[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Planned packages for sarge

On Jul 23, Anthony Towns wrote:
> On Mon, Jul 22, 2002 at 07:38:18PM -0500, Chris Lawrence wrote:
> > On Jul 22, Richard Atterer wrote:
> > > On Mon, Jul 22, 2002 at 12:48:10AM +0200, Tollef Fog Heen wrote:
> > > > And no, sarge will _NOT_ release with boot-floppies.  It will use d-i
> > > > and/or PGI.
> > > ...and that will be the cause for a one-year delay of the sarge release.
> > Free incentive for d-i to be the default for sarge and done on-time:
> > any architecture that can't be installed using d-i will be dropped
> > from sarge N months after d-i is working on i386.
> Sounds a lot more like a threat than an incentive.

Perhaps.  But I've always said nothing concentrates the mind like a
deadline, something sorely lacking most of the time in Debian.  Yes,
we'd get crap from people for missing our deadlines (like we did for
missing the "interpreted-as-a-deadline" May 1 release date), but IMHO
a reasonable, achievable deadline is worthwhile.  And, realistically,
if d-i cannot be ported to a new architecture in N months (where N is
a number like 3 or 4), either d-i has failed as a concept or that
architecture doesn't have enough active developers to be worthwhile to
release for sarge, neither of which is realistically the case for
either d-i or the architectures themselves.

Maybe it is a threat.  (Incidentally, it is the exact same
threat/incentive that you and other RMs have used with individual
developers who have buggy packages that are holding up the release -
hence IMHO it's eminiently fair, just targeted at a different group of
developers.)  But if $ARCH doesn't have anyone (either a developer or
someone in the NM process) who is willing to put in the effort to port
whatever arch-specific stuff needs to be done in d-i from
boot-floppies, $ARCH is an impediment to progress and probably not
worth the project receiving flak until we find someone who can be
bothered to do the work.

BUT, I don't think we should release sarge for just i386 (or any
subset of architectures) when d-i is done.  sarge would be supported
on the arches with a working d-i and be released for all those arches
at the same time.  If any arch got on the ball and ported d-i between
sarge and sarge+1, a point release of sarge (3.1r1) could target that
architecture as well.  If it didn't, it'd still be in sid for when
someone did decide to get to porting (though I would drop any
architecture completely that missed two consecutive full releases).

Chris Lawrence <cnlawren@olemiss.edu> - http://www.lordsutch.com/chris/

Instructor and Ph.D. Candidate, Political Science, Univ. of Mississippi
208 Deupree Hall - 662-915-5765

Attachment: pgp5k5xe7bWwi.pgp
Description: PGP signature

Reply to: