Section 8 Other GAC recommendations [was: Re: [gnso-whois-study] documents for tomorrow's call]

[Eric Brunner-Williams <ebw@xxxxxxxxxxxxxxxxxxxx>]

Liz, other study group members,

I didn't previously comment on the item 8 "Other GAC recommendations".

GAC bullet #1 is only slightly interesting. Whois (ports 43 and 80) 
server operators could provide data, but there will be significant 
artifacts within the data which will limit its utility, as the data will 
reflect the momentary present aggregate of anti-harvesting mechanisms, 
from white lists and rate limiting to anti-robot tests and so on. 
Additional problems arise in the notion that "user" or "users" of 
anonymous servers can be usefully identified or the use to which data 
appears to be used can be corolated to the user(s) who harvested the data.
Upon reflection, this appears to reduce to the non-serious and already 
answered (several times) question -- "Does whois contribute to the spam 
et seq set of problems?"
GAC bullets #5 and #6 are variations on the more specific, but useless 
questions about privacy services. The assumption that a duty to disclose 
data that is known to harm the registrant is conditional upon whether 
the registrant is a person or a non-person flounders on the reality that 
contact data goes to people, and can ultimately be used to force manual 
processing of spam or forged bounce messages or phone calls or postal 
mail or the ever eager MLS marketing associate knocking on the door by 
persons.
GAC bullet #3 assumes a fact not in evidence, that the creation of a 
persistent identifier in a name space creates an unqualified and 
universal 3rd-party right to know the identity of the creator of the 
persistent identifier. Again, some 63 or fewer decimal value 
approximation of e or pi or any other fundamental constant really isn't 
a police matter, when used on paper or when used in the dns. Pity we're 
not discussing policy about trademarks or ongoing criminal enterprises 
associated with well-known felons.
Again, none of these recommendations met my test for useful policy or 
useful data, and all get my "no".
Eric