Re: [ietf-provreg] thursday's meeting

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

> >     Our response to the IESG.  The IESG wants "a standard (as 
> > in the base,
> >     core spec) means for a registrar to tell a registry what 
> > can be disclosed
> >     at fine granularity."
> > 
> >     A means to avoid host objects.
> 
> We have really got to get a grip on the first item.  I thought we were past
> this as of last week, and here we are again.  There's clearly a
> communications issue somewhere...

I agree.

Either a
	session-scope mechanism, which incidently is independent of 
	syntax and schema (it could be morphed into EBNF if RRP has
	not yet reached its sell-by date),
or a
	base schema mechanism, dependent upon syntax and schema, and
	independent of "session" (aka, "privacy" is a property of
	syntax representations, and transport independent, even if it
	is not transported),
or a
	mechanism which subsets the base schema, and constructs some
	syntax and semantics from some subset, and is also independent
	of "session" (see above, and add complexity for subseting in
	general).

That should read as a switch statement, with no fall-through.

Pointing out a specific defect in the <dcp> element's sub-schema would be
really useful.

Eric