Re: [ontology-summit] System Components

[Patrick Cassidy <ontopacas@xxxxxxxxx>]

Re: Doug Foxvog's example:

>> As for the appeal to a natural language
formulation "The president was
>> shot" - is that not broadly similar to
saying "Paul is angry."

 

[DF] >It depends on the meaning of the phrase.  If i say "Hamlet is angry"
> and mean that the character is angry at the current point of the
> play, it is different.  If i mean that Laurence Olivier while playing
> Hamlet is angry, whether or not the character is not at this point
> it is similar.

  This illustrates the point that there are different types of 'roles' that can have different characteristics.  In COSMO, for HumanRoles, such as 'Student', 'PresidentOfTheUnitedStates' or 'BishopOfTrenton', to say that the individual filling that role did something or has some attribute logically implies that the role - in that time interval - also did that or had that attribute.  Likewise for 'Part' roles.  But a 'role' in a drama is a different type of thing.  In COSMO it is a 'PretendingRole', which is a subtype of the generic Role.  In those cases the actor playing that role can have quite different properties from the entity which is the role.  So there needs to be at least two types of 'role filler' relation, depending on whether one can or cannot equate the role filler to the role in a time interval.  In COSMO 'fillsTheRoleOf' can be used for Parts or HumanRoles, but 'playsTheRoleOf' means that the role-filler is actually pretending to be the 'Role', which leads to different implications.   These two types of roles are signaled by different linguistic usage: we usually say 'Joe Smith is the Bishop' or 'This is Pump101' but seldom 'John Gielgud is Hamlet' (except metaphorically), rather 'John Gielgud played Hamlet'.

   In CYC 'role' is a type of Predicate, and in OpenCyc we don't have the axioms that specify the implications, so I don't know how this distinction is made in CYC.

   In DOLCE, using the 'OntoClean' methodology, an instance of Role can never be a PhysicalObject, so I am unclear whether that distinction is made there.  Anyone know?

 

Pat

Patrick Cassidy
MICRA Inc.
cassidy@xxxxxxxxx
908-561-3416

_________________________________________________________________
Msg Archives: http://ontolog.cim3.net/forum/ontology-summit/   
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontology-summit/  
Unsubscribe: mailto:ontology-summit-leave@xxxxxxxxxxxxxxxx
Community Files: http://ontolog.cim3.net/file/work/OntologySummit2012/
Community Wiki: http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2012  
Community Portal: http://ontolog.cim3.net/wiki/     (01)