W3C home > Mailing lists > Public > www-rdf-interest@w3.org > November 2000

RE: A triple is not unique.

From: Dan Brickley <danbri@w3.org>
Date: Mon, 20 Nov 2000 14:40:06 -0500 (EST)
To: "McBride, Brian" <bwm@hplb.hpl.hp.com>
cc: "'Dan Brickley'" <Daniel.Brickley@bristol.ac.uk>, Jonas Liljegren <jonas@rit.se>, Seth Russell <seth@robustai.net>, RDF-IG <www-rdf-interest@w3.org>
Message-ID: <Pine.LNX.4.30.0011201433140.1064-100000@tux.w3.org>
On Mon, 20 Nov 2000, McBride, Brian wrote:

> Hi Dan,
>
> > -----Original Message-----
> > From: Dan Brickley [mailto:Daniel.Brickley@bristol.ac.uk]
> > Sent: 20 November 2000 17:21
> > To: McBride, Brian
> > Cc: 'Dan Brickley'; Jonas Liljegren; Seth Russell; RDF-IG
> > Subject: RE: A triple is not unique.
> >
> >
> > On Mon, 20 Nov 2000, McBride, Brian wrote:
> >
> > > > I am however in the business of trying to make sure all that
> > > > stuff (eg.
> > > > possible rechartering of model/syntax work) reflects the
> > concerns and
> > > > experience of RDF implementors. Specifically, I'd like to better
> > > > understand how the design issues here relate to existing RDF
> > > > implementations and vocabularies. If/when we jump one way or
> > > > the other on
> > > > this issue, current code and systems may break if they've
> > > > made a different
> > > > interpretation of the spec. Right now I'm not sure if most
> > > > implementors
> > > > have for eg tried to remain neutral, with code that could
> > operate in
> > > > either style. I suspect most folk would value resolution
> > of this issue
> > > > pretty highly, and would live with the consequences. What I
> > > > don't know yet
> > > > is how big a disruption this issue's resolution might be.
> > >
> > >
> > > I can see that the disruption caused to current implementations
> > > would be a factor if m&s was ambiguous.  But if the answer lies
> > > in m&s, I humbly suggest the spec takes precedence.  That's what
> > > specs are for.
> >
> > Maybe I missed the appropriate post, but I'm unclear how we square the
> > set-oriented definition of 'Statement' with the syntactic ability to
> > assign various IDs (and hence URIs) to the XML occurances of RDF
> > statements.
>
> Just because an expression is syntactically valid, does not mean it
> is a valid expression of a language.  Consider C, many syntactically
> valid "program" violate other contraints of the language and are
> illegal.

I'd like to think I appreciate that! My concern (see reply to Jonathan) is
that we don't want the status of the expression to depend on the content
of utterly unrelated RDF files elsewhere on the Web which happen to also
encode the same p/s/o content but (via the ID mechanism) assign them a
different identifier. Imagine if your C program became illegal when
someone else far away on the Web wrote C that attempted to do something
that overlapped with your application. <shudder/>

So that's the problem: if there's only one resource of type
rdf:Statement with any given p/s/o, and RDF  syntax is useful enough to
provide a way of ascribing identifiers to those things, whose chunk of
syntax gets to be the lucky markup that names some given triple? How do we
know that the rest are violating some constraint? Or should we be careful
and say they're ALL in voilation?

Dan
Received on Monday, 20 November 2000 14:44:15 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:46 GMT