Re: [ietf-provreg] again with the privacy

[Eric Brunner-Williams in Portland Maine <brunner@nic-naa.net>]

Ed,

A mechanism to disclose the operational practices and policies of operators
of EPP servers with respect to "data collection", aka "privacy", is NOT a
requirement, and section 8.4 of rfc 3375, in particular the final sentance
of para 1, is incorrect.

I can live with being responsible for that train wreck.

Protocols developed under the imprimature of "the IETF", MUST NOT provide
mechanism(s) to disclose the data collection policies, in particular, the
polices to which data which identifies individual persons is subject, of
operators of the protocols.

Sounds grand and very Directorial.

Instead "we're supposed to" come up with some dork that will allow the
naive data we currently have, in the schema we currently have, with the
syntax we currently have, to "twinkle", and to provision, not just any
data collector, but only the registry operator, with this "twinkle"
property, for each provisioned glob of data, and this whole waste of
time is because 954 can't be moved to historic.

I'll be sure and tell my registrar that I want only the octets-pairs that
are prime when interpreted as Mayan digits (base 20) to be "published".

I'm obviously in the wrong IETF. I'm in the one where clue is not held
uniquely by management. The "IETF of Idiots".

I do not see the point in bothering to pretend to write a specification
that is in fact written by someone else. I'll just treat the eventual RFC
as having no more "consensus engineering standards" value than an equal page
count text that comes directly from Verisign. Implement it, and don't think
too hard about it.

I give up. I could use this time better playing with my kids, and making
open, interoperable (but not IETF referencing) product.

Eric