W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > April to June 2005

Re: Name of a graph? and FROM and FROM NAMED

From: Kendall Clark <kendall@monkeyfist.com>
Date: Fri, 3 Jun 2005 14:16:53 -0400
To: jos.deroo@agfa.com
Cc: DAWG Mailing List <public-rdf-dawg@w3.org>
Message-ID: <20050603181653.GC12617@monkeyfist.com>

On Fri, Jun 03, 2005 at 01:34:07AM +0200, jos.deroo@agfa.com wrote:
> 
> [...]
> > I don't see how we can abjure the addition of told triples without also
> > abjuring the addition of inferred triples.
> 
> I'm still one of those who think that inferred triples
> can be sharply described e.g. as sequent proofs and is
> all about explicit assumptions and transparent explanations
> ...

I'm not smart enough to have an opinion about that; but I don't think it's
really to the point. Which is whether a service may (or must?):

       - execute a query against its own RDF dataset, 
       - against only the dataset specified in the request, or
       - against the dataset specified in the request plus some other
       triples or graphs. 

Both inference and an unavoidable service background graph make good
sense in some cases (and less sense in others), but neither of them is
executing a query against *precisely* the dataset specified in the request.
Both of them modify the requested dataset in some way. And, or so I would
argue, whether you are right, Jos, that "inferred triples can be sharply
described", a service provider can transparently and sharply describe its
unavoidable background graph by merely publishing it on the Web!

If that is the crux, inference is only better off than an unavoidable
background graph if you are right about explanation. If there are cases
where inference is hard to explain, then it's *worse* off. :>

Kendall Clark
Received on Friday, 3 June 2005 18:17:27 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:23 GMT