For fun

I'm over at Ian's place at the moment. West Chester, PA, suburbia at its best :) Everytime I come over he makes me do something weird. This time he made me write a bumpmapping algorithm for him. It's mostly used in 3D computer games, since it makes really nice textures. Why did he need it? I'm sure he'll tell you sometime soon in his blog. I'll move the bumpmapping algorithm to kdefx KImageEffect after 3.2. Also if you know any graphics algorithm that you would like to use with QImage's let me know. Either send me the name of the algorithm and the app which implements it, or preferably math behind it. One of my friends pointed me to the following comment by Jamie Zawinski. He says that it's not possible to mix GNOME and KDE widgets. The whole "not possible" thingy never worked too well in Open Source. Hopefully soon I'll release something that proves Jamie wrong. Hold your finger crossed as I'm busy with some other things at the moment.

Like Komposer. Komposer is the new email composer framework that we'll start using after KDE 3.2 will be out. Besides the HTML composing, which for some reason people want so badly, you'll be able to use KTextEditor editors there, write plugins for it, or just enjoy every KDE application using the same composer. I'm putting some finishing touches on KConfigEditor and KCFGCreator which I want to release in about two weeks. KCFGCreator has to be integrated with KDevelop. Personally I of course use Emacs but that never stopped me before. I also play with my web services app, which I use as a testing app for many, many things. It's my toy project so I'm rewritting parts of it on a daily basis. I need to release all my pending stuff before even considering showing the code for that one. There's also KSpell 2 and I'm trying to figure out what to do with it. I'm not looking too favorably at Enchant right now. The idea is there but I don't like the implementation. Also I'm not in favor of having a glib dependency there. People completely miss the point when they say "it's a small library, every distro installs it!". The point is that I do not want to be doing data structures conversion. Period. Especially considering that the core of a spell checker is like 500 lines of code. I don't see a good reason why we would use Enchant for those 500 lines of code. Now sharing plugins would be a good idea. Especially Ispell, but besides it's just silly.