Didn’t Samuel Solve That Game?

  • Jonathan Schaeffer


The road to constructing the world champion checkers player might have been an easy one to follow, except for a brief moment of human fallibility. That one moment of human weakness would continue to haunt me for years to come.


Personal Computing World Championship Checker Program Artificial Intelligence Research Strong Player 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    The first fraudulent chess-playing machine was The Turk, built in 1769 by the Hungarian engineer Baron Wolfgang von Kempelen. During its illustrious career, the machine had the distinction of defeating many well-known players including Napoleon Bonaparte.Google Scholar
  2. 2.
    Shannon’s seminal paper laid the groundwork for chess and checkers programs. Reading the paper forty-five years after it was first published makes one appreciate how important the ideas were then and just how relevant they are today. See: “Programming a Computer for Playing Chess,” Philosophical Magazine, 41, no. 314(1950): pp. 256–275. Alternatively, look at Shannon’s “A Chess-Playing Machine,” Scientific American, 182(1950): pp. 48–51.Google Scholar
  3. 3.
    Excerpt from A Boy From Emporia, the unpublished autobiography of Arthur L. Samuel.Google Scholar
  4. 4.
    There’s a minor error in the quotation. Charles Babbage (1791–1871) was a professor of mathematics at Cambridge University, in England. His difference engine and analytical engine were the forerunners of modern computing.Google Scholar
  5. 5.
    Excerpt of a letter from Samuel given in “Computer Games,” Personal Computing, March 1980, p. 78.Google Scholar
  6. 6.
    Excerpt from A Boy From Emporia. Google Scholar
  7. 7.
    In The Art of Computer Programming, Volume II: Searching and Sorting (AddisonWesley, 1973), Donald Knuth writes (pp. 540–541), “The idea of hashing appears to have been originated by H.P. January 1953....At about the same time the idea of hashing occurred independently to another group of IBMers: Gene M. Amdahl, Elaine M. Boehme, N. Rochester, and Arthur L. Samuel, who were building an assembly program for the IBM 701.”Google Scholar
  8. 8.
    Arthur Samuel, “AI, Where It Has Been and Where It Is Going,” International Joint Conference on Artificial Intelligence, 1983, pp. 1152–1157.Google Scholar
  9. 9.
    Checkers has three categories of players: minor, major, and master. It appears that Nealey was considered to be a major player by all except himself.Google Scholar
  10. 10.
    Arthur Samuel, in Computers and Thought, Edward Feigenbaum and Julian Feldman (eds.), MacGraw-Hill, 1963, pp. 103–105.Google Scholar
  11. 11.
  12. 12.
    John Pfeiffer, “Man vs Machine in the Mechanical Age,” Popular Mechanics, August 1964, pp. 52–57, 172–173.Google Scholar
  13. 13.
  14. 14.
    Fortman is referring to chess rankings, which are (in order): world champion, grandmaster, master, candidate master, class A, class B, class C, class D, and class E (beginner). Statistically, a candidate master should win a sixteen-game match against a B class player by a score of fifteen wins to one. In other words, Fortman doesn’t seem to be impressed with PAASLOW.Google Scholar
  15. 15.
    Burke Grandjean, “A Checker Debate,” Personal Computing, May 1980, p. 83.Google Scholar
  16. 16.
    “Computer Checkers,” Personal Computing, July 1979, p. 88.Google Scholar
  17. 17.
    Letter by Marion Tinsley, published in Scientific American, August 1980.Google Scholar
  18. 18.
    Herschel Smith, personal communication, undated.Google Scholar
  19. 19.
    Eric A. Weiss, “Eloge: Arthur Lee Samuel (1901–1990),” IEEE Annals of the History of Computing, 14, no. 3(1992): pp. 55–69.CrossRefGoogle Scholar
  20. 20.
    Most of the quotes were collected by John R. Gibson and reported in: “No Respect: A Short History of Computer Checkers,” The Keystone Checker Review, January 1993, pp. 476–477.Google Scholar

Copyright information

© Springer Science+Business Media New York 1997

Authors and Affiliations

  • Jonathan Schaeffer
    • 1
  1. 1.Department of Computing ScienceUniversity of AlbertaEdmontonCanada

Personalised recommendations