Re: Primer reification section

On Wed, 2002-11-20 at 06:17, Brian McBride wrote:
> 
> At 08:48 19/11/2002 -0500, Frank Manola wrote:
> 
> >I'm thinking seriously about removing discussion of reification from the 
> >Primer.  Does anyone have any strong feelings about leaving it in?

I dunno about strong feelings, but I prefer to leave it out.
I think the best answer to "how do I use reification?"
is "don't."

> I take it you are referring to
> 
>   http://www.w3.org/TR/2002/WD-rdf-primer-20021111/#reification


I just reviewed that, and it doesn't have any examples
of RDF/XML syntax for reficiation, which is the one
thing we bothered to specify carefully. So I don't
see much value in it as written.


> The vocabulary is defined and we ought to explain somewhere what it 
> means.

I prefer to discuss meaning in the MT and concepts specs.
I prefer the primer to just say how to use it.

Plus, I think we decided it doesn't mean much. Let's see what
exactly we did decide... yes...

"This is a formal way of saying that two reified statements with the
same 
subject, predicate and object can be different resources."

 -- RDF Issue rdfms-identity-of-statements
  From: Brian McBride (bwm@hplb.hpl.hp.com)
  Date: Mon, Mar 11 2002
 
http://lists.w3.org/Archives/Public/www-rdf-comments/2002JanMar/0192.html


(((by the way... whew! that was kinda hard to find...

I saw one issue with reification in the name...
  http://www.w3.org/2000/03/rdf-tracking/#rdfms-reification-required

which was resolved 11 Jan
  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/0095.html

That issue related to syntax (i.e. RDF/XML to n-triples mapping)
but the record shows...

   Desire was also expressed that the WG not keep putting off the larger
   consideration of the reification issues.
   ACTION: 2002-01-11#3: bwm - check that reification is listed as an
	issue ("fix/drop reification")

But I couldn't find the "fix/drop reification" issue, but
googling for 2002-01-11#3 yielded...


[[
I checked the issues list and found

   http://www.w3.org/2000/03/rdf-tracking/#rdfms-identity-of-statements

which is aimed at the statement v statings issue and has been my 
placeholder for defining the semantics of reification.
]]
-- http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2002Jan/0100.html
)))


Hmm... it seems we have dissent on our decision on
rdf-formal-semantics:

"Unfortuantely, the RDF model theory does not (yet) provide an
acceptable
account of RDF semantics.

In particular, there is no treatment of RDF reification or RDF
containers
in the current RDF model theory"
 --
http://lists.w3.org/Archives/Public/www-rdf-comments/2002JanMar/0194.html

Is that dissent still outstanding? or is there news since then?


>  Schema is currently a little bare on the matter, but I suppose 
> Danbri could fix that.

"fix"? What's broken? it seems to quite accurately discuss
what we decided:

"Different individual rdf:Statement instances may happen to have the
same values for their predicate, subject and object properties."

  -- http://www.w3.org/TR/2002/WD-rdf-schema-20021112/#ch_statement

Hmm... I'd suggest a "see also section 2.17 Reifying Statements
from the RDF syntax spec", danbri.

http://www.w3.org/TR/2002/WD-rdf-syntax-grammar-20021108/#section-Syntax-reifying

> Why are you thinking of taking it out Frank?
> 
> I suppose the syntax document refers to it.  We'd be bound to get 
> questions.  What does this bagID stuff do?

Like I said, my answer to that is: if you need to ask that question,
you shouldn't be using this reification design. We kept reification
in for folks that started using it a while ago and consider it
useful. I'd prefer to re-open the issue and have the WG
decide reification syntax is deprecated. But maybe somebody
with an actual use for the present syntax could contribute
something concrete for the primer?

> My inclination is to keep it in.  But we might want to make some 
> discouraging noises about its use.

Discouraging noises... yes, please.


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

Received on Wednesday, 20 November 2002 10:35:25 UTC