“Fatherofhollywood: the book cannot be "full of sweeping claims and generalizations" and have sound arguments. Your post has a logical inconsistency. ”
“I found this book full of sweeping claims, generalizations and is confusing in its presentation. However it made me think. Overall the writer is saying that people independently working on a problem can in a fair vote be more accurate then the smartest individual. He then quotes examples for such behavior and examples of when the crowds got it wrong when they acted not independently but in mass. I suspect that much of his arguments are sound. ”
“Groups outperform experts when there is a right answer. They cooperate & work together more than would be expected. Buyers and sellers find eachother and trade. This does not require consensus, and it can tolerate disagreement and conflict. It does not work when diverse opinions are squelched, or when people worry about what others think. Groups can be inside and/or outside an organizationi. The right conditions are diverse, independent, decentralized & there is a way of summarizing opinions. Things that require a skill, such as flying a plane, are not appropriate. Prediction markets are able to accurately guess election outcomes. Supports Eli Lilly who used internal stock markets & hypothetical pharmaceutical candidates to predict FDA approval. Groupthink makes dissent seem improbable, diversity lets individuals say what they think. Final chapter is on democracy. Missed having an index at the end so there may be an advantage to having a digital version for search. Notes are 21 ½ pages fine-print.”
“The Wisdom of Crowds and Measuring Tacit Knowledge TransferI'm reading this book as background for my consulting work. We're working on measuring tacit knowledge transfer. Specifically, our team is looking at how people can effectively transfer wisdom gained through experience to colleagues/peers/subordinates/successors that lack experience. I’m excited to see what James Surowieki has discovered. ”
“Chapter Five: Shall We Dance? Coordination in a Complex World• Defn Coordination Problem: ubiquitous in everyday operations and life. To solve it, a leader must think about what he believes but what other people think is the right answer. How does the crowd walk down the NYC sidewalk? How does a manager decide on pricing, manufacturing levels, inventory, allocating resources. Another great example is CEMEX that outfitted its cement trucks in 1997 with GPS tracking technology and put the trucks “In orbit” around a Mexican City. Efficiency and revenues skyrocketed. – Independent decision making is pointless• Persuasion versus coercion– In American culture (liberal, free thinking, society) authority has limited effectiveness reach over citizens. Because of this coordination problems require bottom-up and not top-down solutions. • Bottom-up solutions will mean people will voluntarily make their actions fit together in an orderly and efficient manner. • Conventions maintain order and stability of the crowd but they reduce the amount of thinking—you deal w/ situations w/out much thought– First come first serve seating in a subway train (coordination takes place w/out you even thinking about it so you can think about more important things such as reading the paper or drinking your Starbucks on the way to work)– The convention of movie pricing includes a fear of disruption. Why would the theater charge the same amount for Gigali as for the opening weekend of Spider-Man? Movie theaters have Uncoordinated themselves with patrons b/c of fear of disruption. Chapter Six: Society Does Exist: Taxes, Tipping, Television, and Trust? The definition of a cooperation problem is very similar to a coordination problem. However, a coordination problem will be solved even if every individual is acting only in their self-interest. Cooperation problems require people to have a broader definition of self-interest than the one that maximized profits in short term demands. Members of society must do more, with a greater risk of zero return, to solve a cooperation problem. Examples: shoveling snow off the sidewalk, paying taxes, and curbing pollution/litter. • Scientist Robert Axelrod argued that cooperation is the result of repeated interactions (alley neighborhood shoveling snow—everyone always does it). The foundation of an effective cooperation relationship is not trust but the durability (length, frequency) of the relationship. “A stable pattern of cooperation”• Why trust? Capitalism teaches us to trust one another because it is valuable. If you purchase a TV from Best Buy, you expect it to work when you get it home. If it doesn’t Best Buy will refund your money. If this wasn’t the case, Best Buy would loose revenue. Honesty pays.• Faith—example of the Quakers in 18th and 19th Century Britain. Common faith facilitated trust.• People who trust have an advantage because they spend less time protecting themselves and more time in productive/active endeavors.”
“Chapter Four• Current decentralization trends exist in Business, Sciences, and Social Sciences (coincides with the rise of the internet)– Business: reengineering has grasped the idea of self-directed work teams– Sciences: more attention to self-organizing, decentralized systems that are robust and adaptable (beehives, ant colonies)– Social Sciences: recognized the importance of Social Networks with no one person “in charge”• CIA/FBI are decentralized but failed to predict 1993 WTC bombing, Cole Bombing, Kenya Embassy Bombing, and 9-11. Why? – No system of aggregation (grouping and transmitting valuable data efficiently) • Another Example of failure of aggregation is the Iraqi Fedayeen paramilitaries after the Iraq invasion. They were quickly marginalized because they had no method of aggregation of data/knowledge/TTP’s. “All tactics and no Strategy.” By May 2003 they had faded into the background and the mujahideen had taken over as the insurgents. • Tacit Knowledge Transfer--The person closest to the problem has the greatest probability to have a solution (this is the same idea of American Democracy where local governments provide health and welfare – specifically local governments regulate public schools) • The weakness of decentralization is that valuable information (tacit knowledge and problem solving solutions) will not permeate and transmit completely through an organization. ”
“Chapter Three: Monkey See, Monkey Do: Imitation, Information Cascades, and independence.Surowiecki includes a John Mayndard Keynes quote that illustrates he danger of herding. Worldly wisdom teaches that it is better for reputation to fail conventionally than to succeed unconventionally. Herders follow the crowd because that is where it is safest. NFL Coaches don’t go for the first down when its 4th and 3 yards to go on the opponent’s 45 yard line even though Romer has mathematically proven it to be more beneficial to try for the first down. Perception has a lot to do with this. An NFL coach and a money manager make decisions based on the crowd’s perception of their performance. The decisions to punt on 4th and 3 on the opponent’s 45 yard line and the decision to buy a blue chip stock like IBM both seem rational. The fear of seeming irrational causes these types of poor decisions. The practical application in leadership and is that Good leaders must be able to separate herding from decision-making. According to Wikipedia an information cascade “is a situation in which every subsequent actor, based on the observations of others, makes the same choice independent of his/her private signal. In an informational cascade, everyone is individually acting rationally. Still, even if all participants as a collective have overwhelming information in favor of the correct action, each and every participant may take the wrong action.” It is a sequence of uninformed choices because people are copying / imitating those around them. The telecom “dark fiber” bust of the late 1990’s is a great example. [Thomas Friedman’s book, The World is Flat, spends an entire chapter discussing the importance and significance of this mistake.] Not all imitation is bad. Surowiecki discusses Slavish versus intelligent imitation. Intelligent imitation is effective because individuals make correct decisions efficiently. A cascade maintains it momentum b/c people value public information more than their private information. Finally, the more important the decision, the less likely an information cascade will take place. The crux in leadership is getting people to pay less attention to what everyone else is saying. One technique is to ensure colleges/peers are making leadership decisions simultaneously so they aren’t susceptible to herding or an information cascade. ”
“Chapter Two: The Difference Makes: Waggle Dances, The Bay of Pigs, and The Value of Diversity. Surowiecki starts the chapter with the Story of the Olds Motor Works. He describes how the crowds determine who is left "standing and in control of the market" for technology innovation. As time passes, the market (crowds) select the technologies that will flourish. However, the interesting observation is that there is no guarantee that at the end of the process the best technology will necessarily win. The process is "twofold." First, the group uncovers the potential alternatives, second the crowd decides among them. Two aspects make this system successful. First: diversity among the entrepreneurs who are proposing technology solutions. Second: the ability to kill losers quickly. This points to Jim Collin's work in Good to Great. The hedgehog concept immediately comes to mind. Companies must focus only on what they can be the best in the world at. Effective business leadership requires the ability to recognize opportunities as fitting into this hedgehog concept. Returning to Surowiecki: smaller teams and organizations must strive to foster diversity in order to improve group performance. If you achieve this diversity, groups do a better job of arriving at major decisions than a small group of one or two people regardlss of how smart those people are. The best example is Mutual Fund Mangers. Entrusted with hundreds of millions of dollars each, Mutual Fund Managers are very smart people. However, 90% failed to meet the market performance of the Wilshire 5000 between 1984 and 1999. The crowd does a better job of determining market performance. The exception is a "Genuine Expert" such as Warren Buffet. These genuine experts are few and far between. Surowiecki then goes on to discuss the negative aspects of Groupthink. Because a homogeneous group makes dissent improbable, it does not lead to the best outcome/decision. The problem is not censorship. The problem is closing people's minds to creative solutions that lead to the best outcomes. Solomon Asche's experiment of a group that is selecting lines of the same length is referenced. People are not willing to stand out if Group Think is prevalent. Diversity makes it easier for people to express what they really think. Diversity helps to preserve independence. ”
“Self Titled Chapter One: The Wisdom of CrowdsLike most good journalist, Surowiecki is able to relate to the average American reader. He starts out by illustrating the accuracy of group intelligence with the “Who Wants to be a Millionaire” tv show. The “poll the studio audience” option resulted in 91% correct answers versus phone a friend’s 65% accuracy. Surowiecki qualifies this example—it is not a statistically sound study—however it effectively supports his point. The next stand-out case study is the Stock Market and the Challenger space shuttle disaster. Long story short: four main contractors provided technology for the space shuttle. The Stock Market determined the solid-fuel booster rocked was to blame by decreasing its market value by 12% on the day of the crash. The other 3 contractors had only a ~3% decrease in value. Months after the crash, the official investigation stated the solid-fuel cells were to blame. Surowiecki uses this as an example of the wisdom of crowds. The author goes on to talk about other decision markets that measure the public’s decision (stock prices, votes, point-spreads in sports books, Google-like computer algorithms, and futures contracts). The conclusion is the method of the decision isn’t significant. Instead, satisfying the conditions of a smart grout is what’s important:DiversityIndependenceDecentralization”
“Introduction Chapter notes/thoughts: Surowiecki starts off and finished his introduction chapter with a Vignette or Case Study. I am not sure how statistically sound the submarine case study is. The author doesn’t quantify the number of groups that predicted the location of the lost U.S.S. Scorpion. He does quantify the mean of their location guesses was only 220 yards from the actual location of the lost sub. If the n was 3 groups (mathematicians, sub specialists, and salvage men) than the mean could have been a "lucky guess." Decision-Making and Problem-Solving are both touched on in the introduction. These two competencies (or characteristics) spark an interest because they are both crucial to leadership. The brief mention of Bayes Nets is also interesting to me because it can help to model decision making in organizations. A software solution called Netica helps automate this type of modeling. I’ve Netica to help understand how hierarchical government-type organizations make decisions. It is a fantastic tool for war gaming and study how an organization really works. A side-benefit of a Bayes Net analysis is discovering hidden opportunities for improvements. If you understand and graphically display how a decision is made, often you’ll see room for improvements. However, the Bayes net is also limited to the accuracy of the probabilities that are assumptions. If the probabilities are off, the entire model can be severely skewed. ”
