Re: RDFCore last call WD's: Two comments on the RDF documents

Frank,

At 03:04 PM 2/13/2003 -0500, Frank Manola wrote:
> >
>snip
> >
> > So, how am I recommending that you fix things?  Unfortunately, I'm mostly
> > stating what you should NOT do.  I'm claiming that
> > using nested syntax will convey the wrong impression to many readers (e.g.,
> > those that model belief they way I did above), so something like an
> > EXPLICIT quotation needs to be included.  To me, the use of double brackets
> > didn't adequately convey the notion of quotation.
> >
>snip
>
>I will try using a diagram, rather than what appears to be nested
>syntax, since we don't support nested syntax (and I didn't intend for
>what the Primer uses to be interpreted as nested syntax). However, we
>don't really support explicit quotation either, so we can't really
>substitute that.  Whatever is said in the Primer on reification will
>have to be consistent with what is said in the Semantics document, which
>means this is related to your message
>http://lists.w3.org/Archives/Public/www-rdf-comments/2003JanMar/0211.html
>"Statings -- Much ado about nothing".

I agree that figuring out a representation is tough.  I'm used to KIF, where
I can say pretty much anything that I want to.  Its hard even to have an
e-mail conversation about certain aspects of RDF, since the vocabulary just
isn't available.

> > I consider introducing "what we would *like* to be able to do" very
> > dangerous.  It
> > gives the impression that RDF might be used to represent propositional
> > attitudes,
> > when in fact it can't.  I would prefer that the WG be as up front as
> > possible about
> > stating the limitations it has placed on RDF.
>
>I don't think the Primer actually conveys this impression, and I think
>it tries to be up front about the limitations of RDF reification (it
>certainly spends a lot of space talking about them at any rate).
>However, I could see adding a caveat at the beginning along the lines of
>"you might think you're going to be able to do foo, but watch carefully,
>because you can't".  Does this make sense?

Yes.  Again, we have a vocabulary problem.  I've seen the phrase "propositional
attitude" pop up now and again, but the notion of "proposition" is quite 
difficult
to pin down, so that makes it hard to discuss.  But, I will try.

I think of propositions as being the proper objects of belief.  So, one doesn't
believe in a *sentence* "George is a clown".  Rather, one believes in the 
proposition
that that sentence is true.  In KIF, I'm used to seeing nested syntax 
employed when
representing statements about belief.  Thus, if RDF allowed nested 
statements, then
I would use them to represent belief.  But it doesn't, and hence I'm thinking
that an example that *does* employ nesting gives the wrong impression.

However, the following text appears in the Concepts and
Abstract Syntax document:

    > Not every RDF/XML expression is asserted. Some may convey meaning 
that is partly
    > determined by the circumstances in which they are used. For example, 
in English, a
    > statement "I don't believe that George is a clown" contains the words 
"George is a clown",
    > which, considered in isolation, has the form of an assertion that 
George exhibits certain
    > comic qualities. However, considering the whole sentence, no such 
assertion is considered
    >to be made.

I'm still waiting to see an example of an RDF statement that is not 
asserted.  If someone
could illustrate one, that would be a big help.  If, in fact this passage 
is meant to
refer to reified statements, then please write down how this would look in 
RDF using reified
statements.

Note: The text above actually uses term "expression" rather than 
"statement" when it
talks about things being asserted.  RDF terms are expressions, and terms cannot
be asserted, so that makes the
first sentence trivially true.  But I assume that that was not the authors' 
intent.
Suggestion: Replace "expression" by "statement" in the first sentence, if 
that's what's
meant.

My impression is that its impossible to express the sentence
      "I don't believe that George is a clown"
in RDF.  In the OLD RDF, I would have said yes, but not now that we have 
switched
to "statings".  If it is acceptable for a stating to serve as the object of 
a statement
about belief, then I would have to withdraw my objection.  So, can we represent
beliefs in RDF?

Cheers, Bob

Received on Thursday, 13 February 2003 21:16:44 UTC