Re: the mythical RDF inference engine was: Re: What is truth anyways?

patrick hayes wrote:

>
> >
> >But OK, I'll bite. OWL might choose to represent individuals and facts
about
> >individuals _in RDF_. OWL might have a model theory that provides for
> >certain inferences to be drawn given these individuals and a bunch of
> >classes.
>
> OK, stop right there. You said OWL is representing facts in RDF. To
> me that sounds like saying that Im talking English in French: it
> simply doesn't make sense. Which language are you referring to? RDF
> or OWL? Because they are *different*.

To be more precise, we've had this exact discussion in  the context of OWL.
For example, Peter PS has suggested that everything in OWL be dark _except_
individuals. There is a proposed OWL abstract syntax which defines
"individual" and "fact". In that proposal for OWL, the individuals/facts
might be encoded as RDF. So, no, not all OWL consists of simple facts, just
the facts are facts.

>
> >That might be useful. I might describe these inferences as being
> >drawn _on_ the particular facts. Nonetheless we consider such facts as
data
> >from which we draw conclusions. What is the big deal?
>
> The big deal is, you have to say which formal language you are using.

Right. So for the sake of this argument: OWL as Peter has proposed.

>
> Wrong. It is not a datastructure, it is a description in a
> (admittedly rather simple) relational logic. I know that OWLs task
> would be a hell of a lot easier if only RDF *were* just a
> datastructure language, but that isn't what the RDF authors intended,
> or what the W3C spec says.

Either the current RDF M&S is sufficiently precise for your purposes, in
which case it doesn't need revision, or it isn't, in which case the
interpretation remains open to discussion until the current round of RDFCore
documents are issued as W3C RECommendations.

I'm not inclined, personally, to blindly accept the current state of
decisions, and current round of WDs as fait accompli, particularly if they
make life incredibly difficult for other specs which are trying to
"layer" -- whatever that means -- on RDF. The layering of OWL on RDF is also
a W3C charter mandate, so just because it isn't part of RDFCore's worklist,
it _is_ an important architectural issue for the Semantic Web as a
collective.

> >>
> >>  The whole point of having a precise account of meaning in the
> >>  language spec is to sanction some inferences and not others. If we
> >>  can allow anyone to draw any conclusions from an RDF graph, then RDF
> >>  doesn't differ from XML.
> >
> >Well N-triples looks alot different to me than XML. And a triple store
fits
> >into a single relational table with three columns, so that is the
> >different -- perhaps not earth shattering, but what magic are you looking
> >for?
>
> Im not looking for any magic at all. My purpose is to try to make it
> clear to everyone that appeal to magic is right out.

Yeah well having an incredibly precise view of the meaning of an RDF triple
when there is no precise view of the meaning of each URI that forms the
triple is, well, sort of like building a slate roof on a mud hut -- the
problem isn't the roof -- and now you are asking me to add a second floor --
oh great that floor will be made of brick!

Jonathan

Received on Wednesday, 12 June 2002 19:17:45 UTC