Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
[b3-dev] Opening up b3 with Gerrit

Hi all,

This echoes a request that was done earlier on some not-so-related bugzilla ( https://bugs.eclipse.org/bugs/show_bug.cgi?id=485570#c2 ) and Planning Council ( https://wiki.eclipse.org/Planning_Council/February_03_2016#New_Business ).
We (Red Hat) are willing to contribute to Simultaneous Release to make it more agile, ideally driving it to continuous delivery at some point (no real deadline, it's mostly "whenever we can"). We have established some possible plans of actions to improve things progressively and incrementally. We're really thinking in term of evolution rather than revolution, for several reasons: it's easier, less risky, and the most important one is that we believe that b3 is technically quite capable of facing the challenges coming together with continuous delivery; it actually doesn't require important changes.
The improvements we'd like to make are mostly on the b3 editor, in order to empower its users to use more reliable practices. Note that so far, it's all about empowering; then it will be recommending and then maybe constraining if it seems acceptable.

Currently, only David Williams seems to be following changes happening on the b3 project. David is already a busy guy dealing with other more critical things in Eclipse world, and made clear in bugzilla ( https://bugs.eclipse.org/bugs/show_bug.cgi?id=485570#c4 ) that despite he seems interested in the changes we'd like to get merged, it's not something he can take care of as fast as we'd like.
So now we're thinking about how to make the b3 project more agile and open to changes. Concretely, it seems to be all about getting more people contributing to b3. That's why as a new contributor, I would recommend the b3 project to start using Gerrit. Gerrit has proven its effect in increasing the number of contributors to a lot of projects, and in making the contribution process more "trustful" if coupled with Jenkins, leading also to some positive effect on the general quality. Gerrit for Eclipse project is quite simple to set up, and not much intrusive so it doesn't require specific adoption effort. In any case, if one of this effort were be too big for the current b3 committers, we could share this effort and I would be glad to take as much as possible of it.
Does anyone have any idea of the additional effort that would be involved if moving to Gerrit?

Can the b3 project consider enable Gerrit soon? It would mostly be a matter of opening a bug against Eclipse.org admins.

Thanks in advance,
--
Mickael Istria
Eclipse developer at JBoss, by Red Hat
My blog - My Tweets

Back to the top