Burdett ML gap/fit analysis - first cut

I made another pass over Burdett ML recently, and have some
comments about requirements it doesn't currently seem to meet.
I am sure there are things I missed, so consider this only a
possible starting point in the gap/fit analysis.

parallel execution - may be implicitly supported already; what if two
processes have same start state/end state? But may need explicit
support.

decision points - currently, the only indication that a process is a
decision point is that it has >1 transition out of it. Should allow
specification of decision logic? This has been discussed previously -
could be a big issue.

Error vs. normal transitions - it is a requirement to support these.
Should a process have an implicit error transition to an error handler
(probably)? Raises issues of scope/nesting of error handling - cf. BPEL.

Composition - there is an extension mechanism and import - need other
kinds of composition support? Support for subprocesses is one missing
piece IMO.

IMO should simplify things so every state doesn't have to be named (I
have made a suggestion about this earlier - see
http://lists.w3.org/Archives/Public/public-ws-chor/2003Jun/0162.html).
But this is not strictly a requirement.

--Jon

Received on Tuesday, 29 July 2003 17:34:49 UTC