System Design and Ontology

An OCHRE system consists of a central database server and a database "client application" that runs on each user’s computing device and enables him or her to enter, search, and display information. The database struc­ture reflects a highly generic "upper ontology" within which any project’s data can be represented as a collection of many interlinked items of information. Entities of interest and relationships among them are represented by structurally independent database items and by hierarchies of those items and links between them. This ontology is implemented in OCHRE by means of eighteen document types that conform to XML (Extensible Markup Language), a non­proprietary data format published by the World Wide Web Consortium. An OCHRE database consists of a large number of small XML data objects ("documents") that conform to one or another of these eighteen document types, whose structure is defined using the XML Schema standard. The XML objects in an OCHRE database are queried and updated by means of XQuery (XML Query Language).

External Resources

Note that OCHRE distinguishes between "core data" stored in XML format in the central database and "external resources" such as image files, PDF files, 2D and 3D spatial data files, and audio or video files. Each project provides its own Web server to store such resource files, which are not stored in the central database but are simply catalogued there, with URL links to the server locations where they are stored.

CHOIR Ontology

The underlying conceptual ontology, of which the OCHRE database system is one implementation, has been dubbed CHOIR, which stands for "Comprehensive Hierarchical Ontology for Integrative Research." The eighteen XML document types used by OCHRE are collectively called CHOIR-XML because they are the XML version of the ontology. A complementary CHOIR-RDF version that makes use of the World Wide Web Consortium’s RDF (Resource Description Framework) standard is currently being developed using OWL (Web Ontology Language). The CHOIR-RDF version of the ontology will be made publicly available to facilitate the use of this ontology in other interoperable software. Moreover, the OCHRE system itself is being enhanced to allow projects to export their database content as RDF subject-predicate-object "triples" that conform to CHOIR-RDF. This will provide a semantically rich, standards-based archival format for OCHRE data that captures the complex distinctions and relationships found within the database, and it will also enable direct exchange of OCHRE data with other software applications that are based on the same ontology. Note that CHOIR, as a conceptual ontology, is distinct from any particular implementation of it. Even though it is currently implemented in an XML database system (i.e., OCHRE), it is not dependent on XML and could be implemented using a different data model and querying language—for example, in an RDF-graph database system using the SPARQL querying language, or in a relational database system using SQL.