A paradox lies at the center of the Linux desktop today. For all their limitations, reader polls consistently show that KDE is the single most popular desktop, preferred by just under a third of users. Yet at the same time, 40-45% use a desktop that sits on top of GNOME technology, such as GNOME3, Cinnamon, Mate, or Unity.
Even when the numbers vary, the basic relationships never change. You would think that the most popular desktop would also have the most popular applications, but apparently not -- at least at first glance.
This paradox has been often observed. When not denouncing the polls as hopelessly biased, GNOME supporters repeat it endlessly to console themselves that GNOME is still dominant on the desktop. However, in the absence of hard data, it remains unexplained.
Is the paradox a matter of a preference for different design philosophies? The consequence of the user revolts of the last six years? Or is it more apparent than real? While we cannot be entirely sure what is happening, we can make some informed guesses.
The first possibility, that users prefer GNOME's design philosophy over KDE's, can be quickly dismissed.
The philosophies are distinct, with GNOME tending towards minimalism and KDE towards what might be called “completism,” and an argument might be made that GNOME's minimalism is less confusing to new users.
Moreover, GNOME tends to design with usability in mind from the beginning of development, while KDE seems to consider usability in later releases. KDE's tendency has been very obvious in the KDE 4 release series, in which many menus and windows were not organized efficiently for several releases. For example, for many releases, KDE's System Settings window included an Advanced tab that was basically a dumping ground for features whose place in the settings had yet to be properly determined.
Under these circumstances, a preference for GNOME apps might seem plausible, except for one troubling observation -- that preference has never been clearly demonstrated before.
While both desktops have changed in recent years, the design philosophy of their applications is well over a decade old. In previous user polls, the advantage has seesawed back and forth. Until the first of the GNOME 3 release series, GNOME often had a slight advantage, but never a large one.
In fact, when you start thinking of margins of errors -- a concept that readers' polls studiously ignore -- the most accurate analysis of readers' polls over the years before the latest release series would be that the two desktops were so close that neither had a clear advantage.
The difference in design philosophies is evidently such a fundamental one that neither approach is likely to win many converts from the other, and nothing seems to have happened recently to change that.
A likelier alternative is that the paradox has been misidentified. Perhaps the paradox is not technical in nature so much as an unintended consequence of the user revolts against KDE and GNOME and how they were handled.
From a user's perspective, KDE responded most effectively to its user revolt. Doing so was easy, because many of the features that users demanded to have back were already planned to be included in KDE 4.1. However, KDE also responded in a few months, with every sign of interest in readers' concerns.
As a result, KDE seems to have lost 6-9% of its users. There was little time for hostility to settle in, and the only alternative to emerge was the Trinity Desktop Environment, a continuation of the KDE 3 release series that did not even register on the Linux Journal's 2013 poll. KDE might have acted faster, but it reacted fast enough that it kept much of its user base.
However, GNOME's handling of its user revolt was very different. The general perception -- which, needless to say, is not shared by GNOME's supporters -- is that the project expected users to conform to the desktop design imposed by developers. And that it made no concession to complaints for almost eighteen months, when the official acceptance of GNOME extensions finally gave users the flexibility they were demanding.
At the same time, Ubuntu began to develop Unity on top of GNOME. It proved equally unpopular, increasing the demand for alternatives.
At first, users flocked to Xfce, which has always run GNOME applications extremely well. Later, some also turned to Cinnamon and MATE. All of these resemble the GNOME 2 release series -- MATE is actually a fork of it.
What may seem surprising is that few of the ex-GNOME users seem to have considered KDE. KDE might offer much of what they demanded, but the longstanding rivalry was apparently too great for most of the dissatisfied to overcome. If nothing else, switching to GNOME would mean becoming familiar with different applications, and that was not what users wanted. The problem was not design philosophy so much as a preference for the familiar.