some questions

Hi,

I was reading over the task force page and some questions occurred to  
me, which might be relevant for discussion.


For the Model TF:

* How are D2 and D3 related?
** What do people expect to be specified in the "formal (OWL/RDF)  
model"?
** How would this differ from the optional "formal semantics"?
** Should the formal model be developed and implemented first, then  
described informally (my preference), or should the informal spec be  
written first and then formalized?  Or something in between?

For the Implementation/Test case TF:

* Can we start with contributed use cases/test cases based on existing  
languages/models and then migrate them to the standard?
** In particular I remember discussion at the end of the XG of  
features of PML that its developers don't believe are handled well in  
OPM, and vice versa.  Examples would help a lot for those that are not  
expert in either.

* What do we mean by an implementation?  [http://www.mementoweb.org/  
Memento] might provide a good model for an implementation strategy.  
Suggestions:
** A library that reads in PIl, checks its validity, and emits a  
normal form (possibly reading or writing different formats, e.g. for  
migrating existing formats to PIL or exporting PIL as XML or RDF).
** Web server extensions that respond to provenance requests?
** A Web browser plugin that uses PIL (e.g. implementing the "Oh,  
yeah" button)



I added them to the Talk page too:

http://www.w3.org/2011/prov/wiki/Talk:ProvenanceTaskForces

--James

-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.

Received on Thursday, 5 May 2011 15:12:00 UTC