Ivar Ylvisaker wrote:
[...]
IY>Actually, if I had to define intelligence, I'd start with
IY>language. If I or someone can talk with an object, then it
IY>seems reasonable to define the object as intelligent. This
IY>was Turing's definition of an intelligent machine.
[...]
No, it wasn't. Turing was looking at the problem of
establishing that a machine is intelligent. He came up with a
*sufficient*, though not *necessary*, condition under which we
would have to admit such.
The other amusing thing about Turing's original paper is that he
never explicitly changed the rules of the game from a gender
discrimination task when he substituted in a machine for one of
the participants.
Wesley
This archive was generated by hypermail 2b29 : Mon Oct 23 2000 - 10:49:47 EDT