Re: First draft of review of policy must usage
On Wed, 25 Oct 2006 18:54:13 -0700, Steve Langasek <vorlon@debian.org> said:
> On Wed, Oct 25, 2006 at 10:26:44PM +0200, Luk Claes wrote:
>> >> This flows from the Release policy. Not specifying
>> >> /bin/bash
>> >> in scripts is not considered a RC bug.
>> > Oh Gosh! This _is_ a RC bug. If I change the /bin/sh to point to
>> > /bin/dash (and I usually do that) some script will break and this
>> > cannot happen since they said, when defining their shell as sh,
>> > that they're using just POSIX functionality.
>> Indeed, it's a bug, though it's not release critical yet...
>> > That kinda of change in policy can decrease the overall quality
>> > of packages. Release policy is _wrong_.
>> It's an important bug and might become release critical after
>> etch...
> In practice, such bugs were treated as "borderline" RC even for
> sarge, and I don't think I've downgraded any such bugs myself during
> etch. This isn't a good time to make such an adjustment to the RC
> policy without some idea of the impact, but I'd be happy if all
> maintainers treated such bugs in their packages as if they /were/
> RC, if nothing else. (And in general, I think fixes for non-serious
> policy bugs should be considered by the release team during the
> freeze...)
So you would have no objection if this change was reverted in
the draft policy (which is now slated for a post Etch release)?
>> I don't think it's a good idea to change the policy manual for
>> every release to match release RC policy. In current policy it's
>> clear that it's a bug that should be fixed in the package and I'm
>> sure it will become release critical in a next release...
> I would expect the RC policies to differ only incrementally between
> releases, so don't see updating policy for every release being a
> real problem, if this is the direction people think we should go.
I think that once there is agreement on the sanity of policy
dictums, that technical and release policy be modified in lockstep in
areas of overlap.
manoj
--
/* dbmrefcnt--; */ /* doesn't work, rats */ --Larry Wall in hash.c
from the perl source code
Manoj Srivastava <srivasta@debian.org> <http://www.debian.org/~srivasta/>
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
Reply to: