Posted to Robitron
Richard, you must disabuse yourself of the mistaken notion that the judge in the Turing Test is not supposed to know that he or she is selecting a human from a human/machine pair. Turing EXPLICITLY describes a "game" the object of which is for the judge to chose which of a pair is the woman, knowing full well that one of the pair is a woman and the other is a man. That's why the **game** was being played, it was B.T.V (before television), and people had to amuse themselves. Can the Man fool the judge? Can the Woman convince the judge? Let's play the game and have such fun.
Turing EXPLICITLY states:
"The new form of the problem can be described in terms of a game which we call the 'imitation game." It is played with three people, a man (A), a woman (B), and an interrogator (C) who may be of either sex. The interrogator stays in a room apart front the other two. The object of the game for the interrogator is to determine which of the other two is the man and which is the woman. He knows them by labels X and Y, and at the end of the game he says either "X is A and Y is B" or "X is B and Y is A." .... "We now ask the question, 'What will happen when a machine takes the part of A in this game?/ Will the the he interrogator decide wrongly as often when the game is played like this as he does when the game is played between a man and a woman?"
How can this be stated more plainly? It is true that Turing did ** not ** write "and B can be either a man or a woman here, just that it is a human," but some things are just understood. Turing, discussing memory requirements, wrote further on in the article: "I should be surprised if more than 10^9 [binary digits] was required for satisfactory playing of the imitation game, at any rate against a blind ***man***.'
Note - blind ****man**** not blind ***woman*** (as if it mattered).
"As a twist, Turing said, the woman should always tell the truth and the man should always lie."
Turing wrote no such thing. He wrote " The object of the game for the third player (B) is to help the interrogator. The best strategy for her is probably to give truthful answers."
This is not a prohibition, it is merely a comment on strategy. I (and I am sure that Turing) could envision circumstances in which it were best for the woman to lie (eg, a world class runner might misrepresent the time it takes her to run a mile because it might seem improbable that a woman could do so that quickly).
The fact that it is easier to fool a naive judge, one who does not know that s/he is selecting the human from a human/machine pair, is not grounds for stating that that is the test. The test was not designed to be easy to pass, it was designed to be a good.
As a matter of fact, until Loebner Prize 14, the original test had ***not*** been held. It was only on Sunday, September 19, 2004 that a true Turing Test was held. By this I mean that judges were presented with paired comparisons. There were four separate tests. In each test one of which one terminals was controlled by a human, the other by a computer. This was the first time that the method of paired comparisons (the method that Turing described) was actually used.