On Thu, 8 Feb 2007, Ralf Corsepius wrote:
co-maintain them, give other maintainers privileges to rebuild the packages when all agree upon such step?
Yes if I was to do this today I'd say we both make everything up to "make tag", then I shouldv'e waited for the next release mail and after that build first the lib then both dependencies as soon as that build is confirmed.
Having a build-system which requires maintainers to "give explicit clearance before a package is being pushed to the public"?
Yep, the programming equivalent is a "race condition"... I we had e.g. a webpage where you can confirm your (and others) builds to be pushed, we could have spent a week building this lib and its dependent apps, then pushed all of them in batch, breaking nothing.
I believe this is what e.g. Debian use the "unstable" and "testing" repos for... Hm, and I was against that at one point, now I will reconsider that stance.
Linus -- fedora-extras-list mailing list fedora-extras-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-extras-list
