[Modeling] Modeling an Agent Class- register your opinion

Dr. Hong Zhu hzhu@brookes.ac.uk
Fri, 20 Jun 2003 08:55:15 +0100


There is an issue of different levels of abstractions, here. I would
consider class (or agent class) as a language facility that supports the
implementation of a number of more abstract concepts in modelling and
problem analysis. Role is an abstract concept that agent class supports. I
have discussed this in a paper 'Representation of roles in caste'. There is
a close match between the concept of role and the language features of caste
(agent class).

-Hong

----- Original Message ----- 
From: "James Odell" <email@jamesodell.com>
To: "ModelingTC" <modeling@fipa.org>
Sent: Thursday, June 19, 2003 10:46 PM
Subject: Re: [Modeling] Modeling an Agent Class- register your opinion


> On 6/19/03 5:19 PM, "sehl mellouli" indited:
>
> > I see a role as a set of tasks grouped according to a certain semantic.
> > Semantic means that the different tasks forming a role fits together. So
> > I think roles are at a higher abstract level than classes. Can they be
> > represented by classes in the development? I think in that case, we need
> > to  find a definition of a task, and relate it with the definition of a
> > function. Sorry, but I have no formal definition of a task.
>
> If I may play Devil's advocate: Isn't a class "a set of tasks grouped
> according to a certain semantic"?   And if I may play Jim: What is a
> "certain semantic"?  The definition looks a bit circular to me: a role is
> based on a certain semantic, where the semantic is what forms a role.
>
> -Jim
>
> _______________________________________________
> Modeling mailing list
> Modeling@www.fipa.org
> http://fipa.org/mailman/listinfo/modeling
>