- From: Kohei Honda <kohei@dcs.qmul.ac.uk>
- Date: Tue, 22 Mar 2005 13:41:53 +0000
- To: public-ws-chor@w3.org
- CC: Gary Brown <gary@enigmatec.net>, Kohei Honda <kohei@dcs.qmul.ac.uk>
I am sorry I am catching up with this so late. Anyway I have started perusing all mails and found this. in conclusion I fully agree with Gary --- having activity there instead of workunit looks the best and simplest option: appropriate behaviours of finalizer, for which some restricted form MAY be practically enough (and be needed for verification) can come later: now let's set it to be as general as possible. I find introduction of finaliser, completion, isolation and related mechanisms enriching CDL substantially. As far as basic consistency is there, the generality should be preferred rather than restriction. kohei Gary Brown wrote: > Hi, > > Although I have asked Kohei for clarification on this, I actually think > the text in the issue is pretty clear: > http://www.w3.org/Bugs/Public/show_bug.cgi?id=1102. > > "Cannot see why finalizer blocks etc. should not combine in them > corresponding > workunits --- the latter seem rather redundant in semantics (and examples). > As far as I read, no repetition etc. seems possible in these blocks." > > He is saying why is only a workunit permitted inside a finalizerBlock? I > can see no valid reason for this restriction, so I propose that we > change this to be an 'activity'. > > The only changes I can see are in section 2.4.5: > > 1) Syntax to change from > > <finalizerBlock name="ncname" > > WorkUnit-Notation > </finalizerBlock> > > to > > <finalizerBlock name="ncname" > > Activity-Notation > </finalizerBlock> > > 2) Last paragraph in that section - suggested change would be to remove > the sentence that talks about the workunit. > > 3) Obviously the schema would need to be updated > > > The examples in 2.4.10 could be updated, but they are still valid using > workunits. > > > Regards > Gary
Received on Tuesday, 22 March 2005 13:41:56 UTC