FWIW, I think this is a good debate to have here. Every year or so, we have this discussion, and almost always something good comes from it. Hopefully, these discussions can be lifted out of the accusations and personal attacks that they sometimes degenerates into.
IMHO, I do not think XFree86 development is "stuck in a dead end" -- rather I believe the system is working okay today. That's not to say that it couldn't be improved, and that is what I'd like to see discussed. I feel that progress is being made (perhaps not as fast as some people would like, but I think it is a matter of resource availability). In particular, I think that Keith is doing some much needed work on fonts, and has laid out some good proposals for other work (RandR, Render, etc.). David, as he mentioned earlier in this thread, is working on configuration issues in his free time. Many of the other developers are working on various improvements. Hopefully, we can free up resources and get funding to work on some of these projects and proposals. Many, if not most, of the current developers have full time jobs and are thus relegated to working on these projects in our personal time. Many of us wish we had more time to work on XFree86 and push it forward! If companies are interested in seeing further development happen at an accelerated pace, I suggest that they free up the resources (i.e., people and funds) required to let us push the development forward. I see lots of talk about wanting the development to go forward, but very few that put resources toward making it happen! (Please note that this is just _my_ personal opinion). Regarding developer access, I wish we had more motivated people like Jos�. I watched him start out asking questions on the DRI devel list and make first attempts at fixing bugs in the DRI tree. He didn't start out fully understanding how the DRI worked. It took time, energy and effort. Over time, he progressed to the point where he gained the respect of the DRI developers (myself included). He was then granted access to the CVS tree. He's made significant contributions to the DRI development. And, there are several others that have done the same thing in the DRI arena. As far as granting developers access to the XFree86 CVS tree, what I described above for the DRI tree is pretty much how things work in the XFree86 arena as well. People learn about XFree86, make contributions, gain the respect of the other developers and then are granted CVS access. It takes time and dedication. I know that there are others out there who could contribute -- please do so, we could use the help. One problem that I've seen is that there appears to be a huge energy barrier to learning about the internal X Window System and XFree86 architecture. There also seems to be the perception that there is a lack of documentation. I admit there is nothing that plainly lays everything out on a silver platter, but documentation does exist. The main X Window System docs are in xc/doc/hardcopy dir. Dig through the docs there and I think you'll learn quite a bit (e.g., see ddx.PS in the Xserver subdir for some good info). If you want to learn about the libraries, check out the other dirs. Also, tracing through the X server code in gdb (or equivalent) is very helpful in understanding how things work. To do this build a static debug X server, and trace through it. Use the docs (above) to learn about the functions that are called. A good resource for XFree86 is in xc/programs/Xserver/hw/xfree86/doc. In particular, check out "DESIGN". It contains lots of good info on how XFree86 works. Also see the XAA.HOWTO in .../xfree86/xaa. Lastly, when all else fails this Xpert list is a good resource. There are people here with the knowledge to help you get over the stumbling blocks when the documentation has failed. Anyway, I hope this gets people started. If nothing else comes out of this discussion, I hope (1) that everyone realizes that we are doing the best that we can with the resources we have available, and (2) that more people become interested in XFree86 development who have the time/energy required to learn about it so that they can contribute. _______________________________________________ Xpert mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xpert
