Subject: RE: State machine for "empty" agents
From: Rick Reed (Rick.Reed#etsi.fr)
Date: Fri Jun 01 2001 - 06:18:28 GMT
Dear All,
It seems that Reinhard, Thomas and myself are broadly in agreement.
WRT to the options suggested by Reinhard, I prefer the "two line
modification", rather than add a transformation. We need the corresponding
updates to Z.100 text, but I assume that these will be similar to my
suggestions and will go in the Semantics sections.
Rick Reed (at ETSI)
Tel: +33 4 92 94 42 68
mob: +44 79 70 50 96 50
email: rickreed#tseng.co.uk
-----Original Message-----
From: owner-meeting#sdl-forum.org [mailto:owner-meeting#sdl-forum.org]On
Behalf Of Reinhard Gotzhein
Sent: 01 June 2001 07:42
To: thomas.weigert#motorola.com
Cc: SDL Meeting
Subject: Re: State machine for "empty" agents
On 31 May, Thomas Weigert wrote:
> My intuition is that every agent should have behavior. If none is
explicitly
> defined in terms of a state machine, then the behavior is determined by
the
> contained agents (e.g., stops when all contained agents stop) or
immediately
> stops otherwise.
>
> This could be accomplished by having an implicit state machine that just
> transitions into the stop state from the initial state,
... after its substructure has been completely created! If this
stop-transition is executed "too early", i.e., if there is no
substructure yet, then the agent instance ceases to exist right
after its creation.
Reinhard
This archive was generated by hypermail 2a23 : Mon May 05 2008 - 20:30:55 GMT