[gnso-vi-feb10] On harms, the note of 8/16

[Eric Brunner-Williams <ebw@xxxxxxxxxxxxxxxxxxxx>]

There are two notes to the list, one submitted on 8/16, and one 
submitted on 8/9. The note submitted on 8/9 is organized as four 
sections addressing kinds of "Harms", and harms statements from 
Register.Com, Afilias and PIR.
To keep this note to one page I only discuss the 8/16 note.

The 8/16 note is organized as two "Harms to" sections, the first harms 
to consumers, the second harms to businesses, and a section on cost 
and a section on liability.
1. The first section has an "a" part and a "b" part.

The first claim in the "a" part argues "[vi] could [lower prices] 
therefore [structural separation] will [increase prices]". If the 
"will" were replaced by "may" the reasoning would not be necessarily 
incorrect.
The second and third claims of the "a" part are restatements of each 
other, and recite the conclusion of the first claim.
The "b" part makes two unrelated claims.

The first amounts to the observation that at least one registrar will 
not offer all TLDs, a fact pattern which exists presently, and claims 
therefore that the lack of one or more vertically integrated 
registry-registrar combines "may prevent sales of domain names to 
people who want to buy them".
As the fact pattern exists presently, the claimed harm must be present 
also. However, the harm is insufficient to cause policy to reduce or 
remove the present harm, so the claim is insufficient.
The second amounts to the claim that third party rights protection 
under the existing registry registrar system cannot scale, and 
therefore the registry registrar system must end. This is inconsistent 
with the IPC's work since Mexico, the IRT generally. For the claim to 
be sufficient, the IPC must abandon its proposed set of mechanisms. 
Failing that, the claim is insufficient.
2. The second section makes four claims. The first is that business 
conduct will be restricted. The second restates the first. Again, the 
fact pattern exists at present, Verisign and Network Solutions are not 
permitted to merge, so the claimed harm must be present also. And 
again, the harm is insufficient to cause policy to reduce or remove 
the present harm, so the claim is insufficient.
The third claim again amounts to the observation that at least one 
registrar will not offer all TLDs, a fact pattern which exists 
presently, and claims therefore that the registries not structured as 
vertically integrated registry-registrar combines "may find either no 
distribution channel, or a substandard one."
As the fact pattern exists presently, the claimed harm must be present 
also. However, the harm is insufficient to cause policy to reduce or 
remove the present harm, so the claim is insufficient.
The fourth claim invents a registry which does not register domain 
names, and finds an absurdity.
Registrars with profoundly restricted pattern of registrations exist 
at present, e.g., IANA #16 (AOL), IANA #345 (VGRS), IANA #365 
(Educause), IANA #1330 (Microsoft), ...
The "absurdity" claim must relieve these actual registrars of the 
burdens of ICANN Accreditation if it must relieve hypothetical 
registries which do not register domain names of the burdens of the 
registrar function being provided by structurally separate entities 
enjoying equal access.
3. The third section's claim is simple. Policy enforcement has a cost, 
the cost will eventually "trickle down" to consumers.
As ICANN policy is not "lowest price", this claim is without merit.

4. The forth section's claim is also simple. Restriction on 
integration may cause liability risk.
As the US DoJ has not intervened under the present 15% cap rule, this 
claim has no more merit than the claim that the present restriction 
may cause liability risk.
In sum, there is little in the 8/16 note which is useful.


No offense is intended to the authors of the 8/9, or 8/16 contributions.

Eric