W3C home > Mailing lists > Public > www-rdf-interest@w3.org > February 2002

New Model Theory WD + RDFCore update

From: Brian McBride <bwm@hplb.hpl.hp.com>
Date: Mon, 18 Feb 2002 18:22:50 +0000
Message-Id: <>
To: www-rdf-interest@w3.org

I'm pleased to announce the publication of a new Model Theory working draft 
by the RDFCore WG:


dated version at:


In addition there has been a flurry of issue closures:

rdfms-propElt-id-with-dr : Clarify the interpretation of an ID attribute in 
the propertyElt production within a Description element with a distributive 

With the removal of aboutEach and aboutEachPrefix, there are no 
distributive referrants

rdf-terminologicus: The RDF community needs a precise terminology to enable 
it to discuss issues.(Martyn Horner)

The primer will contain a glossary

rdfms-graph: Formal description of the properties of an RDF graph.

The model theory provides a formal account of the properties of an rdf graph.

rdfms-literals-as-resources: Consider replacing literals with resources 
whose URI uses the data: URI scheme.

The WG decided this was too big a change for its charter.

rdfms-literalsubjects: Should the subjects of RDF statements be allowed to 
be literals?

Again, the WG considered this too big a change, but noted that it knew of 
no reason why a future WG should not make this change.

rdfms-uri-substructure: xmlns, uri+name pairs or just uris..? Clarification 

Replacing URI's with pairs is a fundamental change to web architecture and 
beyond the scope of the RDFCore WG.  The WG also advises schema designers 
to end the URI references of schema namespaces with a character such as '#' 
or '/' which cannot be part of an xml name.

rdfms-boolean-valued-properties: Suggestion for a standard way to represent 
boolean valued properties.

The WG decided not to define a vocabulary for boolean's noting that:

   <fred> <chocolatelover> <true> .

can be represented as

   <fred> <rdf:type> <ChocolateLover> .

There was also progress on reification, where the 'statings' view of 
reification was formalized as given two reified statements <s1> and <s2> 
with the same values for their subject, predicate and object properties, 
then from:

   <s1> <foo> <bar> .

one cannot deduce

   <s2> <foo> <bar> .

i.e. <s1> and <s2> may denote different 'statings'.

This may be considered a change to the formal definition of reification 
given in M&S, but there was a strong view in the WG that reification had 
been originally introduced to support provenance, and the form of 
reification defined in the formal model section of M&S does not support that.

Work on data types has been continuing intensely with further progress 
being made.  Thank you to everyone who provided feedback to our earlier 

A unified view which integrates the various proposals into a common 
framework has emerged.  We hope to report back on this shortly; in the 
meantime if anyone is impatient, then the link to the discussion and 
documents can be found in the mail archive:


Received on Monday, 18 February 2002 13:24:13 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:44:34 UTC