W3C home > Mailing lists > Public > semantic-web@w3.org > July 2008

RE: "In Defense of Ambiguity"

From: Booth, David (HP Software - Boston) <dbooth@hp.com>
Date: Wed, 16 Jul 2008 18:17:19 +0000
To: Seth Russell <russell.seth@gmail.com>, "martin.hepp@uibk.ac.at" <martin.hepp@uibk.ac.at>
CC: Alan Ruttenberg <alanruttenberg@gmail.com>, Ian Emmons <iemmons@bbn.com>, "semantic-web@w3c.org" <semantic-web@w3c.org>
Message-ID: <184112FE564ADF4F8F9C3FA01AE50009FCFC4DC962@G1W0486.americas.hpqcorp.net>

> From: Seth Russell [mailto:russell.seth@gmail.com]
> [ . . . ]   Once the spelling of the URI is
> defined, people can add statements to the SW using that URI
> and as they do so its meaning gets fleshed out and even
> changes.  Just like in natural language.

That is one way semantic web architecture could work -- it's what I've called the "competing definitions approach" -- but it would not work very well, because the end result is that the same URI will have different meaning in different contexts, i.e., it causes URI collision:
http://www.w3.org/TR/webarch/#URI-collision

For this reason, it is important that, from an architectural standpoint, each URI has an associated set of "core assertions", and use of that URI in a statement implies acceptance of those core assertions, so those core assertions are common to all uses of that URI, thus giving it a consistent meaning (assuming the core assertions do not change).  This is what I've called the "URI declarations approach".  Other "ancillary assertions" may be made about the URI's denoted resource (beyond the core assertions), and RDF statement authors are under no architectural obligation to accept ancillary assertions.

This paper from the European Semantic Web Conference workshop on Identity and Reference compares these approaches:
http://dbooth.org/2008/irsw/
and these slides further illustrate the ideas (and add a bit more explanation):
http://dbooth.org/2008/irsw/slides.ppt (PowerPoint)
http://dbooth.org/2008/irsw/slides.pdf (equivalent PDF)




David Booth, Ph.D.
HP Software
+1 617 629 8881 office  |  dbooth@hp.com
http://www.hp.com/go/software

Statements made herein represent the views of the author and do not necessarily represent the official views of HP unless explicitly so stated.
Received on Wednesday, 16 July 2008 18:19:36 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 07:42:06 UTC