W3C home > Mailing lists > Public > www-rdf-rules@w3.org > September 2001

Re: What is an RDF Query?

From: Sandro Hawke <sandro@w3.org>
Date: Tue Sep 11 09:00:13 2001
Message-Id: <200109111257.f8BCvEr03080@wadimousa.hawke.org>
To: pat hayes <phayes@ai.uwf.edu>
cc: www-rdf-rules@w3.org
[sorry for any duplicates]

Rather that go into a lengthy reply, can I just say "layering" and
point out that I said "LISP syntax" (atoms and dotted pairs) not "the
LISP programming language" (with lambda and everything).  Is that
enough?  RDF clearly can't talk about the color of my dog without us
defining some terms (that is, extending the language), and the same
goes for talking about rules, queries, ontologies, schemas, and (if
we're as pure as I think we should be in defining the bottom layer)
bags, sequences, statements, and types.

    -- sandro

================================================================

> > > > Yep.   This works, but it ends up more complicated than we need, if we
> > > > just handle existential variables properly.
> > >
> > > I don't understand the need to encode everything in RDF.  Encoding
> > > everything in RDF ends up with a very complicated system, and most of the
> > > semantic import will be in the encodings, which are not part of RDF.
> >
> >The "need" is a perhaps more of an opportunity.  When one is
> >programming in LISP, one tends to use LISP syntax for almost
> >everything because one's mental and software machinery is all geared
> >up for it.
> >
> >Once we've made it easy to represent and work with knowledge in RDF,
> >we might as well do it with all our meta-knowledge, control-knowledge,
> >etc...
> 
> No, we will *not* be able to do that, becuase RDF is intrinsically 
> unable to express most of this stuff. It is a *very* weak assertional 
> language, and can hardly say anything more than simple facts. It 
> definitely is not going to be adequate for the entire semantic web. 
> It cannot represent anything but very trivial control knowledge (eg 
> it can't represent implications or conditionals), and cannot 
> represent true meta-knowledge at all. The analogy with LISP (or any 
> programming langauge) is misleading, since it doesnt take much to 
> have a Turing-universal programming language. But assertional 
> languages are different, and one cannot extend the expressive power 
> of an assertional language by implementing something in it.
> 
> >
> >Except for (1) it is sometimes more confusing and (2) it might have big
> >performance penalties.
> >
> >DAML+OIL has this same issue, right?  The instance knowledge is in
> >RDF, which makes perfect sense.  It's a great language for listing the
> >property/value pairs for objects.
> 
> Its not a GREAT language for anything, but in any case that is *all* it can d
> o.
> 
> > But then the ontology (describing
> >what kind of properties and values are allowed) is also encoded in
> >RDF.
> 
> partially describing; but that is a very strong 'partial'. Even the 
> simple assumption of subclass inheritance can't be said in RDF; it 
> needs RDFS. And the fact that, for example, sequences and bags are 
> disjoint classes can't be said in RDFS; it needs something like DAML.
> 
> > What's more, the characterists of the ontology language are
> >*also* encoded in RDF in http://www.daml.org/2001/03/daml+oil.daml!
> 
> No, they are not. DAML+OIL is much more expressive than RDF, and 
> cannot be translated into, or expressed in, RDF.
> 
> >So why not describe a query of an RDF dataset in an RDF dataset?
> 
> How?
> 
> Pat
> 
> ---------------------------------------------------------------------
> (650)859 6569 w
> (650)494 3973 h (until September)
> phayes@ai.uwf.edu 
> http://www.coginst.uwf.edu/~phayes
Received on Tuesday, 11 September 2001 09:00:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:53:09 GMT