Re: RDF-Based Semantics ready for WG-internal Review

Michael Schneider wrote:
> Dear all, Dear Reviewers! 
> 
> I'm happy to announce that the OWL 2 RDF-Based Semantics is now ready for WG-internal review.
> 
>  <http://www.w3.org/2007/OWL/wiki/index.php?title=RDF-Based_Semantics&oldid=20977>
> 

Huge step forward! Big plus to you...:-)

> However, there are two issues that SHOULD be resolved before LC publication. I regard these issues to go beyond my job as an editor. There are also EdNotes on these issues placed at the beginning of the document, that will only be there within the internal review phase.
> 
> 1) Naming-Issue:
> While the semantics is now consistently referred to as the "OWL 2 RDF-Based Semantics" in the whole document, there are still a lot of occurrences of the word "OWL (1/2) Full" in the document, typically as a prefix of some term. I did some thinking on each of these occurrences, but whichever decision I would take, the outcome would look rather messy to me. For the moment, I have called the old semantics "OWL RDF-Compatible Semantics", as this better mirrors the name "OWL 2 RDF-Based Semantics" and since this was the name of the chapter of S&AS. But I am not a big fan of this renaming, in particular since this chapter actually defines *two* allegedly RDF compatible semantics! Eventually, I decided to stop with my reflection on the names and hope for feedback by the WG on this issue, primarily by my reviewers. But, after all, I am unhappy with the current naming situation for the RDF-Based Semantics, and would prefer to simply call the semantics and the document "OWL 2 Full 
Semantics". This would at least lead to a consistent naming scheme in the document.
> 

I have not reviewed the document yet, but I had a quick look at some
occurrence of the word 'Full'.

- there are many places where you refer to OWL 2 Full ontology in the
syntactic sense. I think that is fine and is in line with the overall
naming approach

- there are some places where you refer to 'OWL 2 Full
Interpretation/Satisfaction/Entailement'). Strictly speaking, this may
not be 100% o.k., but I see the same issue arising in the Direct
Semantics which says, for example, that 'An interpretation I satisfies
an OWL 2 ontology...'. But I do not think it is really a source of major
misunderstandings. Maybe, in both cases, the introduction section should
make a clear statement saying something like

[[[
in this document, the word semantics, interpretation, semantics, etc,
strictly refer to the RDF compatible semantics.  OWL 2 Ontologies can
have other semantics (ref to direct semantics) but this document does
not deal with that alternative except when explicitly mentioned.
]]]

(the same in the direct semantics with the obvious changes)

I let a native anglo-saxon give a formulation.


> 2) Nary Datatypes:
> Recently, I had some discussion with Ivan on the semantics for the n-ary stub. The problem is that simply following the way as it is defined in the Direct Semantics is technically not possible for the RDF-Based Semantics, and it would be a non-trivial task to extend the basic framework of the RDF Semantics to support this (though possible in principle). My first question here is: What is the status of the n-ary stub in the Direct Semantics concerning normativity? Is it necessary for an implementer of a conformant "OWL 2 DL entailment checker" to support, for example, n-ary property restrictions, or n-ary datatype complements? Dependent on the answer to this question, I will try to come up with a solution (maybe with a little help from my friends :)).
> 

I think there are two issues here.

- We have n-ary as optional hooks. I believe the conformance document
should say that conformant OWL 2 are _not_ required to implement those
semantics conditions and even the syntax parsing and mapping

- However, _if_ an OWL 2 tool chooses to implement n-ary then there
should be some standard way of doing that. Ie, I believe both semantics
document must have the appropriate semantic constructs in them to make
interoperability possible. Michael and I had some discussion on how to
do that with n-ary restrictions and it is not hugely complicated, after
all, just one or two more terms to take care of spaces of vectors and
the corresponding extension functions, distinct from what is used for
other terms. It requires some work but nothing compared to what Michael
has already done...

Ivan



> Best Regards,
> 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
> =======================================================================
> 
> 

-- 

Ivan Herman, W3C Semantic Web Activity Lead
Home: http://www.w3.org/People/Ivan/
mobile: +31-641044153
PGP Key: http://www.ivan-herman.net/pgpkey.html
FOAF: http://www.ivan-herman.net/foaf.rdf