[arin-ppml] quantitative study of IPv4 address market

[Owen DeLong]

On Sep 1, 2012, at 20:44 , Joe Maimon <jmaimon at chl.com> wrote:

> 
> 
> Owen DeLong wrote:
>> The goal of ARIN address policy is to place addresses in service where they are needed so long as that is possible.
>> 
>> This is the overarching goal of both the allocation policy _AND_ the transfer policy.
>> 
>> Keeping addresses in inventory when they are needed in implementations is every bit as counterproductive to that goal as would be eliminating the justified need requirement from allocation or transfer policy.
>> 
>> If policy is prematurely driving people to the transfer market because of the huge discrepancy in terms we have created with recent policy changes, then, it is evidence that that discrepancy is harmful.
>> 
>> Owen
> 
> 
> Demand is flexible. Therefore, so is supply.
> 

To some extent... For a finite resource, such as globally unique IPv4 addresses, eventually you reach a point where demand exceeds supply at any achievable price point and the market effectively terminates.

> Looks like things are working the way they ought to.
> 

We can agree to disagree on this point.

> We had the chance to throw it all out there. I made sure of it.
> 

That's not what I am advocating, either.

> What do you think will happen when ARIN free pool dries up? Who will have the addresses then? Will we still need them? If we do, how will we get them? And will we as a whole be better off then now?

1. A variety of things are likely to happen. Some subset of them will.
2. Mostly the people that have them now, plus a few new entrants.
3. Need is a variable term, but certainly many will still want them.
4. If there is availability in the transfer market, then, that will likely be the main if not the only source. If there is no availability in the transfer market, then, we will either do without or some other less useful solution will be deployed (think squatting, global routing table dichotomy, network partitioning, and cats and dogs sleeping together). This period of abrupt and widespread instability will then be followed by a period of rapid IPv6 deployment and will be least disruptive to those that deployed IPv6 early. Eventually, the network will stabilize on IPv6 and things will begin improving again.
5. When we finally get to IPv6, yes. During the period between IPv4 runout and that time, we will have an unfortunate series of upheavals and disruptions and will likely be worse off than we are now. However, we will be worse off during that period no matter what. The question is how to minimize the duration, extent, and impact of that period of upheaval and disruption.

Extending the duration of the ARIN free pool by creating an early artificial shortage through policy is antithetical to that goal.

> I expect you and I have very different answers to those questions.

More than likely. But at least I have shared mine. Care to share yours?


Owen