Nebula 1994 June/NEBULA_SE.ISO/Documents/FAQ/AI-faq/genetic/part2

Internet Message Format | 1993-08-21 | 56.8 KB

Path: senator-bedfellow.mit.edu!bloom-beacon.mit.edu!gatech!howland.reston.ans.net!europa.eng.gtefsd.com!uunet!mcsun!Germany.EU.net!Informatik.Uni-Dortmund.DE!lusty!heitkoet
From: [e-mail] (Joerg Heitkoetter)
Newsgroups: comp.ai.genetic,comp.answers,news.answers
Subject: FAQ: comp.ai.genetic part 2/3 (A Guide to Frequently Asked Questions)
Supersedes: <[e-mail]>
Followup-To: comp.ai.genetic
Date: 20 Aug 1993 13:57:16 GMT
Organization: CS Department, University of Dortmund, Germany
Lines: 1390
Approved: [e-mail]
Expires: 3 Oct 1993 13:57:09 GMT
Message-ID: <[e-mail]>
References: <[e-mail]>
NNTP-Posting-Host: lusty.informatik.uni-dortmund.de
Summary: This is part 2 of a trilogy entitled "The Hitch-Hiker's Guide
     to Evolutionary Computation". A monthly published list of Frequently
     Asked Questions (and their answers) about Evolutionary Algorithms,
     Life and Everything. It should be read by anyone who whishes to post
     to the comp.ai.genetic newsgroup, preferably *before* posting.
Originator: [e-mail] (Joerg Heitkoetter)
Xref: senator-bedfellow.mit.edu comp.ai.genetic:1183 comp.answers:1672 news.answers:11623

Archive-name:    ai-faq/genetic/part2
Last-Modified:    08/20/93
Version:    0.6





FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     [eds  note:  This  is a preliminary version, ie. a "proposal", of the
     forthcoming FAQ to comp.ai.genetic. If you  want  to  contribute  new
     items,  make  corrections, or want to fill in a "[..]" template, drop
     me a mail.]


A5) What about all this Optimization stuff? [preliminary]
     Just think of an optimization problem as a black box.  A large  black
     box.  As  large as, for example, a Coca-Cola vending machine. Now, we
     don't know nothing about the inner workings of  this  box,  but  see,
     that  there  are some regulators to play with, and of course we know,
     that we want to have a bottle of the real thing...

     Putting this everyday problem into a mathematical model,  we  proceed
     as follows:

     (1) we  label all the regulators with x and a number starting from 1;
     the result is a vector x,  ie.  (x_1,...,x_n),  where  n  is  the
     number of visible regulators.

     (2) we must find an objective function, in this case it's obvious, we
     want to get k bottles of the real thing, where k is equal  to  1.
     [some  might  want  a  "greater or equal" here, but we restricted
     ourselves to the visible regulators; we all know that sometimes a
     "kick  in  the  right  place" gets use more than 1, but I have no
     idea how to put this mathematically...]

     (3) thus, in the language some mathematicians  prefer  to  speak  in:
     f(x)  =  k  =  1. So, what we have here is a maximization problem
     presented in a form we know from some  boring  calculus  lessons,
     and   we   also   know  that  there  at  least  a  dozen  utterly
     uninteresting techniques to solve problems presented this  way...

 What can we do in order to solve this problem?
     We  can  either try to gain more knowledge or exploit what we already
     know about the interior of the black box. If the  objective  function
     turns  out  to  be smooth and differentiable, analytical methods will
     produce the exact solution.

     If this turns out to be impossible, we  might  resort  to  the  brute
     force  method  of  enumerating the entire search space.  But with the
     number of possibilities growing exponentially in  n,  the  number  of
     dimensions  (inputs),  this  method  becomes infeasible even for low-
     dimensional spaces.

     Consequently, mathematicians  have  developed  theories  for  certain
     kinds  of  problems  leading  to specialized optimization procedures.
     These  algorithms  perform  well  if  the  black  box  fulfils  their
     respective  prerequisites.   For example, Dantzig's simplex algorithm
     (Dantzig 66) probably  represents  the  best  known  multidimensional
     method capable of efficiently finding the global optimum of a linear,



Version 0.6               Posted: 20 August 1993                         1







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     hence convex, objective function in a search space limited by  linear
     constraints.   (A  USENET  FAQ on linear programming is maintained by
     John W. Gregory of Cray Research, Inc.  Try  to  get  your  hands  on
     "linear-programming-faq"  that  is  posted  monthly  to sci.math.num-
     analysis and quite interesting to read.)

     Gradient strategies are no longer tied to these  linear  worlds,  but
     they  smooth their world by exploiting the objective function's first
     partial derivatives one has to supply in  advance.  Therefore,  these
     algorithms  rely on a locally linear internal model of the black box.

     Newton   strategies   additionally   require   the   second   partial
     derivatives,  thus building a quadratic internal model. Quasi-Newton,
     conjugate gradient and variable metric  strategies  approximate  this
     information during the search.

     The  deterministic  strategies  mentioned  so  far  cannot  cope with
     deteriorations, so the search will stop if  anticipated  improvements
     no  longer  occur.  In a multimodal environment these algorithms move
     "uphill" from their respective starting points. Hence, they can  only
     converge to the next local optimum.

     Newton-Raphson-methods  might  even  diverge if a discrepancy between
     their internal assumptions and reality occurs.  But of course,  these
     methods  turn  out  to  be  superior  if  a  given task matches their
     requirements. Not relying on derivatives, polyeder strategy,  pattern
     search  and  rotating coordinate search should also be mentioned here
     because they  represent  robust  non-linear  optimization  algorithms
     (Schwefel 81).

     Dealing with technical optimization problems, one will rarely be able
     to write down the objective function in a closed form.  We often need
     a simulation model in order to grasp reality.  In general, one cannot
     even  expect  these  models   to   behave   smoothly.   Consequently,
     derivatives  do  not  exist. That is why optimization algorithms that
     can successfully  deal  with  black  box-type  situations  habe  been
     developed.  The  increasing  applicability is of course paid for by a
     loss  of  CONVERGENCE  VELOCITY,  compared  to  algorithms  specially
     designed  for  the given problem.  Furthermore, the guarantee to find
     the global optimum no longer exists.

 But why turn to nature when looking for more powerful algorithms?
     In the attempt to create tools  for  various  purposes,  mankind  has
     copied,  more  often instinctively than geniously, solutions invented
     by nature.  Nowadays, one can prove in some cases that certain  forms
     or structures are not only well adapted to their environment but have
     even reached the optimum (Rosen 67). This is due to the fact that the
     laws  of  nature  have  remained  stable  during the last 3.5 billion
     years. For instance, at branching points the measured  ratio  of  the
     diameters in a system of blood-vessels comes close to the theoretical
     optimum provided by the laws of fluid  dynamics  (2^-1/3).  This,  of



Version 0.6               Posted: 20 August 1993                         2







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     course,  only  represents  a  limited,  engineering  point of view on
     nature. In general, nature performs ADAPTATION, not OPTIMIZATION.

     The idea to imitate basic principles of natural processes for optimum
     seeking  procedures  emerged  more than three decades ago (cf Q10, 3.
     Classics). Although these algorithms have proven  to  be  robust  and
     direct  optimization  tools,  it  is only in the last five years that
     they have caught the researchers' attention. This is due to the  fact
     that  many  people  still look at organic evolution as a huge game of
     dice, thus ignoring the fact that this model of evolution cannot have
     worked:  a human germ-cell comprises approximately 50,000 genes, each
     of which consists of about 300 triplets of  nucleic  bases.  Although
     the  four  existing  bases  only  encode  20  different  amino acids,
     20^15,000,000, ie circa 10^19,500,000 different genotypes had  to  be
     tested in only circa 10^17 seconds, the age of our planet. So, simply
     rolling the dice could not have produced  the  diversity  of  today's
     complex living systems.

     Accordingly,   taking   random   samples  from  the  high-dimensional
     parameter space of an objective function in order to hit  the  global
     optimum  must  fail  (Monte-Carlo  search). But by looking at organic
     evolution as a  cumulative,  highly  parallel  sieving  process,  the
     results  of  which pass on slightly modified into the next sieve, the
     amazing  diversity  and  efficiency  on  earth  no   longer   appears
     miraculous.  When  building a model, the point is to isolate the main
     mechanisms which have led  to  today's  world  and  which  have  been
     subjected  to  evolution  themselves.  Inevitably, nature has come up
     with a mechanism allowing individuals  of  one  species  to  exchange
     parts of their genetic information (RECOMBINATION or CROSSOVER), thus
     being able to meet changing environmental conditions in a better way.

     (Dantzig 66) G.B. Dantzig "Lineare Programmierung und Erweiterungen",
     Berlin: Springer. (Linear pogramming and extensions)

     (Kursawe 94) F.  Kursawe  "Evolution  strategies:  Simple  models  of
     natural  processes?",  Revue Internationale de Systemique, France (to
     appear).

     (Rosen 67) R. Rosen  "Optimality  Principles  in  Biologie",  London:
     Butterworth.

     (Schwefel  81)  H.-P.  Schwefel  "Numerical  Optimization of Computer
     Models", Chichester: Wiley.

A10) Good introductory material on EAs?
 1. Books for absolute beginners
     Dawkins, R. (1976, 1989 2nd ed) "The Selfish  Gene",  Oxford:  Oxford
     University  Press.  [The 2nd edition includes the first unaltered and
     adds two new  chapters,  updated  references,  and  "Endnotes"  which
     update the original text.  One of the new chapters expands ideas from
     his "Extended Phenotype".]



Version 0.6               Posted: 20 August 1993                         3







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     Dawkins, R. (1982) "The Extended Phenotype: The Gene  as  a  Unit  of
     Selection", Oxford: Oxford University Press.

     Dawkins, R. (1986) "The Blind Watchmaker", New York: W.W. Norton.

     Regis, E. (1987) "Who got Einstein's Office?  Eccentricity and Genius
     at the Institute for Advanced Study",  Reading,  MA:  Addison  Wesley
     [eds note: chapters 5, 10 and 12]

     Levy,  S. (1992) "Artificial Life: The Quest for a new Creation", New
     York, NY: Pantheon. [LEVY92] [eds note: read this and you  will  have
     the urge to work in this field]

 2. Textbooks
     Goldberg,  D.E.  (1989)  "Genetic Algorithms in Search, Optimization,
     and Machine  Learning",  available  from  Addison-Wesley  Publishing,
     [phone].

     Davis,  L.  (ed)  (1991)  "Handbook of Genetic Algorithms", available
     from Van Nostrand Reinhold, 115 5th  Avenue,  New  York,  NY,  10003,
     [phone]. ISBN 0-442-00173-8.

     Michalewicz,  Z.  (1992)  "Genetic  algorithms  +  Data  Structures =
     Evolution Programs", Springer-Verlag, 175 5th Avenue, New  York,  NY,
     10010, [phone].

     Koza,  J.R.  (1992),  "Genetic  Programming:  On  the  Programming of
     Computers by means of Natural Selection", Cambridge, MA:  MIT  Press.
     [KOZA92]

 3. The Classics
     Charles  Darwin (1859), "The Origin of Species", London: John Murray.
     (Penguin  Classics,  London,  1985;  New  American  Library,   Mentor
     Paperback)

     Box,  G.E.P.  (1957)  "Evolutionary operation: a method of increasing
     industrial productivity", Applied Statistics, 6, 81-101.

     Fraser, A.S. (1957)  "Simulation  of  genetic  systems  by  automatic
     digital  computers",  Australian  Journal of Biological Sciences, 10,
     484-491.

     Bremermann,  H.J.  (1962)   "Optimization   through   evolution   and
     recombination". In M.C. Yovits, et al, (eds) Self-Organizing Systems.
     Washington, DC: Spartan Books.

     Holland, J.H. (1962)  "Outline  for  a  logical  theory  of  adaptive
     systems", JACM, 3, 297-314.

     Walter, W.G. (1963) "The Living Brain", New York: W.W. Norton.




Version 0.6               Posted: 20 August 1993                         4







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     Fogel,   L.J.,   Owens,   A.J.   &  Walsh,  M.J.  (1966)  "Artificial
     Intelligence through Simulated Evolution", New York: Wiley.

     Rosen,  R.  (1967)  "Optimality  Principles  in   Biology",   London:
     Butterworths.

     Rechenberg,  I. (1973, 1993 2nd ed) "Evolutionsstrategie: Optimierung
     technischer Systeme  nach  Prinzipien  der  biologischen  Evolution",
     Stuttgart:  Fromman-Holzboog.  (Evolution  Strategy:  Optimization of
     technical systems by means of biological evolution)

     Holland, J.H. (1975) "Adaptation in natural and artificial  systems",
     Ann Arbor, MI: The University of Michigan Press.

     De  Jong,  K.A.  (1975)  "An  analysis  of the behavior of a class of
     genetic adaptive systems", Doctoral thesis,  Dept.  of  Computer  and
     Communication Sciences, University of Michigan, Ann Arbor.

     Schwefel,  H.-P. (1977) "Numerische Optimierung von Computer-Modellen
     mittels der Evolutionsstrategie", Basel: Birkhaeuser.

     Schwefel, H.-P. (1981) "Numerical Optimization of  Computer  Models",
     Chichester:  Wiley.  [eds  note:  English translation of the previous
     entry; a reworked edition is currently in preparation for 1994]

     Cramer, N.L. (1985) "A Representation for the Adaptive Generation  of
     Simple Sequential Programs" [ICGA85], 183-187.

     Baeck,  T.,  Hoffmeister,  F.  &  Schwefel, H.-P. (1991) "A Survey of
     Evolution Strategies" [ICGA91], 2-9.

     Rudolph,  G.  (1994)  "Convergence  Analysis  of  Canonical   Genetic
     Algorithms", IEEE Trans. on Neural Networks, Special issue on EP. (to
     appear).  [eds note: unfortunately the publisher denies  distribution
     of  this article, until January '94. I therefore include the abstract
     here: This paper analyzes the convergence properties of the canonical
     genetic  algorithm  (CGA)  with  mutation, crossover and proportional
     reproduction applied  to  statistic  optimization  problems.   It  is
     proved  by  means  of homogeneous finite Markov chain analysis that a
     CGA will never converge to  the  global  optimum  regardless  of  the
     initialization,  crossover  operator  and  objective  function.   But
     variants of CGAs that  always  maintain  the  best  solution  in  the
     population,  either  before or after selection, are shown to converge
     to the global optimum due  to  the  irreducibility  property  of  the
     underlying orginal nonconvergent CGA.  The results are discussed with
     respect to the schema theorem.]

 4. Introductory journal articles
     Goldberg, D.E. (1986) "The Genetic  Algorithm:  Who,  How,  and  What
     Next?".   In  Kumpati S. Narenda, ed., Adaptive and Learning Systems,
     Plenum, New York, NY.



Version 0.6               Posted: 20 August 1993                         5







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     Dawkins,  R.  (1987)  "The  Evolution  of  Evolvability",   [ALIFEI],
     201-220.

     Hillis,  W.D.  (1987)  "The Connection Machine", Scientific American,
     255(6).

     Holland, J.H. (1989) "Using  Classifier  Systems  to  Study  Adaptive
     Nonlinear  Networks".  In: Lectures in the Science of Complexity, SFI
     Studies in the Science of Complexity, D. Stein, (ed), Addison Wesley.

     Brooks,  R.A.  (1991)  "Intelligence without Reason", MIT AI Memo No.
     1293.  Appeared in "Computer's and Thought", IJCAI-91.

     Hillis, W.D. (1992) "Massively Parallel Computing" Daedalus,  winter,
     121(1), 1-29. [HILLIS92]

     Holland,  J.H.  (1992)  "Genetic  Algorithms",  Scientific  American,
     260(9), 44-51. [HOLLAND92]

     Holland, J.H. (1992) "Complex  Adaptive  Systems"  Daedalus,  winter,
     121(1), 17-30.

     Sims,   K.  (1991)  "Artificial  Evolution  for  Computer  Graphics",
     Computer Graphics, 25(4), 319-328

     Spears, W.M., DeJong, K.A., Baeck, T.,  Fogel,  D.  &  de  Garis,  H.
     (1993)  "An Overview of Evolutionary Computation", [ECML93], 442-459.

     Baeck, T. & Schwefel,  H.-P.  (1993)  "An  Overview  of  Evolutionary
     Algorithms  for  Parameter  Optimization",  Evolutionary Computation,
     1(1), 1-23.

     Baeck,  T.,  Rudolph,  G.  &  Schwefel,  H.-P.  (1993)  "Evolutionary
     Programming  and Evolution Strategies: Similarities and Differences",
     [EP93], 11-22.

 5. Introductory Technical Reports
     Hoffmeister, F. & Baeck, T.  (1990,  1992)  "Genetic  Algorithms  and
     Evolution  Strategies:  Similarities  and Differences", University of
     Dortmund, Dept.  of  CS,  SyS-1/92.  Available  via  anon.  ftp  from
     lumpi.informatik.uni-dortmund.de.

     Whitley,  D.  (1993)  "A  Genetic Algorithm Tutorial", Colorado State
     University, Dept. of CS, TR CS-93-103. Available via anon.  ftp  from
     beethoven.cs.colostate.edu.

 6. Not-quite-so-introductory literature
     Davis,  L.  (ed) (1987) "Genetic Algorithms and Simulated Annealing",
     available from Morgan Kaufmann Publishers (address and  phone  number
     below).




Version 0.6               Posted: 20 August 1993                         6







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     Davidor,  Y.  (1991)  "Genetic  Algorithms  and Robotics", Singapore:
     World Scientific. ISBN 9-810202172.

     Forrest,  S.  (ed)  (1990)  "Emergent  Computation.  Self-Organizing,
     Collective,  and  Cooperative  Phenomena  in  Natural  and Artificial
     Computing Networks", [FORREST90], Cambridge, MA: MIT Press.  (Special
     issue of Physica D.)

     Hillis,   W.D.   (1990)   "Co-Evolving  Parasites  Improve  Simulated
     Evolution as an Optimization procedure", [ALIFEII], 313-324.

     Holland, J.H., Holyoak, K.J., Nisbett, R.E. &  Thagard,  P.R.  (1986)
     "Induction:   Processes   of  Inference,  Learning,  and  Discovery",
     Cambridge, MA: MIT Press.

     Holland, J.H. (1992) "Adaptation in Natural and  Artificial  Systems:
     An  Introductory  Analysis with Applications to Biology, Control, and
     Artificial Intelligence, Cambridge,  MA:  MIT  Press/Bradford  Books,
     (2nd edn).

     Serra,  R.  &  Zanarini,  G.  (1990)  "Complex  Systems and Cognitive
     Processes", New York, NY: Springer-Verlag.

 7. Biological Background Readings
     Maynard Smith, J. (1978) "Optimization Theory  in  Evolution",  Annu.
     Rev. Ecol. Syst. 9:31-56.

     Maynard  Smith,  J.  (1982)  "Evolution  and  the  Theory  of Games",
     Cambridge: Cambridge Univ. Press.

     Futuyma, D.J. (1986) "Evolutionary Biology", Sunderland, MA:  Sinauer
     Assoc.   [eds note: the bibliography of this book is truly a treasure
     chest]

     Maynard Smith, J.  (1989)  "Evolutionary  Genetics",  Oxford:  Oxford
     Univ. Press.

     Adams,  D. with Carwardine M. (1990) "Last Chance to see...", London:
     Heinemann.

 8. Videos
     Sims,  K.  (1990)  "Panspermia",  ACM  Sigraph  Review.  order   form
     available                 by                 ftp                 see:
     siggraph.org:/publications/video_review/order_blank  Look   in   that
     directory  for  other  useful information.  Note that "Panspermia" is
     Item 23 of Issue 62 of the "SIGGRAPH Video Review".

     Langton, C.G. (ed) (1992) "Artificial Life II Video Proceedings"  The
     Advanced  Book  Program  of  the  Santa  Fe Institute: Studies in the
     Sciences of Complexity, Addison Wesley, ISBN 0-201-55492-5.




Version 0.6               Posted: 20 August 1993                         7







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     Koza, J.R. & Rice, J.P.  (1992)  "Genetic  Programming:  The  Movie",
     Cambridge, MA: MIT Press. (see GP-faq for an order form, cf Q15)

 9. CD-ROMs
     Carnegie Mellon University is establishing an Artificial Intelligence
     Repository  to  contain  public  domain  and   freely   distributable
     software,  publications,  and  other  materials  of  interest  to  AI
     researchers, educators, and  students.  The  AI  Repository  will  be
     accessible  by  anonymous  FTP  and  Andrew File System (AFS) without
     charge.  The contents of the repository will  also  be  published  by
     Prime  Time  Freeware  as  an  inexpensive  mixed-media (Book/CD-ROM)
     publication.

     For your information, here is a precis of the CD-ROM:

     PTF for AI is a periodic collection of  AI-related  source  code  and
     documentation.   PTF for AI in no way modifies the legal restrictions
     on any package it includes.  The upcoming issue (1-1;  Summer,  1993)
     will  consist  of an ISO-9660 CD-ROM bound into a ~100 page book.  It
     will contain ~600 MB of gzipped  archives  (2+  GB  uncompressed  and
     unpacked).  It will sell (list) for $60 US.

     For  more  information  contact:  Mark  Kantrowitz, Archivist, CMU AI
     Repository, Editor, PTF for AI.  Net:  <[e-mail]>,
     Tel: [phone], Fax: [phone].


A11) Any journals and magazines?
     [eds  note:  comments  on  speed of reviewing and publishing, whether
     they accept LaTeX/TeX format or ASCII by e-mail, etc. may be added]

 1. Dedicated EC Journals:
  Evolutionary Computation
     Published  quarterly  by:  MIT  Press  Jornals,  55  Hayward  Street,
     Cambridge,  MA  02142-1399,  USA.  Tel:  (617)  253-2889,  Fax: (617)
     258-6779, Net: <[e-mail]>

     Along with the explosive growth of the computing  industry  has  come
     the  need  to  design  systems  capable  of  functioning  in complex,
     changing environments. Considerable effort  is  underway  to  explore
     alternative  approaches  to  designing  more  robust computer systems
     capable of learning from and adapting to  the  environment  in  which
     they operate.

     One broad class of such techniques takes its inspiration from natural
     systems  with  particular  emphasis   on   evolutionary   models   of
     computation  such as GAs, ESs. CFS, and EP. Until now, information on
     these techniques has been widely spread  over  numerous  disciplines,
     conferences, and journals.  [eds note: The editorial board reads like
     a who-is-who in EC.]  For paper e-mail submission,  use  one  of  the
     following addresses:



Version 0.6               Posted: 20 August 1993                         8







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



      America:  John Grefenstette <[e-mail]>
      Europe:        Heinz Muehlenbein <[e-mail]>
      Asia:          Hiroaki Kitano <[e-mail]>

      Ed-in-chief:   Ken De Jong <[e-mail]>

     Please  note,  that  submissions  should  be  sent to one of the sub-
     editors.   Grefenstette  and  Kitano  accept  LaTeX   or   PostScript
     submissions.

  BioSystems
     Journal  of  Biological and Information Processing Sciences, Elsevier
     Science  Publishers,  P.O.  Box  1527,   1000   BM   Amsterdam,   The
     Netherlands.

     BioSystems  encourages  experimental,  computational, and theoretical
     articles  that  link  biology,   evolutionary   thinking,   and   the
     information  processing  sciences.  The link areas form a circle that
     encompasses  the  fundamental  nature   of   biological   information
     processing,  computational  modeling  of  complex biological systems,
     evolutionary models of computation,  the  application  of  biological
     principles  to  the design of novel computing systems, and the use of
     biomolecular materials to synthesize artificial systems that  capture
     essential principles of natural biological information processing.

     Topics:  MOLECULAR  EVOLUTION:  Self-organizing  and self-replicating
     systems, Origin and evolution of the  genetic  mechanism;  BIOLOGICAL
     INFORMATION  PROCESSING:  Molecular  recognition,  Cellular  control,
     Neuromuscular computing, Biological adaptability, Molecular computing
     technologies;    EVOLUTIONARY    SYSTEMS:   Stochastic   evolutionary
     algorithms, Evolutionary  optimization,  Simulation  of  genetic  and
     ecological  systems,  Applications  (neural  nets,  machine learning,
     robotics))

 2. Related Journals:
  Complex Systems
     Published by: Complex Systems  Publications,  Inc.,  P.O.  Box  6149,
     Champaign, IL 61821-8149, USA.

     Complex  Systems  devotes to the rapid publication of research on the
     science,  mathematics,  and  engineering  of  systems   with   simple
     components   but   complex   overall   behavior.   Try  finger(1)  on
     <[e-mail]> for additional info.

  Machine Learning
     Published by:  Kluwer  Academic  Publishers,  P.O.  Box  358,  Accord
     Station, Hingham, MA 02018-0358 USA.

     Machine   Learning   is   an  international  forum  for  research  on
     computational approaches to learning. The journal publishes  articles
     reporting  substantive  research  results on a wide range of learning



Version 0.6               Posted: 20 August 1993                         9







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     methods applied to a variety of task domains. The  ideal  paper  will
     make   a   theoretical   contribution   supported   by   a   computer
     implementation.

     The journal  has  published  many  key  papers  in  learning  theory,
     reinforcement  learning,  and  decision  tree  methods.   The journal
     regularly publishes special issues devoted to GAs and CFS as well.

  Adaptive Behavior
     Published  quarterly  by:  MIT  Press  Jornals,  55  Hayward  Street,
     Cambridge,  MA  02142-1399,  USA.  Tel:  (617)  253-2889,  Fax: (617)
     258-6779, Net: <[e-mail]>

     Broadly, behavior is adaptive if it deals successfully  with  changes
     circumstances.   For   example,   when   surprised,  a  hungry  --but
     environmentally informed-- mouse  may  dart  for  cover  rather  than
     another piece of cheese. Similarly, a tripped-up robot [eds note: not
     necessarily built by Sirius Cybernetics Corp.]  could get back on its
     feet  and  accomplish a moonrock-finding mission if it had learned to
     cope with unanticipated lunar potholes.

     Adaptive Behavior thus takes an approach complementary to traditional
     AI.  Now  basic abilities that allow animals to survive, or robots to
     perform their mission in unpredictable environments, will be  studied
     in preference to more elaborate and human-specific abilities.

     The  journal  also  aims  to  investigate  which  new  insights  into
     intelligence and cognition can be achieved by explicitly taking  into
     account  the  environment  feedback  --mediated by behavior-- that an
     animal or  a  robot  receives,  instead  of  studying  components  of
     intelligence in isolation.

     Topics:  Individual  and  Collective  Behavior.  Neural Correlates of
     Behavior. Perception  and  Motor  Control.  Motivation  and  Emotion.
     Action  Selection  and  Behavioral  Sequences. Internal World Models.
     Ontogeny, Learning, and Evolution. Characterization of  Environments.

  Artificial Life
     Published  quarterly  by:  MIT  Press  Jornals,  55  Hayward  Street,
     Cambridge, MA  02142-1399,  USA.  Tel:  (617)  253-2889,  Fax:  (617)
     258-6779, Net: <[e-mail]>

     Artificial  Life  is  intended  to  be  the  primary  forum  for  the
     dissemination of scientific and engineering research in the field  of
     artificial  life.   It will report on synthetic biological work being
     carried out in any and all media,  from  the  familiar  "wetware"  of
     organic chemestry, through the inorganic "hardware" of mobile robots,
     all the way to the virtual "software" residing inside computers.

     Research topics ranging  from  the  fabrication  of  self-replicating
     molecules  to  the study of evolving populations of computer programs



Version 0.6               Posted: 20 August 1993                        10







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     will be included.

     There will also be occasional issues devoted to special topics,  such
     as  L-Systems,  genetic  algorithms, in-vitro evolution of molecules,
     artificial cells, computer viruses, and many social and philosophical
     issues arising from the attempt to synthesize life artificially.

     [eds note: The editorial board reads like a who-is-who in Alife]

  Evolutionary Economics
     Published  quarterly  by:  Springer-Verlag  New  York,  Inc., Service
     Center Secaucus, 44 Hartz Way, Secaucus, NJ 07094, USA.   Tel:  (201)
     348-4033, Fax: [phone].

     Evolutionary  Economics  aims to provide an international forum for a
     new approach to economics.  Following  the  tradition  of  Joseph  A.
     Schlumpeter,  it  is  designed  to focus on original research with an
     evolutionary conception of the economy.   The  journal  will  publish
     articles  with  strong  emphasis  on  dynamics,  changing  structures
     (including technologies, institutions, beliefs, imitation, etc.).  It
     favors  interdisciplinary  analysis  and  is  devoted to theoretical,
     methodological and applied work.

     Research  areas  include:  industrial  dynamics;  multi-sectoral  and
     cross-country   studies   of   productivity;   innovations   and  new
     technologies; dynamic competition and structural change in a national
     and  international  context;  causes  and  effects  of technological,
     political and social changes; cyclic processes in economic evolution;
     the  role of governments in a dynamic world; modeling complex dynamic
     economic systems; application of concepts, such as self-organization,
     bifurcation, and chaos theory to economics; evolutionary games.

A12) Important conferences on EC? Proceedings?
 1. Dedicated EC Conferences:
  ICGA: Int'l Conference on Genetic Algorithms
     Major  international conference held in North America in odd-numbered
     years. Covers all aspects  of  evolutionary  computation.   The  1995
     conference will be held in [..], on [..]  For details contact [..]

     Proceedings of the 1st International Conference on Genetic Algorithms
     (1985) J.J. Grefenstette (ed) [ICGA85] and Proc.  of  the  2nd  Int'l
     Conf.  on  Genetic  Algorithms (1987) J.J. Grefenstette (ed) [ICGA87]
     available from  Lawrence  Erlbaum  Associates,  Inc.,  365  Broadway,
     Hillsdale, New Jersey, 07642, [phone].

     Proc.  of  the  3rd  Int'l  Conf.  on  Genetic Algorithms (1989) J.D.
     Schaffer (ed) [ICGA89] and Proc. of the 4th Int'l  Conf.  on  Genetic
     Algorithms (1991) R.K. Belew and L.B. Booker (eds) [ICGA91] and Proc.
     of the 5th Int'l Conf. on Genetic Algorithms (1993) S.  Forrest  (ed)
     [ICGA93]  available  from  Morgan Kaufmann Publishers, Inc., P.O. Box
     50490, Palo Alto, CA, 94303-9953, [phone].



Version 0.6               Posted: 20 August 1993                        11







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



  FOGA: Foundations of Genetic Algorithms
     Major international workshop focusing on theoretical aspects  of  EC,
     that's  usually limited to some 50 participants and is held somewhere
     in North America.

     FOGA 3 will take place from July 31 to August 3 in 1995, probably  in
     Estes  Park,  outside Rocky Mountain National Park.  Paper submission
     and inquires  to:  Darrell  Whitley,  Dept.  of  CS,  Colorado  State
     University, Fort Collins, CO 80523.  Net: <[e-mail]>

     Foundations of Genetic Algorithms (1991) G.J.E. Rawlins (ed) [FOGA91]
     and Foundations of Genetic Algorithms 2 (1993) L.D. Whitley  [FOGA93]
     available from Morgan Kaufmann Publishers, Inc., P.O. Box 50490, Palo
     Alto, CA, 94303-9953, [phone].

  PPSN: Parallel Problem Solving from Nature
     Major international conference held in Europe in even-numbered years.
     Covers  all aspects of problem solving inspired by natural processes.
     The 1994 conference will be held in  Israel  in  a  Kibbuz  close  to
     Jerusalem,   October   9-14.    For  details  contact  Yuval  Davidor
     <[e-mail]>.

     Parallel Problem Solving from Nature, (1990) H.-P.  Schwefel  and  R.
     Maenner  (eds) [PPSN90] published by Springer-Verlag, 175 5th Avenue,
     New York, NY, 10010, [phone].  Parallel Problem  Solving  from
     Nature 2, (1992) R. Maenner and B. Manderick (eds) [PPSN92] published
     by North-Holland, Elsevier Science Publishers, Sara  Burgerhartstraat
     25,  P.O.  Box  211,  1000  AE  Amsterdam, The Netherlands.  Parallel
     Problem Solving from Nature 3, (1994) Y.  Davidor,  [PPSN94]  (to  be
     published)

  EP: Annual Conference on Evolutionary Programming
     Major  international  annual  conference  held in San Diego, CA, USA.
     Covers all aspects of EC with emphasis on EP related  research.   The
     1994  conference  will  be held in San Diego, in February 24-25.  For
     details contact David Fogel <[e-mail]>.

     Proceedings of the 1st Annual Conference on Evolutionary Programming,
     (1992)  D.B.  Fogel  and W. Atmar (eds), [EP92], and Proc. of the 2nd
     Annual Conf. on Evolutionary Programming, (1993) D.B.  Fogel  and  W.
     Atmar   (eds),  [EP93]  published  by  the  Evolutionary  Programming
     Society, 9363 Towne Centre Dr.,  San  Diego,  CA  92121,  Attn:  Bill
     Porto, Treasurer.

     (EP93:  The proceedings is 215 pages in length and includes 25 papers
     presented at this  year's  conference.  Topics  include  evolutionary
     programming,  genetic  algorithms,  evolution strategies, and various
     applications.    For   more   information,   contact   David    Fogel
     <[e-mail]>)





Version 0.6               Posted: 20 August 1993                        12







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



 2. Related Conferences:
  Alife: International Conference on Artificial Life
     Proceedings  of  the 1st International Conference on Artificial Life,
     (1989) C.G. Langton (ed), Santa Fe Institute Studies in the  Sciences
     of  Complexity,  Proc.  Vol.  VI, [ALIFEI] and Proc. of the 2nd Int'l
     Conf. on Artificial Life II, (1992) C.G. Langton, C. Taylor, J. Doyne
     Farmer  and  S.  Rasmussen  (eds),  Santa Fe Institute Studies in the
     Sciences of Complexity, Proc. Vol. X, [ALIFEII] and Proc. of the  3rd
     Int'l  Conf.  on  Artificial  Life  III, (to appear) C.G. Langton, C.
     Taylor, J. Doyne Farmer and S. Rasmussen (eds), [ALIFEIII]  published
     by Addison Wesley, Redwood City, CA, USA.

     Artificial  Life  IV, will be organized by Rodney Brooks, MIT AI Lab,
     and held at July 6-8, 1994. Proceedings will be edited by  R.  Brooks
     and P. Maes. Papers should be send to: Rodney Brooks/Alife IV, MIT AI
     Lab, 545 Technology Square, Cambridge, MA 02139, USA.  Or  by  e-mail
     to: <[e-mail]>.

  ECAL: European Conference on Artificial Life
     Proceedings of the 1st European Conference on Artificial Life, (1991)
     F.J. Varela and P. Bourgine (eds), [ECAL91]  and  Proc.  of  the  2nd
     European  Conf.  on  Alife:  Self-organization  and life, from simple
     rules to global complexity, (1993) [..] (eds), [ECAL93]  (to  appear)
     published by MIT Press, Cambridge, MA, USA.

  ECML: European Conference on Machine Learning
     Machine  Learning: ECML-93, Proc. European Conf. on Machine Learning,
     (1993) P.B. Brazil (ed), [ECML93] published by  Springer,  New  York,
     NY, USA.

  SAB: International Conference on Simulation of Adaptive Behavior
     From  Animals  to  Animats.  Proceedings  of  the  1st  International
     Conference on Simulation of Adaptive Behavior, (1991)  [SAB90]  J.-A.
     Meyer and S.W. Wilson, ISBN 0-262-63138-5, and Proc. of the 2nd Int'l
     Conf. on Simulation  of  Adaptive  Behavior,  (1993)  [SAB92],  J.-A.
     Meyer,  H.  Roitblat and S.W. Wilson (eds) and Proc. of the 3rd Int'l
     Conf. on Simulation of Adaptive Behavior,  (to  appear)  [SAB94],  P.
     Husbands,  J.-A.  Meyer and S.W. Wilson (eds) published by MIT Press,
     Cambridge, MA, USA.

 3. Pointers to upcoming Conferences:
  The Genetic Algorithm Digest
     Aka "GA-Digest" always starts with a "Calendar of GA-related Events,"
     ie.   a  list of upcoming conferences, covering the complete field of
     EAs. (cf Q15)

  The Artificial Life Digest
     Aka "Alife digest" always starts with a  "Calendar  of  Alife-related
     Events," that lists conferences, workshops, etc. (cf Q15)





Version 0.6               Posted: 20 August 1993                        13







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



  The Evolutionary Programming Digest
     Aka "EP-digest" [..] (cf Q15)

A13) Evolutionary Computation Associations?
  ISGA: International Society on Genetic Algorithms
     The  ISGA  is  a  mostly  fascinating   society:  it  neither  has  a
     membership fee  (which  makes  it  even  more  fascinating),  nor  an
     address.  However,  ISGA  meetings  usually  take  place  during ICGA
     conferences, in so-called business  meetings  (BMs).  [eds  note:  So
     during a conference, aks for BMs, if you want to join; or be ready to
     dart out of a room if you don't...]

  EPS: Evolutionary Programming Society
     Membership  is  $10  per  year.   Address:  Evolutionary  Programming
     Society,  9363  Towne  Centre  Dr.,  San  Diego, CA 92121, Attn: Bill
     Porto, Treasurer.


A14) Available technical reports?
 TCGA Reports
     The Clearing House for Genetic  Algorithms  (TCGA)  distributes  TCGA
     technical reports.

     Contact:   Robert   Elliott   Smith,  Department  of  Engineering  of
     Mechanics, Room 210 Hardaway Hall, The University  of  Alabama,  P.O.
     Box  870278,  Tuscaloosa,  AL  35487, USA.  Tel: [phone], Fax:
     [phone], Net: <[e-mail]>.

 IlliGAL Reports
     The Illinois  Genetic  Algorithms  Laboratory  (IlliGAL)  distributes
     IlliGAL technical reports, as well as reprints of other publications;
     they are available in hardcopy and can be ordered  from  the  IlliGAL
     librarian.

     Contact:  Eric  Thompson,  IlliGAL  Librarian,  Department of General
     Engineering, 117 Transportation Building, 104 South  Mathews  Avenue,
     Urbana,    IL   61801-2996,   USA.    Tel:   (217)   333-2346,   Net:
     <[e-mail]>.

     NOTE: When  ordering,  please  include  your  surface  mail  address!
     IlliGAL  will  soon enter the Internet Age with our own anonymous-FTP
     server, making most of the existing IlliGAL reports, and  all  future
     ones,  available  in  electronic  form.  Until that time, it can only
     offer hardcopy.

 SyS Reports
     The Systems Analysis  Research  Group  (SyS)  at  the  University  of
     Dortmund,   maintains   an   experimental   anonymous   ftp   server:
     lumpi.informatik.uni-dortmund.de (129.217.36.140).  Please give  your
     complete  e-mail  address  as  password.   On lumpi you can find SyS-
     Reports from 1992 on. (Get  "/pub/ls-Ral.Z"  and  look  for  "papers"



Version 0.6               Posted: 20 August 1993                        14







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     folders,  the  server is sorted by EA paradigms, ie. "/pub/GA/papers"
     contains papers related to GAs, etc.).  A strongly  recommended,  and
     quarterly  updated,  report is a list of current applications of GAs,
     EP and ESs; get "/pub/EA/papers/ea-app.ps.Z" (SyS-2/92).

 Bionics Reports
     The Bionics and Evolution  Techniques  Laboratory  at  the  Technical
     University   of  Berlin  maintains  an  anonymous  ftp  server:  ftp-
     bionik.fb10.tu-berlin.de (130.149.192.50).  On  ftp-bionik  you  find
     reports   and   software,  related  to  Evolutionary  Algorithms  and
     Artificial Neural Networks.


A15) Other sources of information?
A15.1) Electronic Digests?
 Genetic Algorithm Digest
     The GA research community exchanges news, CFP's,  etc.  through  this
     digest,  currently moderated by Connie Ramsey and formerly by Alan C.
     Schultz, Naval Research Laboratory, Washington, DC 20375-5000, USA.

     A recently published statistic (v7,i3) states that GA-digest is  sent
     out to world-wide 1800 addresses in 28 countries.

     o  Send submissions to <[e-mail]>

     o  Send administrative requests to <[e-mail]>

     o  Anonymous   ftp    archive:    ftp.aic.nrl.navy.mil    (Info    in
    "/pub/galist/FTP")

     o  Conference announcements "/pub/galist/information/conferences"

 Artificial Life Digest
     The Alife research community exchanges news, CFP's, etc. through this
     digest, edited by Liane Gabora and Rob Collins of the Artificial Life
     Research Group at UCLA.

     o  Send submissions to <[e-mail]>

     o  Send administrative requests to <[e-mail]>

     o  Anonymous   ftp  archive:  ftp.cognet.ucla.edu  (128.97.50.19)  in
    "~ftp/pub/alife"

 Evolutionary Programming Digest
     The digest is intended to promote discussions  on  a  wide  range  of
     technical  issues  in  evolutionary  optimization, as well as provide
     information  on  upcoming  conferences,  events,  journals,   special
     issues, and other items of interest to the EP community.  Discussions
     on all areas of  evolutionary  computation  are  welcomed,  including
     artificial  life,  evolution strategies, and genetic algorithms.  The



Version 0.6               Posted: 20 August 1993                        15







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     digest is meant to encourage interdisciplinary communications.   Your
     suggestions and comments regarding the digest are always welcome.

     o  Send        administrative       requests       to       <ep-list-
    [e-mail]>

     To   subscribe   to   the   digest,   send    mail    to    <ep-list-
     [e-mail]>  and include the line "subscribe ep-list"
     in the body of the  text.   Further  instructions  will  follow  your
     subscription.   The  digest  is  moderated  by  N. Saravan of Florida
     Atlantic University.

A15.2) Electronic Mailing Lists?
 Genetic Programming Mailing List
     The GP community uses this list as a discussion forum, news  exchange
     and  FAQ  distribution channel, edited by John Koza and James Rice at
     Stanford.

     o  Send submissions to <[e-mail]>

     o  Send    administrative    requests    to     <genetic-programming-
    [e-mail]>

     o  Anonymous   ftp   archive:   ftp.cc.utexas.edu   in  "pub/genetic-
    programming"

     o  Contains a lengthy, but "mostly interesting" FAQ by James Rice  on
    GP related subjects.

 Tierra Mailing List
     Thomas  Ray's  Tierra is discussed elsewhere (cf Q4.1); here's how to
     obtain Tierra electronically and get in contact with other users.

     o  Send submissions to <[e-mail]>

     o  Send       administrative       requests        to        <tierra-
    [e-mail]>

     o  Anonymous   ftp  archive:  tierra.slhs.udel.edu  (tierra,  almond,
    beagle, etc.)

A15.3) Electronic Access to Research Institutes?
 The Santa Fe Institute
     The Santa Fe Institute Studies in the Sciences  of  Complexity  (SFI)
     issues   a   recommended  series:  SFI  Studies  in  the  Science  of
     Complexity, published by Addison Wesley and maintains  a  well-sorted
     ftp server with EC related material.

     o  Send administrative requests to <[e-mail]>

     o  Anonymous ftp archive: ftp.santafe.edu in "/pub"



Version 0.6               Posted: 20 August 1993                        16







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     Additionally,  you  can  telnet bbs.santafe.edu and login as "bbs" to
     get access to the Santa Fe Institute's bulletin  board  system.  It's
     administrator  Scott  D.  Yelich  <[e-mail]> is actively
     seeking information on any complex system information  for  the  BBS.
     This includes CAs, GAs, Alife or almost anything else related to this
     field. You will also find infos on SUMMERSCHOOLs held by the SFI.

 The Australian National University (ANU)
     The Bioinformatics facility at Australian National University has set
     up  an  anonymous  ftp  server,  that  contains  EC related material,
     maintained by David G. Green.

     o  Send    administrative    requests    to    David     G.     Green
    <[e-mail]>

     o  Anonymous        ftp       archive:       life.anu.edu.au       in
    "/pub/complex_systems/alife"

     o  Gopher protocol: Besides direct access to all FTP information, the
    gopher  server  offers  on-line access to relevant newsgroups, on-
    line  databases  and  direct  links  to   relevant   international
    services.

      Name=Complex systems
      Host=life.anu.edu.au
      Type=1
      Port=70
      Path=1/complex_systems

     o  World  Wide  Web protocol: Besides access to all of the above, the
    hypermedia server offers  introductory  tutorials,  preprints  and
    papers     on-line.     The    URL    for    this    service    is
    "http://life.anu.edu.au/complex_systems/complex.html" or link  via
    the servers home page "http://life.anu.edu.au/".

A15.4) Relevant Electronic News and FAQs?
     Besides   the   obvious   comp.ai.genetic,  there  exist  some  other
     newsgroups that sometimes carry EC related topics:

     o  comp.ai (FAQ in news.answers, comp.answers)

     o  comp.ai.fuzzy (FAQ in news.answers, comp.answers)

     o  comp.ai.neural-nets (FAQ in news.answers, comp.answers)

     o  comp.robotics (FAQ in news.answers, comp.answers)

     o  comp.theory.cell-automata (no FAQ)

     o  comp.theory.dynamic-sys (no FAQ)




Version 0.6               Posted: 20 August 1993                        17







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     o  comp.theory.self-org-sys (no FAQ)

A15.4) What about all these Internet Services?
 Access Tools
     Most of the recently published  books  about  the  Internet  describe
     these  tools  in  detail. Kehoe (1992), the first book to appear, was
     offered first in a free electronic version  over  the  Internet.  The
     first  edition,  January  1992,  is  still available online from many
     anonymous ftp  archives  around  the  world,  in  a  directory  named
     something like pub/zen/; read files stored there for help producing a
     printed document. Krol (1992) has received excellent reviews.

     All the tools mentioned in this section share the quirk that they are
     actually two things: a "server" or "daemon" program that runs all the
     time on a host computer and accepts  requests  to  connect  over  the
     Internet,  and  a  "client"  program that people use to connect to or
     access these servers. A few host computers  in  this  FAQ  allow  the
     public  to  telnet  to  the  host,  and then use the host computer to
     access servers via either gopher  or  WAIS.  These  arrangements  are
     offered  as  a courtesy to those people who do not have the necessary
     client software on their own computers, and want to try  these  tools
     before  going  to  the  trouble  of  installing  the  client software
     themselves.

 Telnet
     Telnet is a protocol whereby someone who is logged onto an account on
     a computer with full Internet access can access another computer over
     the Internet and login there, assuming he or she has login privileges
     on that computer as well. Anonymous telnet sessions are generally not
     permitted, but occasionally usernames  are  created  with  restricted
     privileges, for use by the Internet public.

 Anonymous Ftp
     Ftp  is  a  protocol  for  file transfers between computers with full
     Internet access. Ftp allows people to transfer files to  and  from  a
     remote  computer  on  the  Internet, assuming they have privileges on
     both the local and remote computers. "Anonymous ftp"  indicates  that
     people may login to the remote system with the userid "anonymous" and
     an arbitrary password. By convention,  anonymous  ftp  users  provide
     their  e-mail  addresses when asked for a password. This is useful to
     those archive managers who must justify  to  their  bosses  the  time
     spent  providing  this  free (but not cheap) service.  Some computers
     set restrictions on when transfers may be made from  their  archives,
     and  most prefer that large transfers be made only during off- hours.

     Bitnet does not support telnet or ftp sessions, but many Bitnet nodes
     are  also Internet nodes, and so do support telnet and ftp. For those
     people who  only  have  access  to  computers  on  Bitnet,  Princeton
     University    offers    a    file   transfer   service   by   e-mail.
     <[e-mail]> will send a help file in response to the message
     "help". There is also a BITFTP server in Germany: <bitftp@dearn> from



Version 0.6               Posted: 20 August 1993                        18







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     within BITNET/EARN/NetNorth or <[e-mail]> from the  Internet.
     This  server should be used only for ftp requests involving transfers
     within Europe. Those people who have neither full Internet access nor
     an  account  on  a  Bitnet node can retrieve files from anonymous ftp
     archives by e-mail courtesy of <[e-mail]>,  which  will
     send  instructions  in  response  to  the  words "help" and "quit" on
     separate lines of an e-mail message.

     Here is an example of how to send for a document on  the  archive  at
     rtfm.mit.edu.  This example will return a list of all the FAQs stored
     in the directory pub/usenet/news.answers/, and a copy of this  guide.
     Send the message

      send usenet/news.answers/index
      send usenet/news.answers/ai-faq/genetic/part1
      send usenet/news.answers/ai-faq/genetic/part2
      send usenet/news.answers/ai-faq/genetic/part3

     by e-mail to <[e-mail]>.

 Gopher
     Gopher  is a user-interface program that makes ftp and other types of
     connections for computer users when they select an item in a menu. It
     is  an  easy way to get stuff off the Internet without having to know
     where the stuff lives. Gopher is free, and there  are  nice  versions
     for  most  types of computers, especially Unix workstations and Macs.
     It was invented at the University of Minnesota; current versions  can
     be  retrieved  via anonymous ftp from boombox.micro.umn.edu. The name
     is a clever pun on the "go-for" person who runs  errand  for  people,
     and on the burrowing rodent, which pops down a "hole" in the Internet
     and comes back up who-knows-where.   comp.infosystems.gopher  is  the
     newsgroup  for  gopher-related  issues  in  general. The FAQ for this
     group    is    stored     on     rtfm.mit.edu     in     the     file
     "pub/usenet/news.answers/gopher-faq".   There is an entire chapter on
     gopher in Krol (1992).

 Archie
     Archie is a program that helps people locate software in any  of  the
     thousands  of  anonymous ftp archives around the world. A copy of the
     software can be retrieved via anonymous ftp from any  archie  server,
     including  ftp.cs.widener.edu,  in  the  /archie/clients/  directory.
     There are versions of archie for all sorts of Unix systems,  as  well
     as  VMS and PCs running a variety of network software. Good places to
     look for help installing archie on any  computer  are  from  <archie-
     [e-mail]>  or bionet.users.addresses. Documentation is available
     via anonymous ftp on ftp.ans.net in the  /pub/archie/doc/  directory,
     or by e-mail from <[e-mail]>.

     Archie  can  be  used  via  e-mail,  by sending e-mail with a list of
     commands to <[e-mail]>. For details, send the  command  "help".
     Due to the high demand for this service, the archie administrators on



Version 0.6               Posted: 20 August 1993                        19







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     ans.net prefer that requests be made via e-mail.

 Veronica
     Veronica  is  the  Very  Easy  Rodent-Oriented  Net-wide   Index   to
     Computerized  Archives.  The  reason Veronica's name is so long is so
     that it can play on the concepts of both gopher and archie. (Remember
     the  comic  book couple Archie and Veronica? Veronica does for gopher
     what archie  does  for  anonymous  ftp.)  Veronica  searches  through
     hundreds  of gopher holes looking for anything that matches a keyword
     supplied by the user, and assembles a list  of  gopher  servers  that
     contain  items of interest.  Note: Veronica checks *titles* of gopher
     items only, not their contents.  Veronica was only invented  in  late
     1992, but already seems very promising.

 WAIS
     WAIS  stands  for  Wide Area Information Servers. The idea is to make
     anonymous ftp archives more accessible by indexing their contents and
     making  those  indexes searchable with software distributed to anyone
     on the Internet. The user interface is simple so far, but the concept
     is  so  powerful  that every person with an anonymous ftp archive has
     spent part of this past year building WAIS indices of  all  available
     material  (software,  data,  documents and other information). In the
     course of all this effort a great deal of  information  has  suddenly
     become publicly available all in the past year.

     Good  WAIS  client  programs for the Mac (WAIStation) and PC (PCWAIS)
     are available on the anonymous ftp  archive  at  think.com.  If  your
     computer  has  full  Internet  access, you can try out WAIS on a Unix
     system,  courtesy  of  Thinking  Machines  Corp.,  by  telnetting  to
     quake.think.com. Use the username "wais" and give your e-mail address
     as the password. See the  newsgroup  comp.infosystems.wais  for  more
     details. The FAQ for this group is stored on rtfm.mit.edu in the file
     "pub/usenet/news.answers/wais-faq/getting-started".

 The World-Wide Web (WWW)
     The  World-Wide  Web  is  yet  another  tool  for  gathering   useful
     information  from  the  Internet.  It  was  invented  at the European
     Particle Physics Laboratory (CERN), Switzerland. The Web looks like a
     document  that users can open and read, but clicking on certain words
     causes other documents to be retrieved and opened for inspection. The
     most  powerful  aspect  of  the Web at present is the ease with which
     seamless, attractive online documentation can  be  created,  that  is
     easy  to  find and browse, no matter where on the Internet the actual
     documents are. You can try the  Web,  courtesy  of  CERN:  telnet  to
     info.cern.ch   (no  username  needed).   WWW  features  a  remarkably
     beautiful X interface called xmosaic(1).

     References

     Kehoe, B.P. (1992) "Zen and the Art of  the  Internet:  A  Beginner's
     Guide  to the Internet", 2nd Edition (July). Prentice Hall, Englewood



Version 0.6               Posted: 20 August 1993                        20







FAQ(2/3)                          ANSWERS                         FAQ(2/3)



     Cliffs, NJ. 112 pages. The 1st Edition, (February)  is  available  in
     PostScript  format via anonymous ftp from ftp.cs.widener.edu and many
     other Internet archives.

     Krol, E.  (1992)  "The  Whole  Internet:  Catalog  &  User's  Guide".
     O'Reilly & Associates, Inc., Sebastopol, CA. 376 pages.

     LaQuey, T. and J.C. Ryer (1992) "The Internet Companion: A Beginner's
     Guide to Global Networking". Addison-Wesley Publishing Co.,  Reading,
     MA. 208 pages.

     Smith,  Una  R. (1993) "A Biologist's Guide to the Internet."  Usenet
     sci.answers.  Available  via  anonymous  ftp  from  rtfm.mit.edu   in
     pub/usenet/news.answers/biology/guide. ~20 pages.








































Version 0.6               Posted: 20 August 1993                        21