I think it's a good idea to keep old release branches for security fixes and such. Otherwise, mainline trunk development seems to make sense to me.
On 13 July 2014 10:22, Gary Gregory <garydgreg...@gmail.com> wrote: > FYI http://semver.org/ > > Gary > > > -------- Original message -------- > From: Gary Gregory > Date:07/13/2014 09:59 (GMT-05:00) > To: Log4J Developers List > Subject: RE: Versioning/branching after 2.0 release > > I think we just work as usual in trunk and release when ready and use > semantic versioning. No need to maintain branches unless absolutely > necessary. API breakage is only for major versions. We need to decide if > that is just for the API module or all modules. > > Gary > > > -------- Original message -------- > From: Remko Popma > Date:07/13/2014 09:39 (GMT-05:00) > To: Log4J Developers List > Subject: Versioning/branching after 2.0 release > > After the 2.0 release, how are we going to do versioning? > > One idea is to have 2.0.x releases with bugfixes only, where new features > and any API changes would go into a 2.1 release. > > Another way of doing this is to simply have a 2.1 release next containing > both bugfixes and new features as well as API changes if any. > > And perhaps there are other ways... > > I don't have a strong preference about this but it may be good to discuss > this since it may determine how we do branching going forward. > > Thoughts? > > Remko > -- Matt Sicker <boa...@gmail.com>
