If we want to be serious about following the SemVer spec, there can be no API changes in 4.0.x releases and basically only bug fixes not any real feature-improvements. Furthermore, there can be no breaking changes without an increase in the major number. This would be a change compared to how previous releases have been handled, where breaking changes was allowed even when only the second number changed.
Anyway, regarding current branches: 3.3.x release series. Latest is 3.3.4. I don't think we should bother doing any more 3.3.x releases. 3.4.x series. Current is 3.4.0. Looking at NuGet statistics compared to NH4.0 so far, it seems interest is low. I don't think we should bother much with features or improvements here. We can fix regressions and important bugs (especially if there is no workaround). 4.0.x series. This is still the master branch. We could let 4.0.x remain as master for a little more time and have some more fixes and do a 4.0.1, before we branch it off and start work on 4.1.0 on master. Delaying the branching means a little less merging between releases. Time-wise I'm thinking something like: 4.0.1 in september 4.1.0 in november or december Thoughts? /Oskar -- --- You received this message because you are subscribed to the Google Groups "nhibernate-development" group. To unsubscribe from this group and stop receiving emails from it, send an email to nhibernate-development+unsubscr...@googlegroups.com. For more options, visit https://groups.google.com/d/optout.
