Robin Hanson comments on another mind change, this time the disappointment of psychiatrist Randolph Nesse, described as a Darwinian medicine pioneer, with expertise.

I used to believe that you could find out what is true by finding the smartest people and finding out what they think. However, the most brilliant people keep turning out to be wrong. . .

I also used to believe that you could find out what is true by relying on experts — smart experts — who devote themselves to a topic. . .

I used to believe that truth had a special home in universities. After all, universities are supposed to be devoted to finding out what is true, and teaching students what we know and how to find out for themselves. Universities may be best show in town for truth pursuers, but most stifle innovation and constructive engagement of real controversies, not just sometimes, but most of the time, systematically. . .

by a process of unintentional selection, diversity of thought and topic is excluded. If it still sneaks in, it is purged. The disciplines become ever more insular. And universities find themselves unwittingly inhibiting progress and genuine intellectual engagement. University leaders recognize this and hate it, so they are constantly creating new initiatives to foster innovative interdisciplinary work. These have the same lovely sincerity as new diets for the New Year, and the same blindness to the structural factors responsible for the problems.

Hanson recommends Idea Futures.

Our policy-makers and media rely too much on the "expert" advice of a self-interested insider's club of pundits and big-shot academics. These pundits are rewarded too much for telling good stories, and for supporting each other, rather than for being "right". Instead, let us create betting markets on most controversial questions, and treat the current market odds as our best expert consensus. The real experts (maybe you), would then be rewarded for their contributions, while clueless pundits would learn to stay away. You should have a free-speech right to bet on political questions in policy markets, and we could even base a new form of government on idea futures.

These thoughts are loosely related to those of Kevin Kelley discussed in Old School. The problem each grapples with is the lack of heuristic diversity even when there are collaborations. In Nesse's words: "most stifle innovation and constructive engagement of real controversies, not just sometimes, but most of the time, systematically". There are group dynamics that invariably bring this sort of situation about except in fanciful situations where a wise philosopher king dictates other outcomes. There are problems with that model too, in addition to the difficulty of finding and empowering these exceedingly rare kings.

I see more hope in ad-hoc teams of heuristically diverse members. The task orientation and short duration of such groups prevents some pathologies, and the diversity of methods as well as knowledge improves problem solving by jolting semi-somnolent minds out of well worn grooves.

Hanson's idea markets have an additional virtue of being more persistent, and so able to deal with some larger and longer term problems as new data appears and old interventions are evaluated. Repeated assembly of ad-hoc groups could do something similar. I'm undecided about the relative virtues of these approaches. Processing . . .

Update: This seems relevant.

The cheeky YouTube video that explains Carmun describes the service as "Wikipedia meets Facebook". But essentially its a social network for people who have to write term papers. There's lots of tools for asking peers questions, managing bibliographies, and some del.icio.us-esque tools for saving online sources.

This is one of a dozen or so micro-startups orbiting the Betaworks incubator in Manhattan. With all the attention focused on MySpace and Facebook, sometimes I feel that these lightweight sub-100,000 user social sites are the real Web 2.0 story. They take weeks to months to launch, are perfectly tailored to their audience, evolve in a sort of perpetual beta, and don't suffer from the "eveything but the kitchen sink" disease the big, broad sites do.

How's this for ad-hoc, task-oriented teams of heuristically diverse members?