More good videos from MIT: Stephen Wolfram speaks on his widely discussed and sublimely idiosyncratic tome A new kind of science. Exciting stuff, though I am still only halfway through and need to think carefully about what I hear – the transition from understanding snowflakes as modelled by cellular automata to more abstract examples takes some thinking.

Update two days later: This really is a great video, best absorbed in small increments. The book is fascinating, first from the viewpoint of "Wow!, how could someone spend 10 years of his time doing only this?" (aside from being CEO of a company), but a lot gets answered in the video – Stephen Wolfram is no crank, for one thing, and does not claim to have found the answer to anything. Only a new and very interesting branch of something between formal mathematics and computer (or, at least, computational) science. Makes me want to get Mathematica and start playing with patterns…

The flip side, of course, you can see from reading the reviews at Amazon, which alter between five stars and one, the latter claiming he is taking credit for lots of papers and ideas that others have produced before him. It seems to me that the five stars are by people who are relatively new to the subject, and the ones are by people who have studied some of it – meaning that in a sense, Stephen Wolfram may be the Bill Gates of computational science – not the one to come up with the idea, but certainly the one that managed to pull off the instantiation that made the difference. I guess history will be the judge. I for one found this fun – and until the chips fall down, enjoy the ride…

Side note: My wife, who does knitting and quilting, found the book fascinating because of the many interesting patterns it describes. Which got me thinking about whether it is possible to reprogram a knitting machine to do cellular automata.