Ron,
> I was referring to Ugo's request that
we seek to prove that the WS approach to modelling subprocesses
> couldn't handle all use cases. I
regard this approach as biased, since it puts an unreasonable burden
> of proof on one side of the
discussion, but not the other.
Sorry, this has nothing to do with being biased.
The burden of proof is asymmetric because the problem with start with is
asymmetric. In other words, we are
not starting with two possible approaches both of which would have to be
developed from scratch. We are starting with one approach being already
available (except for possibly minor modifications), and a second one that would
require to be developed.
If you are saying that
the existing approach is not sufficient, and are asking us to do additional
work, I think it's very legitimate to request use cases that show how
that is so and why we should go through the extra work. (You are certainly free
not to give such use cases, but I am equally free not to vote in favor of your
proposal if I don't find it
convincing).
Ugo
