Re: Nested Swarms and Schedules

[Marcus G. Daniels]

>>>>> "B" == Benedikt Stefansson <address@hidden> writes:

B> ActionGroups are by
B> definition unordered sets, but the mental models typically would
B> involve an ordered set of events etc. 

ActionGroups are by definition ordered sets.  There was once a
protocol called DefaultOrder that was adopted by ActionGroup for
indicating the desire for Sequential/Concurrent/Randomized orderings,
but I took it since it didn't do anything.  In the next release it
will be back in order to support Randomized ordering.

B> But the natural question then is: If schedules are really only used
B> for the "outer simulation loop" in practice, why bother with the
B> activity library in the first place? Wasn't the activity library
B> supposed to provide a more natural language to express simulation
B> time, virtual time, multiple clocks etc? 
B> answer to this question.

The main thing multiple schedules are for is compartmentalizing all
the agents' respective work plans within the agents themselves.  The
main point of Swarm is that you can overlap the local experiences of
many agents and agent subcomponents into a uniform temporal space.  This
is useful because it means you only have think about the logic of a
particular agent type at time.  Further, you can do this without
worrying about the scalability of your model because Swarm
consolidates schedules into absolute timeline and thus concurrency is
evident to the simulator.

It would be useful to have different time grains for different kinds
of agents (it's in the queue), but unless the ratios are constant, it
will make reasoning about the composability of the model very
difficult.

                  ==================================
   Swarm-Support is for discussion of the technical details of the day
   to day usage of Swarm.  For list administration needs (esp.
   [un]subscribing), please send a message to <address@hidden>
   with "help" in the body of the message.