To hear more of Marvin Minsky’s stories, go to the playlist:    • Marvin Minsky  The amazing videophon...  

The scientist, Marvin Minsky (19272016) was one of the pioneers of the field of Artificial Intelligence, having founded the MIT AI Lab in 1970. Since the 1950s, his work involved trying to uncover human thinking processes and replicate them in machines. [Listener: Christopher Sykes; date recorded: 2011]

TRANSCRIPT: [CS] Right, so you... you were haunted by the fact that you couldn’t make this work for any old triangle.

Right, but it seemed, if it was true for a... equilateral one and a right angle one and a pentagon... triangle in a pentagon, then it must be true for all of them, but I couldn’t prove it and then... somehow I got a message... I don’t remember how I learned but... that this fellow Dyson had extended this and managed to prove either that or a similar thing about four points on a big circle, a square. Anyway, it was clear to me that, if it was true for these three kinds of triangles, it must... there’s an extra degree of freedom here, that it’s true for a whole class of things and so maybe it’s also true for a particular set of four points instead of a lot of different sets of three points and Dyson proved that in a paper that I actually retrieved yesterday and...

But there was another mathematician who proved something similar around the same time. But anyway, when I read Dyson’s paper I was so impressed that anybody could actually produce such an amazingly complicated proof of something at all and my reaction was not to learn how to do that, but to say: 'Oh he has a different skill set from me and I don’t have that particular skill set so I’ll stop trying to push this anymore because there’s somebody in the world who’s better at it.' So I’ll... I'll get... I’ll print out that part of Dyson’s... Dyson actually mentioned me and Kakutani in his paper of... as having started this thing.

And... so... anyway, it was one of my early experiences of saying: 'Oh look, I would like to do this kind of fixed point topology, but since there’s someone better at it, why...' It’s the cowardice principle: Why bother to do something if somebody else is already better at it? And so I’ve always been excited to see what Dyson does next and I’ve followed his career quite a lot so I did read some of the early Orion things, but I never actually saw the final result of the theory and the politics that must have gone into getting it dismissed. Of course once you’re on the moon you can do it, because there’s nobody to kill, but it’s pretty hard to get a lot of people up on the moon.