[uos-convene] Modularizing ontologies in preparation for mapping

[Dagobert Soergel <dsoergel@xxxxxxx>]

This is perhaps obvious and may have been stated before.

In preparation for mapping one could dissect an ontology into ever
smaller pieces, possibly down to individual axioms.  Each piece
would have to be packaged with the minimal context needed for its proper
interpretation, where context would include basic assumptions, scope
(pragmatic assumptions, as pointed out particularly by Doug Lenat),
axioms or other ontology elements on which the given element depends, and
perhaps more.  These contexts could in turn be arranged in modular
packages for ease of specification.

Mappings between two ontologies would than mappings between these pieces,
establishing correspondence at the highest level possible.  For
example, two ontologies could be the same in the way they deal with
time.  Mapping at that level would imply that all elements included
in the time modules would align between the two ontologies. 
Sometimes a mapping (or a statement of discrepancy) would have to go to
the lowest level.

DS

Dagobert Soergel
College of Information Studies
University of Maryland
4105 Hornbake Library
College Park, MD 20742-4345
Office: 301-405-2037     Home:  703-823-2840        Mobile: 703-585-2840
OFax:   301-314-9145        HFax: 703-823-6427
dsoergel@xxxxxxx     www.dsoergel.com