Re: "private" Element Attribute

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

> I'm not sure how or why you concluded that we settled on (c) in the context
> of this thread.

I think we're discussing as the authors of -07 and -05, some new external
requirement statement, which may or may not be techncally sophistcated.
Our drafts, -07 in particular, provide a mechanism for (b), and for those
who deploy EPP implementations in ICANN private-law, or public-law, TLDs,
that means a mechanism that allows expression of policy consistent with
applicable law, (c).

> My understanding of the IESG comments is that they were focused on (a).

That's their problem. Randy wouldn't be the first menber of the IESG to
assert he knows more about the problem domain than I do, but I don't think
he will.

What (a) implies is that a registrant can create enforceable policies on
registries. Now we look to the space created by all values of policy and
all values of policied objects.

Actually, I can live with the IESG making the "Internet Standard" (or ICANN)
protocol hideously baroque. The ccTLD and delegated registries can ignore
the damage and route around it.

Having consulted to the _other_ IAB on choice and its consequences, and to
MicroSquash on the same subject, whether in IE or in MS's CDN, I can say
it is as endearing as RSVP and keeping arbitrary amounts of arbitrary state.

Element-wise semantics, but not necessarily distinct semantics for each
element.

I'm thinking.

Goodnight.
Eric