[arin-ppml] Draft Policy 2009-4: IPv4 Recovery Fund

Leo Bicknell bicknell at ufp.org
Tue Mar 24 16:49:13 EDT 2009


In a message written on Tue, Mar 24, 2009 at 12:54:59PM -0700, Matthew Kaufman wrote:
> Why don't we combine 2009-4 and 2009-2? ARIN could just use its 
> remaining stock of address space and sell it to the highest bidder via 
> the policy in 2009-4.

I'm not sure I'm going to come at this from the same direction you
were thinking, but you raise an interesting point.

Dan (original author of 2009-2) and I share one concern.  We don't
want someone large to make a request right at the end and have it
filled by ARIN with all of the drips and dregs left around.  It's
not hard to imagine someone asking for a /12 at the right (wrong?)
time and having ARIN try to fill it with 2 /16's, 14 /17's, 9 /20's,
241 /21's, ....

The obvious bad effect, deaggregation, isn't the real problem.
That's going to happen if many smaller companies get the space.
The real problem is that it causes everyone to hit the wall at the
same instant.  With that one large request ARIN is out, period game
over.  While we can't make the landing really soft, it seems like
having some small amount of spread biased towards the smallest
players with the least buffer makes sense.  "Please save the
children", if you will.

Dan attempted to take this problem on directly in 2009-2 by limiting
requests once we fell below some minimum.  I attempted to take it
on indirectly by implementing the recovery of space as specific
size prefixes run out.  If you ask for a /12 and ARIN has no more
you must start to bid, even if ARIN still has a lot of smaller
blocks out there (which would be given out normally until they are
gone).

In any event, they aren't really compatable (doing both as-is doesn't
make a ton of sense), but they aren't really incompatable (as they share
many of the same motivations, I think).  If the community likes both
concepts I think we would have to hammer out a compromise that made
sense but I stand ready to do that and I think Dan does too if it is
what people want.

> Clearly as the numbers run out, the prices will get higher and higher 
> and since "the highest bidder" is going to be who gets IPv4 space 
> *after* the runout, we could all get experience now with who does and 
> doesn't end up with space with that as the allocation policy.
> 
> I'm sure ARIN could find something creative to do with the excess funds 
> that might be generated.

My reading of the room is there is a relatively strong sentiment against
ARIN generating "windfall profits" off the exhaustion, and thus I'm
tempted to say your idea is a non-starter.  That said, there have been
several folks express concern over the amount of cash ARIN may have to
tie up in the 2009-4 program.  I think it's a fairly small amount of
money and risk, but perhaps there would be support to start 2009-4 with
a /small/ amount of reserve left such that a buffer can be created.
It's not something I had considered before and I wouldn't add it unless
there was strong support, but it is an idea.

-- 
       Leo Bicknell - bicknell at ufp.org - CCIE 3440
        PGP keys at http://www.ufp.org/~bicknell/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
URL: <https://lists.arin.net/pipermail/arin-ppml/attachments/20090324/fe5eda04/attachment-0001.sig>


More information about the ARIN-PPML mailing list