Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

RE: multiple inheritance




Dear Dietrich,
 
Clearly "country" is not a primitive concept, though it is an important one.
It is critical to distinguish between importance and primitiveness. We have
found a lot of difficulty in developing models where people wanted to see
the things that were important to them, rather than what was
basic/primitive.
 
The point we got to is that we distinguished between basic and derived
concepts. A derived concept is one that can wholely be expressed in terms of
more primitive concepts, e.g. red apple, or country (by your definition).
 
A lot of important concepts are not primitive. It can be useful to represent
them explicitly in the model, because it gives them a formal definition. It
saves reinvention.
 
This may sound as if it is against a pure cannonical form of basic concepts,
but in my experience the main problem with really pure concepts is that they
have no immediate applicability, it is actually (some of) the combinations
that are useful. The utility of the base concepts is simply to know that you
have arrived at the end of your analysis (hopefully).

Regards 
      Matthew
============================================
Matthew West
Operations & Asset Management
Shell Services International
H3229, Shell Centre, London, SE1 7NA, UK.
Tel: +44 207 934 4490 Fax: 7929
Exchange e-mail: Matthew.R.West@is.shell.com
Compuserve e-mail: Matthew_West@compuserve.com
http://www.shellservices.com/ <http://www.shellservices.com/> 
============================================ 

 

-----Original Message-----
From: Dietrich Fischer [mailto:fischer@DARMSTADT.GMD.DE]
Sent: 16 June 2000 19:54
To: standard-upper-ontology@ieee.org
Subject: RE: multiple inheritance


Christopher Spottiswood wrote: 
While I myself find it difficult to imagine a congenial conceptual world
without 
multiple inheritance, .... 

Thompson, John A wrote: 
...Being able to partition a class into subclasses along more than one
dimension 
is basic to every large ontology that we're trying to merge (I'm guessing). 
For example, we may want to partition the class Person into Male-Person 
and Female-Person, and also into Adult-Person and Child-Person. 
Then to create the class Girl we'll want to make it a subclass of both 
Female-Person and Child-Person.  Without this ability, we'd have to drop 
one of these partitions (say, Adult-Person and Child-Person), and stay with 
a strict hierarchy where Male-Person has subclasses Man and Boy and 
Female-Person has subclasses Woman and Girl.  But then any properties 
of adults have to be repeated for both Man and Woman, and any properties 
of children have to be repeated for both Boy and Girl. 
If you repeated these kinds of duplications throughout a global ontology, 
it would become a nightmare and a straight-jacket. 
But we've all known this for years, so why are we even thinking of doing 
without multiple inheritance?? 


---- 
May I abstract from John A Thomson's example the following 
formal argument, which tries to minimize resort to pragmatics?: 


If the ontology contains a concept A and a concept B, 
should it not also be able to contain the defined concept C := (A AND B) 
as a defined concept? 
Then C has both A and B as direct superconcepts,  if 
neither (A subconceptOf B) nor (B subconceptOf A), - 
a constellation which (as far as I see) is called a case of 
"multiple inheritance" in this discussion. 


If it makes sense to assert that an undefined concept D is a subconcept of A

and is also a subconcept of B, then these two assertions can be replaced by 
(D subconceptOf (A AND B)), i.e. a primitive (undefined or partially
defined) concept 
in principle needs only one direct superconcept, however, that may be a
defined concept 
which needs "multiple inheritance". 


The only case of "multiple inheritance" which remains in the final example 
taxonomy of the Guarino/Welty paper (referred to by N. Guarino) is "Red
apple", 
which seems to be equivalent to ('Red Thing' AND 'Apple'). 
Not yet having followed their strain of thoughts, I still wonder whether a
methodology 
is attractive which advises against the assertion (for the "standard
context"), 
that every country (as a primitive concept) is 
(possibly among others) both a geographical region and a social entity. 

Dietrich H. Fischer