From owner-mcg-talk@localhost Mon May 12 15:48:53 1997
Return-Path: owner-mcg-talk@mcwg
Received: (from majordom@localhost) by localhost (8.8.5/v3.2) id PAA19789 for mcg-talk-outgoing; Mon, 12 May 1997 15:48:53 -0300
Message-Id: <199705121848.PAA19789@localhost>
Date: Mon, 12 May 1997 15:48:38 -0300 (EST)
From: "416720" <mcg-talk@mcwg>
To: MCG <mcg-talk@mcwg>
Subject: Towards MC Standards
Sender: owner-mcg-talk@mcwg
List: 
Status: RO
X-Status: 

Since our very first postings, it has been important not only to try to
exactly define the terminolgy we are using (e.g., our discussions on
Domain-Space, Image-Space, Personas, Anonymous, etc.) as well as to keep
them in correct use.

How do Meta-Certificates certify? That is answered in the slide document:
MCs use an archetypical trust model.

This means that in the same that your son is certified as belonging to the
human race and a fruit-fly is certified as a strain of "drosophila
melanogaster", a private-MC is certified as an offspring of a particular
MCC, which belongs to the Meta-Certificate family (MCAC).

(A certified private-MC is just a mature son that is able to speak and
recognize his parent MCC by itself, so to say. Before he matures, when he
is just a private-MC, his "older brothers" -- the public-MCs -- are used
to show him who his parent MCC is, so as not to confuse his true parent
with other MCCs.) 

Simply put: "A MC is a signed object, that follows inheritance rules."

MCs are neither identity nor attribute certificates. 

MCs are object certificates. MCs certify identities, attributes *and*
methods. 

Thus, it would be incorrect to try to apply other nomenclatures to define
what is already differently defined.

The MC trust model is not peer-to-peer (such as PGP), hierarchical (such
as some X.509), linked (such as other X.509), direct, indirect, etc. It is
archetypical, in the sense defined above. 

The MC archetypical trust model can, in turn, link to other peer-to-peer,
hierarchical, etc. systems, in close interoperation.

When that is done, MCs become what one could call a "combined" trust
model. 

So, I would ask the list to keep the already defined terminology, as we
work towards a MC Standard. 

You can refer to MCs either as archetypical or combined, as a function of
use (stand-alone or in interoperation).

Yours,

Ed Gerck
______________________________________________________________________
Dr.rer.nat. E. Gerck                        ed@gerck.com
http://mcwg
P.O.Box 1201, CEP13001-970, Campinas-SP, Brazil  - Fax: +55-19-2429533