The evolution and complexity discussion from GECCO 2007 featuring Richard Dawkins, Lewis Wolpert, and Steve Jones, yields more interesting stuff.

Turns out that around the 1-hour mark there's a question from Ken Stanley, the developer of NEAT, who's a very nice and smart guy, and who has helped me immensely in my own line of research. Anyway, he asks:

How is it possible for 30,000 human genes to encode the connectivity of a 100 trillion-connection brain? And how is it possible for evolution to search a space of such astronomical complexity to discover such a structure?

Great question, and one I'm obviously incredibly interested in. Steve Jones gives the answer that we really don't know, that the definition of a gene may be fundamentally flawed, and that there may be much, much more information in the DNA than we currently think. He also says that you can get vast complexity from simple rules, and cites Mandelbrot sets.

Lewis Wolpert didn't impress me much with his answer. He did at least admit that we don't understand the process very well. But this was a funny bit: "Come on, you're all mathematicians here. If I've got 20,000 genes, how many possible combinations can I specify? It's billions."

Dawkins likens embryological processes to the computer code analogy. He basically says that the answer has to do with a massive amount of reuse. His answer was probably the closest to an actual answer, but I wasn't sure any of them did a really good job with the question.

It is a hard question, though.