On Sat, Dec 09, 2006 at 10:00:02PM +0100, Thorsten Leemhuis wrote: > > So regular non-rawhide users are really not affected at all. > > The last release maybe. The situation might be different with other > releases. The numbers show that this was the case since Fedora's birth. Why should the future be different? And if history is not enough, then check the usage of disttags: Currently 97.3% of extras packages have a disttag, which means that these packages will be rebuilt no matter what from release to release. So again, there is no drawback for non-rawhide users. We will have these rebuilds between releases anyhow, so let's make a better use out of this situation and use mass rebuilds more effectively, e.g. both in a better automated fashion as well as for doing QA on test releases. > BTW: There was a strong opposition when we did the first mass-rebuild > for FE (and that was really needed iirc). Yes, because it was and is an unneccessary waste of man power if done non-automatically. > >>> In fact frequent mass rebuilds would disclose bugs that have > >>> gone by unnoticed, e.g. when a package's direct or indirect > >>> build dependencies change and the maintainer missed to notice > >>> that it affects his package indeed. > Sure. But is the rebuild one really better, even if there was no > reason for a rebuild? I would prefer a known good rpm over a > rebuilded one when there was no need for a rebuild. If the rebuilt rpm is worse then what you had before there is a bug somewhere, right?. Don't you want to find that bug? It will surface later someday anyway, so better to let it show up when it entered the system. -- Axel.Thimm at ATrpms.net
Attachment:
pgpnuchhu4KXR.pgp
Description: PGP signature
-- fedora-extras-list mailing list fedora-extras-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-extras-list
