November 16, 2020
"A major influence on Jimmy Wales’s conception of [Wikipedia] was an essay by Friedrich Hayek called 'The Use of Knowledge in Society'..."
"... published in 1945, and Hayek is virtually the father of postwar neoliberalism.... Hayek’s argument about knowledge is... markets are self-optimizing mechanisms.... This theory of knowledge is not unrelated to the wisdom-of-crowds scenario in which a group of people are guessing the number of jelly beans in a jar. The greater the number of guesses, the closer the mean of all guesses will come to the true number of jelly beans. A crucial part of crowdsourcing knowledge is not to exclude any guesses. This is why Wales, in his role as Wikipedia’s grand arbiter, is notoriously permissive about allowing access to the site’s editing function, and why he doesn’t care whether some of the editors are discovered to be impostors, people pretending to expertise that they don’t really have. For, when you are calculating the mean, the outliers are as important as the numbers that cluster around the average. The only way for the articles to be self-correcting is not to correct, to let the invisible hand do its job. Wikipedia is neoliberalism applied to knowledge...."
Subscribe to:
Post Comments (Atom)
Posts
33 comments:
"The only way for the articles to be self-correcting is not to correct, to let the invisible hand do its job."
Poorly phrased. Articles can't be "self-correcting." Editors' hands are always visible.
Wikipedia is not a market-like mechanism. It does not use knowledge in a Hayekian way. It can be gamed.
But Hayek was right about the limits of knowledge, in particular any individual participant's knowledge.
"For, when you are calculating the mean, the outliers are as important as the numbers that cluster around the average."
In order for this to work, wouldn't the outliers all have to have the same dedication to making corrections? Is there any evidence they do?
If a majority of the bean counters are acting in bad faith you can throw all of that invisible hand crap out the window. The number of jelly beans will be what they say it is.
Charles Mackey in Extraordinary Popular Delusions and the Madness of Crowds might disagree.
Wikipedia is generally fine for largely apolitical topics. But the second a subject becomes remotely political is when their reliability and objectiveness goes down the toilet.
You can depend on people to behave honestly as long as lying doesn't benefit them.
I see that the obvious objections have been stated in the first three comments. I will add, that Hayek was concerned with prices as a mechanism for communicating information about the scarcity of resources. In that system, the communicators have a very strong incentive to provide the best information available, because doing anything else costs them money.
The greater the number of guesses, the closer the mean of all guesses will come to the true number of jelly beans.
Doubntful. You'd want the median. The high guess is likely to be a joke, like a trillion. In particular the extremes are garbage.
If you digitally measure the temperature of the basement, you're likely to get a series of numbers like 62 65 63 61 62 999999999 64 63, and processing has to take care of it.
..."markets are self-optimizing mechanisms"... "is not unrelated to the wisdom-of-crowds scenario in which a group of people are guessing the number of jelly beans in a jar"
It is almost wholly unrelated. The self optimizing mechanisms of a market are a result of the communication of knowledge in interactions and transactions of the participants. While that communication is far from perfect, at scale it tends toward optimization.
Guessing the number jellybeans in a jar is taking place with a fairly equal lack of knowledge among the participants.
If you don't believe me, check out this research found on Wikipedia
Read the article, and I have absolutely no idea what Wikipedia and Jeopardy have to do with one another.
Someone once said that nothing is as bad as a good idea taken too far. Was it Chesterton? It sounds Chestertonian.
So what is bad about free markets? In ideas, or ice cream, or education, or anything else?
A free market is an ideal rarely seen in actuality. The people who are making money off of a non-free market, or who are in a free market & want to make more money for themselves constantly attack and degrade free markets. It is a human constant. This behavior is not going to go away.
The more free a market is, the more pressure it will be under to be non-free, because that is where the greatest marginal profit lies.
The day Xi crashes the internet, I'll continue to know Crime and Punishment was released in 1866.
"Someone once said that nothing is as bad as a good idea taken too far. Was it Chesterton? It sounds Chestertonian."
According to Wikipedia, it was Barack Obama.
I am an editor of exactly one Wiki page, but I haven't done an edit in over a decade now. Just lost interest in the subject when I retired. I was viewing the page just now, seems up to date, and the subject is completely non-political- Cathepsin S.
I knew a group of British teens that manned a certain politician’s wiki like it was warfare for years. They had assigned shifts.
Wonder if people can guess which wiki it was...
Damn they are thirty years old with kids now so they probably finally surrendered.
Scott Patton said...
"The self optimizing mechanisms of a market are a result of the communication of knowledge in interactions and transactions of the participants."
Good point. In a market, individuals only provide information about matters where their information is very high quality. The engineer trying to decide whether to use titanium or stainless steel probably does not know much about current realities in the mining industry. He doesn't need to. He looks at the price, which was set by people who are extremely knowledgeable about those realities.
It's interesting that Wales is utterly clueless about Hayek's theories. A stopped clock.
I stopped trying to correct Wikipedia articles about 10 years ago. An article on a famous American historical figure was full of left-wing lies. I kept revising it, and quoting the facts & including the sources, and some Leftist would come along and change it back. So, I just gave up.
when it comes to history, wikipedia is good for things like when someone was born or died, or what office he held. but the more details they provide the worse it is.
"[P]rices [are] a mechanism for communicating information about the scarcity of resources. In that system, the communicators have a very strong incentive to provide the best information available, because doing anything else costs them money."
Absolutely. Free markets are designed to encourage people to maximize their own benefit, within the limitations imposed by other people openly trying to do the same thing. You can see that the system doesn't work in a tyranny, because tyrants have no use whatever for maximizing the communication of information, or at least, any communication not controlled by themselves. There's nothing in it for them. Corollary: people who are hostile to a free flow of information are motivated by the opportunity for bullying rather than consent.
Utter nonsense. An innumerate idiot that reads just enough to be dangerous.
“"Someone once said that nothing is as bad as a good idea taken too far. Was it Chesterton? It sounds Chestertonian."
According to Wikipedia, it was Barack Obama.“
He should know. That was his modus operandi.
Wikipedia has its uses--I go there a lot. I could care less about correcting anything-- long ago I even had some fellow who didn't like some of my comments on a newsgroup, put my name and some nonsense into a Wikipedia article on a topic of interest at the time.
What point he thought he was making, was beyond me, but it's actually a sort of template for one way in which Wikipedia can be useless.
I have a younger friend who was dedicated to maintaining a clean, factual Wiki article on the social theorist Roberto Unger. I don't know if he still plugs away at it or not, but it seems a Sisyphean task to me.
Narr
Like commenting here, sometimes
Actually I've had some bad experiences with Wikipedia editors on topics that are entirely apolitical. At one point several years ago we were trying to get a Wikipedia listing for a national model plane organization that flew model airplanes designed in the 1930s and 1940s. The organizations was incorporated under New Jersey state law and had tax exempt determination letters from both the State of New Jersey and the IRS.
The desired listing was simply the name of the organization and a statement of its purpose. Whatever self appointed and anonymous editors Wikipedia had would not accept the listing--wanted a series of edits etc. It got to the point of being ridiculous so we simply gave up. The organization can be found on the Internet with its own website--we're still not on Wikipedia.
Yes, assuming a principled premise, signal diversity, or large-scale diversity (individual, not class). However, people are not jelly beans, not even colorful clumps of cells, and the progression toward a mean does not imply improved accuracy. Case in point: democracy, or an ensemble of divergent climate models. It does imply a feedback that will mitigate unqualified progress.
no idea what Wikipedia and Jeopardy have to do with one another
The game of Jeopardy is played with an authoritarian model, where the answers are preconceived and there are only correct and wrong questions. Apparently, Wikipedia is an enterprise, detached from human society, from the natural world, rising to the hallowed heights of ivory towers, and Jeopardy stages, operating in that mode with that perception.
Is Menard really that stupid? I agree with Jupiter and others who dispute the idea that Hayekian observation of price information generated by actual market actors (with skin in the game) that is in any way comparable to the self-interested (and virtually costless) manipulations of text of politically loaded entries on Wikipedia.
My favorite crowdsourcing is when scientists ask reddit users to help them solve how puzzles fold. This is seriously cool...
Recently, Trump used Twitter to crowdsource his fraud claims. Through an "open call," or a tweet ("I was robbed!"), he asked Twitter's giant community for evidence of voting fraud. There have been many insightful and even surprising results. We now know that the election was won in the suburbs, where White men overwhelmingly and emphatically chose Biden for President. The Democrats lost ten to fifteen House seats but it was not a landslide for the Republican presidential candidate. And Trump won all of the bellwether counties.
Wales's dream is long dead, and everyone knows it. Wikipedia has famously been captured by a small coterie of editors who strictly police the ability of outsiders to edit the site. They are, of course, quite far to the left, so their influence on political articles is easily seen, but they also enforce their preferences on a wide variety of other subjects as well.
wisdom-of-crowds scenario
The evil white man Galton "discovered" that effect and called it the "trustworthiness of a democratic judgment".
Rabel said
“If a majority of the bean counters are acting in bad faith you can throw all of that invisible hand crap out the window. The number of jelly beans will be what they say it is.”
Or if as is more likely they have a financial interest in the outcome of the counting.
Free markets work perfectly only if one assumes perfect information and perfect rational behavior, neither of which actually ever exists. Yet it is still the optimal way to allocate resources and discern truths. One cannot let the perfect be the enemy of the best.
"Is Menard really that stupid? "
I'll take "Rhetorical Questions" for $400, Alex.
Post a Comment