RDF datatypes

[John Madden <john.madden@mac.com>]

Hi guys,

I looked a little into the "datatypes in RDF" question. I think the answer is in the RDF Abstract Semantics, section 4.3

4.3 A Note on rdfs:Literal

Although the semantic conditions on rdfs-interpretations include the intuitively sensible condition that ICEXT(I(rdfs:Literal)) must be the set LV, there is no way to impose this condition by any RDF assertion or inference rule. This limitation is due to the fact that RDF does not allow literals to occur in the subject position of a triple, so there are severe restrictions on what can be said about literals in RDF. Similarly, while properties may be asserted of the class rdfs:Literal, none of these can be validly transferred to literals themselves.

Note the comment that "...there is no way to impose this condition by any RDF assertion or inference rule." What this means, as I take it, is that in effect literals (including datatyped literals) are a "special feature" of RDF, and their semantics DOES NOT map to any "plain-vanilla" RDF model. They have their own special semantics, which is explicit (see http://www.w3.org/TR/rdf-mt/#dtype_interp), but is not specifiable in RDF.

What this means for us is, that the meaning of m:datatype is NOT that it generates any new triples, but rather that is restricts the value space of the interpretation of the typed literal to which it applies in accordance with http://www.w3.org/TR/rdf-mt/#dtype_interp .

While datatype restriction cannot be expressed as RDF triples, nevertheless the RDF spec does provide a standard notational device for indicating that such datatype restrictions apply , namely the "^^" notation, as in 

"125"^^xsd:integer

"20070101"^^xsd:date

"3.14159"^^xsd:float

Consequently, when an ODF metadata parser encounters an m:data-type attribute, the response of the parser should NOT be to produce any additional triples, but SHOULD be to append the appropriate ^^-suffix to the literal to which the m:data-type attribute applies (which will be either the value of the associated m:data-value element; or in the absence of such an element, the value of the XMLLiteral that is the child of the xml element carrying the datatype restrcition.

Elias, does this sound reasonable? 

john