managing multiple Swarm models...

[cgl]

In his last message to swarm-support, Mattie Hare suggests as one topic
for discussion at SwarmFest:

> whether and how to provide automatic output analysis tools for swarm 
> experiments. e.g. Tools for looking at multiple runs of a simulation 
> experiment and deciding whether or not the population data tend to be 
> cycling/falling/rising etc.

  In this context, everybody might want to take a look at the latest
version of the tutorial application that was just put up on the Swarm
web-site (tutorial-1.0.1), besides fixing a few bugs, I've added a 
new application (simpleExperBug) which demonstrates one way to build
an Experiment Swarm, which manages creating, running, analyzing, logging,
and etc. a sequence of model invocations. It makes use of a Parameter
Manager object that will handle sweeps through parameter space as
it initializes the various model incarnations....

  Let me know if this looks like a useful way to go about addressing
some of the issues Mattie raises - there are a number of ways that
such a capability could be implemented in Swarm (or outside of
Swarm, as the Drone package from Michigan does...) I think that
this issue should serve as the basis for one of the discussions
we have about Swarm, and modelling in general. So, check the
tutorial app out to see the basic idea for one approach to the problem..

Chris Langton