April 19, 2012
to be posted on silentnursery.com and nobabies.net
Robert M. French
Universite de Bourgogne
Dijon, FRANCE
Robert.french@u-bourgogne.fr
Dear Robert French,
As a first year medical student I looked at my anatomy book and counted the knuckles on the fingers and the number of tendons leading to them. There were not as many as two tendons per joint so there had to be actions that were not possible. I was able so to contort my fingers as to discover a few but none so elegant as the one you describe (Robert M. French Dusting off the Turing Test SCIENCE vol. 336 no. 6078 April 13, 2012 page 163) as something a human could find out but a computer would not be likely to unless a human found it first.
I very much liked your article, but in the particular instance, it seems to me that a computer might indeed be able to discover the same thing that a desultory medical student calculated.
I had taken a look at the question of whether anybody ever came up with a new idea, or whether a computer could ( http://nobabies.net/Open%20letter%20to%20Barry%20Cooper%20about%20incomputable%20insight.html and http://nobabies.net/More%20thoughts%20on%20Alan%20Turing.html) do a computation that resulted in an increase in information. The topic is a bit moot. I have (as recounted on the same site) been inviting people to consider the proposition – obvious to me and supported by massive evidence – that in the long run normal fertility requires moderately close kinship. Few can grasp the concept. Thanks for a good read.
Sincerely,
M. Linton Herbert MD
There have been 51,566 visitors so far.