On 24.07.2009 20:01, Steve Borho wrote: > On Fri, Jul 24, 2009 at 12:26 PM, Adrian Buehlmann<adr...@cadifra.com> wrote: >> On 24.07.2009 17:58, Steve Borho wrote: >>> Now that 0.8.1 is out the door, it's time to concentrate on 0.9. >>> >>> The first large steps will be happening soon. I am going to apply >>> names to the two existing lines of development. The 0.8 line of >>> development, which is present in both the stable and crew >>> repositories, will be given a branch name of '0.8'. The 0.9 line of >>> development, which is only present on the crew repository, will be >>> given a branch name of '0.9'. >> I think setting a branch name for 0.9 *now* is rather bad idea, IMHO. >> This should simply be the default branch now. >> >> Main development ("trunk") should happen in default branch. >> >> This also fits with what you get when you do a fresh clone: Mercurial >> updates to the tip of default branch. >> >> Feature freeze of 0.9 should be the earliest birthday of the 0.9 branch. > > There do seem to be two schools of thought about branches. One is for > release "trains" that start immediately after the current branch goes > stable. The other is to have a main trunk for continual development > and for branches to be created when you want to make a release. > > In practice, the only difference seems to be that the development > leading up to the each release is not on any branch. Do you prefer > the development trunk approach for the semantics of branch == "bug > fixes only" and default/trunk == "where features get added"?
I fail to see the point in answering this question. What's the problem in simply continuing on default branch for non-0.8 stuff right now? What's the default branch intended for? ------------------------------------------------------------------------------ _______________________________________________ Tortoisehg-discuss mailing list Tortoisehg-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/tortoisehg-discuss