Meeting Notes

Dan C joined about 1:15 PST
	
Perhaps put off user defined types until later

Mike: Using full name or #shortname?  Get this on the table encourage
  XML schema group to allow the abreviation and if they dont then
  define a transformation

Dan: please scream at the W3C XML schema group for using the short name.

Mike: will draft an email on behalf of the comittee about wanting the
shorter names.

If they don't, we can do DAML Equivalent to statements to make the
   shorter ones work.  Not something we want to do.


Talking about the Peter and Ian revised proposal.

Mike sent an email that said that any language that doesn't allow you
     to say size 14 is going to be a hard sell.  But our language does
     allow you to.  You can use size (integer), or size decimal
     (decimal), or size numeral (string).

Mike: Peter enumerated three things ... first two are acceptable and
       third was a bit not easily acceptable.

       1. consensus 
       2. mostly consensus ... but how do you do this?
       3. 
       
...

Dan C says the axioms are in http://www.w3.org/2001/01/ct24

Mike: If someone designs an ontology and uses something as a float
   ... and all users have to tag things as float ... that is a lot of
   overhead for the user.  Can someone describe why we need this so
   everyone will say "aha, that's why we need this"

Peter(?): think daml+oil should make lots of sense without the schema
	  as xml is. This is a big bonus for xml!

If the classes are disjoint, then you can't have something that is an
   instance of both.  This means I can't make my short to a long, for
   example (?)

Dan: worried that you will parse something, and never know that you
did it wrong.

There are numbers and numerals, thats just how it is, better to model
this.

Dan's proposal solved Mike's problem but causes problems for Ian?  

URL From Ian:
http://kogs-www.informatik.uni-hamburg.de/~mwessel/report4.ps.gz

Two issues:
    What something can do
    Expressing the result of that doing

Jim:  This has become too asthetic.  

Ian:  If they dont say int everywhere, it is assumed it is a string.

Idea: Publish this, say it sucks, we know, if you have a better idea,
let us know.

Mike:  Want use cases

Apparent consensus to put something out there, that may not be perfect, 
	 

Dan: Propose that Ian and Frank reve the spec, integrating the "one by
     Frank and Ian" that includes examples about constrainting integer
     (etc.) and by the way this sucks ....

Frank: What still needs to be integrated?

Need update to reference docuemnt.
Need update to walkthrough.
Need axiomatic semantics. Dan C volunteered to work with Deb ...

Release notes??

Get this out fast so that we can have some impcat on the rdf interest
group meeting next week.

Feb release might not have axiomatic semantics ... yeah we can handle
not having it for an unspecified period of time.