Re: does RDF require understanding all 82 URI schemes?

> How is this different from, or better than, every one having an SSN
(Social
> Security Number for non-US subscribers)? And given that most humans have
never
> made a phone call, isn't this getting ahead of ourselves somewhat?

This comment was tongue-in-cheek, more than anything else :).  Perhaps we
should body-bar-code at birth our own URIs?  An SSN isn't so good -- it's
too hard to administer on a world-wide basis.

>
> :The metamodel itself must be subject to and derived from the
> : web of trust.  As an entity I declare those other entities I wish to
trust.
> : An entity can make statements about other objects.  An entity can also
have
> : a locally defined metamodel, which can be built on top of the metamodels
> : published by other, trusted entities.
>
> More's the point, why do we need to architect trust in the network at all?
Peers
> can model, engage and disengage in trusting relationships among
themselves. I
> believe architecting trust is certainly one hubris the network can do
without.
>
> Trust isn't a requirement.

I believe it is.  Unless the protocols by which trusted relationships can be
created and destroyed are well-known, they can't be engaged in on a
widespread basis.  This is the equivalent of web browsers that don't support
http, where each browser has its own protocol.

Anything can be modeled by a set of peers; the question is whether the
protocol is of sufficiently general use that it belongs in a well-defined
standard.  Trust is the wheel of the internet; it should not be re-invented.
Let's get it right (or at least good).

The key point here is the notion of a trusted metamodel.  Trust is a fuzzy
thing, and so is the universal metamodel.  Current RDF thinking is binary
with regards to a statement.  Statements also extend into time and trust
dimensions.

My metamodel should be different from yours.  I might start with the CERN
metamodel, blend in the W3C metamodel, and add the metamodel of the
Washington Post, but only where it did not conflict with W3C.  You could
choose otherwise.

>
> : We should worry less about the representation of knowledge, and more
about a
> : stable and robust set of rules for resolving conflict and compositing
> : multiple knowledge and meta-knowledge sources.
>
> That would be a body of law.

This is law that needs to be written (if someone hasn't done it already).  I
think there is a clever information-trust recombining algorithm waiting out
there to be discovered.  It is not obvious to me if there is a single best
way to do it.

>
> : I wonder what patterns of information would survive and prosper in this
> : memetic greenhouse.
>
> Untruths and mistruths.
>
> Bill de hOra

Sadly, this is likely to be true.  In a correctly functioning information
environment with peer-based trust and review, the signal-to-noise ratio is
proportional to the number of good-willed to evil-intended citizens of that
infosphere.

RJ

Received on Monday, 12 February 2001 16:40:51 UTC