Reinventing Discovery_ The New Era of Networked Science - Michael Nielsen [121]
p 66 a collaboration needs to know what , aration knows: This observation, often in different guises, seems to have been made many times. I first fully appreciated it after reading [28].
p 67 “If anything in my life that I’ve participated in . . .”: This quote is from a comment made by commenter AdmiralBumblebee [30] on the website reddit. It’s worth mentioning that the comment was stimulated by an early version of the material that opens chapter 2 of this book, which AdmiralBumblebee felt reflected “commercial hype” and a sponsor’s view of the game. My account is, however, not based on information from the sponsor, Microsoft, but primarily on the firsthand accounts of Kasparov and Krush, and corroborated by several other sources.
Chapter 5. The Limits and the Potential of Collective Intelligence
p 69: The Stasser-Titus experiments are described in [204], which contains many more details than my abbreviated account. A review of work following up these experiments is [203]. An informative broader summary of the way collective intelligence can fail is Sunstein’s book Infotopia [212].
p 75 the stronger players on the World Team could usually agree on which analyses were best: There was a significant exception to this, which is that early in the game Microsoft asked the World Team advisors not to consult with one another, and so they did not have the opportunity to come to agreement. But many of the stronger World Team players were in close contact, and they were frequently able to come to agreement.
p 78: On the limits to collective intelligence, and problems such as groupthink, information cascades, etc., see [99, 212, 213, 214], and references therein.
p 79: Regarding the rapid acceptance of Einstein’s ideas, it helped that leading scientists such as Lorentz and Poincaré arrived at similar conclusions at about the same time. But although Einstein’s formulation of relativity was even more radical than the formulations of Lorentz and Poincaré, it quickly became accepted as the correct way to think about relativity.
p 79: On the discovery of DNA, and Pauling’s error, see Watson’s memoir, The Double Helix [234].
p 80 “If Feynman says it three times, it’s right”: [72].
p 84: My thanks to Mark Tovey for help constructing this example on optical illusions and cognitive science.
p 85: On collaboration markets, see also [246] and [146].
p 85: The discussion of topological quantum computes is inspired by [22]. Topological quantum computers were originally proposed in a remarkable article by Kitaev [111].
Chapter 6. All the World’s Knowledge
p 91: Swanson’s discovery of the magnesium-migraine connection is described in [215], and reviewed in [216].
p 92: An interesting question about the migraine-magnesium connection is why it wasn’t discovered by, say, scientists working on epilepsy, some of whom were presumably aware of the connection of epilepsy to both migraines and magnesium deficiency. Speculating, it seems likely that the reason this connection went unnoticed is that (1) those scientists were focused mostly on understanding epilepsy, not other conditions; and (2) a single connection linking migraines and magnesium deficiency isn’t enough of a pattern to infer anything. Epilepsy is connected to many different conditions, most of which have no direct relation to one another.
p 92: On Swanson’s procedure, there is, of course, nothing new about inferring undiscovered knowledge from existing scientific knowledge. It’s standard practice in fields like my own field of theoretical physics. But Swanson’s systematic computer-mediated application of this idea in medicine was new, and foreshadowed an explosion in the use of similar data mining techniques in many areas of science.
p 93: The notion of the extended mind has been discussed