Re: [Swarm-Modelling] Something New.. ?

[Simon Bland]

I guess I'm not too good at explaining it yet, I've only had a few weeks 
looking at the topic so far.

The original proposal was this:
Create a simulation environment that didn't revolove around providing a 
method for creating modules specifically for it, but instead around 
providing a method to use existing "agents" to provide parts of the 
environment.

Consider a situation where you have legacy agents you want to use as 
part of a more complex model, but don't have access to the source, or 
the task of converting the agent, not only to a new language but to 
match the API of a given framework, is prohibitive for whatever reason.

Providing a framework that could use the existing agent(s) without the 
need to modify or redevelop it would be nice.

After looking around a little for existing simulation frameworks, none 
of the systems I found had anything to handle the above situation, but 
the idea of extending an existing framework to take advantage of the 
ability to develop modules using it's framework, and add to it the 
functionality of using legacy agents, was raised as a better approach to 
the problem.

Swarm was presented as a well developed framework with a lot of active 
support and development, so I thought I should put it to the Swarm 
community to see if anyone had worked on or seen any similar work.

I'm aware that any form of translation layer between a legacy agent and 
the rest of a system would take a big performance hit, but if the 
simulation isn't time critical that may often be a better option than 
totally reworking the legacy agent (at least in the short term).

So as I was originally asking, feedback on this idea would be greatly 
appreciated, hilighting important things to consider before I go chasing 
down some blind alleys.

Regards,
Simon Bland.

Bill Northcott wrote:

> I saw this on the other list but did not really understand it.
>
> What is your motivation?
>
> The communication between agents and the rest of the simulation code is 
> pretty intense and my first thought is that using *NIX i/o instead of 
> Objective-C messages would be horribly slow.
>
> Bill Northcott
>
> _______________________________________________
> Modelling mailing list
> address@hidden
> http://www.swarm.org/mailman/listinfo/modelling



--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.308 / Virus Database: 266.7.1 - Release Date: 9/03/2005