jme-contril will host code, libraries and utilities for the JME 3d engine. It will function like xith-tk for xith allowing jme users to contribute code such as model loaders, scene managers etc that cannot be hosted to jme’s repository because of strict access requirements. The startup of the project will contain cal3d and md5 loaders and any other code supplied by jme community.
Even if I don’t use JMe I gave a +1, because I know from the xith-tk, that such is really useful.
A very good idea, to bypass the jme-core cvs-access!
Great idea kman! 
I’m not sure about jme-contrib yet - I did not vote, but:
I was the first making an actual ‘poll’ here : (instead of just a topic) and it proved to be a bad idea! There are 6 votes on the counter but only three comments. I’d suggest to remove/lock the vote to force people to post comments instead…
I thought the MD5 loader was never contributed to jME. It was just made by ChaosDeathFish with (IIRC, a Copyright notice on it). Did you get his approval for this ?
This whole project voting thing is shambles anyway, no point is saying any opinions regarding it. Its not like you can have different read/write permission on different modules, that would just be stupid wouldn’t it ?
DP
Yes you’re right about the copyright. The Md5 loader i’m speaking is a new version that has nothing to do with ChaosDeathFish’s code and it is under devlopment now.
For the jme-contrib project it will provide code that will add extra functionality for jme and will have more flexible permissions for jme community members and a central place to find such user contributions.
- bump *
I voted yes. theKman talked to me about a Cal3D loader and suggested I could do Xith support, and I’m interested. Now we probably have to think of a way having common loaders and adapters for each lib. It would be benefit for the two libs and much less maintaining pain. We could even adapt Croft’s COLLADA loader to jME. Collaboration’s really important, or java game development will stay/become under-represented.
I think is a great idea as it will ease to differentiate the current core + basic modules from other add-ons, along providing a ‘place’ where to look for those extras without having to find out by yourself through deep-searching on the forums and so 
Contributions will be more up-to-date with the latest jme versions that way, and if an author decides not to continue with his/her own contributions others can continue developing them without problem. Also everyone can get help improving those contributions too keeping all in synch.
Also, small contributions like utility classes of every sort could get their place here. Sometimes you end up writing lots of code that some else did as you encountered the same situations. So the possibility of posting on a centralized place those sort of things could be great, and would help everyone to reduce development times. Those utility classes don’t fit on the current jme repository, as they are just for some kinds of situations. For example, a i18n-enabled resource manager, an auto-detecting format converter for models, several sorts of in-game debugging panels, etc…
So, I hope this gets done, as it would really help everyone a lot
who the hell voted “no”? speak up bozo! >:(
I did, and you’re the bozo.
First of all, all the things you wanted to do with it can be done using a little bit of elbow greese and modifications of the CVS server to make a “contrib” module in the jme CVS. In that module, you can modify permissions easily and make them more flexible. I.e. the contrib people who have write access to the contrib module can’t have write access to the core module and so on and so forth…
I really, and I mean really, dont see a point in 3 jme reps for the heck of it…
DP
“for the heck of it” = java.net offers extremely limited visibility of the permissions for CVS and though I haven’t attempted to do this I don’t believe it is possible. I have been wrong before * insert shocked gasps here *, but please explain how to do this in Java.net if it is possible.
-Matt Hicks (aka darkfrog)
Since the existence of xith-tk for xith i believe DPs arguments are invalid. From the beginning of jme there are several contributions with code that exists only in jme forums.
I really dont see whats the point for not having a jme-contib project hosting utility code and other stuff supported by developers who dont work on jme’s core . And since DP is an ex- jme - contributor who had several aruments against the way jme is heading i really dont care about his negative vote( both here and in jme-physics) I dont need cvs access to a user contibuted jme repository . I’m making a proposal for project that adds extra functionality for an existing java.net project.
Since i am working with model loaders maybe i will submit a proposal for 3d-model loaders and maybe i get DP’s positive vote 
That was a low blow. Me being an ex-developer has nothing to do with this. I have my own projects these days and they use modules in CVS/SVN…
I did not know there was precedence to this as I dont follow the Xith community very well, but since there is precedence now, go for it. While your at it, have jme-shape, jme-renderer, jme-ai, jme-frustum, jme-math, jme-entity, jme-soundrenderer, jme-sounddecoder, jme-blah, jme-somemoreblah…It all makes sense from your perspective.
I wonder how all other projects in the world managed with just 1 rep, they must be ridiculously stupid. Yes, that must be it…they are all stupid…
DP
If you read the first two posts you’ll notice about xith-tk and the concept of this proposal.
This projects main goal is to host all the things you mention and it will be governed by developers working on stuff like this without restrictions existing in the base jme project.
so if you dont like a java.net project like this i’ll setup it on mine servers and when the codebase will reach some certain levels of functionality i will repost a proposal with no jme in its name or description and perhaps then i will get your positive vote ( another low blow ).
theKman,
I’m gonna have to side with DP, that was a low blow. Lets focus on the facts and try not to reduce this discussion to personal attacks. DP has not done so and from his perspective at least has raised some valid questions. Lets focus on appeasing his questions.
DP, if you can provide a legitimate means to provide modularity in java.net’s CVS repository that may allow us to do this without another project, but if you cannot we must create an additional project for such tasks.
-Matt Hicks
I do not know how to do it under java.net, however, SF.net has had that functionality for a very long time. Maybe consider sending an RFE to the java.net people to supply this functionality (i know sf.net allows python scripts to govern such actions)…
DP
let me tell this one more: since xith-tk is a java.net project providing means for adding extra functionality that is hosted and supported out of xith’s repository i don’t see any reason for not doing the same for jme. i just want this project to be a part of java.net since i believe it is the best for this project and the community. if i am wrong please forgive me. In any case jme-contrib will be hosted somewhere. I’m not going to continue trying to persuade people with other opinions. DP sorry for low blows. at the current time my work concentrates on model loaders with skeleton systems and i think that jme-contib is a good place to host such code.
I dont have anything more to say, and I dont care anymore about it either. Ive written what i needed to say and all I got was abuse for it…shows the level of professionalism this project will have.
End of discussion from me
PS. you have two negative votes, im only one of them.
DP