Annotation of axioms

All

Apologies if I am out of the loop.  A problem has just come to my  
attention, that I hadn't properly appreciated.

As I understand it, the current spec calls for comments on individual  
axioms but not annotations of individual axioms.
This worries me, although I am not sure how much difference there will  
be in practice, given how weak OWL annotations are.  However, I  
presume that implementations are at liberty to ignore comments and not  
annotations, and that they may have lesser status in other ways.

In setting up a system for collaborative editing, ontology evolution,  
and re-use, I have as much interest in annotating individual axioms -  
as I do classes, individuals or properties as a whole.  Each axiom  
represents a "statement" - either a fact or a restriction.  They can  
be made by different individuals, for different purposes, in different  
modules, on different authority, etc.  I need the same range of tools  
to annotate them as I do for the complete entities.   To have two  
different mechanisms is going to be very awkward indeed.

in simple applications to date, we have got by annotating just the  
entities as a whole, but as we move to "real" applications, it is  
obvious that this at best leads to ugly non-standard kluges.  My  
informed guess is that it will become just plain unworkable at scale.

I presume annotating individual axioms creates difficulty for the RDF  
serialisation (unless we use something like named graphs),  but I  
think it is a very serious problem if we are putting forward OWL as a  
serious KR language for distributed use in which different statements  
will be made by different folk and trusted to different degrees.

Regards

Alan

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

Received on Friday, 6 June 2008 15:04:01 UTC