Sarah's Chess Journal my journal, blog, web log, blog.....about The History and The Culture of Chess |
Turing Test March 21, 2004 |
|
Bletchley Park
The Bletchley Park Mansion was owned by financier, Sir Herbert Leon from 1883 and 1926. During this time he renovated and enlarged the mansion in a hodgepodge of Gothic, Tudor and Dutch Baroque architectural styles.
During WWII, the British government set up a code-breaking network based at Bletchley Park. It's crowning achievement was the breaking of the German ENIGMA code.
Even before 1939 the Germans had already developed an encryption devise that could scramble a message in 150,000,000,000,000,000,000 ways using an invention called the ENIGMA machine. The Nazis were so confident in the impossibility to break this encryption that when it was finally cracked, they remained totally unaware of the fact. The original ENIGMA machine was invented around 1923 by Dr. Arthur Scherbius, who had hoped to sell it to German businesses to secure their own transmissions. By 1928, the German navy had developed it's own version and by 1933, the German air force had theirs. Over the years the Germans made the machine increasingly more sophisticated. The British were aware of it's existence since 1931 when a German spy named Hans Thilo Schmidt produced a photograph of the machine's operating manuals.
The ENIGMA machine looked a bit like a typewriter sitting in a wooden attache case.
How it worked The keyboard sent an electric current through a set of rotors and a plugboard to light up the 'code' alphabet. The Germans changed the order of the rotors, their starting positions and the plugboard connections daily. In order to decipher a message sent using the ENIGMA, it was necessary to know all these details.
How it was cracked During the 1930s, even before the war broke out, Polish cipher experts [ Marian Rejewski, Jerzy Rozycki and Henryk Zygalski ] had been secretly trying to crack the code. They were able to obtain drawings and from them build models of the ENIGMA machine and pass this information to the British and French Intelligence. This information provided a necessary key or insight into the German code system. The British Intelligence referred to the code breaking activity as Ultra.
In 1940, Alan Turing invented a machine to facilitate the code breaking work. He called it a bombe, a term already used by the Poles. His device was then referred to by others as a Turing bombe. Bombes were huge, noisy, electro-mechanical machines which could check through combinations of letters far quicker than a human being could. When the bombe stopped, this meant that the code-breakers' guesses were right. Hundreds of such machines were utilized at Bletchley Park. The Turing Bombe was, in fact, a proto-type computer.
Other factors, such as the recovery of an actual ENIGMA machine along with the code book were essential developments in the code-breaking success. But breaking the code itself was just an element of the job. These codes had to be not only deciphered daily, but messaged had to be intercepted, usually through Morse Code, with zero tolerance for error, deciphered, then translated and evaluated. Almost 3,000 messages a day passed through Bletchley Park. At it's height, 12,000 people were employed there, living and working in cramped, hastily constructed, small wooden or concrete buildings known as Huts. The work was tedious, exacting, repetitive and difficult. It was also top-secret.
The importance of the role of Bletchley Park has been debated over the years. Probably the outcome of the war wouldn't have been different without it's existence, but certainly the duration of the war as well as the outcome of many operations would have changed significantly. Bletchley Park undoubtedly saved many lives, both Allied lives and, indirectly, German lives.
And what does all this have to do with chess??
It has to do with Alan Turing and the first chess computer.
Alan Mathison Turing
Alan Turing conceived of the idea of the modern computer already in 1935 while at Cambridge University, . His concept model consisted of unlimited memory with a scanner constantly reading the memory and building on what it finds. Turing's stored-program concept in effect is a self- improving program. Turing's computing machine of 1935 is now known simply as the universal Turing machine.
Turing applied his ideas, as well as those of Babbage and others, into making the bombes. It was during his time at Bletchley Park that Turing first expressed publicly his basic ideas on computers and artificial intelligence:
Today's artificial intelligence programs make major use of both machine learning and heuristics, but computers then were hardly imaginable much less formulations on their nature. Turing also expressed his ideas using chess as a measure of machine intelligence. While, in principle, a chess-playing computer could play by searching through all the available moves, in practice this is impossible, since it would involve examining an astronomically large number of moves. To reduce and refine the search. heuristics are necessary. Turing experimented with two heuristics that later became common in artificial intelligence programming:
In 1950 Turing developed the test for computer intelligence. It's name, the Turing test, has gained general acceptance in everyday speech. As described by Turing, it requires three participants: an interrogator, a human and a computer. By asking questions, communicating only by keyboard and screen, the interrogator must determine which of the two is human and which is a computer. The interrogator may ask any question and the computer is programmed to fool the questioner, even lying. The human must try to convince the interrogator that he is indeed human.. A variety of people play the roles of interrogator and human, and if a sufficient number of interrogators is unable to distinguish the computer from the human, then it is to be concluded that the computer is an intelligent, thinking entity. In 1991 Hugh Loebner, a New York businessman, created the annual Loebner Prize competition with a $100,000 prize for the first computer program to pass the Turing test. $2,000 is awarded each year for the best effort. No artificial intelligence program has come even close to passing an undiluted Turing test.. During the same year that he devised the Turing Test, Alan Turing wrote the first computer chess program.
So, Turing wrote the instructions - the program- for his theoretical computer and did the math just as the computer would if it existed, acting for all practical purposes as the CPU. He spent about a half hour computing each move. Turing's program was named TurboChamp and it lost it's only published game - against a colleague of his, Alick Glennie, who wrote the first computer compiler, a high level language called Autocode.
The game
Turing never finished his chess program. 1952 proved to be the peak - and the end - of his career According the an article in Wikipedia:
Back to chess
In 1939, the Chess Olympiad was held in Buenos Aires. During the contest Germany invaded Poland and war was declared in Europe. The British team, composed of Hugh O'D. Alexander, Harry Golombek, Stuart Milner-Barry, Sir George Thomas and Baruch Wood and who had just qualified from the six-game preliminary section to the finals of the Hamilton-Russell Cup, withdrew from the tournament and caught the first ship back to England. Alexander, Golombek and Milner-Barry were quickly recruited to serve under Alan Turing at Bletchley Park to help crack the German naval ENIGMA codes.
England was the only team to bow out of the Olympiad. Germany won the Hamilton-Russell Cup just a half point ahead of Poland whom they had just invaded and conquered. Many players decided not to return to their countries after the tournament. The entire Polish team, all Jews, stayed in South America. Austrian masters, Albert Becker and Erich Eliskases, who were on the German team, also remained as did the Swedish master, Gideon Stahlberg.
Germany won the tournament, but lost the war.
Resources and references:
http://www.bletchleypark.org.uk/ http://www.cl.cam.ac.uk/Research/Security/Historical/hinsley.html http://www.iwm.org.uk/online/enigma/enigma7.htm http://www.chessbase.com/columns/column.asp?pid=102 http://www.geocities.com/SiliconValley/Lab/7378/comphis.htm http://en.wikipedia.org/wiki/Alan_Turing http://www.alanturing.net/turing_archive/pages/Reference%20Articles/what_is_AI/What%20is%20AI03.html http://www.cs.usfca.edu/www.AlanTuring.net/turing_archive/index.html |
links personal Sarah's Serendipitous Chess Page The Life and Chess of Paul Morphy Sarah's Chess History Forum chess - general Chesslinks Worldwide chess - history Mark Week's History on the Web Chess Journalists of America Chess History Newsgroup Hebrew Chess Chess Tourn. & Match History Super Tournaments of the Past La grande storia degli scacchi Bobby Fischer Bil Wall's Chess Pages |
comments |