Re: rdf as a base for other languages

From: Sandro Hawke <sandro@w3.org>
Subject: Re: rdf as a base for other languages 
Date: Sat, 02 Jun 2001 10:22:21 -0400

> 
> pat sez:
> > My point was that even with a (full) understanding of how to 
> > interpret the RDF, it still wouldnt be a translation of disjunction. 
> > Disjunction is not expressible in RDF.
> > 
> > (I might add that I would not keep going on about this if people 
> > (like Sandro) didnt keep saying the opposite.)
> 
> I'm not saying disjunction is expressible in RDF [caveat below].  I'm
> saying, of course, that disjunction is expressible in a language which
> can be layered on RDF, and talking about how we might work with that
> layering (which is why I was getting into performatives).  Peter's use
> of a vocabular[y] term ("OR") which is not in RDF meant that his example
> text was (for the most part) meaningless without outside knowledge of
> the meaning of that term.   To respond to his message I explicitely
> inferred (using human knowledge) what he meant by the term; of course
> a computer would not be expected to do that -- it would only
> understand his RDF if the term OR had been defined for it.
> 
>     -- sandro
> 
> * caveat: I mean RDF in just the sense of positive ground triples.
>   Disjunction may be expressible with RDF as specified in M&S 1.0
>   using "alt", but I'm trying to ignore that.

I would have fewer problems with the use of RDF for encoding languages
while at the same time having ground facts in the encoded language be RDF
triples if it was the case that the triples were, indeed, meaningless.
However, RDF gives meaning to every triple, which means that the OR triple,
and the other triples used in the OR have a different meaning from my
intended one.

peter

Received on Saturday, 2 June 2001 15:02:51 UTC