[ppml] Legacy users and ARIN duties

[John Curran]

Good afternoon Dean -

At 1:50 PM -0400 7/28/07, Dean Anderson wrote:
>If address space is a problem, the very _first_ thing that should happen
>is to slow down the assignment of new resources.  In contrast, ARIN is
>_accelerating_ the delegation process. I'm a little dubious of this
>combination.  The conjuction has a lot in common with the effects of
>people trying to hoard up the remaining space, and people at ARIN
>helping them do so.

"People at ARIN" implement the policies that come through the
public process.   "ARIN and the Internet community" is likely a
better antecedent.   "We" (including all those on this list and in
the meetings) is probably the shortest, most common term.

>The transition will be a lot more "interesting" if ARIN blindly just
>allocates space until it is all gone one day.  "Stewards" are expected
>to avoid abrupt changes.

Community consensus on policy changes prior to depletion will
determine the nature of the transition.   Some advocate for a
status-quo situation till depletion as businesses rely on predicable
policies, and some recommend changes so as to affect the nature
of the transition.   The ARIN Board resolution made it clear what
we'd like to see in the 7 May 2007 resolution:
<http://www.arin.net/announcements/20070521.html>.
In particular, we'd like to have policies that encourage IPv6
transition, and have asked the Advisory Council to consider
this issue.  That has resulted in quite a few policy proposals
and lots of community discussion and all of that is a good thing.

>I'm glad you brought up Enron. The Enron Board had to give back a lot of
>money for their mismanagement.  Bechtel just returned a Billion dollars
>to Boston/MA/Feds for their BigDig engineering and construction
>failures.  I think an abrupt 'oops we're out of space' is going to
>result in some very definite legal challenges to ARIN and its
>management.

An interesting assertion, but I do not believe there is a valid claim
that this will be an abrupt change, unless the community fails to
act responsibly in light of all of the preparation.  We started the IPng
task force in the early nineties, have had a stable Draft standard for
IPv6 since 1998, and many vendors have been shipping production
code for 5 years. Now, it is true that we have 2 or 3 years left for
general availability of large IPv4 blocks, and that it would be wise for
all to include IPv6 connectivity for their public facing servers in next
years budget.  If your an ISP, you've likely got more work to do, but
that isn't news to anyone.

Some may claim that our formal announcement of the forthcoming
changes IPv4 block availability hasn't given them enough time to
transition, but that's a judgement call which I personally feel we've
balanced very well.

/John