[arin-ppml] A Redefinition of IPv4 Need post ARINrun-out(was:Re:Against2013-4)

Mike Burns mike at nationwideinc.com
Tue Jun 18 17:16:14 EDT 2013

Hi Owen,

As for examples of Whois inaccuracy, how exactly would they be substantiated?
I can tell you that I am aware of several transactions which have occurred in this manner, but I am not able to divulge the participants.

However, I was a bidder in the Nortel/Microsoft deal,  and I can tell you that not a single one of the 38 blocks transferred had the name Nortel on it in Whois. Particularly with legacy space, which has no contract assigned which mandates any particular form of transfer, the sale or lease of address rights to another party is legal.  This is one of the very few public transactions which could be mined for proof of Whois inaccuracy, yet it yields evidence.

You have agreed in the past that it is possible to advertise addresses for which the name in Whois does not match the name of the advertiser. When you put the facts together- no legal obstacle to the transaction involving legacy space and no operational obstacle to the transactions- along with the pecuniary factors, you have a recipe for off-the books transactions which would be invisible to the ARIN community.

>From my perspective, there are absolutely some buyers and sellers who are willing to engage in off-the-books transactions, although I will concede that they are in the minority.  

Thanks for considering support for an experiment in removing the needs test for transactions of a /22 or less per year or in aggregate.  I do consider that size too small to have much of an effect on the market, but even that kind of policy change might encourage a seller to consider slicing off chunks of his block and engaging in many transactions with multiple buyers, because one element of transactional uncertainty-the buyer’s needs test- would be removed.

I began this discussion off-list with as an attempt to take the community’s temperature in regards to a potential policy proposal to remove needs-testing of transfers, at least in a limited way. I hoped that my experience as a broker of addresses would be informative, and that potential movement in RIPE away from needs-testing transfers might elevate consideration of such a proposal here.

Also, can you send me (off list) anybody you are aware of who can consult on ARIN justifications? Thanks!


From: Owen DeLong 
Sent: Tuesday, June 18, 2013 4:14 PM
To: Mike Burns 
Cc: arin-ppml at arin.net 
Subject: Re: [arin-ppml] A Redefinition of IPv4 Need post ARINrun-out(was:Re:Against2013-4)

On Jun 18, 2013, at 7:14 PM, Mike Burns <mike at nationwideinc.com> wrote:

  Hi Jason,

  1. It has been argued that the larger ISPs have the prior advantage of holding highly valuable alienable assets which they received for free, which provide them with a competitive advantage over less endowed entities seeking to purchase addresses at a much higher relative price. 

Yes, it has been argued. It hasn't necessarily been substantiated, nor has anyone raising said argument provided any real evidence to support it.

  2. It has been argued that larger ISPs have greater experience and resources required to navigate the justification process, which provides them with a competitive advantage over less experienced smaller entities.

Again, argued, but not neither of the above statements has necessarily been substantiated.

For example, a number of smaller ISPs have hired consultants to help them navigate the justification process. These consultants cost much less than the FTEs employed by the larger ISPs in support of the process, so, it could be argued that smaller ISPs have a competitive advantage because of that reduced cost of expertise.

I'm not sure that argument holds water, either, but I don't completely buy into all the "competitive advantage" arguments about the larger ISPs. I think there are tradeoffs at any size and that while different sizes bring about different tradeoffs, the ARIN process is, generally, about as fair as it can be.

   3. Other registries have enacted policies restricting access to their last /8s to provide an advantage to newer and smaller companies through their /22 maximum restrictions.  Applying your logic, this restriction allows the tiny ISP some years of planning, but larger entities only a few days, so I assume you also reject these policies for reasons of fairness.

Those policies are actually to provide a disadvantage to any size established company in favor of nonexistent entities.

  Your post could be read as a plea to remove needs testing transfers altogether, for both the large and the small, in in the interests of fairness, which I would also support. I believe the duration of the planning horizon should be a matter of each business to decide on its own. As of now, that duration is mandated by ARIN policy, which I believe is unfair and arbitrary. Unfair and arbitrary for transfers, but not un-necessarily unfair and arbitrary for free pool allocations.

I do not think that it is unfair or arbitrary to have a time-horizon limit on the amount of resources one can remove from availability to others with need at one time. In fact, I think it is quite unfair to eliminate that limit.

I believe that remains true regardless of whether those resources are being made available from a 3rd party or from the registry. 

   Remember the cap on needs-free transfers is designed to free up the market to incentivize more transactions, each of which presumably entails the move of addresses from lower-use states to higher-use states, while providing some protection for market manipulations. I don’t believe that market manipulation is a real threat, my discussion of the cap is in the context of providing some protection for those who do think it is a threat.  I do think un-booked transactions are a real threat, and the lifting of needs-testing transfers is designed to protect the integrity of Whois.

I challenge you to defend that presumption. I remain utterly unconvinced that money alone can define the "higher-use" of addresses. Further, I see no reason to believe that if the protections put in place (including needs basis) in the current transfer policy are removed that it would be at all unlikely to see addresses purchased purely for purposes of speculating on the value of addresses in the future, thus moving them from under-utilized to perhaps un-utiliized.

I don't know whether market manipulation is a threat or not. If you reduced the non-needs based cap to /22, I would be willing to accept the experiment. At /12, it's utterly inadequate protection and I cannot support it.

I don't believe that un-booked transactions are as likely to occur as the anti-regulation zealots have made them out to be. For the most part, it is quite clear that those purchasing addresses want the addresses registered as a condition of purchase, so sellers are forced to go through the RIR system in order to complete the sales. As such, I'm a lot less worried about whois integrity (which so far has not been a problem) than about an unregulated market (which throughout history has proven time and again to be problematic at best in a multitude of contexts).

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20130618/4c0c1ddb/attachment-0001.html>

More information about the ARIN-PPML mailing list