Re: Classes as values

My 2 (euro) cents... ;-)

AR: "I think the above indicates that of all the problems we could have 
started for on best practice, 'Classes as Values' is the perhaps most 
difficult and problematic.

I agree but IMHO some of these are not directly relevant to "Classes as 
Values" and should really be out of the scope of the design pattern, 
possibly referenced in a "see also section". In particular vii) and ii)

AR: " ii)   'Born Free' a book about a particular lion"
This is a classical case of linking two individuals with no special 
problem as you underline latter in your message.

AR: "vii) That 'John' is credited as the author of the class Lion in a 
KB (As a statement about John). (...)  NB we don't have a distinction 
between 'class' and 'concept represented by the class' available to us. 
  The author of the class and the person who first described the species 
are clearly different."
My opinion is that this is a different problem: this statement is not 
about the class but about the statement declaring the class. Therefore I 
would rather expect something like:

<rdf:Statement rdf:about="#LionDeclaration">
  <rdf:subject rdf:resource="&example;#Lion"/>
  <rdf:predicate rdf:resource="&rdf;type"/>
  <rdf:object rdf:resource="&rdfs;Class"/>
  <dc:creator rdf:resource="&company_staff;#John"/>
</rdf:Statement>

AR: " v)  A reference to the 'Lion Genome' - (plausible for Dog, Cat, 
etc. and no doubt eventually for Lion)"

Here I think the problem is rather at the inference level: there is a 
book referencing an object "Genome" that belongs to the Specie "Lion", 
then we are missing an inference saying that when a document is about an 
object directly attached to a species then the document is about the 
species.

AR: "(...) the semantics of any negated queries (there is no explicit 
negation in RDF) will be closed world"
I am by no mean an expert here but my opinion is that the "not" operator 
should may be split in two: a "naf" operator implementing "negation as 
failure" i.e. the "there does not exist a statement saying that..." 
(with all the non monotonic problem that comes with it) and the full 
negation operator "not" saying that it can be proved that the statement 
is false for instance if "Lion" and "Whales" are disjoint classes then 
it can be proved that a given instance of lion cannot be a whale.

AR: "(...) A is immediately in OWL-Full, yet neither A nor B expect to 
perform any inferences using the information in C's ontology.  If the 
rest of A's ontology conforms to the restrictions on OWL-DL, it seems 
perverse not to apply OWL:DL semantics and reasoners."

I think I agree with you here but I am probably misunderstanding your point.
I personally don't quite understand the danger in referencing an OWL 
Class as long as you are not talking about the class (ex: adding to its 
description) etc. If the reference is only used to qualify an instance 
(ex: give the subject of a book) then in what respect is it more 
dangerous than the classical typing relation which also qualifies an 
instance with reference to a class?

Fabien.
-- 
"Stories are our dreams. And our dreams are our life."
                                  -- Tim Burton.
  ____________
|__ _ |_  http://www-sop.inria.fr/acacia/personnel/Fabien.Gandon/
|  (_||_) INRIA Sophia Antipolis - ph# (33)(0)4 92 38 77 88

Received on Thursday, 29 April 2004 13:22:14 UTC