[arin-ppml] Q2: on Address Transfers - Overkill on the freezeperiod?

[Milton L Mueller]

> -----Original Message-----
> From: Robert E. Seastrom [mailto:rs at seastrom.com] On Behalf Of Robert
E.
> 
> Note that these are both restrictions on the transferor, not the
> transferee.

Yes, I understand that. And if your target is truly speculation, the
restrictions seem to be targeting the wrong side of the transaction. 

Note that true speculators are likely to be, indeed must be, first
recipient and then sources of addresses. 

You can't speculate without first buying and then selling, and the pure
speculator wants to be able to control the time at which they sell to
capture the highest price. So a long-term restriction on what the
recipient does with transferred addresses is much more on target than a
restriction on the source. 

> The goal here is to create a situation where organizations *who are
> not using their address space* are incented to give up some or all of
> it.  

Indeed. So restrictions on the transferor (source) are more likely to
interfere with the incentive to release address resources in a way that
undermines the goal of the policy. 

As Scott noted, the only thing you want to do with the source is make
sure that they don't sell their assignment and then come back to ARIN
for more for free. 

The only other reason you have cited to impose a prior 2-year restraint
on sources is to prevent "fabricated requests" and "creating a run on
the market." 

This argument shows a rather interesting lack of faith in the vaunted
"needs-based assessment." Are you suggesting that you don't think ARIN
staff can distinguish between real and fabricated requests? If it can,
this is no problem, because unnecessary requests will not be rewarded.
If it can't, then the run on the bank will occur even if there is no
transfer policy of any kind. 

> I would be
> an enthusiastic proponent of expanding the timeouts on both sides of
> the event to three years or whatever the time between "right now" and
> the best guesstimate of IPv4 ARIN-exhaustion-of-final-block is,
> whichever is less.

This attitude reveals a lack of awareness of the cost-benefit tradeoffs
that will define how the policy actually works in practice. If the
timeouts on sources are too long you undermine the incentives to
release. So if you recognize the need for a transfer policy, as you seem
to do, reluctantly, you can't just expand the timeouts indefinitely.