michael_nielsen

This is a fun article, and there's interesting truth to it. But I wish it engaged seriously with the other viewpoint. Is there ever a good time to turn down a von Neumann for your math-focused venture? Einstein for a physics venture? Etc.

If the answer is just "You get experts whose domain expertise is more appropriate [than Einstein etc] for the particular problem under consideration" then the article becomes true but obvious and uninteresting.

The likelihood that you will get someone who understands the math _and_ the institutional politics _and_ the local specificities of Pennsylvania politics is obviously very low. Genius travels - but nearly always within a specific ambit and with a specific bag of tricks.

So I think you need to integrate very different kinds of expertise if you want to answer these questions. Though I can’t remember how/if you talk about these specific kinds of cognitive crossover in Reinventing Science …

I gave a few very examples - there are some particularly nice ones in Kasparov versus the World where amateurs contributed important ideas that the experts didn't think of.

Yes, I like that example a lot. I do disagree with the sharp division you draw between science and politics in that book (science seems more political to me, and politics at least a little more concerned with search for knowledge than you say, as best as I recall it).

I was trying to make a very specific point about the conditions under which knowledge can build. I'd be a bit more measured today, particularly understanding some of the problems around replication - many fields of science seem to barely deserve the term.

By contrast, there are parts of mathematic & physics (& a few other fields, though spottier) where it's possible to make tight arguments thousands of pages long, integrating the results of tens of thousands of people. I know of nothing remotely like that in politics.