W3C home > Mailing lists > Public > www-rdf-comments@w3.org > October to December 2001

Re: RDF Model Theory Working Draft: Comment

From: Arjun Ray <aray@nyct.net>
Date: Mon, 8 Oct 2001 16:02:13 -0400 (EDT)
To: <www-rdf-comments@w3.org>
Message-ID: <20011008145455.I62411-100000@bsd1.nyct.net>

 [Would it be possible to send text/plain messages only?  Thanks.]

On Mon, 8 Oct 2001, Pat Hayes wrote:

> As to the issue you raise, I confess that this interpretation
> simply had not occurred to me, but I agree that the wording you
> cite does invite such an interpretation. The problem is in the
> wording, however, rather than in the pictures. The intention is
> *not* that all three of a triple's terms should be 'nodal'. [...]
> The graph contains precisely one edge for each triple, and one
> node for each 'nodal' label in the N-triples document, ie each
> label which occurs in a subject or object position in the document
> somewhere. (However, a URI label which occurs only as a property
> name in the document will not be a node label anywhere in the
> graph; that is where the text you cited is incorrect, and needs to
> be rewritten more carefully.)

Okay, this changes things considerably.  [Now, a different part of the
RDFMS spec is broken, or at least, very poorly formulated:)]

> It may be germane to point out that with the current MT, nothing
> would be gained, speaking semantically, by including a node for each
> edge label,

Graham Klyne's "moot node" clarification, okay.

> [On a different "gestalt":]
> The reason that reification would be unnecessary in this scheme is
> that this *is* reification.

Yep:-) [With the difference from RDFMS-style reification that the
'type' arc is redundant when the arc labels in this scheme are all
fixed ('subject', 'object', 'verb', 'literal', 'uriref', etc.) the
type being effectively inferable from the outbound arcs.  The benefit
that I was personally prefering - and hoping that RDF would eventually
have - was the ability to simply describe a property (or synthesize
one) without also having to name it explicitly for the purpose of
using it.  Currently, bNodes aren't allowed in the property slot, and
IMHO that loses a lot of the semantic appeal of viewing "properties"
as (a subset of) "resources".  What happens instead is a whole bunch
of factitious URI inventing - handwaving.  BTW, I don't subscribe to
the official W3C dogma that everything is, can be and/or should be a
URI reference of some sort, much less in the HTTP scheme.]

> The model theory can be extended to reification in several
> different ways; it was omitted from the current version because
> the proper way to handle reification is still under discussion.

This will be very interesting, I'm sure!

> PPS. There *are* some places where the correspondence between graph
> syntax and XML-RDF becomes rather blurry,

A very delicate understatement, if I may say so.  The so-called XML
syntax is a complete mishmash of (rather clueless) ad hoc taggery
"developed" (and I use that term advisedly) with only how to keep it
safe for Netploder in mind.  A huge raft of entailed kludgery (like
colonified names) followed to shore it up, and it still stinks.  That
is my NSHO:-)

I sincerely hope that the WG works out a good MT before even thinking
of XML syntax.

Arjun
Received on Monday, 8 October 2001 16:00:22 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:15:16 UTC