who gets a /32 [Re: IPV6 renumbering painless?]

Sun Nov 14 19:40:55 UTC 2004

> > ...
> 
> Actually, the policy also specifies that you must not be an end-site.

well, you sure caught me this time.  in august 2002 when the /32 in question
first came to isc, i had not read the policy.  so i don't know if it was
different from the current policy.  i assume it was, because i know that
we qualified, officially, under the rules at the time the /32 came to us.

> I'd be particularly interested in knowing what ISC said who would be their
> 200 other organizations who they intended to allocate the address space
> (their employees?), and how ISC would not be an end-site.
> 
> This is a more generic issue, of course.

of course.  in august 2002 there were no v6 isp's.  isc is multihomed, so
it's difficult to imagine what isp we could have taken address space from
then, or now.  we do allocate /48's to various open-source organizations
who get their transit from us, but it could take us some years to add 200
of these.  if arin's allocation policy for ipv6 does not take account of
multihomed non-allocating enterprises then either that policy will change,
or the internet exchange point business model will be dead.  speaking as
a co-founder and former president of PAIX, i don't like that idea at all.
speaking as someone who's had too much coffee today, it seems possible that
the preponderance of arin's membership could prefer a pure transit world
to a mixed transit/IXP world.  but since woody has been on the arin board
for a while now, i don't think that such a decision could have been taken
quietly.  i hope that if nothing else, this proves that the issues are
more complex than which policies arin followed when allocating 2001:4f8::/32.

(and note, as before, that i'm not speaking for arin.)