Re: Comment on Last Call Working Draft of RDF Syntax document concerning blank node identifiers

My view is that the current Syntax document is incorrect, and incorrect in
an important manner.  I have spent quite a bit of effort to get this
changed, starting last year, well before Last Call, but the issue is not
yet resolved to my satisfaction.  

The last communication I received was on 30 January 2003.  I had been
waiting for the rest of the promised changes, but to my recollection I have
not yet recieved any.  

Peter F. Patel-Schneider
Bell Labs Research
Lucent Technologies


From: Brian McBride <bwm@hplb.hpl.hp.com>
Subject: Re: Comment on Last Call Working Draft of RDF Syntax document concerning blank node identifiers 
Date: Wed, 14 May 2003 14:30:23 +0100

> At 08:55 14/05/2003 -0400, Peter F. Patel-Schneider wrote:
> 
> 
> >I view the following message sent to www-rdf-comments@w3.org as a
> >substantive comment.  However, I believe that it has not generated an
> >entry on the last call comments issues list.  Please add it to this list.
> 
> Peter,
> 
> You state in your message that the intent of the last call WD is 
> clear.  I'm minded therefore to classify this as an editorial 
> comment.  These are tracked by the editors, not on the last call comments 
> list.  Dave proposed amendments to address your concern some time ago (30th 
> Jan).
> 
> Brian
> 
> 
> >Peter F. Patel-Schneider
> >Bell Labs Research
> >Lucent Technologies
> >
> >
> >From: Dave Beckett <dave.beckett@bristol.ac.uk>
> >Subject: Re: Comment on Last Call Working Draft of RDF Syntax document 
> >concerning blank node identifiers
> >Date: Thu, 30 Jan 2003 15:15:30 +0000
> >
> > > >>>"Peter F. Patel-Schneider" said:
> > > >
> > > >
> > > > The handling of blank nodes is still problematic in the LCC version 
> > of the
> > > > RDF Syntax document.
> > > >
> > > > The intent is clear.  Each nodeElement that does not otherwise get a
> > > > subject is given a blank node identifier as a subject.  The 
> > string-value of
> > > > this blank node identifer is to be different from the string-value of 
> > every
> > > > other blank node identifier resulting from the parsing of the RDF/XML
> > > > document.
> > > >
> > > >
> > > > However, the document does not follow this intent.
> > >
> > > <snip/>
> > >
> > > I feel it does and although we've already discussed this in earlier
> > > messages, I propose to address this by adding the following
> > > clarifications in the sections you mentioned:
> > >
> > >
> > > 5.2 Identifiers - Blank Node Identifiers
> > > 
> > http://www.w3.org/TR/2003/WD-rdf-syntax-grammar-20030123/#section-Identifiers
> > >
> > > I will try to improve the second paragraph to make it clearer how the
> > > algorithm used for generating / constructing concrete blank node
> > > identfifiers must not result in erroneously merged blank nodes in the
> > > graph.  If you have specific wording suggestions, they would be
> > > useful for me to consider.  I cannot work on the exact new set of
> > > words at this time, but will look at it in a few weeks.
> > >
> > >
> > > 6.1.7 Blank Node Identifier Event
> > > 
> > http://www.w3.org/TR/2003/WD-rdf-syntax-grammar-20030123/#section-blank-nodeid-event
> > >
> > > Add a note that the generated blank node identifier may be not be the
> > > exact concatentation here but may be generated by any algorithm as
> > > discussed in 5.2, already pointed to here.
> > >
> > >
> > > 6.3 Grammar Notation
> > > 
> > http://www.w3.org/TR/2003/WD-rdf-syntax-grammar-20030123/#section-Infoset-Grammar-Notation
> > >
> > > Add a pointer from the definition of bnodeid in the notation to point
> > > directly to the 5.2 blank node identifiers section, to be amended as
> > > described above.
> > >
> > > Dave

Received on Wednesday, 14 May 2003 09:40:48 UTC