Newsgroups: comp.ai.alife
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news4.ner.bbnplanet.net!news3.near.net!paperboy.wellfleet.com!news-feed-1.peachnet.edu!gatech!howland.reston.ans.net!math.ohio-state.edu!caen!hearst.acc.Virginia.EDU!murdoch!viper.cs.Virginia.EDU!ccb8m
From: ccb8m@viper.cs.Virginia.EDU (Charles C. Bundy)
Subject: Re: Emergent behavior in lossless systems
X-Nntp-Posting-Host: viper-fo.cs.virginia.edu
Message-ID: <DBM9v7.ICI@murdoch.acc.Virginia.EDU>
Sender: usenet@murdoch.acc.Virginia.EDU
Organization: University of Virginia Computer Science Department
References: <DAuq7r.I55@oakhill.sps.mot.com> <Ted.Belding-0507952119090001@pm056-00.dialip.mich.net> <3tqtpc$4kg@laplace.ee.latrobe.edu.au>
Date: Wed, 12 Jul 1995 18:54:43 GMT
Lines: 47

In article <3tqtpc$4kg@laplace.ee.latrobe.edu.au> khorsell@ee.latrobe.edu.au (Kym Horsell) writes:
>In article <Ted.Belding-0507952119090001@pm056-00.dialip.mich.net>,
>Theodore C. Belding <Ted.Belding@umich.edu> wrote:
>>In article <DAuq7r.I55@oakhill.sps.mot.com>, nahas@particle.sps.mot.com
>>(Michael Nahas) wrote:
>>
>>> Has anyone done research on Emergent behavior in lossless system,
>>> specifically in cellular automata?  By a lossless system, I am talking
>>> about a system where the initial state can always be determined given
>>> the current state. 

"lossless" as defined above seems to refer to information theory, not
physical entropy that we deal with daily.  This I grasp, what I don't
grasp is the leap to "bit flipping using no energy."  Would someone
please explain how you maintain information taking into account entropy
with no energy expenditure?  I'm not even talking about "processing"
I just want to know how you fight decay from heat death at the storage
level?

>>
>>You might want to read the chapter on "ballistic computation" (chap. 18) in:
>>Toffoli, Tommaso, and Norman Margolus. 1987. Cellular automata machines.
>>MIT Press. pp 209-220. ISBN 0-262-20060-0 
>>
>>Their intent is to build a reversible computer, with the idea that this
>>might allow us to avoid the heat caused by erasing information, as well
>>as the power that is needed to supply fresh signals.
>
>This might be a slightly different thing I think. 

I think it is a chasmic (ouch the pun) difference :)

>But certainly it is possible to make a non-reversing process/machine
>from reversing components (as I guess you're alluding to).
>
>Reversible computation allows processing using no lower-bound on the 
>energy to "flip a bit". If the machine operates near absolute zero

Well kiddies the closer it operates at "absolute zero" the slower
your computation until viola at abs 0 you aren't computing anything
and hence not expending energy :)  QUIZ Why do they call it absolute?

>(and actually quite a bit above is OK too) you can compute for no 
>energy cost -- if it's reversible.
>
Charles Bundy
ccb8m@preferred.com
