On Thu, 2007-02-15 at 14:19 -0600, Tom 'spot' Callaway wrote: > On Thu, 2007-02-15 at 15:12 -0500, Jason Corley wrote: > > So compatibility should only be maintained when someone at Red Hat > > says it's worth keeping. > > Well, I wouldn't go that far. If a small group of people are reliant > upon a program's bad behavior, should you fix the behavior to be sane > and make life much simpler for the majority of users, or ignore the bad > behavior to keep the small group happy? > > In this case, fixing the bug so that the default buildroot is not / is > trivial for people dumping binary bits into RPM.
The point is not that we won't ship with a default buildroot, but that it should not be hard coded within the source. I don't have any real objection to rpm.org releases shipping with a buildroot defined in macros, but I do think we should avoid overly putting policy in the code base. We can have a released default, just please not in the code. Paul _______________________________________________ Rpm-maint mailing list [email protected] https://lists.rpm.org/mailman/listinfo/rpm-maint
