[arin-ppml] A Redefinition of IPv4 Need post ARIN run-out (was: Re:Against 2013-4)
mike at nationwideinc.com
Tue Jun 11 21:22:36 EDT 2013
> There were issues with your proposal as writen, but I think most of them
> could have been worked out. Some of the ideas in your proposal have been
> incoporated in to policy already.
Actually my draft proposal became policy with essentially the single change
which was the removal of the /12 exclusion for needs tests.
Hence my original email to you, which was to pose the question of whether
you might be willing to reconsider this exclusion in the light of your
discussion with Mr. Ryerse, who has trouble receiving an allocation from
> I think the more important issue is an appropriate criteria on the
> lower-end and for new enterants, the current slow-start for IPv4 isn't
> going to work, post-ARIN free pool. Yes, I know eliminating need
> alltogether eliminates that problem, but I'm not sure I can get myself all
> the way there. I'd like to see some minimal technical criteria that
> entitles someone to be able to buy up to between a /16 and a /12 and more
> than just that they have the money to do so. Maybe its just as simple as
> demonstrating efficient use of at least a /24. If you can't do that then
> you can only buy a /24, then you utilize it and you qualify for bigger
Why not consider a proposal which you would acknowledge solves Mr. Ryerse's
If you can't get all the way towards dropping needs tests (for transfers
only), why not consider a half-measure?
Allow needs-free transfers, but cap them so as to protect against hoarding
> The point is that's my opinion, what we need is to develop a community
> consensus. This require comprimise on both sides. I think a much
> liberlized defintion of need for IPv4 is possible. But, when people
> continually call for the elimination of need completely, the majory of the
> community circles the wagons and we get nowhere. My question to you and
> others, is a signifigantly liberalized defintion of need for IPv4 good
There are obviously two PPML camps, the no-needs camp and the needs camp.
So in an attempt to bridge the gap in quest of the consensus you describe, I
am directing my original question to the list at-large.
What about a needs-free transfer cap?
This would be a compromise which provides some of the free-market
functionality the no-needs camp loves,
And it would limit the threat of hoarding and market manipulation which many
in the needs camp feel is a requirement of reasonable market regulation.
Looking at the transfers on record, /12 is a very large size. Most transfers
are much smaller, so most address-rights transactions will flow more easily
from the perspective of buyers, sellers, and brokers, if I may presume to
speak for them. For "corner cases" like Mr. Ryerse, this would solve the
problem. For all but the largest of those whose planning horizons go beyond
24 months, this would solve the problem. For those who wanted to keep a
reliable inventory of available addresses at nearly every block size, this
would solve the problem.
I am aware of a /8 which has been on sale for years, yet remains unsold. It
is hard to argue that possession of a /12, or even several of them, would be
enough to corner the market to the extent that price could be manipulated.
Lest we forget, unlike other commodities, IPv4 address values will certainly
go to zero if they are held forever, and nobody really knows how fast the
IPv6 dominoes will fall when critical mass is reached. Nobody knows how many
sellers there are, who they are, or how many addresses could be available.
This uncertainty provides a natural limit to speculation, but we can impose
a hard limit through the negotiation of an appropriately sized cap.
> My suggestion is we keep talking about what such a signifigantly
> liberalized defintion of need for IPv4 looks like. I think you and I have
> similar ideas probablly not exactly the same, but you and I are not
> enough, we need to develope more consensus around the idea before were
> ready for a proposal.
Agreed that we should keep talking and trying for consensus.
I understand that you are suggesting some kind of liberalization or change
to policies related to IPv4 need to address issues like access for new
entrants and smaller enterprises.
I propose that removing the needs tests for transfers under the cap will
lead to a vibrant market at the low end, where Bill Herrin correctly pointed
out that finding and buying a /24 is very hard. Because of the needs
requirement for all transfers, buyers are looking for relatively small
blocks for which they think they will get approval. We get many requests,
mostly from Asia, for blocks of /20 and smaller. But there are few sellers
willing to shave off small blocks, due to the transactional costs and
uncertainties. If I have a /16, why not hold out and wait for a single
buyer, negotiate a single contract, and make one trip through the
ARIN/APNIC/RIPE justification tunnel? After all, the sellers have never done
this before, it will usually be a once in a corporate-lifetime kind of
transaction. Shepherding a single transaction through to completion in this
kind of environment is difficult enough, but a series of small ones, with
overseas buyers, all dependent on a third party approval via a needs test?
That's a tough sell.
On the other hand there are entities who are willing to enter this market,
bear the risks of multiple smaller transactions, strive to standardize and
normalize the market, and if successful, profit from their endeavors. Of
course if they are successful, new entrants and small companies would have a
reliable market from which to acquire addresses. And sellers of addresses
would be able to offload their /16 to a single buyer in a single
transaction, and get back to their core business. What is preventing this
business venture in IPv4 address distribution is the needs-test for
I have tried to point out some of what I think could be the benefits of a
cap for needs-free transfers, and of course I am from the no-needs camp.
Maybe someone from the needs camp will keep the conversation going, or
propose something else to try to address the points held by both sides?
More information about the ARIN-PPML