Comment on Eric's doc:
rdfs:range is (currently) defined by
http://www.w3.org/TR/2000/CR-rdf-schema-20000327/#s3.1.3
"An instance of ConstraintProperty that is used to indicate the class(es) that the values of a property must be
members of. The value of a range property is always a Class. Range constraints are only applied to properties.
A property can have at most one range property. It is possible for it to have no range, in which case the class
of the property value is unconstraint"
This definition is addressed by a rdf:core wg decision from 2nd August 2001
[http://www.w3.org/2000/03/rdf-tracking/#attention-developers], which allows for multiple ranges with conjunctive
semantics. No other change of semantics recorded.
(Multiple domains were allowed already in the candidate recommendation, but interpretation was changed to conjunctive (!)
semantics as well.)
------------
One use the document makes of rdfs:range is to declare a range Class for all dc-[element]s and specify the
Class as [element]Scheme.
That is (from the above) the admissible values for dc:subject for instance must be members of the Class SubjectScheme.
The document defines the Class SubjectScheme as a set of subject encoding schemes.
That is: The members of the Class are subject schemes and NOT items taken from a specified subject encoding schemes.
We only could use dc:subject in the metadata of resources, which explain ABOUT subject encoding schemes.
Huch! There is a slight chance i missing something.
Minor comments address the way the relation to the underlying DCQ rec is declared and the use of dcq as abbreviation -
and missing stuff from dc:usage decisions.
rs
|