Kevin Kofler wrote: > Stephen Warren wrote: > >> Horst H. von Brand wrote: >> >>> And presumably you (and everybody else) would wait out the "until known >>> good" period; and as nobody tried it before, get to keep the pieces of >>> the resulting breakage... >> If that is true, then it would mean there's nobody who wants bleeding >> edge. That in turn would mean that Fedora should be redefined to not be >> bleeding edge, because nobody wants it that way... > > The problem is that users are asking for contradictory/impossible things: > they want new versions as soon as possible, i.e. the day upstream releases > them, but also updates tested for weeks. > Fedora currently has a good > compromise (new versions normally get 1-2 weeks of testing, and major > changes known to break things are only pushed to Rawhide), In theory. However, does anything/one enforce that? I guess it'd be difficult to do that programatically in Bodhi. Is this theory emphasized enough to maintainers? I don't remember reading that it should work this way, although it's been a while since I read the packaging wiki thoroughly. I'm sure there are plenty updates that have gone straight to devel, F $lastest, F $latest-1 at the same time. > people who need > something more conservative should be using a more conservative > distribution. > > And there's also a Prisoner's Dilemma problem here: users moving to the > conservative update stream => fewer testers for updates-testing and updates > => more breakage => more users moving to the conservative update stream and > the vicious circle is complete. > > Kevin Kofler > -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
