Newsgroups: comp.ai.alife
Path: cantaloupe.srv.cs.cmu.edu!bb3.andrew.cmu.edu!newsfeed.pitt.edu!scramble.lm.com!news.math.psu.edu!news.cse.psu.edu!uwm.edu!newsfeed.internetmci.com!in2.uu.net!omega.metrics.com!news.maplesoft.on.ca!news
From: Terry Van Belle <vanbelle@maplesoft.on.ca>
Subject: Re: Reproduction
Content-Type: text/plain; charset=us-ascii
Message-ID: <31878F7A.770@maplesoft.on.ca>
Sender: news@maplesoft.on.ca (News File Owner)
Nntp-Posting-Host: garnet
Content-Transfer-Encoding: 7bit
Organization: Waterloo Maple Inc.
References: <4lit47$ekt@news.magi.com> <4lln6h$t6a@soleil.uvsq.fr> <Pine.SOL.3.91.960425014622.21856B-100000@exp2.wam.umd.edu> <9604291731.AA15788@pluton.robot.uvsq.fr.cleopatre> <Pine.SOL.3.91.960430004308.2888B-100000@exp2.wam.umd.edu>
Mime-Version: 1.0
Date: Wed, 1 May 1996 16:21:14 GMT
X-Mailer: Mozilla 2.01Gold (Win95; U)
Lines: 59

Keith Wiley wrote:
> 
> > >> >After reading all the documentation I could, I still have this basic
> > >> >but fundamental question: is the reproduction of artificial organisms
> > >> >a natural tendancy or is it an "order" given by the programmer of its
> > >> >genetic algorithms?
> > >>
> > >> To answer to Guillaume` s question, I would say that if artificial creatures
> > >> reproduce it is because someone told it to do.
> > >> For now, artificial creatures are still not living things and cannot reproduce
> > >> or do whatever you want without human will behind.
> > >> For genetic Algorithms, the problem is quite simple, you will find in all
> > >> genetic programs a function which says "reproduce" or equivalent.
> > >
> > >You are obviously not aquainted with Tierra, which is only the most
> > >obvious example of truly reproducing computer code.  The tendency of
> > >tierra organisms to replicate within a computer's memory is an inate in
> > >the physics of the situation as of a human reproducing.  There is no
> > >"reproduce" function.  Tierran organisms don't suddenly enter the "copy"
> > >phase and run the sub-routine for copying code.  It is inherant in their
> > >genetic structure that they will, indeed, copy their code throughout a
> > >computer's memory.  So, basically, you are totally wrong.
> >
> > Well, I think you missunderstood my words when I wrote"..someone told it todo."
> > I meant that the reproduction phase has to be written somewhere in some place
> > to do its work.Even if enclosed in the code of the entity, it is written first
> > then the first entity can reproduce.If no code is written to reproduce,
> > I see no reasons for the code to live more than one generation.
> 
> No.  It's not written anywhere, unless you're refering to the ROM which
> understands how to execute a mov command or a jmp command, and that's
> just analagou to the physics of our universe.  Even though various laws
> of chemistry rule or reproduction, our code for reproducing is still a
> truly reproducing algorithm within the confines of the physics, and the
> same is true for Tierra..

You guys are talking around each other.  There are two separate issues
here.  The first is one of representation:  Is the process of reproduction
represented in the "physics" of the world?  The answer for Tierra is
yes.  The reproduction abilities of each "organism" are coded in
the lower level building blocks of the Tierra universe, and are subject
to change and improvement.  There is no "reproduce" function in the
(C?  C++?  Basic?) code that Tom Ray wrote.  Tierra itself is a simulation
of a whole lot of little "reproduce" functions.  For that matter,
every living organism on earth is a reproduce function.

The second issue is one of origins:  Does the first reproducing organism
arise spontaneously, or is it created?  In the case of Tierra, the first
organism, the one that starts off the whole process, is hand-coded
by the programmer.  Thus, Tierra isn't a completely satisfactory model
of abiogenesis, although it shows evolution nicely.

This is not to say that we couldn't extend Tierra to satisfy the second
constraint as well.  If you were patient, you could randomly change
memory locations and run random parts of the universe until you came
across a self-reproducer, after which the normal process of evolution
would take over.

Terry
