While this book is full of interesting anecdotes and research, it feels a bit thrown together.

Some of the research seems to contradict his main thesis -- that groups of people are often smarter than the smartest individual among them. He seems to conclude that the only way to avoid traffic jams is to put people in smarter cars.

Equally as frustrating, one of the anecdotes he uses throughout the book -- the story of the Scorpion submarine being (nearly) located by the average of several experts' predictions -- seems to be a one-off. It is a non-repeatable experiment, yet he returns to this again and again.

Nevertheless, the book is worth reading for what he has to say about the best conditions for working in groups. The chapter on diversity is definitely worth reading.

It's also worth finishing the book. He addressed most of the questions I had towards the end of the book. Not always to my satisfaction, but at least he acknowledged that they existed.

I finished this book as a more of a skeptic about group-think that I was when I started it. If I learned anything it was that groups can make incredibly smart decisions under the right conditions, but getting it right is incredibly difficult.