[arin-ppml] DRAFT POLICY ARIN-2011-1: GLOBALLY COORDINATED TRANSFER POLICY (Legecy space)

[Milton L Mueller]

> Milton -
>
>   What is your basis for this assertion? ('"needs based" does
>   not and cannot work when there is no free pool')

Very simple, really. If there is no free pool you can have N number of claimants who can prove "need" for a resource, whereas there are only enough resources to fulfill N-X of those claims. 
By virtue of that fact, the ultimate allocation is based not on need but on something else: competitive bidding, lottery, political pull, whatever. 

The immediate point is that it makes little sense to invoke RFC 2050 in the current environment as if it were decisive. 

The other problem is that the meaning of need changes under scarcity. Operator A could claim they "need" a /16, someone else could claim that A really only needs a /24 if they reconfigure their network or use NAT. What you think you need often depends on what you have to pay. 

>   Are you actually suggesting that a market simply cannot work
>   if there are conditions on who can participate?  It would be
>   useful if you could cite a reference for this to help educate
>   others.

Markets are not purely dichotomous, either working perfectly or not working at all. Sure, you can have a very constrained, expensive and thin market for a few players. But if one discovers most of the trades going outside of that constrained market it tells you that willing buyers and willing sellers prefer to go elsewhere, and possibly there's something wrong with the "conditions"