OK here's a quick sum-up of what drives me to start this thread :
- It seems that Xith3D has been developed very quickly (I'd say, in a "rush") by David Yazel for his game, Magicosm
- Some people have been interested and added features (Yuri VI. Gushchin, Java Cool Dude, Artur Besiadovski and others)
- There was no code conventions, nor real reviews/testing/validation of the added features
- Some features have been broken along the releases (ex : collision system, xith gui)
- Small packages have been added everywhere
- Some classes aren't javadocumented at all
- Xith3D's architecture is now "hard" to understand, and you need a real work to know how the whole system works
So from what I understood the ARB (Architecture Review Board) is a bunch of competent people that review any changes of the project, feature adding, and who looks for the best way to implement something, that maintain documentation, that ensures that changes are conform to the original vision of the project.
With my very little experience and my pretty small knowledge, I think it's a good thing to have, but heh we have to group the competent people together, right ? And my questions are : "How is it organized ?", "How do the ARB communicate ?", "What's the procedure to add a feature ?", and all that things..
To me, some names comes to my mind for the ARB composition :
- William Denniss
- Jens Lehmann
- Yuri Gushchin
- Kevin Glass
- Artur Biesiadowski
- Abdul Bezrati
- Arne Müller
- David Wallace Croft
I wouldn't mind participate to it but it would be for learning purpose and it's to you to judge if I would be useful or not.
About the JSR (Java Specification Request), it's a part of the JCP (Java Community Process, see http://www.jcp.org
) it seems to me the "formal" way to make a good lib ^^ Of course it doesn't avoid to work a lot, but I think Xith would gain more visibility and professionalism.
Do you think it's a good idea to send a JSR for a game-oriented scenegraph ?