gtkpool/gperiodic

[jfortier@gani.aging.umanitoba.ca]

I'm the author of gtkpool, about which Roger was enthusing a while ago.  I'm
sorry it took me so long to get on to this, but I've been a little busu
recently.  I recently got enough time to sit down and think about Roger's
suggestion, and I'm not sure if what he is suggesting will work. 
Essentially, he wants me to hook up gtkpool with GPeriodic, so that one can
use gtkpool to simulate the behaviour of a gas, which would be selected from
GPeriodic.
	My main problem with this is that I am not sure how interesting it would
be possible to make this without a lot of work.  There is also the problem
with gtkpool's performance -- on my computer at least, once most of the
screen is being updated at once, the animation gets incredibly sluggish.  As
well, once too many "molecules" were introduced, I suspect that the physics
engine would bog down a lot.
	Getting back to the first issue, the way I see it, my program would just
be showing a bunch of molecules bouncing around, going at different speeds
depending on their temperatures.  This would look the same no matter what gas
were used.  Things could be made more interesting if more than one gas were
introduced, but then people would wonder why they weren't reacting.  Other
things like gravity and histograms of average, maximum and minimum
temperature might also make things interesting, but I suspect would get a bit
repetitive.
	I guess what I'm really working my way towards is a simulated chemistry
set.  I think this would be really neat, but also really complex and time
consuming.  However, I don't know enough about chemistry to know how much can
be realistically (or quasi-realistically) simulated how easily.  I hope that
someone who is a bit more knowledgeable could shine some light on this.

			Jacques Fortier