[ppml] Policy Proposal: Resource Reclamation Incentives

[Stephen Sprunk]

Thus spake "Keith W. Hare" <Keith at jcc.com>
> I've seen a lot of discussion over the last couple of days about
> legacy address space reclamation and whether or not legacy
> address holders are paying their fair share to support ARIN.
>
> The thing I haven't seen are any numbers.
> ...
> This discussion of reclamation incentives has also generated
> a lot of discussion about legacy address holders who have
> not signed an RSA and do not pay ARIN anything.  However,
> I have not seen anything about how many legacy address
> holders exist.

Per the presentation at ARIN XIX, there are 31,386 legacy direct 
registrations to 20,501 organizations, and 2,277 of those orgs have signed 
an RSA.

Currently no fees are collected for those 31,386 legacy blocks; however, the 
orgs that have signed an RSA are likely paying for other, non-legacy 
resources.

Also, only 44% of the blocks appear in the routing tables, and only 54% have 
been updated since Dec 97.  That means a sizeable fraction of the blocks are 
likely abandoned.

Since ARIN (per a response to an off-list query) doesn't know whether legacy 
blocks are "assignments" or "allocations", it's not possible to determine 
how much revenue would be generated if all of them were subject to fees.

Worst case, ARIN would collect around $1M/yr if the active blocks were all 
determined to be "assignments" and the registrants were paying the $100/yr 
maintenance fee -- an increase of about 10% to ARIN's revenue.  OTOH, if 
most of the blocks were "allocations", the increase could be 20+ times that. 
Nobody knows.

> How many legacy address blocks exist that are large enough
> to be useful to ARIN?

All of them are potentially useful.  However, I haven't seen any stats that 
break out how many of the registrations are of the various sizes.  Common 
sense says that most will be /24s, but there's gobs of /16s out there as 
well, and a few /8s.

S

Stephen Sprunk      "Those people who think they know everything
CCIE #3723         are a great annoyance to those of us who do."
K5SSS                                             --Isaac Asimov