tierra.doc

地球模拟器

Whole Earth Review (Steven Levy, USA) Fall 1992: ``Artificial Life, The
Quest for a New Creation.''  Reprinted from a book of the same name published
by Pantheon Books.

Science News (John Travis, USA) August 10, 1991: ``Digital Darwinism:
Electronic Ecosystem.  Evolving `life' flourishes and surprises in a
novel electronic world''.

Nature (Laurence Hurst & Richard Dawkins, UK) May 21, 1992:
``Life in a test tube.''

Scientific American (John Rennie, USA) January 1992: ``Cybernetic Parasites...
Tierra... has been hailed as the most sophisticated artificial-life program
yet developed...''

New Scientist (Roger Lewin, UK) February 22, 1992: ``Life and death in a
digital world.  No one can turn back the evolutionary clock, but we can
follow the fate of a rich menagerie of artificial organisms as they evolve
in a model world.''

The Economist (Anon, UK) January 4, 1992: ``The meaning of `life'.
In order to understand the origin of life, scientists are switching from the
chemistry set to the computer.  In the process, they are beginning to
understand what it means to be alive.''

Release 1.0 (Esther Dyson, US) April 28, 1992: ``Artificial Worlds: A
Field Scientist in Tierra Cognita.''

Guardian (Jocelyn Paine, UK) January 9, 1992: ``Unravelling the loop in the
primordial soup.  Tierran machine code is so adaptable it survives.  Jocelyn
Paine charts the evolution of artificial life within the computer.''

Asahi (Katsura Hattori, Japan) September 15, 1992: Title in Japanese
characters.

Actuel (Ariel Kyrou, France) April 1992: ``Visite Guidee Aux Extremes De
La Science: La Vie Artificielle.  Etes-vous pr\^{e}ts \`{a} entrer dans
l'univers vertigineux de la vie artificielle?  Un champ scientifique tout neuf
sur lequel se penchent les grosses t\^{e}tes et les Nobel de labos
am\'{e}ricains.''

The Chronicle of Higher Education (David Wilson, USA) December 4, 1991:
``Approaching Artificial Life on a Computer.  Survival-of-the-fittest
electronic organisms dramatically illustrate Darwinian principles.''

Mikrobitti (Pekka Tolonen, Finland) November 1991: ``Olemmeko humanoiden
biologinen koe?  Tierra simuloi el\"{a}m\"{a}\"{a}.''

Europeo (Giovanni Caprara, Italy) September 1991: ``Anche il computer ha
fatto un figlio.  Un biologo americano ha creato un software capace di
elaborare programmi che si evolvono da soli.''

GenteMoney (Riccardo Orizio, Italy) November 1991: ``Cos\`{\i} ho dato
la vita al software.''

Computerworld (Michael Alexander, USA) September 30, 1991: ``Tierra adds to
evolutionary studies.  A computerized world created on an IBM PC could
have real-world benefits for scientists.''

Sueddeutsche Zeitung (Konrad Peters, Germany) October 21, 1991:
``Die Evolution im Computer.  `K\"{u}nstliches Leben' hilft Biologen und
Informatikern auf die Spr\"{u}nge.''

Super Interessante (Anon, Brazil) November 1991: ``A vida dentro do
computador.''

Technology Review (Susan Scheck, USA) April 14, 1991: ``Is It Live Or Is
It Memory?''

Corriere Della Sera (Giovanni Capara, Italy) August 28, 1991: ``Pronto in
USA il programma che si riproduce.  Il computer `padre' crea vita
informatica.''

Fakta (Tom Ottmar, Norway) March 1992: ``Den Lever!  En `skabning', der
best\aa r af nuller og \'{e}nere, er vokset ud af indamaden p\aa \ en
computer og er blevet en videnskabelig sensation i USA.''

Associated Press (Theresa Humphrey, USA) October 1991: ``Bringing life to
computer.  U of D biologist's program is self-replicating, shows evolution.''

Hovedomr\aa det (Jakob Skipper, Denmark) December 6, 1990: ``Kunstigt liv.
Nu kommer det kunstige liv.  En voksende gruppe af dataloger, biologer,
fysikere, psykologer og mange andre forskere efterlinger p\aa \ computer
det naturlige liv.''

3)  RELATED SOFTWARE (IMPORTANT)

     The Tierra simulator is the central piece of a growing set of programs.
The accessory programs aid in observing the results of Tierra runs.  You will
want to select one or both of the following:

  3.1) The Beagle Explorer which graphically displays the output that Tierra
   saves to disk.  Beagle runs only on DOS systems, and while the heart of
   the source code is available, Beagle uses the Greenleaf DataWindows
   interface, and that source can not be distributed (available from
   Greenleaf Software, 16479 Dallas Parkway, Bent Tree Tower Two, Suite 570,
   Dallas, Texas, 75248, phone: [phone]).  Beagle would normally be
   distributed in the executable form.  Once you have the executables, you
   will be able to run them on your PC, and you will not need Greenleaf.
   You only need Greenleaf if you plan to modify the source code.  Beagle is
   currently configured only for the CGA or VGA graphics modes, but can be
   extended on request, in time.  Beagle executables are now available on disk
   from Virtual Life, or in the same ftp site as the Tierra source code, in
   the directory: /beagle.  If you pick up the Beagle executables from
   the ftp site, remember to trasfer the files in binary mode.

  3.2) The ALmond Monitor which displays activity in a running Tierra
   simulator.  ALmond runs as a simultaneous but independent process (from
   Tierra) on the same or on a seperate machine, and establishes network
   socket communication with Tierra.  ALmond can be attached to and detached
   from a running Tierra simulator without interrupting the simulation.
   ALmond runs only on unix systems supporting X Windows.  The entire ALmond
   source code is available.  Almond was developed and tested on Sun 3 and Sun
   4 machines.  It was developed under X11R4 by Marc Cygnus.  Dan Pirone
   has now taken over ALmond development. 

4)  QUICK START 

     The steps required to run the system on DOS and UNIX are slightly
different, so there are two sets of instructions listed below.

  4.1) DOS QUICK START

     If you obtained the Tierra software on disk, the installation program
will take care of steps 1 - 3, so you can skip to step 4.  If you
obtained the software over the net, start with step 1. 

      step 1)  You should have a directory containing the executables and
source code and five subdirectories: td, gb1, gb2, gb3, and gb4.  The td
directory is where a record of births and deaths will be written.  The gb
directories contain the initial genomes used to innoculate the soup and the
opcode maps.  The genebanker will save new genomes to the gb directories.
There is a gb directory for each of the four instruction sets.

      step 2)  You must compile the assember/disassembler, arg, and the
simulator, tierra.  We include the two Turbo C V 2.0 project files:
tierra.prj and arg.prj.  If you are using a more recent version of the
compiler, such as Borland C++, you must use the Borland project tool to
create a binary project file.  Just list the files listed in the two ascii
project files that are provided.  Compile these projects using the large
memory model.  Put the executables in the path.

      step 3)  You must assemble the initial genomes, as binaries are not
portable.  To do this, go into the gb1 directory and type:

arg c 0080.gen 80 0080aaa.tie

This will create the binary file 0080.gen which contains a creature that you
can use to innoculate the soup, the ancestor 0080aaa.  You can check to see if
this worked by disassembling the genome, by typing:

arg x 0080.gen aaa

This will create the ascii file 0080aaa.  Compare it to the original,
0080aaa.tie (it will not be exactly the same).  Before you start a run, copy
0080.gen to 0080gen.vir, in order to have a virgin copy for use later when
you start another run.

copy 0080.gen 0080gen.vir

     You can do the same for each of the four gb directories (gb1, gb2, gb3,
and gb4).  Be sure to assemble the genomes listed at the ends of the
corresponding soup_in files (soup_in1, soup_in2, soup_in3, soup_in4).

      step 4)  Go back to the source code directory and examine the file
soup_in1.  This file contains all of the parameters that control the run.  It
is currently set up to innoculate the soup with one cell of genotype 0080aaa,
and to run for 500 million instructions in a soup of 50,000 instructions.  You
will need a text editor if you want to modify this file.  If you use a regular
word processor, be sure that you write the file back out as a plain ASCII text
file.

      step 5)  Run the simulator by typing: tierra

      step 6)  When the run is over, if you want to start a new run, you
should clean up the genebank, because the simulator will read in all genomes
in the genebank at startup.  The best way to do this is to use the batch files
that are provided for this purpose: clr1.bat, clr2.bat, clr3.bat and clr4.bat.
Since we are discussing a run of instruction set 1, use clr1.bat, by typing
clr1 to the DOS prompt.  The batch file will take care of the cleanup.

     If you wish to use a cumulative genebank in successive runs, use the
corresponding cumulative clear batch files: cclr1.bat, cclr2.bat, cclr3.bat,
and cclr4.bat.

  4.2) UNIX QUICK START

      step 1)  You should have a directory containing the source code and five
subdirectories: td and gb1, gb2, gb3 and gb4.  The td (tiedat) directory is
where a record of births and deaths will be written.  The gb (genebank)
directories contain the initial genomes used to innoculate the soup and the
opcode map, and the genebanker will save new genomes to these directories.

      step 2)  You must compile the assember/disassembler, arg, and the
simulator, tierra.  There is a Makefile included to perform the compilation.
This Makefile needs to be edited to comment in the lines for your particular
hardware.  It has been tested on Sun 3, Sun 4, IBM RS6000, Silicon Graphics
Personal Iris and Indigo, DEC DS5000, and NeXT.  If you can use the Makefile,
type: make, and follow instructions.

      step 3)  You must assemble the initial genome, as binaries are not
portable.  To do this, go into the gb1 directory and type:

../arg c 0080.gen 80 0080aaa.tie

This will create the binary file 0080.gen which contains a creature that you
can use to innoculate the soup, the ancestor 0080aaa.  You can check to
see if this worked by disassembling the genome, by typing:

../arg x 0080.gen aaa

This will create the ascii file 0080aaa.  Compare it to the original,
0080aaa.tie (they will not be exactly the same).  Before you start a run,
copy 0080.gen to 0080gen.vir, in order to have virgin copies for use later
when you start another run.

cp 0080.gen 0080gen.vir

     You can do the same for each of the four gb directories (gb1, gb2, gb3,
and gb4).  Be sure to assemble the genomes listed at the ends of the
corresponding soup_in files (soup_in1, soup_in2, soup_in3, soup_in4).

      step 4)  Go back to the source code directory and examine the file
soup_in1.  This file contains all of the parameters that control the run.  It
is currently set up to innoculate the soup with one cell of genotype 0080aaa,
and to run for 500 million instructions in a soup of 50,000 instructions.

      step 5)  Run the simulator by typing:
           tierra
       or: tierra > /dev/null &  (to run it in the background
			         a Log file can be created by setting
			         the soup_in variable Log = 1)

     In order to run tierra in the background, you must compile it with:

#define FRONTEND STDIO

     If you will run Tierra in the foreground, we recommend that you use:

#define FRONTEND BASIC

     These definitions are made in the configur.h file.

      step 6)  When the run is over, if you want to start a new run, you
should clean up the genebank.  The best way to do this is to use the
Unix script files that have been provided for this purpose (clr1, clr2,
clr3 and clr4).  You must make the clr# files executable by changing their
protection:

chmod +x clr1

     Then all you have to do is type ``clr1'' to the prompt, and the
shell script will take care of the cleanup.

     If you wish to use a cumulative genebank in successive runs, use the
cumulative clear files (cclr1, cclr2, cclr3, cclr4).  You must also make
sure that they are executable:

chmod +x cclr1

5)  RUNNING TIERRA

     This section has the following sub-sections:

  5.1) Startup
  5.2) The Assembler/Disassembler
  5.3) The Birth-Death Output
  5.4) The Genebank Output
  5.5) Restarting an Old Run
  5.6) The User Interface
    5.6.1) The Basic Interface
      5.6.1.1) The Basic Screen
      5.6.1.2) The Menu Options
        5.6.1.2.1) The Size Histogram
        5.6.1.2.2) The Memory Histogram
        5.6.1.2.3) The Genotype Histogram
        5.6.1.2.4) The Size Class Information Display
        5.6.1.2.5) The Virtual Debugger
        5.6.1.2.6) The Genome Injector
    5.6.2) The Standard Output
    5.6.3) The tierra.log file

  5.1) Startup

     The first steps in running Tierra are described briefly above.  One must
place the genomes and the opcode map in the gb directory, and one must have
created the td directory to receive the output of birth and death data.  The
genome files are supplied in the form of ASCII assembler code files.  These
must be assembled into binary form to be able to execute on the virtual
machine.  If you type arg, the assembler will give you a brief listing of
assembler options.  More complete documentation of the assembler follows:

  5.2) The Assembler/Disassembler