Re: issues with respect to the test cases overall manifest document

> FWIW - and my non-chairs opinion only - I know of a number of groups 
> working successfully from the current version, using automated tools 
> to handle the running of the tests (Euler, E-wallet, Pellet, 
> OwlLisaKb, Owlet - at the very least) - I would oppose any attempt to 
> change the format at this time when so many people have already 
> invested the time in using the LC version


I don't think this proposal is a substantitve change just an editorial 
clarification.

The current all-test manifest is produced with this algorithm

1: read owl.rdf
2: read the issue list document and transform some parts into RDF
3: read the rdf test schema and the OWL test ontology
4: read all the single test manifest files
5: combine all these as a single RDF graph and output it whichever way it 
comes

It's fine for the RDF literate, but:
- is very messy
- and includes lots of irrelevant stuff

The suggested approach would contain all the single manifest files (part 4), 
none of the other stuff (which is more stable), and be more readable as a 
plain text file, or an XML file, rather than just as an RDF graph.

Jeremy

Received on Wednesday, 23 July 2003 14:53:59 UTC