Re: TCDL Review

Le ven 31/10/2003 à 20:46, Dimitris Dimitriadis a écrit :
> Here follows my review of the TCDL WG note 
> (http://www.w3.org/QA/WG/2003/10/tcdl-20031013.html).

Some random remarks on my own, inspired also from Dimitris' review:
- we should make sure we clearly distinguish the definition of the data
model from the actual markup (and schema technology) used to anchor it
- I agree with Dimitris that the test case should be the atom, and the
versioning question (in a broad sense, including levels hierarchy,
changes from errata, etc.) should be done at another level; e.g.:
<testCase id="foo">
 <description>Testing that foo is bar</description>
</testCase>
<conformanceObject id="foo-1-parser">
 <name>FooML Level 1 Parser</name>
 <testCasesList>
   <li href="#foo"/>
 </testCasesList>
</conformanceObject>
- many aspects of it make me think that using RDF to model this would
make it easier for many things; but that doesn't need to be a
requirement

Dom
-- 
Dominique Hazaël-Massieux - http://www.w3.org/People/Dom/
W3C/ERCIM
mailto:dom@w3.org