Re: Robustness Check and "A growing body of ad-hoc analysis solutions"

[Marcus G. Daniels]

B> I would be happy and eager to enter into discussion about the
B> standardization part (i.e. "growing body of ad-hoc analysis
B> solutions").

I think Ferdinando Villa's IMA proposal would be a good foundation for
standardization work. 

http://kabir.cbl.umces.edu/~villa/IMA

Given a IMA description of an experiment (in XML) and IMA-compatible
tools, a modeler could feasibly use more tools.

In some cases, it would be appropriate to use a tool like Drone
(http://www.pscs.umich.edu/Software/Drone) or MPI
(http://iee.umces.edu/~villa/svp) to perform a experiment.  One reason
might be practical -- a thought-out interface to a lot of common
tasks.  Another reason might be because such a tool was intended to
paradigm-neutral and expressly implemented to manage other tools.

In other cases an experiment might be more tightly focused on an ABM
approach, and it would be logical to make all processing
Swarm-centric; loading the XML description of the experiment directly
into Swarm.

Roughly speaking, the goal of IMA is to figure out what objects,
methodologies, data types, and procedures, are common across the board
(e.g. the notion of a treatment) and what should be extensions
(e.g. visualizing distance geometry, ABM, a mammal).  Then, with some
of the conceptual work in hand, go ahead and invent a language for
describing this taxonomy in XML, implementing standard parsing
libraries for using it in real tools like Swarm.

Note that a ABM description type could have different tiers as well,
with fine-grained distributed (Swarm) discrete event scheduling as
one subtype.

(IMA was a presentation at SwarmFest '99.)


                  ==================================
   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.