Individual vs Individual-Thing

• Mail folder: Ontolingua Mail
• Next message: Mike Uschold: "Enterprise Ontology now available"
• Previous message: Mike Uschold: "Role Classes -- e.g. Resource, Teacher"

From: Mike Uschold <mfu@aiai.ed.ac.uk>
Date: Fri, 28 Jul 95 17:43:21 BST
Message-id: <18926.9507281643@subnode.aiai.ed.ac.uk>
To: ontolingua@HPP.Stanford.EDU
Subject: Individual vs Individual-Thing
Cc: Piet-Hein.Speel@2488TAUX.urlnl.sprint.com, martin_king@uk.ibm.com,
        moralee_s@urpsl.co.uk, tcsyzz@aie.lreg.co.uk

Ontolinguists:

It appears that for the practical purpose of building an ontology, the 
class Individual is utterly useless.  It exists purely for theoretical
reasons of avoiding nasty paradoxes.

In particular:

* the only difference between an Individual and an Individual-Thing is that
  only the former may contain Unbounded entities.

* the extension of Unbounded is empty

Therefore, one should never create an instance of Individual, but rather
of Individual-Thing.  Yet, I see in various ontologies that people DO
define instances of Individual.  However, it seems that:

* this causes no harm;

* those instances could be instead of Individual-Thing and this would be 
  equally correct logically (and possibly identical in meaning?);

* the only benefit of using Individual-Thing instead of Individual is
  one of conceptual tidiness; i.e. it hardly matters.


Two questions: 

1. Is this analysis correct?

2. Is there any other reason why one should prefer to 
   create instances of either Individual or Individual-Thing?

Mike
----


Mike Uschold,                 	AI Applications Institute,                    
INTERNET: M.Uschold@ed.ac.uk    The University of Edinburgh,
Tel: (031) 650 2732             80 South Bridge, Edinburgh EH1 1HN 
Fax:       650-6513		Scotland          
				WWW: http://www.aiai.ed.ac.uk/