W3C home > Mailing lists > Public > public-owl-wg@w3.org > May 2009

RE: JC5 and possibly replacing owl:subject, owl:predicate, and owl:object

From: Michael Schneider <schneid@fzi.de>
Date: Wed, 13 May 2009 22:32:58 +0200
Message-ID: <0EF30CAA69519C4CB91D01481AEA06A001393AFA@judith.fzi.de>
To: "Peter F. Patel-Schneider" <pfps@research.bell-labs.com>
Cc: <public-owl-wg@w3.org>, <bparsia@cs.man.ac.uk>
>-----Original Message-----
>From: public-owl-wg-request@w3.org [mailto:public-owl-wg-request@w3.org]
>On Behalf Of Peter F. Patel-Schneider
>Sent: Wednesday, May 13, 2009 8:11 PM
>To: bparsia@cs.man.ac.uk
>Cc: public-owl-wg@w3.org
>Subject: JC5 and possibly replacing owl:subject, owl:predicate, and
>owl:object
>
>One proposal that is being floated is to use some other name for
>owl:subject, owl:predicate, and owl:object.

Yes, and one rational was that this is for backing the argument that we do
*not* have our own OWL'ish reification vocabulary (we basically do not
provide a generic means for making statements about statements, i.e. meta
statements), but only provide a means to solve the technical problem that
there are cases where a multi-triple-style annotation of an axiom (or
another annotation) is inevitable. 

I notice that in OWL 2 DL, there are clear limitations on when and how the
annotation vocabulary can be applied, and Peter has listed these cases in
his mail. And even in OWL 2 Full, where no one can be stopped from using the
annotation vocabulary for whatever they want, I would consider it abuse to
apply this vocabulary for anything else then for annotating. In contrast, at
least in OWL 1 DL, where the use of RDF reification was still permitted, no
limitations on the use did exist, AFAIK. And in OWL 2 Full, as in RDF, the
use of reification for all sort of talking about statements is also
considered ok, AFAIK (apart from the fact that many people would like to see
RDF reification being banned).
 
But currently, some people who see "owl:(subject|predicate|object)" seem to
believe that this is just the RDF reification vocabulary put into a
different namespace. At least this sort of confusion can be avoided by
renaming the annotation vocabulary. By having names that clearly show the
intended use of the vocabulary AND are significantly different from the RDF
vocabulary terms people will hopefully in general refrain from using the
annotation vocabulary for other stuff than creating annotations.
 
Now, looking at your proposals

>We could, perhaps use something like
>  owl:annotatedSubject
>  owl:annotatedPredicate
>  owl:annotatedObject
>but this seems to be rather prone to misinterpretation.
>
>We could also use some "technical names"
>  owl:aSubject
>  owl:aPredicate
>  owl:aObject
>in honour of "annotation" and "axiom".

I would rather prefer

    owl:annotationSource
    owl:annotationProperty
    owl:annotationTarget

since by this

* the names clearly show the intend that this vocabulary is about
annotation;

* there is no mentioning of the "dangerous" terms "subject", "predicate",
"object" anymore;

* the naming is similar for all of OWL 2's "triple-description"
vocabularies, because the terms "source", "property" and "target" are used
in the Negative-Property-Assertion vocabulary as well;

* if Jeremy, or someone else, then sais that this renaming doesn't change
anything technically (which is true), we can again, as in LC1, point to the
fact that we *HAVE TO* provide *SOME* form of multi-triple encoding in order
to be able to annotate *EVERY* axiom. So the best we can do is choosing
names that clearly show the intend and avoid confusion with other
vocabularies. 

I think that this sort of action, removing sources of confusion and better
clarifying the intend of the feature, is something that is both justified
and sufficient for this LC comment. 

And it makes *me* happy, too, for those who care! ;-)

Cheers,
Michael

--
Dipl.-Inform. Michael Schneider
Research Scientist, Dept. Information Process Engineering (IPE)
Tel  : +49-721-9654-726
Fax  : +49-721-9654-727
Email: michael.schneider@fzi.de
WWW  : http://www.fzi.de/michael.schneider
=======================================================================
FZI Forschungszentrum Informatik an der Universität Karlsruhe
Haid-und-Neu-Str. 10-14, D-76131 Karlsruhe
Tel.: +49-721-9654-0, Fax: +49-721-9654-959
Stiftung des bürgerlichen Rechts, Az 14-0563.1, RP Karlsruhe
Vorstand: Prof. Dr.-Ing. Rüdiger Dillmann, Dipl. Wi.-Ing. Michael Flor,
Prof. Dr. Dr. h.c. Wolffried Stucky, Prof. Dr. Rudi Studer
Vorsitzender des Kuratoriums: Ministerialdirigent Günther Leßnerkraus
=======================================================================


Received on Wednesday, 13 May 2009 20:33:42 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:42:12 UTC