Path: utzoo!attcan!uunet!timbuk!cs.umn.edu!uc!tut.cis.ohio-state.edu!ucsd!pacbell.com!pacbell!m5!tma
From: tma@m5.COM (Tim Atkins)
Newsgroups: comp.object
Subject: Re: How would you model this situation?
Message-ID: <3937@m5.COM>
Date: 28 Oct 90 06:19:04 GMT
References: <16734@rouge.usl.edu>
Reply-To: tma@m5.UUCP (Tim Atkins)
Organization: Object Sciences Corp., Menlo Park, CA
Lines: 56

In article <16734@rouge.usl.edu> pcb@cacs.usl.edu (Peter C. Bahrs) writes:
>The Domain: Object-oriented with concurrency if needed, objects have
>     local state, behaviors
>
>The Problem:
>  Model an application where two things(people) ara each applying a 
>force(pulling) on another thing (ends of a rope).  The acted upon
>thing (the rope) breaks.
>
>So suppose I tell person1 to pull X amount.  Should I model all of
>the interactions at an application level and then send the rope and 
>person2 reactionary messages.  Or should person1, who knows about a
>rope (globally) tell the rope to adjust it self, which in turn must
>tell person2 to adjust hisself.  So if the rope breaks, how are the
>references in person1 and person2 updated?  Does the existing rope
>references become forwarders to new part-rope objects?  In summary,
>are interactions managed external to objects or should this be
>migrated inside the objects.
>
>Please post responses to the net.

Well now, I would model such a thing a bit differently.  

Each person model might contain:
	what force is being applied to (reference to rope);
	what force is being applied (with direction);
	what force is being experienced (counter pull);
	perhaps acceleration detector stuff and correction if over some limit
	or some fixed distance from a known point is traversed.

The rope model would contain:
	end data on agents attached and length of rope starting at end;
	total length of rope;
	force applied to each end;
	some model of rope stress limits.



So, each agent would be instructed:
	grab rope 
		set first model variable and inform rope of attachement;
	apply force OVER time
		set force variables for persons which get transmitted to rope;
		rope calculates stresses and resultant forces and transmits
		each end force to opposite end holder as long as it is intact;
	on separation 
		rope computes break point and sets internal state;
		transmits change of force messages to end holders;
		end holders respond with acceleration and therefore velocity
		changes which they can respond to;
	each person or the rope itself could be queried directly or indirectly
	for how big their piece of rope is, there current acceleration, etc.


Does this model seem reasonable?

- Tim