Re: "ObjectProperty" and "DatatypeProperty"

Thanks for your messages concerning the relationship between classes and
individuals.  I've taken the liberty of revising and rearranging your
messages to just select the relevant parts for the comment.

> Subject: "ObjectProperty" and "DatatypeProperty"
> From: "Kevin D. Keck" <kdkeck@lbl.gov>
> To: public-webont-comments@w3.org
> Date: Mon, 7 Jul 2003 13:13:50 -0700

> Lost in all this is an apparent ambiguity in the spec. Despite having no
> formal definition of the term "object", section 5.2 of the AS&S
> (http://www.w3.org/TR/owl-semantics/rdfs.html#5.2) asserts (in the second
> table, "Characteristics of OWL classes, datatypes, and properties") that
> "Class instances are all OWL objects."  However, section 5.4
> (http://www.w3.org/TR/owl-semantics/rdfs.html#5.4) insists that IOT and
>  IOC must be disjoint. Thus, clearly "OWL objects" must refer to the union
>  of IOT and IOC, right? Only, returning to 5.2, interpretations of
>  owl:ObjectProperty asserted to be contained in IOTxIOT. Well, which is
>  it? Are Classes Objects, i.e., is IOC a subset of IOT, or not?
> 
> And if not, then what type of Property are rdf:type, rdfs:range,
>  rdfs:domain, etc. in the OWL universe?

> Subject: Re: "ObjectProperty" and "DatatypeProperty"
> From: "Kevin D. Keck" <kdkeck@lbl.gov>
> To: Jim Hendler <hendler@cs.umd.edu>
> Cc: public-webont-comments@w3.org
> Date: Fri, 18 Jul 2003 14:09:24 -0700

> Sure, I'll move the naming discussion to the rdf-logic list, but that was
> only the first part of my comment.

> The second was that I, at least, was confused about whether (in OWL-DL)
> owl:Class'es are owl:Thing's. I thought the idea is that they're not, but
> then I came across the "Class instances are all OWL objects" statement
> (AS&S, Sec 5.2, see below). What does this mean? Does this mean
> owl:Class'es _are_ owl:Things? Or does it mean they're both in the range
> of owl:ObjectProperty (contrary to the range given)? And if not, then
> what subclass of Property includes rdf:type? Some class there's no name
> for?

I agree that the use of ``object'' here is misleading.  I've made several
editorial changes to the document, which you can see at
http://www-db.research.bell-labs.com/usr/pfps/owl/semantics/, to replace
several uses of ``object'' with ``individual''.  This results in the use of
object (by itself) for only the objects of triples.  Individual is then
used to refer to elements of the domain of discourse that can be instances
of OWL classes.  Instances of ObjectProperty are referred to as
individual-valued properties, which is longer but more precise, and
consistent with usage from Section 2.

Not changed is ObjectProperty itself, which is somewhat of a misnomer, but
is the result of considerable working group discussion.

Changes have been made to Section 2.3.1.2, 2.3.2.3, Appendix A.2.  The most
changes, however, have been made to Section 5.2.  In particular, 
``Class instances are all OWL objects.'' has been replaced with ``Instances
of OWL classes are OWL individuals.''   This should make it clear that it
is not OWL classes that are OWL instances in OWL DL, but that instances of
OWL classes are OWL individuals.  In OWL DL individuals (IOT) and classes
(IOC) are disjoint.

Please reply to public-webont-comments@w3.org indicating whether this 
resolution is satisfactory.  

Peter F. Patel-Schneider
Bell Labs Research
Lucent Technologies

Received on Tuesday, 22 July 2003 14:30:11 UTC