Hello everyone, as noted at the DDF and https://jira.opendaylight.org/browse/TSC-207, I would like to transition the controller project to being MRI.
There are multiple reasons for doing this: - it is mandated by our goverernance since day one, it's just that our processes were not up to par in the early days. That is no longer the case, as the history of odlparent/yangtools/mdsal clearly shows. - it will lower our infrastructure resource consumption (autorelease, distcheck jobs) - our codebase is very stable, with good mechanisms to support versioning going back as far as Lithium, there simply is little pressure to get changes out immediately - being part of MRI cycle will improve ergonomics for downstreams, as we will be entering the MRI version bump with controller fully ready Unlike the previous MRI split-offs of odlparent/yangtools/mdsal, I do not believe controller is ready to also transition to strict SemVer rules -- we have not consolidated our codebase yet and we will need to sweep our codebase at some point for defunct/unused remnants. I would hate to see that effort being hampered by artificial rules. That having been said, we should (and will, if I can help it) maintain reasonable compatibility where our downstreams are concerned. I am therefore raising the following vote for controller project committers, please vote with usual +1/0/-1: Do you agree to controller transitioning to being a Managed Release Integrated project? Thanks, Robert, voting +1 :)
signature.asc
Description: OpenPGP digital signature
_______________________________________________ controller-dev mailing list controller-dev@lists.opendaylight.org https://lists.opendaylight.org/mailman/listinfo/controller-dev
