[arin-ppml] Draft Policy 2011-1 - Inter-RIRTransfers -Shepherd's Inquiry

[Owen DeLong]

On Jun 23, 2011, at 1:45 PM, Mike Burns wrote:

> Hi Owen,
> 
> You've been saying this stuff for years, but where is a section 12 reclamation of legacy space you can point to where the reclamation took place due to the original registrant selling his addresses?
> 
I don't know whether such a reclamation has occurred yet or not. Frankly, even if it has not, that does not mean that it will not.

> Where is any reclamation of legacy space which is not associated with actual hijacking, voluntary return, or dissolution? Spamming or child porn? Any examples?
> 

Just as you have no examples of transfers that have occurred outside of policy, I have no examples of this. I am not privy to the information and those that are would be unable to disclose it.

> The bankruptcy judge, need I remind you, found that Nortel had the exclusive right to transfer addresses which were allocated to other companies, and for which no ARIN transfer had ever taken place, and this was after all the companies involved were bankrupt. I find this much more convincing than allusions to RFCs and the writings of Jon Postel. Now if Mr. Postel had actually created an agreement which could be legally construed as a contract, you might have something. Absent that, you have only bluster which grows more outdated each month.
> 

That's your right. It does not change my opinion.

> As a test example, I was back at my alma mater last June and I spoke with one of the campus computer engineers, who told me although he had 16 million ip addresses to work with, most of the campus is Natted. Do you think I should open up a ticket and inform ARIN that they should institute a section 12 review of MIT? Would you like to confront ARIN with that knowledge and see if it nets the free pool a sizeable chunk of a /8?
> 

No. MIT is still using the addresses for the purpose for which they were issued and the efficient utilization of those addresses is not part of the requirements that existed at the time of issue. However, if MIT were to abandon the /8 altogether and/or to transfer it to a third party outside of NRPM 8, that third party would have no rights to the space in ARIN's eyes and unless MIT chose to continue to assert their utilization, ARIN would, indeed, be able to if not obliged to reclaim it.

> ARIN controls Whois, I'm pretty sure that is what John Curran will say. ARIN is within its legal rights to change Whois data in whatever way it pleases.
> If ARIN wants to make Whois even more irrelevant, they can take the course you prescribe, but since they have not done this, ever, the threat of reclamation of legacy space is a nonstarter.
> 

I am not convinced that they have not done this ever and even less convinced that they would not do so ever. I believe that ARIN has reclaimed abandoned resources in the past which non-entitled entities have claimed they received through transfer or have claimed they are the successor organization. I believe that ARIN has also completed 8.2 transfers and possibly 8.3 transfers when they have found successor organizations that have some legitimacy under those policies making such transfers retroactive (as I believe happened with Nortel, actually).

> John Curran referencing ex post facto transfer requests demonstrates that unreported transfers have occurred where the IP addresses worked fine after they were transferred to another party.
> If ARIN has no control over routing, and no legal ability to stop ip address sales, their leverage is little. If the benefits of the transfers outweigh the little costs which ARIN can apply, the transfers will happen.
> 

And nothing I have said precludes this practice continuing. Yes, there are transfers that occur without ARIN being notified and nothing in policy will change that fact. In most of the cases that result in ex post facto transfer requests, it is the result of the organizations in question being unaware of ARIN or its role and choosing to cooperate in that role upon becoming aware of it.

Owen

> Regards,
> Mike
> 
> :
> 
>> Hi Owen,
>> 
>> I don't think ARIN will be confronted with anything, I think they will be ignored.
> 
> When I say confronted, I mean becomes aware of. There are numerous ways in which it is very likely ARIN will become aware of such things.
> 
>> And I don't think ARIN has any rights to revoke or reclaim legacy space unless they have an agreement with the rights holder which gives them that right.
> 
> You are incorrect, sir. ARIN is the successor registry to the earlier registries for this region. Those registries ALL issued the address space on the
> basis that it was issued to a particular organization for a particular purpose and should be returned when one or both of those criteria was no longer
> valid.
> 
> 
>> Clearly they have no additional rights under section 12 to reclaim legacy space, that's explicit in the NRPM.
> 
> The words in NRPM 12 were chosen very carefully. The word additional is just that. We added no ability to reclaim legacy space, but, the existing
> abilities remain.
> 
>> So you would have to demonstrate some other source of that reclamation right.
> 
> Q.E.D. See above.
> 
>> It's not in any agreement between ARIN and the legacy holder, and it's not part of any agreement the legacy allocants had with ARIN's precursors.
>> Where does it come from?
>> 
> 
> It is actually part of policies documented in earlier RFCs and in the writings of Jon Postel as described above.
> 
> Owen
> 
>> Regards,
>> Mike
>> 
>> 
>> 
>> 
>> ----- Original Message ----- From: "Owen DeLong" <owen at delong.com>
>> To: "Mike Burns" <mike at nationwideinc.com>
>> Cc: "Kevin Kargel" <kkargel at polartel.com>; "arin ppml" <ppml at arin.net>
>> Sent: Thursday, June 23, 2011 3:39 PM
>> Subject: Re: [arin-ppml] Draft Policy 2011-1 - Inter-RIRTransfers -Shepherd's Inquiry
>> 
>> 
>>> 
>>> My argument is that we cannot abandon the whole IP market thing. It is out of our control, because the limited power we have is in Whois and reverse naming.
>>> We can't force network operators to route or not to route certain addresses, and I see no stomach at ARIN for challenging legacy transfers by revoking and reissuing their space.
>>> 
>> Like speculation, just because you don't see it does not mean that it is not there.
>> 
>> I believe that confronted with a legacy transfer that happened off the books ARIN will go through the following steps (John, feel free to correct me if I'm wrong):
>> 
>> 1. Attempt to contact the original holder of the resources
>> If they acknowledge the transfer, see if the recipient and original holder will work to
>> complete the transaction under NRPM 8.
>> 
>> If they do not acknowledge the transfer, let them know their space has been hijacked
>> and encourage them to take appropriate action.
>> 
>> If they are unreachable, work with the recipient to see if sufficient documentation
>> exists to complete an 8.2 or 8.3 transfer and attempt to do so.
>> 
>> 2. If the original resource holder is unreachable and a transfer cannot be completed
>> as described above, I believe ARIN will reclaim the space as abandoned and take
>> the appropriate steps to clean and reissue it.
>> 
>> 3. If the original resource holder is unreachable or uncooperative and the "recipient" refuses to cooperate,
>> I believe ARIN is also likely to proceed with reclamation under NRPM 12.
>> 
>> 
>>> So to think we have any significant power to prevent the rise of an IPv4 market is naive. What we should do is acknowledge it and take the steps to transition ARIN from an arbiter of who gets addresses to a title-agency whose routing authority is respected enough to handle the challenges facing registration in the post-exhaust world.
>>> 
>> 
>> Can we prevent the rise of a market? Probably not. Can we apply a needs-basis test to that market and regulate the redistribution of addresses through that market by applying reasonable stewardship policies governing the source and recipient conduct in the market? Yes.
>> 
>> There is a lot of room between the extremes of "ignore the market altogether" and "acknowledge the market and let it become a free-for-all with no rules".
>> 
>> I realize that you favor the latter extreme. I favor something towards the former, but much closer to the middle.
>> 
>> Owen