On Mon, Feb 13, 2012 at 11:10 AM, Martin Heidegger <m...@leichtgewicht.at> wrote: >> Martin, Nick brings up something here that we need to be very careful of. >> If we hope to graduate from the incubator then we need to be sure we're >> doing things the "Apache Way". If we start working over on github I think >> there is a very real danger that there could be two code conversations at >> which point I believe we would be putting this project in jeopardy.
I've been following the conversation and right now I don't see a problem. I'd place the following parameters around working on things at github: * all discussion about the code should happen on list * github is an acceptable way for non-committers to work on patches. They will need to be submitted as Jira tickets to be included in our main repository. * long, drawn-out work efforts should not occur on github and then be brought to the community. Let us know what you are working on, work on it a bit, then bring your work back here so we can all see it. * committers should favor doing work in branches, whiteboard, etc. in our svn tree. Yes it's harder for non-committers to contribute, but when a non-committer starts to contribute he/she should not be a non-committer much longer :-) To summarize, I don't have a problem with people using github as a place to share ideas. But those ideas should be discussed on-list and developed in our repo as much as possible. As a disclaimer, there's been *tons* of discussion about git and github at Apache. My views may be way off of what the overall foundation position as a whole is, so this is subject to change. But I think we're ok as long as we focus on sharing work here on-list and getting stuff into our svn repo as soon as possible. Greg
