OSNews

Well, keeping with the subject of the article KOffice is a fine piece of work. When you consider the lack of developer resources these guys have, and a lack of time, it's incredible that they're managing to keep a word processor, spreadsheet, presenter, visio-like app, database application, image application, chart and drawing applications going and pushing them forwards you can see the power of KDE's development framework and infrastructure. Krita, in particular, and the pace of its development is incredible when you consider that the GIMP has been around for ten years now and still isn't a Photoshop replacement.

Now, the other thing. From a general user perspective I really don't care about KDE or Gnome. There could be fifteen desktops for all I care. However, when I'm looking at the pros and cons of each of those desktops, and looking at doing what people have only been talking about for the last five years, namely getting businesses to use desktop Linux, I need a desktop that will give me the minimum amount of grief possible and be an adequate Windows replacement. Because of the quality of the development infrastructure (and that matters, because it comes out in the applications for an end user), and hence the quality of the applications, that desktop is KDE. I can also be confident that that desktop development infrastructure will help all developers, giving them the support they need, and because of that I can then place more trust in the desktop itself.

With Gnome, yes, it is a reasonable desktop for many end users. However, for what I referred to before (getting Windows users in businesses to use it) my experience has not been good. Keep in mind the target audience here. Too many things are not right with Nautilus, such as handling large numbers of files and folders (how long has this been in development and how much did Eazel spend on it?), printing infrastructure, shaky media players and applications, UI responsiveness, slowness with lots of Gnome/GTK applications open etc. At one time there was an accusation that KDE didn't have many applications, but when you look at the relative quality of KDE applications today (amaroK, Kontact, Krita etc.) versus GTK and Gnome there is no contest. Look at the variety as well - remote desktop, a wide assortment of games and check out the kde-edu module for educational apps! These things make a difference. Too many Gnome/GTK apps just don't hang together. The ones that just about do are Nautilus, Evolution, GAIM, GIMP etc. but they tend to be large applications with a lot of development weight, and in some cases, millions in investment - and that's free software you can use Qt to develop for nothing as well! What's happened with Evolution is a bit shocking (curiosity finally killed me and I went off and had a look). Evolution was a good e-mail app at one time, but it just seems as though since they've added features (groupware, calendaring etc.) the thing has started to drop to pieces. What's that phrase some Gnome devs tend to use? "Just Works"(tm). Right. I would say that the development base they're using is just not scalable, not necessarily that the developers are not talented.

Considering that we've got people like Dave Neary and others who have rabidly pushed Gnome as a corporate desktop, and tried all ways to scare people off KDE (licensing, blah, blah, blah), the fact that they've got a desktop that just doesn't hang together isn't funny. No wonder there are many "Linux not ready for the desktop" comments. At the very least, the KDE people are not trying to sell something that doesn't exist. I don't have a problem with Gnome itself, but the continuous marketing crap is just totally tiring now.

Is KDE perfect? Hell no. Is any desktop? The things I hear consistently from KDE naysayers tend to involve the words 'clutter', 'options', 'usability', blah, blah, blah. Well excuse me, but I and other users can put up with a bit of clutter and some slightly confusing dialogues for the sake of a working, solid desktop that "Just Works"(tm) and a desktop you can have confidence in for the future. Waiting twenty seconds for Nautilus to catch up with itself or Evince's printing ability is not anyone's idea of usable. Now, if some people just can't accept all that within the context of what I am talking about then that's just tough luck. You can sell me a pair of the glasses you use, because they must be good!

Does KDE need tidying up and to have more focused usability work done? Yes, it does. Has Gnome taken a lead in this area and inspired others? Yes, it certainly has. However, this stuff doesn't just happen and its about more than getting developers to follow a HIG. In KDE devs won't need to follow it all anyway because of the common base they can all use. When the HIG changes, whoila, you change your base and all the apps inherit all of the stuff they have in common.

KDE also seem to have some pretty nice usability experts on board now, but it's going to take some time in the run-up to KDE 4.0 to see the results of that. Getting usability in the process is hard work on any project. I also hope to see KOffice 2.0 there as well with its incredible array of applications and functionality, and with the steady pace of development (and good development infrastructure!) that should be a nice release.

Finally, thanks to Inge Wallin and Anne-Marie Mahfouf for doing more promotion and making the KOffice site nicer!