Monday, May 14, 2007

The Problem With Wikipedia

The problem with Wikipedia is not biased information or lack of accuracy. Wikipedia is not perfect of course, but it is remarkably accurate, much more so than one might expect from such a project. The real problem is much more subtle. It is really part of a greater problem with the Internet as a whole, but Wikipedia magnifies this problem greatly.

Wikipedia is not anywhere close to accumulating all the world's knowledge. But remember how in Oregon Trail when you would kill a buffalo that weighed 900 pounds, you would only be able to carry back 200 pounds of meat? It's like that, but imagine that the buffalo weighed a million tons. 200 pounds is only a small fraction of 1 million tons, but it's still a lot of meat, and it's more meat than you are going to find anywhere else (except perhaps at university libraries and other libraries like the library of congress; there might be a little more meat at libraries, but it tends to be separated by lots of bones so it is difficult to find the meat that you want). And with Wikipedia, if you bring back 200 pounds of meat, you might wake up the next morning to find that you now have 202 pounds of meat.

But the knowledge in Wikipedia is too large and too easily accessible for our own good. If one has a source of knowledge like Wikipedia that is no more than a few minutes away, incentive for knowing stuff ourselves decreases. If Wikipedia can know just as effectively as you, why know at all? Knowledge has become vulgar and devalued. Since so much of it is now so easily accessible, people forget how valuable and important all knowledge is. Consider this passage from Foucault's Pendulum by Umberto Eco, where Casaubon, the protagonist, invents a profession for himself:
A sudden illumination: I had a trade after all. I would set up a cultural investigation agency, be a kind of private eye of learning.

Instead of sticking my nose into all-night dives and cathouses, I would skulk around bookshops, libraries, corridors of university departments. Then I'd sit in my office, my feet propped on the desk, drinking, from a Dixie cup, the whiskey I’'d brought up from the corner store in a paper bag. The phone rings and a man says: "Listen, I’'m translating this book and came across something or someone called Motakallimun. What the hell is it?"

Give me two days, I tell him. Then I go to the library, flip through some card catalogs, give the man in the reference office a cigarette, and pick up a clue. That evening I invite an instructor in Islamic studies out for a drink. I buy him a couple of beers and he drops his guard, gives me the lowdown for nothing. I call the client back. "All right, the Motakallimun were radical Moslem theologians at the time of Avicenna. They said the world was a sort of just cloud of accidents that formed particular shapes only by an instantaneous and temporary act of the divine will. If God was distracted for even a moment, the universe would fall to pieces, into a meaningless anarchy of atoms. That enough for you? The job took me three days. Pay what you think is fair."

That is what the quest for knowledge should be like. Today that translator would just open up Firefox and type "wp Motakallimun". The quest for knowledge is shortened so greatly, that you don't appreciate how valuable the knowledge you receive is. If you have a lime tree in your back yard and grow your own limes and make your own lime juice, you probably appreciate lime juice more than those who go to the store and buy lime juice in a bottle. Except that the appreciation of knowledge is more important than appreciation of lime juice. Honestly, who even buys lime juice?

If this all seems a bit elitist, that's because it is. I recognize that, at the same time, greater accessibility to knowledge is a good thing. No one should be priced out of knowledge. The cost of knowledge should not be a trans-Atlantic crossing to read an ancient book located only in some European library. No monetary expenditure should be required, the poor should have access to knowledge also. Casaubon shouldn't be able to make a living charging for information. But knowledge cannot be too easy to get either. How to resolve this dilemma? Scramble Wikipedia. Keep all the information, but make everything difficult to find. Imagine. You pull up Wikipedia, but you find that you have been redirected to the front page of Ukranian Wikipedia. After a few more tries, you manage to get Wikipedia speaking English again. You search for Bartolomeo Vanzetti, but instead you end up with the article on the 1948 Philadelphia Athletics. You realize immediately that the Search function has been tampered with and that you will get nowhere using that. You formulate an alternate plan of attack. Starting on the current page, what chain of links will take you to your destination fastest? How about American League->New York Yankees->Joe DiMaggio->Famous Italian-Americans->Vanzetti? Thwarted again when the link to American League actually takes you to List of Defunct Korean Automakers. A thought occurs to you. Though it is highly improbable that you will ever need to know the name of a single defunct Korean automaker, much less 15 of them, the chance is non-zero, and at the time when you need the knowledge the most, you know it will be impossible to get back to this page. You print out the page just in case and file it under 'L', for 'List of Defunct Korean Automakers.' Twenty years later you get into an argument with a Korean diplomat about the relative reliability of Korean and American cars. You know that there is a Korean automaker that went bankrupt after its vehicles were shown to be prone to massive explosions. You rush down to the filing cabinet in the basement, hoping to return upstairs triumphant, shouting out the name of Taisoon or Wajbai or whatever the name of the company was, knowing that that one word will win the debate for you. You look around madly in your filing cabinet, first under 'K' for Korean, then under 'A' for Automobile and 'C' for cars, but you can't remember where you filed it twenty years earlier. You dash over to your computer, hoping desperately that Wikipedia will cooperate, but your search for Korean Cars only turns up an article on Nikola Tesla. You return upstairs, dejected, having no answers to the Korean's argument and forced to accept defeat. But back to the present. You are still searching for Vanzetti. You give up on finding a link to the page within Wikipedia and resort to Google. Your search results include a link to Wikipedia, which you click. Wikipedia notes your perseverance and offers a compromise. If you agree to rewrite two articles on the Spanish Civil War, it will let you read the article you desire. Too worn out to fight Wikipedia anymore, you agree to its terms. You will now truly appreciate whatever information you eventually get about Vanzetti and Italian anarchism.

No comments: