On Sunday 10 June 2007 08:17:03 Hans de Goede wrote: > Well I'm the most active member of the Games SIG and testing my own updates > doesn't feel right (ofcourse I test them, but I shouldn't be the only > tester). > > As for Keving, thats hardly a procedure one can count on for future > updates. > > So I guess I'll just push the mark as stable button then. I don't think we're ever going to be able to come up with a policy that will work for every package / every update. How one gets a package from -testing to stable is really up to that person and/or group that the package resides in. This gives the maintainer and SIG the flexibility to define their own policy, hopefully with some overall project level guidelines. Overall like "please use updates-testing first" and "please address bugs / comments filed during -testing". As for games, couldn't the games SIG come up with a strategy for making game updates to a released platform? One that is geared specifically for games and makes use of the people in the SIG? Sometimes you as the maintainer /are/ the best person to test things, and part of that is verifying that A) it fell out of the buildsystem correctly, B) it has the right information in the update ticket and the email content got generated correctly, C) it upgrades cleanly from the previously shipped one through the public repo. If you've been able to verify all of this, and the application itself tests out fine, I see no problem with marking your own update as stable. By being in updates-testing you've given others the opportunity to verify this and more as well. -- Jesse Keating Release Engineer: Fedora
Attachment:
pgpg0oPCyhX4m.pgp
Description: PGP signature
-- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
