[OWL] Annotations on subclass axioms etc.

[Alan Rector]

All

One minor point that I made to Ian but didn't get a chance to bring  
up in the regular discussion under the heading of 'enriched  
annotations'.

For detailed annotation of provenance and attribution to modules, we  
need to be able to put an annotation on each axiom, and each use of a  
restriction.

USE CASE: We need to know who asserted that subclassOf(B,A) and why,  
its editorial status, etc.  Particular subuse case, markup to  
differentiate inferred from asserted axioms.  We need this  
information at a much finer grain than the definition of a class or  
individual, because the complete definition might be assembled out of  
numerous different submodules.

Likewise, we need a solution, presumably annotation, for indicating  
which restriction usage came from which ontology when importing an  
ontology.  The namespace mechanism works fine for the names of  
classes and individuals, but not for the axioms themselves nor for  
anonymous classes.

I think a standardised/agreed annotation vocabualry for the purpose  
of identifying the module from which an axiom came would be a big  
step forward in modularisation. As it is, as I understand it,  if I  
have three modules, M1, M2, MJoin and  M2 imports M1, and MJoin  
imports M2 (and possibly M1),  I have no way of knowing if an axiom  
on a class in M1 that uses a class or property in M2 was created by  
M2 and imported by MJoin or added by MJoin.

As I read it, the abstract syntax doesn't currently allow annotations  
on subclassOf or typeOf axioms; I am off line and can't check but I  
don't think it allows them on individual applications of restrictions  
either, but only on the classes.

Regards

Alan




-----------------------
Alan Rector
Professor of Medical Informatics
Department of Computer Science
University of Manchester
Manchester M13 9PL, UK
TEL +44 (0) 161 275 6188/6149
FAX +44 (0) 161 275 6204
www.cs.man.ac.uk/mig
www.clinical-esciences.org
www.co-ode.org