[Zope-PTK] Member Publishing
Michael Bernstein
webmaven@lvcm.com
Thu, 14 Dec 2000 09:58:54 -0800
Renato De Giovanni wrote:
>
> But sub-workflows... Maybe I'm wrong, but I see them as a mere grouping of
> states/transitions from a big workflow according to some criteria - in other words,
> sub-workflows would happen "inside" generic "super-states" (Am I right here?? Someone
> please correct me if I'm wrong).
> I'd like to see a real example where sub-workflows are really needed. Otherwise we'll
> be introducing unnecessary complexity to this issue.
You're right in that any state machine with
sub-state-machines can be represented by a single
state-machine.
However this does not make it an unneccessary complexity.
The larger that your state-machine gets, the more difficult
it is to get that 'birds-eye view', and you need to
encapsulate groups of states to hide the underlying
complexity.
HTH,
Michael Bernstein.
P.S. I consider workflow to be an application of a generic
state-machine tool, but not the only such application. They
would be good for many other types of quick TTW logic
prototyping, and would be more accessible to programming
newbies.