W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > December 2002

Re: rdf:li ?? in primer

From: Dave Beckett <dave.beckett@bristol.ac.uk>
Date: Mon, 09 Dec 2002 17:24:35 +0000
To: pat hayes <phayes@ai.uwf.edu>
Cc: w3c-rdfcore-wg@w3.org
Message-ID: <20647.1039454675@hoth.ilrt.bris.ac.uk>

>>>pat hayes said:
> Dave said:
> >I think you are a bit confused Pat.
> 
> Yes, I was.
> 
> I have to say, this idea of a 'syntax name' that isn't actually a 
> name *is* very confusing, particularly when it sure LOOKS like a 
> name.  Im still puzzled about how it's supposed to actually work. 
> Seems to me that the actual graph formed from some RDF/XML which uses 
> rdf:li can vary from moment to moment at the whim of the parser, 
> since it depends on some ordering that isnt specified in the syntax.

An (XML) namespace we've been using as a set of names.  Some of these
names are not used in triples, just for rdf/xml; I've tried several
ways to say that, and syntax name is about as snappy as I've got.

The rdf:li syntax in rdf/xml does emit a different property as it is
used in the document, but the exact rdf:_n property emitted is well
defined and the same for each mapping of rdf/xml to triples.  So, you
don't need to worry about it :)

> >  rdf:li is a syntax name; it has
> >never been an RDF property, class or whatever.  We have thus never
> >deprecated or forbidden it.
> 
> Well, its ours, right? I mean, its in our namespace? It starts with 
> 'rdf:', after all.

Yes, we have control over that.

> >Just to be clear; RDF as we have revised has changed two names in the
> >RDF namespace
> >   rdf:aboutEach rdf:aboutEachPrefix - syntax only devices, now removed
> >
> >(we've added some of course)
> >
> >The definitive words for rdf: things:
> >
> >   5.1 The RDF Namespace
> >   http://www.w3.org/TR/rdf-syntax-grammar/#section-Namespace
> 
> OK, thanks.
> 
> >
> >including:
> >   [[Syntax names - not concepts
> >
> >     RDF Description ID about bagID parseType resource li nodeID
> >     datatype
> >   ]]
> 
> Now Im puzzled about something else. I've been using rdf:Datatype as 
> a class name. Is that a mistake??

Note the case & namespace difference
   rdfs:Datatype rdf:type rdfs:Class

     See RDF Vocab
     http://www.w3.org/TR/rdf-schema/#ch_datatype

   rdf:datatype - just a bit of rdf/xml syntax used to give the URI
     for Typed Literals

     See RDF/XML Syntax
     2.9 Typed Literals - rdf:datatype
     http://www.w3.org/TR/rdf-syntax-grammar/#section-Syntax-datatyped-literals

> 
> Pat
> 
> PS. BTW, you say there that 'any other names are not defined and 
> SHOULD generate a warning when encountered' which seems a bit strong. 
> That means for example that any OWL vocabulary is going to generate 
> an RDF warning. In fact, any RDFS vocabulary is going to generate an 
> RDF warning.

I said "... not defined" in the context of the RDF namespace.  That
is so that if someone uses a name we don't list, such as rdf:foo, it
gives a warning.

Other vocab will be in a different namespace (RDFS, OWL, ...)

Dave
Received on Monday, 9 December 2002 12:27:11 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:54:50 EDT