[arin-ppml] DRAFT POLICY ARIN-2011-1: GLOBALLY COORDINATEDTRANSFER POLICY (Legecy space)

[Owen DeLong]

On Apr 12, 2011, at 2:17 PM, Milton L Mueller wrote:

> 
> 
>> We're tangling up a bunch of issues, including -
>> 
>> A. Can a current legacy IP space owner / holder / whatever sell the IPs at all?
>> 
>> B. Can a current IP owner sell IPs to a speculator?
>> 
>> C. Or only to a legitimate user with need who failed to get an
>> allocation from a RIR?
> 
> 
> The answer to A is, imho, unambiguously Yes, and the authority for that statement is ARIN's own legal filings. ;-)
> Not to mention common sense: ARIN and this community have no authority over what someone not under contract to them does with assets they hold.
> 
Can you cite the filings to which you refer or produce copies of them for the list? I think it would
be very interesting to see you back up this claim.

> As for B, if you mean by "current owner" someone under the RSA, then the answer is unambiguously No.
> 
> As for C, if you follow the party line, someone who failed to get an allocation from a RIR is, by definition, lacking justified need and therefore should not get it under current policy.  
> 
I don't believe that is an accurate statement. I believe the more accurate statement is that as of today, if
you can justify need, it is unlikely that getting the addresses somewhere else would cost less than getting
them from an RIR. Therefore, it probably makes little sense to acquire addresses elsewhere if you have
justified need. Once the RIR(s) run out of available IPv4 addresses, it will be possible to qualify with
an RIR for justified need which the RIR cannot satisfy. In such a case, you would have justified need
while failing to get an allocation/assignment from an RIR and a transfer may be the best/oinly way
to obtain resources if it is possible at all.

Please do not misconstrue my statement above to indicate that I think you should be able to acquire
addresses anywhere without justified need. I believe allowing people to do so is contrary to the
good stewardship of the address resources and should not be permitted in any case.


> Just in case people still don't understand this: I believe that the transition to IPv6, if it ever happens, is going to be long and messy, and that the demand for ipv4 addresses is going to increase for quite a few years before it decreases. Therefore, an overriding priority of policy must be to facilitate rapid and efficient transfers of v4 blocks from holders who need them less to holders who need them most (and I mean "needed" in the economic, not technical sense). 
> 
I disagree. I think the transition to IPv6 (and the state of IPv4 services) in a relatively short time will become
so completely messy that people will rapidly move to IPv6 as the only possible way to make things
at all less messy. I will be surprised if IPv4-only internet access is widespread or if single-stack IPv4
servers/services are commonplace for another 5 years. I will be even more surprised if there is
readily available residential IPv4 internet access from most providers in about 10 years.

Yes, the next 5 years are going to be messy, with a sharp increase on the left side and a more bell-
curve shaped right side. (The left side may even look parabolic). The opportunity to avoid messy
ended about 5 years ago when we still had the ability to push vendors and deploy IPv6 over the
course of the last 5 years of IPv4 free pool.

Owen