[ppml] Policy Proposal: 2007-12 IPv4 Countdown Policy Proposal

Owen DeLong owen at delong.com
Wed Mar 21 18:41:06 EDT 2007

On Mar 21, 2007, at 3:04 PM, Ted Mittelstaedt wrote:

>> -----Original Message-----
>> From: Owen DeLong [mailto:owen at delong.com]
>> Sent: Wednesday, March 21, 2007 2:43 PM
>> To: Ted Mittelstaedt
>> Cc: Leo Bicknell; ppml at arin.net
>> Subject: Re: [ppml] Policy Proposal: 2007-12 IPv4 Countdown Policy
>> Proposal
>>>> - Reclamation of unused address space.  It doesn't matter if we do
>>>> this
>>>>  or not, all predictions are we still run out of address space.
>>> This is an extreme simplification that is essentically  
>>> incorrect.  If
>>> relamation were to exceed everyone's estimates then it might push  
>>> the
>>> runout date so far in advance that it would become theoretical.  I
>>> agee the chances of this are small but the are not nonexistent - so
>>> in fact, reclamation does have a place in the discussion.
>> And, if the aerodynamic coefficient of monkeys could be modified
>> sufficiently, then, they could fly, perhaps even out of my butt.  Get
>> real.  The odds of any reclamation effort succeeding to such an
>> extent are so close to zero as to not even be good theory.
> :-)
> Is there some usage of the phrase "extreme simplification" that  
> escapes
> you?  I did like the monkeys image, though.
The primary point was that the likelihood of getting more than 3 or so
years out of reclamation efforts is so low that calling the exhaustion
date theoretical is absurd.

>>>> - Are the predictions of when we run out correct?  Same problem,
>>>> doesn't
>>>>  matter if it's 2010, 2020, or 2050, the question is what do we do
>>>> when
>>>>  it happens.
>>> If it is 2050 then we are setting policy prematurely if the  
>>> policy is
>>> not going to come into effect for another 43 years.  You and I will
>>> certainly both be retired, very likely both dead of old age.  We do
>>> not
>>> have the moral right to dictate policy to our children for a
>>> community problem that will arise after we are dead of old age.
>>> We only have the right to set policy that we are going to live by.
>>> I also have the same objection to the continual
>>> immoral lengthing of copyright terms, by the way.
>> Wrong again.  We have the right and, indeed, the obligation to
>> set appropriate policy for likely circumstances now.
> Now <> 2050.
True, but, IPv4 exhaustion, by all rational estimates, is much closer to
now than to 2050.  Since there is a likelihood of IPv4 exhaustion within
the next 10 years (not 40), there's no reason not to set whatever policy
we think is appropriate for that event now, even if we're not sure the
event will occur.
>> No matter
>> what policy we set now, it will likely be changed by our children
>> and perhaps their children between now and 2050, so, this is no
>> reason to avoid setting policy now.
> Hmm..  So we should set policy that we think is likely to be changed.
Yes.  If we have half a clue, then, we realize that all policy is  
to change.  Even the constitution can be (and has been) amended.
Change is inevitable, and, if we wait for the perfect opportunity and
the perfect policy, then we'll never do anything.  A common term for
such a situation is paralysis by analysis.
>>> You might as well write policy now for the runout of IPv6.
>> Personally, I think it should be the same as the current policy
>> regarding
>> the runout of IPv4 (which I also think is perfectly fine),
> You mean the nonexistent policy for IPv4 runout that we have now?
I think we have a perfectly fine policy.  ARIN will continue to allocate
space until there isn't any to allocate.  When/if more space becomes
available through reclamation or other mechanism, then, they will
begin allocating again.  I don't see a problem with that policy.

I also would have no problem with the idea if someone wanted to
propose a policy to address IPv6 exhaustion.  I'm not saying I would
necessarily support the policy, as it would depend on what the poilcy
was, but, I would support the idea of discussing the policy and running
it through the process, just as I did with 2007-12, even though I do
not support that proposal.


More information about the ARIN-PPML mailing list