[arin-ppml] Policy Proposal: IPv4 Recovery Fund

Eliot Lear lear at cisco.com
Mon Nov 24 16:56:49 EST 2008


Thank you for clarifying your position, and attempting to support it.  
The crux of the matter is really in the analogy below:

On 11/24/08 10:06 PM, Kevin Kargel wrote:
> It is rather like your home budget when you are laid off..  income is
> reduced but you are pretty much at the same expenditure level, you can
> stretch things out by robbing Peter to pay Paul, but in reality all you
> are doing is digging a deeper hole.  The thing you really need to do is
> find a new job (IPv6).  Of course you have the option of reducing
> expenditures, but that really does little to prepare for the future when
> unemployment insurance (IPv4) ceases to come in.

I would not argue that there is an alternative to IPv6 in the long run.  
However, to carry your analogy further, were I to become unemployed I 
probably would eat less fancy food, take less expensive vacations (if 
any), and perhaps if necessary look for a lower paying job that might 
not satisfy my every desire.  In this case there are a plethora of lower 
paying jobs, one of which is multiple-layers of NAT (e.g., peanut butter 
& jelly sandwiches).  As a technologist I would argue that multiple 
layers of NAT are technologically inferior to both our existing v4 world 
and  IPv6 for all manner of reasons.  But when we look at what is 
easiest, and what costs least, I am forced to look at multiple layers of 
NAT, unless there is a cheaper alternative.  Will an address market be a 
cheaper alternative?  Perhaps.  One really doesn't know prices until 
we've seen trades.

And so this leads to the key question that requires careful examination: 
will the policies that have served us well in the past continue to serve 
us well in the future?  In answering that question we have to understand 
the underlying assumptions in those policies.  And so I will ask: what 
underlying assumptions argue for a first-come-first-serve approach with 
IPv4, and do those assumptions continue to be valid?  Secondly, are 
there other factors that weigh against continued use of FCFS?

Finally, there is a tone on this list in many of the threads about this 
subject that people and organizations who do not move to IPv6 somehow 
deserve what they get.  Policies cannot afford to take such an approach, 
because the people and organizations we are talking about serve and are 
part of the community.  I doubt that many government agencies throughout 
the world have even begun to consider conversion.  That includes 
hospitals, social security systems, and, just imagine departments of 
motor vehicles throughout the U.S.  Thus an argument for optimal 
policies that cause least pain through transition.  And so another 
question; what impact will each policy (both current and proposed) have 
on the community?

Consider this last one carefully: government agencies are MOST put at a 
disadvantage with a black market because they are often precluded from 
participation, both by law and by the fact that black marketeers 
generally don't deal with governments.


More information about the ARIN-PPML mailing list