Re: [sig-policy] prop-057-v001: Proposal to change IPv6 initial allocati

To: Leo Vegoda <leo.vegoda at icann dot org>
Subject: Re: [sig-policy] prop-057-v001: Proposal to change IPv6 initial allocation criteria
From: Randy Bush <randy at psg dot com>
Date: Sat, 26 Jan 2008 10:33:20 +0900
Cc: sig-policy at lists dot apnic dot net
Delivered-to: sig-policy at mailman dot apnic dot net
In-reply-to: <C3C040F1.E644%leo.vegoda at icann dot org>
List-archive: <http://mailman.apnic.net/mailing-lists/sig-policy>
List-help: <mailto:sig-policy-request@lists.apnic.net?subject=help>
List-id: APNIC SIG on resource management policy <sig-policy.lists.apnic.net>
List-post: <mailto:sig-policy@lists.apnic.net>
List-subscribe: <http://mailman.apnic.net/mailman/listinfo/sig-policy>, <mailto:sig-policy-request@lists.apnic.net?subject=subscribe>
List-unsubscribe: <http://mailman.apnic.net/mailman/listinfo/sig-policy>, <mailto:sig-policy-request@lists.apnic.net?subject=unsubscribe>
References: <C3C040F1.E644%leo.vegoda@icann.org>
User-agent: Thunderbird 2.0.0.9 (Windows/20071031)

I don't know the numbers but it would be interesting to know whether there
is considerable pent-up demand that would be released with a wording tweak.


<rant>

i love the tradition of speaking for those not present. one can claim

anything! personally, i know that, if we painted the prefixes cerulean

blue, that 2,319 more users would deploy ipv6.

but looking at actual measurements, e.g. foil 3 of the study of actual

announcement, <http://rip.psg.com/~randy/071022.v6-use.pdf>, we see that

allocations of ipv6 space within the apnic region continues on a steep

curve. this would hint that the barrier to *allocation* is negligible.

on the other hand, on that same foil, note that actual bgp announcement,

i.e. *use* of ipv6 allocations has been flat. i.e. people get the space

and do not use it. this should give us a very big clue.

at best, removing the 200 limit will allow a few hoarders to pick up

ipv6 space cheap which we will later complain about just as we complain

about legacy ipv4 holders today. cheap thrills.

so sitting here tweaking some trivial words is a lazy and useless path.

as is spinning fantasies about ipv6 saving battery life.

what is needed is to make ipv6 *usable* for the isps' customers. this

is not so easy. i know as i have been putting some effort into it.

one is just documenting getting some servers and services converted. i

have been keeping notes at <http://rip.psg.com/~randy/ipv6-westin.html>.

i solicit and would appreciate further contributions and collection of clue.

second is planning an educational experiment at nanog, apricot, ietf,

ripe, ... where ipv4 transit is shut off for an hour or so, and we show

that pure ipv6 does not really work, what different ipv6 crutches are

needed to actually deploy at an enterprise-like end site, and how to

deploy those crutches so isps can do this for their customers. see the

working notes at

<http://www.civil-tongue.net/clusterf/wiki/NANOG%20IPv6%20Hour%20-%20February%202008#NANOGIPv6Hour-February2008>.

again, contributions and clue would be appreciated. the engineering

design team for this will actually be meeting monday in the bay area.


randy