TAC Ontology needs to extend the STIX 2.1 specification with respect to "Extension Definition"

[rhohimer]

The mantra of the TAC-TC has been to "embrace and extend". The TAC extensions have incorporated the object oriented approach of subclassing and taxonomic classification. However, STIX 2.1 does not have the notion of classes and subclasses. Therefore, we need to understand the relationship between STIX Extension Definition and the class hierarchy of the TAC ontology.

I'm contemplating the notion that if a Extension Definition is asserted that it may be used to assert a "subclassOf" property.

[rhohimer]

In conversation with @Vasileios-Mavroeidis we formed an approach to dealing with the STIX Extension Definition. We concluded that a SEMANTIC extension mechanism is different from the STIX extension mechanism.

The TAC Ontology uses a formal ontological language (OWL) to represent concepts such as classes and properties. With a formal ontology we have available the syntax to create a subclass and subproperty. This is an embedded inherent capability in the OWL language. Because of this capability it is straight forward (easy) to create extensions.

The Extension Definition provided in the STIX specification is not robust or expressive enough to express a subclass relationship or subproperty relationship. It does allow the writer to add properties to existing objects (or create new objects) but the mechanism does not assert any relationships between the new class/property and the old.

New SEMANTIC Extensions should be given there own namespaces during development. This implies that the new extensions will be in their own ontology files. These files will be imported into the stix-semex ontology (stix-semex.owl).