From: Heikki J. J. <hj...@gm...> - 2009-08-25 11:33:18
|
2009/8/25 Chris Cannam <ca...@al...> > On Tue, Aug 25, 2009 at 6:45 AM, Heikki Johannes Junes<hj...@gm...> > wrote: > > I do not want to commit if the code does not work properly. That is a > dilemma. > > That is the nub of the problem, but I'm not sure that the right answer > is (effectively) to make it easier for people not to commit publicly > at all. It is the right answer for me. So, it is not the wrong answer. > Surely, the right answer is somehow to get people to be more > relaxed about publishing "provisionally" their work into a branch even > if they are not yet happy with it. > My answer for this problem is to start using git, or hg. If you really feel a strong need to make local commits, why not just > run a local git repo and use git-svn to commit from it to the central > repository? It's a well enough established method. That's true. It is a working method. Although it is a bit lame, because you replace partly svn with git, because svn is fundamentally broken in this sense. Anyway, I try to experiment also git-svn. -- Heikki |