Encouraging return of legacy space WAS Re: [ppml] ARIN Policy Proposal 2002-9

[Jim Fleming]

----- Original Message ----- 
From: "J. Scott Marcus" <scott at scottmarcus.com>
> Nearly a quarter of all IPv4 space, and nearly a 
> half of all allocated IPv4 space, is tied up in 
> blocks 003/8 to 057/8.  These seem to me to 
> represent low hanging fruit - if memory serves, 
> =====
> 
> Does this make sense?
> 
> Do people see either positive or negative 
> incentives that ARIN could use to encourage the 
> return of large, low utilization IPv4 address 
> blocks?
> 

In theory, ARIN does not control (own ?) the address blocks most likely underutilized.
ARIN only has certain /8s, which supposedly come from the Wizard of IANA (aka ICANN).
http://www.iana.org/assignments/ipv4-address-space

In practice, the administration of IN-ADDR.ARPA could also be viewed as control or
ownership of all of those address spaces. Apparently, ICANN does not play a very active
role in managing that. The large /8 holders also apparently pay nothing to ICANN for that.
ARIN pays a token amount, nothing compared to the $168 million per /8 per year which
would match the market value.

If ICANN starts sending bills to /8 holders for $168 million per year, one would likely see
plenty of address space returned (or abandoned and reclaimed).

For some, reclamation is not a high priority because selling new space brings in money to
ARIN. Reclaiming space from bankrupt companies with no money is not an interesting problem
when an organization operates to generate revenue to simply pay the people who collect it.
If the entire system was automated with some simple web interfaces and data bases, the
costs of the administration would drop dramatically. Non-profits have no incentive to keep
costs low, just profits low. They do that by paying out everything they make. The money has
to go somewhere. Just watch the new .ORG Registry try to deal with that next year. People
will be lining up to help absorb the cash flow. Projects will have to be invented to do that.
The customer base is already there and it does not cost much to run the core Registries.


Jim Fleming
2002:[IPv4]:000X:03DB:...IPv8 is closer than you think...IPv16 is even closer...
http://www.ietf.com
http://www.iana.org/assignments/ipv4-address-space
http://www.ntia.doc.gov/ntiahome/domainname/130dftmail/unir.txt
http://ipv8.dyndns.tv
http://ipv8.dyns.cx
http://ipv8.no-ip.com
http://ipv8.no-ip.biz
http://ipv8.no-ip.info
http://ipv8.myip.us
http://ipv8.dyn.ee
http://ipv8.community.net.au