On Wednesday 28 October 2009, Zach Welch wrote: > I am becoming fairly discouraged by the ongoing high-pace use of the > central repository, seeing this type of commit pushed without sufficient > review period. Are we trying for 0.3.0, or not?!?
I was kind of expecting you to cut an RC some time ago; so I've been wondering the same thing ... On the other hand, nobody exactly signed up to be the release manager for the 0.3 cycle, and that's likely why it hasn't been happening. > > I would like to see: > > > > - 0.3 out of the door. Shut down sf master for anything but > > bugfixes as of friday 30. to wed 4.? Cut release nov 5.? > > At present, I do not see what the point is. We might just as well cut > the current master. I think the quality will hardly be unchanged, until > we implement better processes. My notion was to cut an RC, and have a one week "please everyone, test the **** out of this" cycle. That'd be an effort at quality improvement. And I can't think anything _less_ than that would really affect quality for _any_ release whatever. That's why I fired up my Beagle the other day, and fixed the nasty bugs that turned up, for example ... If we don't have any RC cycle at all, yes -- just tag the current stuff, call it 0.3.0, and maybe the quality experiments would involve backporting bugfixes to a new 0.3.1 branch. > > - that we can agree that major features are developed in public > > branches on repo.or.cz and then applied to sf when they > > are as ready as they are going to be. Communication can be > > via the mailing list and explicit posts. > > I think that I would rather see us dump SF.net as the primary repository, > switching to a site that does not _require_ central administration for > clones of the repository. I continue to be adamant that pushes to the > main tree should be done with machine-like rigor (i.e. scripts only). This is git; *NOTHING* requires central administration like that. Ever, under any circumstances. Every developer now has a clone... Scripts-only is IMO unrealistic. It's always going to come down to people; who's going to run the scripts, resolve conflicts, etc? > > W.r.t. testing I believe that we have to push changes to sf master > > before they are tested beyond the few developers working on > > a particular feature branch. > > That might be true, but the voice of those "few" developers deserve to > be heard before unleashing new patches on the public. The "public" is more likely to use the 0.3 code whenever that eventually happens... - Dave _______________________________________________ Openocd-development mailing list Openocd-development@lists.berlios.de https://lists.berlios.de/mailman/listinfo/openocd-development
