Re: EPP statuses and other questions

[Michael Graff <Michael_Graff@isc.org>]

Stephane Bortzmeyer <bortzmeyer@nic.fr> writes:

> On Tue, Nov 26, 2002 at 10:32:44PM +0000,
>  Michael Graff <Michael_Graff@isc.org> wrote 
>  a message of 55 lines which said:
> 
> > (1)  A handle (like FOO1-ISC) is not self-describing.  Is that a contact
> >      handle, a domain handle, or what?
> 
> I see it a a "registry policy" issue. Some will have a global
> namespace for handles (with the risks you explain) and some will have
> separate namespaces for contacts and hosts.

According to the current draft, this isn't possible.  For hosts and for
domains, sure.  However, for clients, the registrant (not even the registrar
according to how people say they'll be used!) chooses the local name,
and therefore the GLOBAL name as well, since the global ROID is defined
as the local_part-REGISTRY

> > (2)  Part of the handle namespace is client-chosen, part is registry-chosen.
> >      On contacts, the local (and thus the global) identifiers are chosen
> >      by the registrant, 
> 
> Which is bad, IMHO. It may require several round-trips before the
> registrar (<pc>the client</pc>) finds a free handle. Why is there no
> provision for registry-generated contact handles? (Or should we assume
> the ROID will have this role?)

See above, the ROID is derrived directly from the local part.

So, we have:

        entity local name       ROID            ROID Chosen by
        ---------------------   ------------    --------------
        foo.com                 FOO-ISC         Registry
        ns1.foo.com             NS1_FOO-ISC     Registry
        mg2                     MG2-ISC         Registrant

I don't see why using what I see as a far more sane method of using URNs
here wouldn't work.  After all, that's more or less the expected thing in
an XML encoding, and it solves the problem without any need for ROIDs.
Baring that, I think the ROID issue is serious enough, and badly spec'd
enough, that it needs to be fixed before the draft makes it to RFC status.

--Michael