LANG: Moving issues 5.6 and 5.14 forward

I think an important part of resolving issues 5.6 and 5.14 forward is to
decide whether or not things like (owl:Ontology, owl:imports,
owl:backCompatWith, etc.) should be in the domain of discourse (this is
one of the four major questions from [1]). At last week's telecon and on
this mailing list, I heard a number of people say that they shouldn't
even be in the RDF domain of discourse (which is what I originally
proposed), and I don't recall anyone disagreeing with this position.
Therefore, I am going to make a proposal that treats these features as
"magic syntax" that is outside of both the RDF and OWL domains of
discourse.

Background:
------------
Peter had proposed that imports simply be an attribute on the RDF tag
[2]. 

<rdf:RDF owl:imports="foo:bar">
    ...
</rdf:RDF>

One problem with this approach is that if there are multiple imports you
have to use space to delimit them. I think this is a bad idea. Not only
is it forcing multiple data values into a single "field" (generally
considered a bad idea in databases), but I think it will make it more
difficult to use standards like XPath and XSLT to point to portions of
and transform OWL documents.

Mike had proposed putting the imports information outside of the scope
of RDF [3], by doing something like:

<owl:include import="uri1" />

<rdf:rdf ...> 
 ...
</rdf:rdf>

In a separate message, he had suggested replacing the rdf:rdf tags with
owl:Ontology tags.

My proposal was influenced by each of these ideas.

The Proposal:
---------------
I propose that imports and versioning information be declared outside of
RDF. In particular I propose the we have an owl:Ontology tag that wraps
all OWL ontologies. The content of this tag is import and versioning
tags followed by the RDF content of the ontology. This has the nice side
effect, that the content of an ontology is finally nested within an
ontology element.

Here's an example:

<owl:Ontology xmlns="...">
   <!-- multiple imports are allowed in separate elements -->
   <owl:imports resource="..." />
   <owl:imports resource="..." />
   ...
   <owl:priorVersion version="..." />
   <owl:backCompatWith version="..." />

   <rdf:RDF>
        <!-- all of are class and property statements go here -->
        ...
   </rdf:RDF>
</owl:Ontology>

Thus, RDF parsers can just ignore the surrounding XML content and still
get "partial understanding" of it. OWL parsers can use this extra
information in whatever ways are deemed fit for owl.

Please let me know what you think of this proposal. If possible, I'd
like to find out how many like it, how many can live with it, and how
many can't live with it.

If we can get agreement on this, then I think we can start tackling the
issue of how do we define the meaning of imports.

Jeff

[1] http://lists.w3.org/Archives/Public/www-webont-wg/2002Sep/0272.html
[2] http://lists.w3.org/Archives/Public/www-webont-wg/2002Sep/0166.html
[3] http://lists.w3.org/Archives/Public/www-webont-wg/2002Sep/0300.html

Received on Tuesday, 24 September 2002 10:34:56 UTC