[At-Large] Depository (was Re: Privacy and domain abuse vs the IP constituency)

[Eric Brunner-Williams]

Karl,

I accept that in your view Depository is attempting to innovate and 
experiment.

I don't think it is.

Absent the suggestion by McTim that this Advisory Group should form a 
view on the ASO process, or the appeal to an untested authority 
inherent in ICANN through "hierarchy" offered by Mike Burns, the 
substance and process are appropriately addressed on the ARIN-PPML 
list, in which you are free to participate.

Turning to the history of RIRs, two mechanisms intended to scale as 
the price of network adapters and of network attachment dropped have 
been adopted.

I proposed a regional scheme, based upon the ITU's X.121 
Recommendation, specifically the zone, represented as the first octet 
of a Data Network Identification Code (DNIC) to create a small number 
of regional registries. The ISO 3166 scheme was adopted instead, 
creating two orders of magnitude more registries, for name to address 
mappings.

The model of a regional scheme was adopted for address allocation 
registries, as much in recognition that no specific competency existed 
in North America which did not also exist in Western Europe and Japan, 
as for territorial diversity, and the attendant jurisdictional 
diversity. If consolidation were seriously proposed, it would be as 
arbitrary now as it was earlier to assert which of {ARIN, RIPE, and 
APNIC} is the first RIR to be eliminated.

I don't think the characterization "For no technical reason, but for 
every political reason ... created two more RIRs" is correct.

There is technically defensible reason for delegation in technical 
institutional development, a course pursued with diligence by the 
Network Startup Resource Center, and others.

There is technically defensible reason for non-revocable delegation of 
address resources, as Geoff Huston's projections show, the AFNIC and 
LACNIC regions will be capable of general allocating v4 resources for 
several years after the APNIC, ARIN and RIPE regions transition to 
exhaustion allocation policies.

The aggregation is an enemy of regional registries claim would be more 
credible if, as a general rule, access networks were predominantly 
trans-regional. As they are not, for pricing and tariff and other 
jurisdictionally scoped policy reasons, this is a weak claim of 
non-necessity or inefficiency.

Finally, assuming all of the initial claims of Depository for the 
purposes of argument, there is the specific choice of replicating the 
domain market structure, in the address market.

There are two final claims to consider: first, that the domain market 
structure is good, and the second, that innovation and experiment are 
actually less useful than replication of an existing structure. I 
don't think either of those are reasonable claims, or capable of 
proof, though I understand they are offered as proved by assertion by 
Depository, in its own pecuniary interest.

Eric