[soac-newgtldapsup-wg] Report on the Temporary Drafting Group (Legal) call of 1/27

[Eric Brunner-Williams <ebw@xxxxxxxxxxxxxxxxxxxx>]

Colleagues,

Sorry to have missed today's 1300 (8am EST) call, I'd not updated my 
schedule for Friday SOAC calls, which occurred at 1500 (10am EST).
I don't know if, or when, a transcript of yesterday's Temporary 
Drafting Group (Legal) will be available.
The agenda was re-ordered and the Continuity Instrument issue was 
first on the revised agenda.
Kurt Pritz spoke for several minutes on the thinking that went into 
the formation of the COI draft, more than a year ago.
The first intervention was by Jeff Neuman of NeuStar. In his example 
NeuStar has several own-or-client applications.
Is one (1) COI sufficient?

There was discussion, and ICANN legal staff was somewhat supportive of 
reducing the COI requirement for NeuStar's scenario.
I commented that applications are evaluated independently, the source 
of several "bundling" requests, e.g., for IDNs, for fee reduction 
proportionate to non-duplication of evaluation cost, etc. and that a 
"shared" COI, while attractive, violated the application evaluation 
process model which has each application evaluated independently.
Related was the issue of actual cost of continuity. The ICANN staff 
member who spoke to that issue, Francisco Arias, argued that there is 
some incremental cost for each continuity. I offered the opinion that 
platform operators would set the incremental internal cost to zero and 
for competitive reasons would "include the COI at little or no cost" 
when soliciting registry backend technical support business from 
would-be applicants.
ICANN legal staff indicated that they would reflect on the question of 
multiple COIs for a single multi-applicant.
Aside: Bundled COIs through registry backend technical support vendors 
could help needs-qualified applicants, but it is likely that a 
precondition would be sole-source registry backend technical support 
contracts. See the Names Council's charter and its limitation on 
non-fungible forms of assistance.
The second intervention was by myself. In prior mail you have seen the 
issues and suggestions I've circulated. Karen Lentz of ICANN legal 
provided responses via email to two sets of issues an hour prior to 
the meeting. I made the point that the registrant protection did not 
go far enough, as it did not preserve the registration policy, 
allowing .travel like repurposing (from travel agents to unrestricted).
ICANN staff responded that the respective communities and public 
administrations for community-based and geographic registries "in 
continuity" will have an opportunity to comment on the designation of 
the continuity operator.
Aside: I don't think this is sufficient, but it is better than nothing.

Continuing my intervention, I mentioned designation of regional ccTLD 
operators, or other operators, as an alternative to a financial 
instrument.
Here the Staff response was surprising. First, the abandonment of the 
"Certified Registry Backend Provider" program was offered as a 
rational for limiting the means of continuity to a funded instrument 
only, though ICANN never offered a public explaination for the 
abandonment of that program (I was involved), and the contract between 
the applicant and the designated continuity provider might be fictive.
Aside: Elsewhere in the agenda the contractual obligations upon third 
parties, in particular subcontractors was discussed at length, and the 
ICANN legal staff position was that these are not fictive and may be 
enforced. This is worth a follow-up on both points.
Continuing my intervention, I mentioned applicant formed and funded 
insurance pools.
Here again the Staff response was surprising. The staff hypothesized 
that all of the applicants in a risk pool might fail simultaneously, 
leaving all of the failed registries lacking any form of continuity.
Aside: Not the subject of any agenda item, but present within the 
overall contractual envelope, are insurance requirements for the 
applicants. This too is worth a follow-up.
The remainder of the issues were without direct substance to the cost 
to needs qualified applicants, and so I omit them in this report. Of 
only potential interest was the final item, the methodology for SLA 
measurement, which up until now has been at the point of service (at 
the registry operator public network interface), and is now 
conceptualized (somewhat vaguely) as from an arbitrary point on the 
net. I will follow this for pure geekishness reasons, and if there is 
a means to prevent qualified applicant cost from increase, or cause 
its decrease, I'll update.
Overall, the Staff awareness of the rational basis for lower cost than 
originally conjectured has increased, and their willingness to 
consider alternatives that would facilitate applications from less 
developed economies has increased.
Cheers,
Eric