On Thu, 2005-02-03 at 19:02, Gilbert Sebenste wrote: > On Thu, 3 Feb 2005, Dominic Hargreaves wrote: > > > (ie 1.5 years after Red Hat EOL). I would say that we should drop it no > > later than the introduction of FC2 support in March, though, because > > three distros is already fairly unmanagable. I'd be happy to drop it > > earlier, though. > > I agree. Given very limited resources right now, everyone should be at RH > 9 or later. And since every 6 months you're going to have a new OS to do > patches on, the EOL's are probably going to have to be reduced unless > additional funding/programmers are brought on. if there are people willing to do 7.3 patches i don't see the need to drop it. if people are willing to actively support any EOL'd RedHat release they should be able to do that via Fedora Legacy. the important point is that versions that do not have as much interest should not slow down updates for versions that have more interest. in other words, i think we should have a process that says we release updates for versions as they make it through QA- rather than queuing updates until all versions are ready. this would make it easy to judge what versions people are interested in. also, this would eliminate the problem of setting an EOL date. also, too much time is spent getting BuildRequires correct for mach. getting upstream FC-4+ packages to have the correct BuildRequires would dramatically cut the time required for many updates. does mach support x86_64, ppc, and more? if not we will have to rethink our buildsystem for the future. perhaps we can get more mileage out of our limited resources if we streamline our build process. i'd like to see legacy cvs trees, perhaps even on cvs.fedora.redhat.com. does anyone already have scripts that can take an srpm and make the appropriate modifications, adds, deletes and commits to the cvs repository? sorry for rambling. rob. -- fedora-legacy-list@xxxxxxxxxx http://www.redhat.com/mailman/listinfo/fedora-legacy-list
