[Modeling] comments on AUML diagrams

[James Odell]

Dear Mellouli,

If we all had the time to do both, I would agree.  But, most of us do not.
So, we decided to work on the language as a first priority.   According to
the workplan that was approved by FIPA, our charter is to "Focus on problems
and notations that are deemed necessary to support modeling of autonomous
agents systems."  This charter can be changed, but only if we have the
participation of several people.  Would you organize this?  If you have the
time and bandwidth to do this, you would be most welcome to help in this
area. 


Cheers,

Jim


On 4/14/03 7:11 PM, "sehl mellouli" indited:

> I think that at the same time that we define AUML diagrams, we need to
> think about the implementation issue (how to define a higher layer
> language to develop MAS easily). This language has to be defined based
> on MAS concepts: BDI, roles, agent, social relationships, cooperation,
> communication, etc...
> 
> For example, objects in OO implementation, cooperate by procedure or
> function call. But cooperation concept in MAS is more than a procedure
> or a function call. In agent cooperation, many elements intervene such
> as the agents beliefs or plans. So, it is important to porpose, for a
> designer, a way to implement cooperation (this must be proposed by AUML).
> 
> IMO, agent aspects have to be considered in designing and implementing MAS.