Collective Mind - Myths and Reality

This article is a detailed answer to this article . naz2 , in my opinion, didn’t understand well the question he raised. Here is what Shuryevsky says ( original ):

In order for a crowd to be smart, it must meet the following conditions:

1. Diversity. The opinion of specialists from several applied areas will be more accurate than the opinions of specialists from one area. A team of programmers, designers, managers, users and copywriters will make the site better than each of them individually (if at all separately they will do =)).
2. Independence. That is, the designer should not impose his opinion on the programmer and vice versa. All opinions must be summed up by one person, for example, an art director.
3. Decentralization. “Strength is inherent not only to centralized structures, where decisions are often made on the basis of some of their opinions and local specifics, and not on global and looking to the future.” Open Source is an excellent example of decentrolization in action.
4. Aggregation (if you honestly did not know the meaning of this word, if you too, then read here ). As I wrote all the opinions should be summarized. This is done by managers, art directors and the like.
')
If there are translators I will be glad of your help in accurate translation.

I will add from myself: in order for this model to work, there should be no doubt about the professionalism of the specialists, that is, they should really be experts. With these theses, I fully agree. They give a clear picture of the method and its limits. In Google, too, the campaign agrees : in the last paragraph it is written that they use their employees for these purposes (I think few who have doubts about the professionalism of the Googlers). For the exact translation I would also be grateful to the translators =)

Let's return to the author of the topic, with whom I decided to argue. Here is his interpretation of the book: the idea is simple: if you take a large group of people and ask each one a specific question, and then calculate the average answer, then, with a high degree of probability, it will be closer to the truth than the answers of the experts. It can be noted that either the author did not understand the ideas of Shuryevsky, or in the books the ideas of the author differ from his comments about the same book. Maybe Shurevsky is cunning and he wrote a book for the sake of money, and not for the sake of promoting his ideas, maybe the translator has gotten bad, maybe it's all nonsense, maybe a brilliant theory. But the book alone is not enough to clarify this issue.

Any attempts to count nuts, volume or linear dimensions are meaningless. This pseudo method cannot achieve great results. Google and other large companies using this approach probably have a complex algorithm that no one will reveal. And this is only average arithmetic. Conducting such experiments can not achieve anything. A theoretical base and serious research are needed.

Lyrical digression:

Suppose they find a way to predict various events with great accuracy. Immediately disappear all interest in the area where everything can be predicted. Why, do something if the result is known? I immediately recall the Gattaka movie. There were two categories of people in the film: normally born and with a modified genetic code. Everything was predicted (not entirely predicted, but an analogy can be drawn) on the basis of this gene. code. Naturally, normally born this code was worse, so they were considered outcasts. There was an ugly duckling in the film, he was born without any gene changes and he was predicted that he would not live to be 18 years old (I don’t remember exactly), but no, our duckling not only survived the years it was supposed to, but also had a system - despite all predictions surpassed many with the correct genetic code, proving the futility of this theory. So it will happen in reality - it is human nature to go beyond (not the frame of the picture, but the framework of the generally accessible, generally accepted). Does it make any sense to create these predictions?

Now, about the prediction markets - if you believe the wiki, then on the world's largest online bookmaker online exchange, in 2007 the trade turnover amounted to 28 billion US dollars. Here the situation is the same as with gambling. People want to believe that they will be lucky and they will break the bank. Therefore, play. I can not stand gambling (casino), and similar exchanges, too. I do not understand how you can hope for it. We must plow and earn money, and not hope to win a million. Therefore, I will not discuss the topic of such exchanges further.

In conclusion, I would like to wish success to naz2 and allow myself to give advice: if this topic is interesting, then it is necessary to consider it from all sides, subject all theses, theories and assumptions to constructive criticism. If they withstand criticism, then they are viable. naz2 not deep enough in this direction. We must look for the source. Google has blog graphs, graphs are built using formulas, why don't you have these or similar formulas? Find primary sources, support the theory with scientific evidence and go ahead =)

PS One head is good, two is better.

Google Docs took an active part in creating the topic, for which many thanks to him =)