Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

cell component ontology

Larry.Hunter at uchsc.edu Larry.Hunter at uchsc.edu
Wed Nov 2 09:36:58 PST 2005


Peter,

I know you weren't responding to me, but our emails were only moments
apart, so I will respond to you....

You are clearly right that an ontology designer has a range of choices
about which concepts will be modeled as instances and which as
classes, and that the correct choice depends on the the uses for which
an ontology is designed.  As you say in the excerpt below, you were
intended in make a particular point in your set of choices.  

However, when contributing an ontology to the public, it seems to me
that it is important to obey the principle of least commitment.  That
is, an ontology provided to the public with the intention that it be
used in many different ways should be constructed to make the fewest
commitments (in formal semantic terms) compatible with the authors
intent.   

In this particular case, that principle suggests modeling all concepts
as classes.  If a user desires to model certain classes in such an
ontology as instances, a trivial addition to the ontology (not
requiring any chances to the provided ontology itself) provides such
instances.  However, the inverse is not true.  If you provide the
ontology with certain concepts as instances, then it is not possible
for a user who wishes to model those concepts as classes to do so
without modifying the provided ontology.

I believe that principle of least commitment is the driving force
behind the fairly standard practice of providing public ontologies
defined entirely as classes (without instances). 

[And don't forget my second question from the previous email asking
CCO folks to providing information about the relationship between the
CCO and GO Cell Component...]

Finally, I want to make sure that the importance of your making the
CCO public, and all the related effort that has gone into the BioCyc
resources, is acknowledged.  You really have been a leader in this
area. 

Regards,

Larry


Peter Karp writes:

 > Yes, of course we could have chosen to make "nucleus" be a class
 > instead of an instance, however, for the purposes for which this
 > ontology was designed, we wished to emphasize in our model the
 > notion of cellular structures as the base-level objects of discourse.
 > You will note that our ontology is consistent in this sense.
 > 
 > My point, again, is that there is rarely if ever an objective
 > definition of what is "a particular object in a domain" --
 > it is virtually always a function of the modeler's perspective,
 > and the uses for which the ontology is designed.


--
This message is from the GOFriends moderated mailing list.  A list of public
announcements and discussion of the Gene Ontology (GO) project.
Problems with the list?           E-mail: owner-gofriends at geneontology.org
Subscribing   send   "subscribe"   to   gofriends-request at geneontology.org
Unsubscribing send   "unsubscribe"  to  gofriends-request at geneontology.org
Web:          http://www.geneontology.org/



More information about the go-friends mailing list