[ppml] FW: No transfer policies are needed

[Paul Vixie]

rbf+arin-ppml at panix.com (Brett Frankenberger) writes:

> My view:
> ...
> There's an IPv4 black market now.

can you describe it?  how many transactions, what dollar volume, what
average price, is it all sales or also leasing, how do they manage whois,
is it done by M&A of shell companies, when did it first begin operations,
is volume growing or shrinking, how many participants?  things like that.

> ARIN keeps the black market very small by undercutting the price of the
> would-be black market sellers.  The only reason there's a black market
> at all is for the population that wants addresses and is willing to pay
> a lot for them but can't justify them.

can you describe the reasons someone who wants addresses mightn't qualify?

> Again:  The transfer policies do nothing to prevent a black market. 
> ARIN keeps the black market small by undercutting the prices.

can you say whether subdivision is occurring following transactions?  that
is, does the black market contribute to global routing table growth?

> When ARIN has no more addresses, they won't be able to undercut the
> market price for IPv4 addresses.  And there will be a free and open
> market; addresses will be sold to anyone who can pay market price. 

do you believe that current participation in ARIN by many of its members is
one of benign tolerance, and that when the pool is depleted, most of those
folks will welcome "freedom and openness" even if looked like the wild west
in terms of competing claims, bullying, safety coming only from strength,
or in our case, in immediate explosive deaggregation and a lack of ability
to figure out what route is supposed to be coming from whom?  (worse than it
is now, that is?)

> (Who's going to want to go through the trouble of justifying address
> space if it costs the same without a justification?  And what seller is
> going to reduce the price he gets by selling only to buyers approved by
> ARIN?)
> 
> Policy won't be enough to stop that; it will be ignored.

you paint a very grim picture.  what's your proposed better alternative?

> Refusing to record such transfers in ARIN's database won't be enough to
> stop that; new databases will be developed.  (And providers will honor
> the information at first becasue it gives them a compettitive advantage
> over providers that won't, and then later because not doing so will put
> them at a compatitive disadvantage to the providers that are honoring
> transfers.)

this is the most puzzling to me of all the things you said.  ARIN's policies
are controlled by the community.  are you saying that the community who could
choose whether ARIN's database did or didn't record such transfers, is not
the true community who would simply set up a database "more open to transfer"?

if so, can you explain why this true community doesn't help govern ARIN?  i'm
sensing in you a suspicion that the things folks are willing to say or do in
public today are meaningless niceties, and the real game won't play out until
later.

if that's so, then i want to go on record as not sharing that suspicion -- i
think that ARIN will be told by its community of interest to operate the
prefix/route/owners/whois/rpki/mumble database that will be needed, if one is
needed, and if ARIN isn't told to do it, then that's proof that one isn't
needed and/or wouldn't be used.

> I'm not saying an open IPv4 market would be a good thing or a bad
> thing.  I'm saying it's an inevitable thing.
> 
>      -- Brett

any advice?
-- 
Paul Vixie