experts--what are they good for?

Via Marginal Revolution I discovered this pair of posts by Daniel Drezner regarding experts and the value of their predictions. Here's the bottom-line: experts are often worse than non-experts when it comes to predicting future events within their area of expertise. This effect increases as the fame of the forecaster increases and is worse among specialists and thinkers who have a single all-encompassing theory or idea that informs their views than among generalists and those who don't look at the world through any single dominant philosphical framework.