On Wed, Jul 07, 2004 at 12:45:50PM +0100, Martin Michlmayr - Debian Project Leader wrote:
> * Josselin Mouette <joss@debian.org> [2004-07-06 20:58]:
> > you still haven't spoken a word about the absence of reaction of the
> > FTP masters regarding the inclusion of AMD64.
>
> There are 3 issues which have to be clarified and taken care of before
> AMD64 can move into the archive:
>
> - A general port inclusion policy: there are a number of pending ports
> (s390x, powerp64 and various BSD ports), and therefore it is
> important to have a clear policy saying which criteria a new port
> has to fulfil.
>
> - The mirror situation: the archive has grown quite substantially
> and some mirrors have problems keeping up with our increasing disk
> space requirements. The plan is to have only a common set of
> architectures mirrored by default, and mirrors can opt-in to carry
> additional architectures. The tools for this exist now; what is
> left is creating a policy about this and switching the mirrors to
> the new system.
>
> - Some technical AMD64 questions: ftpmaster had some specific
> questions about the AMD64 port they want to see answered. Also,
> an LSB person recently expressed some technical concerns (see [1]).
>
> As to the timeframe of these 3 points being addressed, I'm not sure.
> I have recently seen a draft policy about port inclusion which, in my
> opinion, looks pretty good. What is left is for ftpmaster (plus some
> other concerned parties, such as the security team) to agree on that
> policy and write a rationale for each point. (From what I can tell,
> the AMD64 port should not have a problem with most of the points in
> the port inclusion policy.) I'm not sure about the status of the
> other 2 points, but I hope that something will be posted soon.
Of the three points, I (as one of the folks involved with the BSD ports,
at least) am quite willing to grant that amd64 is worthy of special-case
treatment because of what it is, and where it's going. This doesn't require
the Amazing Kreskin to predict, and with the speed of our release cycle,
failing to figure out some reasonable way to support it *before* Sarge+1
(whether that be Sarge, or Sarge+=.1) is just going to make us look even
more ridiculous than we already do in the eyes of our users.
Point 2... I would suggest, if it's that critical, picking one of the ports
with a vastly lower popcon score. First come first serve is *not* a sane
way to handle this, if total size is the problem. People will scream "not
mine!" - but then, the amd64 folks are saying the same, and to me they're
just a little bit louder.
Point 3 I cannot address, since the ftpmaster concerns do not appear to be
visible in public; at least, I see no URL in your message saying where, and
one of the themes of the thread so far is the lack of visibility...
As for the latter... I think the policy is a great idea. I think letting
the policy cripple releasing the fourth most popular architecture out of
(proposed 12) is, however, in a word... insane.
--
Joel Baker <fenton@debian.org> ,''`.
Debian GNU/kNetBSD(i386) porter : :' :
`. `'
http://nienna.lightbearer.com/ `-
Attachment:
pgpC64huwiOYs.pgp
Description: PGP signature