This is a re-statement of the schemas for [RDF] and [RDFS].
see also: rdfs.n3 in some work on vocabulary evolution around May 2001.
@@I wish we'd divided the RDF M&S 1.0 namespace between syntactic things (ID, about, RDF, li) and properties/classes (type, subject, property, object)
"The most general class".
"The concept of Class"
RDF 1.0 says: "A name of a property, defining specific meaning for the property". RDF schema says "The concept of a property."
"Properties used to express RDF Schema constraints." @@what does that mean?
"Resources used to express RDF Schema constraints." @@what does that mean?
"A triple consisting of a predicate, a subject, and an object."
RDF Schema says "This represents the set of reified statements."
"This represents the set Containers."
"This represents the set of atomic values, eg. textual strings." @@not a subclass of Resource? @@TODO relate to XML infoset for parseType="Literal", relate to XML Schema datatypes for integer etc. "Note: We expect future work in RDF and XML data-typing to provide clarifications in this area. "
RDF 1.0 says "Identifies the[@@definite description on purpose?] Class
of a resource." @@not a sentence. RDF Schema says "Indicates membership
of a class" @@not a sentence. called instance-of
in the Classes
and Individuals section of OKBC. hmm... eq(x, Class) <- type(x, x)?
type otherwise acyclic?
"Identifies the principal value (usually a string) of a property when the property value is a structured resource." @@not a resource.
"Identifies the property used in a statement when representing the statement in reified form." @@not a sentence. @@acyclic?
"Identifies the resource that a statement is describing when representing the statement in reified form." @@not a sentence. @@acyclic?
"Identifies the object of a statement when representing the statement in reified form." @@not a sentence. @@acyclic?
(@@omitting french labels)
"This is how we associate a class with properties that its instances
can have"@@not a sentence
@@acyclic? I think the spec says so, but that makes it useless for
declaring synonyms.
same semantics as OKBC
:DOMAIN? i.e. this rule?
@@OKBC domain has these rules; do we need them?
type(s, Property) if domain(s, c)
type(c, Class) if domain(s, c)
@@TODO: study relationship to algernon's idea of domain.
"Properties that can be used in a schema to provide constraints"@@not
a sentence. @@what does that mean?
called slot-value-type
in OKBC (@@their rule has more preconditions related to the domains. Are
those necessary?)
@@compare with algernon, oil
Indicates membership of a class@@not a sentence. @@weak! same as type!
@@acyclic?
called subclass-of
in the Classes
and Individuals section of OKBC.
this rule isn't a complete definition. @@state the only-if part.
"Indicates specialization of properties"@@not a sentence. @@acyclic? I think the spec says so, but that makes it useless for defining synonyms.
compare with OKBC's subset-of-values: subPropertyOf(?specl, ?genl) => (subset-of-values ?specl ?f ?genl), I think.
These have no impact on inferencing:
"Use this for descriptions" @@complete sentence, but doesn't include the predicate 'comment' nor a typical subject and object. @@relationship to dc:description?
"Provides a human-readable version of a resource name." @@not a sentence
"Indicates a resource that provides information about the subject resource." @@not a sentence
"Indicates a resource containing and defining the subject resource."