Well, it's been an interesting few weeks since I last wrote anything here.. A lot has happened in both Debian and QuakeForge and not all of it has been good. Not all of it has been bad either, but it has seriously got me thinking about how open development projects do things, and what seems to be wrong with projects as they gain notariety and attempt to scale upwards.
If I get a chance, I'll try to write down my thoughts on this matter later.
It completely sucks when you're trying to prevent a flamewar and someone who has no stake in anything comes in with the intention of essentially stirring shit. The person in question was (I thought) just trying to express his issues with the attitudes expressed by some of the QuakeLives people. Not so, he even admitted he was actively trying to start WWIX in the irc channel. *sigh*
Being a foolish optimist, I'm still holding out on the hope that the current problems with them will be resolved. Not because the team wants them resolved - it's clear they don't care. But in the end they need users to be willing to play using their client. And the users have been saying in droves that they don't have to have a bunch of different clients just to play on their favorite QW servers. Enough people remember John Carmack threatening to sue them over GPL violations (they still haven't put up source, but they currently don't have anything downloadable at all) that I should think they're not likely to ignore the overwhelming demand. They don't exactly have the best reputation here at the moment.
On a more technical matter, I am debating what the hell I'm going to do about ARB multitexture in the GL targets. Most cards and drivers support it just fine these days. The one major exception is ATI which is not exactly improving my outlook on that company's products. It seems that ATI's cards support only one extra texture - a lightmap. That's completely useless when the goal is to add a second texture layer to emulate "fullbright colors" in the software renderer or other features that involve multitexture. I'd be curious to know how Q3A deals with this issue given that it relies on heavy doses of (multi-)multitexturing.
I'm considering working on some GPL'd quake tools.. They'd make my life a little easier and some of them would be good projects, especially since I want to pick up some X programming. There are issues of course such as language and toolkits for the GUI versions. I can probably make "better" code in C++, but interfacing C++ to gtk is a nightmare because gtk tries to invent objects in a language that doesn't support them. =p The code will be GPL'd so Qt is out of the question. I guess that leaves wxWindows which I've been meaning to look at anyway. Most people would prefer I just use C I suppose but a procedural design is just clumsy for dealing with object-oriented data and interfaces IMO. It's a style point I guess.
It's finally here: spring break. Yeah our school schedule is strange. The good news is that I have about a week and a half to seriously write a serious amount of serious code. (Seriously! And no, I haven't had anything to eat yet so excuse the excessive weirdness..)
Job's been kinda lame lately. They haven't been needing me when I don't have class and I've had to choose a few times work over school because if I don't work I can't afford school. (Can't get financial aid for some oddball reason...) It's seriously hurting my grades in all but one of my classes - programming of course, in which I am probably the most advanced student allowing me to attend or not at will and still make the grade and more.
I think I need to look into a more serious job for the summer with just a bit more pay than I'm getting for ISP tech support. =p I realize though that if I get a job I'll probably be moving and probably will be taking only a class or two at a time from then on because I wouldn't simply work three months and then quit.
Anyway, I said something about coding over the next week and a half. We have a number of very odd bugs in QuakeForge that need to get squished if possible. The catapult entities in DyerMaker's map only seem to work once before they get out of sync. There's still some weirdness with the X11 target and I have no idea what the hell we did to the sound engine. Fixing these things and getting fs_*path cvars and (for unices) /etc/quakeforge.conf parsed is all that stands between here and a release. Simple right?
Yeah, of course, simple. Don't Bet On It.
New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.
Keep up with the latest Advogato features by reading the Advogato status blog.
If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!
