W3C home > Mailing lists > Public > semantic-web@w3.org > December 2007

Re: Ontology for describing conformance criteria?

From: Yoshio Fukushige <fukushige.yoshio@jp.panasonic.com>
Date: Tue, 18 Dec 2007 13:15:50 +0900
To: Karl Dubost <karl@w3.org>
Cc: fukushige.yoshio@jp.panasonic.com, semantic-web@w3.org
Message-Id: <20071218120138.215E.FUKUSHIGE.YOSHIO@jp.panasonic.com>

 Hi Karl,

Thank you for the pointers.

On Tue, 18 Dec 2007 11:24:14 +0900
Karl Dubost <karl@w3.org> wrote:

> Do you have a practical example to understand a bit more?

Yes. 

#Please forgive me if I use terms in an inappropriate way
(That's what I want to avoid by introducing the vocabulary ;-)

---- Background from here ----
I've been building a vocabulary for describing information appliances
(in a Japanese government funded project),
and to keep it being developed by multiple contributors,
we have published a guideline for defining(adding) such vocabulary
and are now publishing another guideline for publishing such vocabulary
on the Web.

And in the new guideline, we are thinking about making a list of
conformance criteria, with a level of normativeness attached to each
criterion.
---- Background till here ----

So, a RDF description of a criterion would be like:

<earl:test rdf:about="#criterion_1">
 <rdf:label>Criterion for character-encoding</rdf:label>
 <dc:description>
  The ontology file MUST be encoded in UTF-8
 </dc:description>
 <ex:normative rdf:resource="&ex;Required"/>
</earl:test>

<earl:test rdf:about="#criterion_2">
 <rdf:label>Criterion for human readable documentation</rdf:label>
 <dc:description>
  The ontology file MAY have accompanying human readable document
 </dc:description>
 <ex:normative rdf:resource="&ex;Optional"/>
</earl:test>

#We've not decided yet if we use the earl vocabulary or define our own.

My intension is to say that the criterion_1 is a mandatory criterion
for a conformant ontology to pass, and the criterion_2 is an optional
criterion.

And the property ex:normative and the classes ex:Must, ex:Optional
are the ones I'm searching better (existing) alternatives.

The level of normativeness should include at least three levels:
Required, Recommended, Optional (representing the concepts described in RFC2119)

Best,
Yoshio
fukushige.yoshio@jp.panasonic.com

> 
> > By "vocabulary for describing conformance criteria", I mean
> > Classes and properties for asserting some resource is a conformance
> > criterion, etc, among which I badly need Classes to assert whether
> > a criterion is normative or optional.
> 
> Not completely related
> * Test Metadata
>    http://www.w3.org/TR/test-metadata/
> * Test Metadata in RDF
>    http://lists.w3.org/Archives/Public/www-qa/2006Mar/0023
> * Test Case Metadata
>    http://esw.w3.org/topic/TestCaseMetadata
> * Test Case Description Language 2.0
>    http://bentoweb.org/refs/TCDL2.0.html
> 
> 
> --
> Karl Dubost - W3C
> http://www.w3.org/QA/
> Be Strict To Be Cool
> 
> 
> 
> 

-- 
Yoshio Fukushige <fukushige.yoshio@jp.panasonic.com>
Network Development Center,
Matsushita Electric Industrial Co., Ltd.
Received on Tuesday, 18 December 2007 04:16:23 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 21:45:19 GMT