I think the moral of this story is that the input to the process is fallible. Shit always happens. Automated systems that filter or delay the 'happening' should be backed up by statistics to show that they help... Otherwise, when they filter and delay attempts to fix problems by people who are trying to help, they will just cause frustration. -Cam On Fri, Oct 1, 2010 at 11:56 PM, Matthew Garrett <mjg59@xxxxxxxxxxxxx> wrote: > On Sat, Oct 02, 2010 at 12:45:14AM +0200, Kevin Kofler wrote: >> Matthew Garrett wrote: >> > "Some packages were pushed to stable before they should have been, >> > therefore we need to make it easier to push packages to stable"? >> >> Yes! Sure, this sounds paradoxical, but my premise is that NO MATTER how >> strict you make the requirement for pushes to stable, there will ALWAYS be >> the possibility that "sh*t happens" and thus a need to be able to rush out >> fixes to stable as quickly as possible. > > And my premise is that we should be making harder for shit to happen, > and the cases where it *does* should be examined carefully to determine > the best way forwards. "Force this untested package into stable" isn't > the best way to do things. > > -- > Matthew Garrett | mjg59@xxxxxxxxxxxxx > -- > devel mailing list > devel@xxxxxxxxxxxxxxxxxxxxxxx > https://admin.fedoraproject.org/mailman/listinfo/devel > -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel
