On woensdag, apr 30, 2003, at 02:52 Europe/Amsterdam, J. Noel Chiappa
wrote:
The Multi6 WG just went over geographic addressing again, for about
the 17th
time in the IETF (I seem to recall previous go-rounds on it in CIDRD
and
IPng), and finally left off talking about it, on the grounds that
there were
enough people against it that it was extremely unlikely there would
ever be
rough consensus for it.
I wouldn't exactly characterize the discussion this way.
Yes, we discussed it, but I don't feel my draft got a fair hearing.
I've seen very little evidence that people actually read it, most of
the arguments are of the quality "my aunt Bertha connects to two ISPs
in different continents so geographic aggregation can never work".
We sorta kinda agreed that we need a draft to put the issue to bed for
a good long time, but nothing concrete here.
I suggest the same calculation ought to occur in this forum, i.e. that
there's no point discussing geographic addressing, since there's never
going
to be rough consensus for it.
This sounds awfully similar to what the French said about a new
security council resolution about Iraq: regardless of the content of
your proposal, we'll veto it.
For those of you who want to form your own opinion, this is the
addressing draft that uses a /16 for the entire world (just for
multihomed networks, others continue to use PA space):
http://www.muada.com/drafts/draft-py-multi6-gapi-00.txt
And this draft outlines the aggregation procedures inside ISP networks
(so no announcing of aggregates between ISPs):
http://www.muada.com/drafts/draft-van-beijnum-multi6-isp-int-aggr-00.txt
(They both expire this month so I'm not sure if they're still in the
IETF drafts repository.)
Any questions are probably best handled off-list.
Iljitsch van Beijnum