[arin-ppml] 2014-2 8.4 Anti-flip Language

[Owen DeLong]

On Feb 23, 2014, at 6:32 PM, David Farmer <farmer at umn.edu> wrote:

> On 2/23/14, 14:38 , Steven Ryerse wrote:
>> This is an example of how policies penalize legitimate organizations
>> needing to do legitimate transfers.  In my opinion the Polices have
>> swung so far towards preventing abuse they impact legitimate transfers.
>> In the publicly traded company world, each quarter is like a year and a
>> year is like 4 years to them since they have to publish their quarterly
>> results 4 times a year.  For them a year is an eternity.  Also, the
>> Internet itself has allowed business functions that once took months or
>> years to take days or weeks which is todays reality.  Some of y’all
>> think 12 months or even 18 months is a short time but that doesn’t align
>> with the reality of today’s business world that the Internet has helped
>> foster.
>> 
>> Although I don’t like abuse either, I am definitely AGAINST raising the
>> hold period to be longer than a year.  If it has to be a year then at
>> minimum, there should be a procedure defined in the policy that an
>> organization can appeal to the ARIN CFO (or whoever) to get an exception
>> for a shorter timeframe – even as short as 30 days.  If what the
>> organization is doing is legitimate and the ARIN CFO will approve it,
>> then the 12 month hold rule should be waived.  My 2 cents.
> 
> 
> I've been thinking about this maybe the restrictions for anti-flipping don't belong in section 8 at all.  Maybe they belong in section 4 as they are intended to protect the ARIN IPv4 free pool.

I disagree. I don’t want to see flipping become a tool for speculation in the market post-exhaustion, any more than I want to see it become a tool for draining the free pool. In fact, I think that the former might be significantly more harmful than the latter at this point.

> The restrictions were needed because we enabled Inter-RIR transfers, so they were included with the Inter-RIR transfer policy.  I'm beginning to think this may have been a mistake.  Tactically we allowing IPv4 allocations to remain liberal and restricting transfers.

I don’t see a problem with that. I have no desire to encourage transfers as a primary choice. I think it is, in fact, just bad policy to do so. Transfers should, IMHO, be viewed as a last resort when free pool options have been exhausted.


> It may be more effective to be more restrictive on the allocation of IPv4 resources and more liberal on transfers.

I see no benefit whatsoever from this approach. I think it would do more harm than good. cf. deck chairs.

> I think part of the reason we didn't do this is that we were trying to minimize the changes to allocation policy because of run-out and not wanting to have uncertain effects on the end-game of run-out.  So we've placed the restrictions on transfers as the new policy items rather than the the allocation policies.
> 
> I'll note, that several people have advocated liberalizing transfer policies for a while now.  However, they have not also advocated the accompanying restrictions that would be necessary on the allocation policy side.  I'll also note, that at this point it may be to late for this type of radical change.

I’ll also note that several people have been arguing against significant liberalization of transfers and I have not seen any indication that there is anything approaching consensus towards such liberalization.

Owen