[ppml] Policy Proposal: IPv4 Transfer Policy Proposal

[Scott Leibrand]

Randy Bush wrote:
> all in all, this proposal and much of the discussion feels to me as if 
> it has an underlying subtext of "this is bad and the people doing it are 
> bad people, but we know we can't stop them so we're going to impose all 
> the obstacles and conditions we think we can get away with."
>   

I know others feel that way, that a transfer market is a bad thing, but 
we have to do it anyway.  I personally don't share that view: I think 
that a transfer market is the best way to minimize the cost to the 
community of the IPv4 to IPv6 transition, and therefore it's a good 
thing.  I believe that the conditions in the proposal can be justified 
based on sound principles, such as the ones you mention below and a few 
others.  If there are conditions or other aspects that aren't justified, 
I'll be more than happy to discuss removing them.  (In fact, I've been 
having some other discussions and will probably propose removing one of 
the transferor conditions shortly.)

> aside from global engineering issues of routing impact minimization and 
> operational issues of ensuring fair and valid transfers, why do we need 
> to place silly restrictions such as "you can only get one bathroom pass 
> in a given class period," "you can't share your lunch with a foreign 
> friend," etc?
>   

The "one bathroom pass per class period" restriction (one transfer every 
6 months) is directly intended to minimize routing impact, by ensuring 
that a transferee gets all the space they need as a single block, not by 
vacuuming up lots of smaller blocks, thereby forcing other blocks to be 
split to meet the needs of smaller orgs.  The "can't share your lunch" 
condition is meant to keep the impact of the policy within the scope of 
our policy-making authority, so we don't have to do a globally 
coordinated policy.  As I've expressed earlier, I think we should create 
a simple way to allow transfers between regions once two or more RIRs 
have passed transfer policies.

> address space transfer is going to be an ever-increasing part of normal 
> life; get over it.  our job is to make internet operations simple and easy.
>   

In your opinion, which conditions unnecessarily complicate 
life-after-free-pool without commensurate benefit?

Thanks,
Scott