On Wed, Oct 04, 2000 at 08:24:05PM +1100, Hamish Moffatt wrote:
> On Tue, Oct 03, 2000 at 07:28:54AM +1000, Anthony Towns wrote:
> > On Tue, Oct 03, 2000 at 08:15:48AM +1100, Hamish Moffatt wrote:
> > > The problem is that our definition of release critical is flawed.
> > > A package should only hold up the release process if it has
> > > an important (or higher severity) bug report, and the package itself
> > > is important.
> > This is largely what happens anyway, though, no?
> In effect, yes. Perhaps they could be left in the distribution
> though; a grave bug in an extra package is not a big deal,
Errr. A grave bug could be due to, say, a /tmp race that lets any random
user on a student.cs.some-u.edu randomly trash any other student's thesis
work. Or it could randomly trash your entire home directory when run
with --help as its only argument. Leave it in unstable, maybe, but not
in stable.
(We've switched from "It doesn't work for me, therefore no one must
have it" to "It might work for someone, therefore everyone must have
it", hmmm...)
Cheers,
aj
--
Anthony Towns <aj@humbug.org.au> <http://azure.humbug.org.au/~aj/>
I don't speak for anyone save myself. GPG signed mail preferred.
``We reject: kings, presidents, and voting.
We believe in: rough consensus and working code.''
-- Dave Clark
Attachment:
pgpRcwxgWv6vs.pgp
Description: PGP signature