Re: summary of reification semantics issues (material for discussion).

On Wed, 2003-03-12 at 15:54, Tim Berners-Lee wrote:
> Thank you Pat for this summary.  I find in it
> a conclusion that de-re semantics for reification are useful,
> which is not the conclusion to which I come.

I don't think they're useful either; but I don't think
that's all that relevant to the WG's position.

The fact of the matter is: there's deployed code that
handles reifictation syntax, and it can only be
reconciled with the de-re semantics.

cf

2.17 Reifying Statements - rdf:bagID and rdf:ID
http://www.w3.org/TR/rdf-syntax-grammar/#section-Syntax-reifying

in particular,
  http://www.w3.org/TR/rdf-syntax-grammar/example20.rdf
and
  http://www.w3.org/TR/rdf-syntax-grammar/example20.nt


we have

  <http://example.org/> <http://example.org/stuff/1.0/prop> "blah" .

and its reification includes

  <http://example.org/triples/#triple1>
<http://www.w3.org/1999/02/22-rdf-syntax-ns#subject>
<http://example.org/> .

by analogy...

[...]

>   This can be dramatized by asking: is the 
> > rdf:subject of the triple
> >
> > ex:Mary ex:had ex:littleLamb .
> >
> > a girl or a uriref  ?

... this says that the rdf:subject is a girl; i.e.

	<#marySentence> rdf:subject ex:Mary.

no quoting around ex:Mary there. :-{

That's the way (all?) the deployed RDF parsers work.


The WG considered 3 options:

  (a) write a spec for the way the deployed code works
  (b) write a de dicto spec, and try to get that deployed
  (c) punt

and we chose (a). I argued for (c) or (b), but we ultimately
chose (a). I abstained.

Or at least... that's the way I remember it; I can't
seem to confirm from the records.

We have
   ACTION: 2002-01-11#3: bwm - check that reification is listed as an
	issue ("fix/drop reification")
   http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/0095.html
but I can't find that issue in the issues list.

It was item 8 on a subsequent agenda...

  RDFCore WG minutes for the Telecon 2002-02-01
  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Feb/0010.html

... and here's Jan's re-statment of the 3 options, which made
it clear that the quoting issues in the de dicto style were
tricker than they looked...

  Proposals? Re: use/mention and reification
  From: Jan Grant (Jan.Grant@bristol.ac.uk)
  Date: Thu, Jan 24 2002
  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/0236.html

but I can't find any actual decision on the matter. Odd.
Aha! Those long action ids are golden... googling yields...

  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/0100.html
  ==>
  http://www.w3.org/2000/03/rdf-tracking/#rdfms-identity-of-statements

... which is a closely related issue, but it's not
an issue about de re vs. de dicto per se, as per
the options in Jan G's message.

So it looks like the WG never decided to adopt the de re
semantics as opposed to de dicto; the spec just ended
up de re as clarification of status quo.

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/

Received on Thursday, 13 March 2003 01:03:49 UTC