[arin-ppml] Policy Proposal 2008-6: Emergency TransferPolicyfor IPv4 Addresses - Last Call

Tom Vest tvest at pch.net
Thu Jan 1 15:37:52 EST 2009

Hi Milton,

On Jan 1, 2009, at 12:52 PM, Milton L Mueller wrote:

>> How are transfers going to affect
> (drum roll...)

I guess we can consider what follows to be the jangling noise that  
conventionally follows the drum roll ;-)

>> registration data quality,
> Seems it would improve it by requiring transfering parties to go  
> through ARIN and secure their title via accurate registrations

I'll grant you that transfers might indeed improve registration data  
quality for those transferring parties *who choose to* go through ARIN  
because (a) they care about securing their title, (b) they think that  
securing it is best accomplished by accurate registrations, and (c)  
the think that accuracy is best achieved by registering with ARIN (or  
any other monolithic public registry). So for every transfer  
transaction, six independent variables (three each for both the  
parties involved) must be aligned properly in order for things to turn  
out as you suggest. Perceptions of overall registration data quality  
tend to fall exponentially as individual registry entries lose the  
qualities of timeliness, completeness, and completeness, so while it's  
anyone's guess how/how much overall quality would have to suffer  
before address resource registration data came to be regarded as "as  
useful as route registry data", my prediction is that it would not  
take all that much.

>> the risk of (intentional and/or
>> unintentional) address collisions
> ditto the above


>> the continuing viability of cross-
>> jurisdictional IP networks/services,
> has anyone on this list noticed that RIPE-NCC has passed its  
> transfer policy? shouldn't this be changing the complexion of the  
> debate somewhat, or do we all assume here than only the US matters?

How do you think that it should change the complexion of the debate  
Milton? Are you assuming here that one choice by one region is enough  
to settle all aspects of the matter globally?

>> the future likelihood of IPv6
>> adoption
> Most likely, _if_ v6 migration goes slower and a transfer market  
> drives up the price of v4 blocks it will increase the incentive to  
> move to v6. A lot of "ifs"

When the price of gas goes up, most US consumers don't jump into their  
electric cars or switch to public transportation, because the US  
economy has not presented most consumers with those options. Nor do  
you see a lot of home tinkerers building their own personal electric  
cars because the market has yet to deliver. If the price of IPv4  
blocks goes way up, no doubt many incumbents will make use of  
alternatives address resources -- that has never been disputed. But  
with the fully loaded (even if imperfectly/unevenly realized) public  
IP addressing standard effectively gone, there's no reason to assume a  
priori that individual enterprise-level addressing strategies are  
going to somehow sum up to something like a new (equivalent if not  
better) public IP addressing standard -- i.e., something that, at the  
very least, would enable new entrants to participate the market  
without securing permission from a rival.

Sure, there are always a lot of "ifs" in any assumptions or claims  
about the future. That's another master fact that doesn't excuse us  
from applying historical knowledge and common sense to plan for where  
to go from here.

>> or any other TCP/IP-related development?
> a bit of a vague question, no?

Not so much vague as all-encompassing -- but regardless I will try to  
be clearer. Lots of people are critical of IPv6 for lots of reasons.  
The statement was intended to cover IPv6 or any other future public IP  
addressing standard. But even more generally, there seems to be some  
(growing?) interest in the IETF community about why so many ostensibly  
useful new developments and standards have failed to be implemented  
over the years. One result of that interest is RFC5218 (http://tools.ietf.org/html/rfc5218 
). The text attempts to cover the problems specifically associated  
with IPv6 (e.g., in 2.1.2), but the RFC was really written from the  
standpoint of IPv4-based incumbents. In order to make it applicable  
for the coming era, the list of failure conditions will have to be  
expanded to include "lack of public IPv4 addresses," and all of the  
correlates of success will have to be amended with the clause,  
"Possess sufficient IPv4 addresses AND..." whatever notionally  
objective success condition was listed before.

>> When the amateurs leave the field, are the professionals
>> going to take over, or will "policy" simply wither away
>> altogether, as earlier utopians have repeatedly predicted?
> Policy will never go away, and the creation of markets heightens the  
> need for making (good) policy decisions.

The creation of a new market is itself a policy decision, and one that  
should take into full account what is known about the conditions that  
can be reasonably expected to result from the introduction of market  
incentives and all of their likely/common manifestations, given (no  
less, but no more than) reasonable efforts regulate or limit those  
manifestations. Some people still argue the market for credit default  
swaps could have been (& could still be) perfectly successful had  
there been sufficient institutional transparency and disclosure -- but  
these tend to be the same sort of people that staunchly defend the  
freedom *from* transparency and self-disclosure for both natural and  
artificial persons, under the right to privacy. You can't have it both  
ways, Milton, unless you're willing to just accept whatever comes as  
the best that one could have done, by definition -- a.k.a. "fatalism."

> As for the "amateur" charge, I would say as someone with 25 years of  
> (often professional) experience in policy, law and regulation of  
> communications that I would not be so dismissive of the Internet  
> technical community's perspicacity. One often finds a shocking  
> historical and institutional ignorance in this community, some  
> bizarre ideological prejudices (as one finds in any community),  
> coupled with a keen sense of the implications of policy decisions on  
> the technical infrastructure in the here and now. It's a decidedly  
> mixed bag, and it's best to maintain civil dialogue that fosters  
> mutual learning.

Hear hear! A civil dialogue and mutual learning sounds like another  
excellent idea for 2009 :-)



More information about the ARIN-PPML mailing list