[arin-ppml] ARIN-prop-176 Increase Needs-Based Justification to 60 months on 8.3 Specified Transfers

[Owen DeLong]

Sent from my iPhone

On Jun 28, 2012, at 9:36, <jeffmehlenbacher at ipv4marketgroup.com> wrote:

> Owen, I appreciate your continued concern for the health of our business
> model.  However, your suggestion to wait for the free pool to run out
> (http://www.potaroo.net/tools/ipv4/index.html) and then wait another
> 6-12 months to view partial statistics within ARIN and ignore all other
> statistical evidence outside of ARIN's statistical capture isn't
> terribly compelling is it?  Projecting free pool depletion and adding
> your outer limit of another 12 months might suggest we not touch 8.3
> justification duration until sometime 2014.  One might be accused of
> "head in the sand" thinking if the community embraced such a position
> while inter-RIR transfers are afoot between ARIN and APNIC in July 2012
> followed by ARIN and RIPE in September 2012.
> 

If such transfers create enough of a market to have some statistical viability, I'm fine with looking earlier. My point is that the current lack of a vibrant market does not in and of itself represent a policy failure.  A vibrant market this early would, in fact indicate that the policy was too wide open. 

> Again, in the here and now, I state based on our extensive dialogue with
> prospective buyers that there can be two categories of planners: those
> that will systematically continue to go to the 90-day well for free
> allocations and those, that if properly motivated by increasing the
> justification from 24 months to 60 months, will enter the 8.3 Specified
> Transfer market under ARIN policy.  If efficiently allocating unused
> IPv4 legacy resources via policy improvements on 8.3 specified transfers
> prolongs IPv4 utilization and defers IPv6 adoption, I empathize only
> with the people and organizations who base their business models on
> evangelistic trumpeting of all things IPv6 and comment on IPv4 transfer
> policy proposals on that basis.   

I don't doubt that. The question, however is whether that would be good for the Internet.  Geoff Huston has an excellent presentation where he explains the dangers inherent in RIR runout skew. (different runout dates in the regions)

Indeed, a primary goal of the last 5 policy was to reduce that skew and narrow the time gap between various RIR runout dates. Unfortunately, hat policy had the somewhat opposite effect, compounded by the early trigger date in the three Month policy. 

Making that even worse by encouraging the behavior you describe above could be very harmful, making CGN and NAT64 effectively permanent structures of the Internet. 

Current local NAT is bad enough. Let's not deliberately make the situation even worse. 

Owen

> 
> Jeff Mehlenbacher
> 
> 
> Let's see what the transfer statistics look like 6-12 months after ARIN
> free pool exhaustion. At that point, if there are still not many
> transfers,
> then I will accept that 8.3 is not working as intended. Until that time,
> I would argue that increasing the number of transfers by skewing
> policy to give undue advantage to market transfers over free pool
> allocations/assignments is contrary to good stewardship.
> 
> 
> -------- Original Message --------
> Subject: Re: [arin-ppml] ARIN-prop-176 Increase Needs-Based
> Justification to 60 months on 8.3 Specified Transfers
> From: Owen DeLong <owen at delong.com>
> Date: Thu, June 28, 2012 7:14 am
> To: <jeffmehlenbacher at ipv4marketgroup.com>
> Cc: "Scott Leibrand" <scottleibrand at gmail.com>, arin-ppml at arin.net
> 
> 
> On Jun 28, 2012, at 3:05 AM, <jeffmehlenbacher at ipv4marketgroup.com>
> wrote:
> 
>> I understand your desire to sit tight and assess statistical evidence
>> before suggesting a longer justification period is required. My concern
>> with such a strategy is the decided lack of comprehensive transfer
>> market data. We have only the ARIN Specified Transfer Listing Service
>> (STLS) and ARIN Statistics on 8.3 transfers (# Requested / # Completed).
>> Is there another source you would consider useful to assess the
>> appropriateness of extending the 24 months justification period?
> 
> I consider that adequate. (primarily the latter and not the former).
> 
> 
>> The STLS would suggest there is no market at all. In 2012, there was an
>> update in February (a Facilitator was added) and two updates in June (a
>> Facilitator was removed and another added). The last and only Seeker
>> (Needer/Buyer) was December 2011 and it was a post-facto pre-approval of
>> a public transaction in progress. Only two Listers (Sellers) have ever
>> posted since November 2010...the most recent Lister being October 2011. 
>> It may be safe to say the STLS is a vehicle that doesn't attract
>> interest and thus cannot be used as a reliable statistical measure for
>> the purpose of monitoring the appropriateness of the current
>> justification period. Now, if we implemented 60 months justification
>> for a 1-year period and monitored uptake on the STLS for
>> Listers/Seekers, that might be evidence that longer justification
>> periods a) bring unused resources (listers) to market and b) bring
>> organizations with need (seekers) forward.
> 
> The fact that transfers have been completed would suggest that there is
> a market at a time when rational policy would render one irrelevant or
> at least nearly non-existant. So long as there is a free pool, policy
> which
> makes transfers more attractive is contrary to the good of the
> community.
> 
> Judging anything about specified transfers and policy effect therein is
> premature until after free pool runout.
> 
>> ARIN Statistics on 8.3 transfers further suggests the market is far from
>> vibrant: 24 requested in 2011; 12 requested 2012 year-to-date. The
>> issue with these statistics is the limited insight one can glean. Yes,
>> the number of completed transfers is published but it is a running total
>> and doesn't correspond directly, on a month to month basis, with the
>> number of transfers requested. Further, we do not know anything about
>> the recipient organizations including block size requested, previous
>> ARIN customer or new ISP, ISP or End-User, size of last allocation from
>> the free pool...and most importantly, whether 24 months was too short a
>> window to obtain transfer approval.
> 
> Far from vibrant is the desired and rational state prior to runout as
> far as
> I am concerned, so I am not seeing the problem. I realize this doesn't
> necessarily help your business model, but, I think it is in line with
> the
> intent of the community. The market is currently competing against
> essentially free allocations and assignments (after all, you have to pay
> registration fees after you transfer through the market anyway, so,
> there's no price advantage there). This means that the only advantage
> to purchasing through the market at this time is the convenience of
> a longer allocation or assignment period.
> 
> This should result in a small and lack-luster market until free pool
> exhaustion and as near as I can tell, the result is in line with that
> intent.
> 
>> So what statistical evidence will constitute success or failure of the
>> current 24-month justification period? 
> 
> Let's see what the transfer statistics look like 6-12 months after ARIN
> free pool exhaustion. At that point, if there are still not many
> transfers,
> then I will accept that 8.3 is not working as intended. Until that time,
> I would argue that increasing the number of transfers by skewing
> policy to give undue advantage to market transfers over free pool
> allocations/assignments is contrary to good stewardship.
> 
> Owen
>