KDE and the desktop future

OSNews features a nice article this evening about the future of the desktop. While the article does not try to predict the futre it is a quite nice overview of the state of available desktops in these days. It looks at all important and/or interesting desktops, and tries to give a basic idea where the problems are.
If you have to spend some time you could read it although you will not get to much new information if you already have a quite good overview.

But there is one thing which I like to discuss here:

As a side note the KDE developers seems to also spend time re-inventing the wheel. Gimp is already there yet Krita is heavily developed these days. OpenOffice is alive and kicking, yet Koffice reminds us that KDE needs everything to be done the KDE way.

I couldn’t disagree more!
Sure, we already have the mentioned applications – but there are two reasons why I think that KDE is on the right way here:

First the more obvious one: The *nix desktops have the opportunity to grow through friendly competition. They already used thsi opportunity, and I think it is working. The same is true for applications: we already have digikam and amarok, but f-spot and banshee are nice applications, and it is not bad that they are written – competition will bring up new ideas, and these new ideas will bring us new innovations – and these innovations are how the Free Software can win against the proprietary world (although I do not think that amarok can be topped😀 ).

But here is the second reason, the longer reason, and the more difficult:
I like KOffice and Krita, because Gimp and OpenOffice are suffering from serious problems, and there is no solution in sight!
Gimp was nice in former times, but it hasn’t changed to much in the last month and years. The Gimp users are still waiting for 16bit support and for CMYK-support. It was promised years ago, but nothing happened. And although they’ve announced the planned use of the GEGL-libraries there are no news about this, no beta version or something comparable.
And OpenOffice is just a huge, huge bunch of code – nothing which you can use if you are running a laptop in powersave mode. Than you are faced with slow menus, with startup times of almost 20 seconds, and so on, and so on.
It’s even hard for me to advice the use of OpenOffice to my friends who use Windows because they are than faced with a huge preloader or real poor performance. And: sure, you could start arguing now that this is partly because OpenOffice does not use the native Windows libraries like MS Office, but even if you start MS Office in wine under Linux it starts faster than OOo (ok, that test was a long time ago and against OOo 1.x, but I don’t think that so much changed).

Koffice in contrast is a welcomed alternative: it is fully integrated, it uses the same file dialogs, it can take use of the very important KIO-slave, and it is light – it starts almost immediatly.

So if you wonder that KDE is writing such applications, ask yourself why they do it before you start blaming them. OpenOffice.org 2.0 was a huge disappointment for me due to several reasons, and there is no movement at the moment to get rid of the major flaws at the moment. It reminds me of Mozilla before it was replaced by firefox, btw. Probably it would be the best to give OOo a new UI-engine, and split the processes as much as possible to get out of the “bloated” state.

And, btw.: one reason why several apps are rewritten for “the other desktop” is the problem of native integration – that is something that has to be changed soon, but the address to blame here is the LSB and probably freedesktop.org, not KDE or Gnome.

So, blame them, send them e-mails and tell them, that the community urgent need such things as ative integration independent from the used toolkit🙂