OWL version of schema.org
Contact: Holger Knublauch (updated 2016-08-15)
See also: SHACL version of schema.org
This version of the schema.org namespace has been created with the
goal of supporting the use of the schema.org data model with semantic technology tools based on OWL and RDF.
It uses the same identifiers as the official version but applies the following conventions that
are hopefully well suited for many RDF/OWL tools:
- schema:Thing has been mapped to owl:Thing
- schema:name has been mapped to rdfs:label
- schema:description has been mapped to rdfs:comment
- schema:url has been deleted - the URI of the subject should be used instead
- XSD datatypes are used instead of schema datatypes
- The types owl:Class, owl:ObjectProperty and owl:DatatypeProperty are used throughout
- Legacy properties (e.g. in plural form) have been deleted
- schema:Class, schema:Property and schema:additionalType have been deleted
- rdfs:labels have been un-camelcased
- Properties that had multiple schema:domainIncludes statements use a single owl:unionOf domain.
- Properties that had multiple schema:rangeIncludes statements use a single owl:unionOf range.
- Properties that either allowed numbers or strings have been mapped to xsd:float.
Properties that either allowed xsd:string or URL have been mapped to xsd:anyURI.
There is an alternative "single-range" version that has the
following treatment of property ranges:
- Every property has exactly one rdfs:range statement. If the property originally had multiple
range classes, then an artificial named superclass has been generated.
Properties that allowed objects or strings allow objects only.
The goal of this convention is to make sure that data can be
consistently entered and that queries better know what to expect about the datatypes.
While the original schema.org design focuses on making it easy for content providers to use
it (with a great deal of flexibility), the main use case of this OWL version is for RDF-based
data entry and storage of data. In a nutshell, people should be able to create instances of
the schema.org classes with off-the-shelf ontology and instance editors (such as
TopBraid Composer and
TopBraid Enterprise Vocabulary Net)
and then store the data in an RDF triple store.
From the triple store it then becomes possible to generate HTML web pages with embedded RDFa or
Graph URI (for owl:imports): http://topbraid.org/schema/