including one year of postponements to the original schedule -- a project with more experience in marketing might have tried harder and earlier to dispel rumors. As things are, GNOME's official marketing efforts may come too late to counter user's expectations and the misinformation that has been floating around the Internet.
At any rate, despite the occasional delusions of marketing managers everywhere, the most expert campaign imaginable is not enough to completely counter what the audience considers unacceptable.
Despite GNOME's efforts to publicize what it has been doing for the last two years, no one seems to have made a concerted effort to learn what users actually wanted in a desktop.
True, GNOME has consulted usability experts, and tried to apply the theory to the building of a new and modern desktop. As Canonical may find in its development of the Unity shell, usability theory in software is often resisted by user's practice.