IBM To Linux Desktop Developers: 'Stop Copying Windows' (InformationWeek)

You're absolutely right. Doing usability by removing features (or not
writing them, due to lack of resources, then claiming "it's usability") is
a weak and ineffective strategy. It just leaves you with a less capable
system. You might make it easier to learn, but you'll also ensure the loss
of efficiency.

And you're right in claiming KDE 4.x right now does both things wrong: in
some areas we rewrote so much, we have too few features. In other area's,
we're not ready doing usability work and still have that overly-complex
KDE-3 interface.

But we're working on it. There are area's where we're getting there: not
too empty, not too complex, but powerful AND usable. It's what we're aiming
for, and we're doing it. That was what I meant with my first post - we're
aiming for something much more appealing than KDE 3 (too many features) or
Gnome (too dumbed down).