[arin-ppml] ARIN-prop-151 Limiting Needs Requirements for IPv4 Transfers

[John Springer]

Hi Mike,

Brrrr, this is growing massive already. I'll try to edit judiciously.

On Tue, 24 May 2011, Mike Burns wrote:

>>> 
>>>> What I am saying is that needs testing was a requirement for free pool 
>>>> allocations, and that is why we imposed needs tests on applicants for 
>>>> address space. Actually some kind of constraint was required, as it 
>>>> always is, for the distribution of unpriced but valuable resources.
>>> 
>>> One thing that I have seen a lot of lately is asserting that something 
>>> happened _for_a_particular_reason_, and then using that _reason_ as a 
>>> launching pad. WRT the above, I agree that needs testing was imposed, but 
>>> saying that it was a requirement, which is why we imposed it seems 
>>> circular.
>
> I have tried to show that having some constraint was an absolute minimum 
> requirement, and that a needs-based justification was the steward's softest 
> touch which would meet the goals of conservation and aggregation.

Was and is. Needs basis remains despite suggestions that recent events 
have obliterated it.

>>>> So the stewards chose needs-testing of applicants, with the idea that we 
>>>> want allocated addresses put into actual and efficient use, and this 
>>>> method of constraint would serve those goals.
>>> 
>>> Check.
>>> 
>>>> Now, however, we do not need needs-testing to drive addresses towards 
>>>> efficient use, because we are allowing them to be priced, and that is a 
>>>> fundamental difference.
>>> 
>>> Not so much check. We could restate this in reverse with equally valid 
>>> effect: We have needs testing to drive addresses toward efficient use, so 
>>> we don't need to be allowing them to be priced.
>
> You could say exactly that.
> Indeed, we did have needs testing to drive addresses toward efficient use, so 
> we did not need to price them.

And we still do and we still don't.

> Except that only works for allocations from the free pool, and we have 
> already decided to price them for transfers.

Not proven and not you and me, surely!

> My proposal is meant to be considered in an environment with priced 
> addresses, like our current one.

The proposal is being considered in a place that Gibson called "a 
consensual hallucination". Statements like the above may or my not have 
less semantic content than you think here, at least yet.

>>>> Tragedy of the Commons tip, cant delete that!
>>>> Now, however, whether with or without my policy, the stewards at ARIN 
>>>> chose to allow individuals to buy and sell addresses, and thus for 
>>>> addresses to have a price.
>>> 
>>> Don't think so. I know you keep saying so, but IIRC the STLS allows the 
>>> right to transfer the right to use, presumably, but not necessarily for a 
>>> consideration. Seems to me there is a lot of care exercised to be sure 
>>> that the phrasing does not give even the appearance of sell and buy.
>
> I agree the text looks like the tortured work of a group nearly evenly split 
> between allowing transfers for money and not allowing them.
> Yes, clearly there is no requirement for payment for transfers.
> But we already have a $7.5 million dollar example of an ARIN-blessed 8.3 
> transfer.
> Which I take to mean that we are already in the trading-for-money world.

When did we leave? I thought I talked to you about drawing conclusions 
from the MS/NN deal.

>>>> The price is the constraint on wasteful use that the needs test used to 
>>>> be.
>>> 
>>> Would be. Maybe. Among other things. Happy to talk about it.
>
> OK, my assertion is that holding other conditions constant, raising the price 
> of a commodity leads to more efficient use of it.
> Like, I figure that there is less left uneaten on the plate at Ruth's Chris 
> than there is at Sizzlers, in general, because people don't like to waste 
> expensive things.
> (There must have been some psychological test run at some point to test my 
> hypothesis, I will see if I can find one.)
> (And I suppose the taste of a Ruth's Chris steak is not constant with the 
> taste of a Sizzlers steak.)

And you chide me on my analogy/illustration/metaphor/whositz.

>>>> So maintaining the needs test is not a requirement for stewardship, as 
>>>> other forces outside of our control will work towards the same ends of 
>>>> efficient usage that the needs test was designed to incentivize.
>>> 
>>> Maybe not, if we were absolutely determined that at all costs the needs 
>>> test had to go. But we have not so determined. You propose that this is 
>>> desireable and are experiencing some discussion. Not everyone agrees.
>
> I guess I should say if  you accept the general notion that increased value 
> leads to decreased waste,

It might be time to slow here because I don't necessarily accept general 
notions. It is hard to tell the context of this conjecture from the 
proceeding text.

> then address space will trend towards meeting the 
> goal of conservation as it obtains monetary value.

This is almost poetry, and I almost understand what you mean, my soul 
almost begins to soar, but no.


>>>> I think some stewards have gotten it into their heads that their role is 
>>>> to decide the best, or at least better, use of addresses, but I argue 
>>>> that was never a goal of stewardship.
>>> 
>>> Here again: some stewards _MAY_ be considering that it might be better if 
>>> existing uses of addresses be prefered to proposed and possibly 
>>> undesirable uses. I would think YOU have to justify convincingly the 
>>> advantages of previously prohibited practices, not just assert repeatedly 
>>> those advantages.
>
> Yes, I agree the proposer of a policy change has the burden of having to 
> justifiably convince others before any change should be effected.
>

Well, good. I'll just note that Owen disclaims any such getting it into 
their heads.



regards


John Springer