Re: X.500, Naming and the Internet

[[e-mail]]

> >I think we're using the word "listing" with two different meanings.
> 
> No. I use "listing" as the joint decision of a data base server and a client
> to "list" information on the client, including its "registered name", in the
> server's data base
> 

We are talking about the same thing then. Sorry for the misunderstanding.

> You are saying that the names shall be determined by the registration
> authority, and I am agreeing to that: we shall use registered names.
> Then, you are adding that the "natural"
> registration instrument is the legal registry of companies held by cities,
> chambers of commerce, tax office -- whatever, it changes with national laws.
> At this point, I somewhat disagree: this is just one way to register.
> The important point is that registration should guarantee uniqueness;
> there are many ways to achieve this, including ad-hoc registration
> authority. For example, INRIA can validly claim the OID { iso member
> afnor(250) organisations(1) inria(2) }. That is as good a name as
> anything else.

With the caveat that you separate INRIA the registration authority from
INRIA the Directory operator, I agree with you. However I'd like to point
out that the French government might have something to say about
INRIA claiming the ability to perform registration in the civil
infrastructure :-). That's the very subtle point in listing vs.
registration: the listing model reflects registrations from the
official authority allowed to make such registrations. INRIA (in
this case) can only perform registrations in the civil infrastructure
(as opposed to the tree under INRIA) iff INRIA is the recognized
authority (or one of the recognized authorities) for civil registrations
in France (effectively meaning that the registrations INRIA makes
have a scope beyond the Directory, but can actually be used in the
"real world").

I'm not very sure what INRIA's OID assignment has to do with this.
I don't doubt that somebody at INRIA has the right to assign
things under the INRIA OID tree, but I don't see what this has
to do with assigning RDN values. You're not suggesting we use
OIDs in RDNs are you?


Wengyik