Thursday, July 27, 2006

The New Yorker on Wikipedia

There's a good article about Wikipedia in this week's New Yorker. Favorite part:
Wikipedia has become a regulatory thicket, complete with an elaborate hierarchy of users and policies about policies. Martin Wattenberg and Fernanda B. Viegas, two researchers at I.B.M. who have studied the site using computerized visual models called "history flows," found that the talk pages and "meta pages" - those dealing with coordination and administration - have experienced the greatest growth. Whereas articles once made up about eighty-five per cent of the site's content, as of last October they represented seventy per cent. As Wattenberg put it, "People are talking about governance, not working on content." Wales is ambivalent about the rules and procedures but believes that they are necessary. "Things work well when a group of people know each other, and things break down when it's a bunch of random people interacting," he told me.

For all its protocol, Wikipedia's bureaucracy doesn't necessarily favor truth. In March, 2005, William Connolley, a climate modeller at the British Antarctic Survey, in Cambridge, was briefly a victim of an edit war over the entry on global warming, to which he had contributed. After a particularly nasty confrontation with a skeptic, who had repeatedly watered down language pertaining to the greenhouse effect, the case went int arbitration. "User William M. Connolley strongly pushes his POV with systematic removal of any POV which does not match his own," his accuser charged in a written deposition. "His views on climate science ar singular and narrow." A decision from the arbitration committee was three months in coming, after which Connolley was placed on a humiliating one-revert-a-day parole. The punishment was later revoked, and Connolley is now an admin, with two thousand pages on his watchlist - a feature that enables users to compile a list of entries and to be notified when changes are made to them. He says that Wikipedia’s entry on global warming may be the best page on the subject anywhere on the Web. Nevertheless, Wales admits that in this case the system failed. It can still seem as though the user who spends the most time on the site - who yells the loudest - wins

Connolley believes that Wikipedia "gives no privilege to those who know what they're talking about," a view that is echoed by many academics and former contributors, including Larry Sanger, who argues that too many Wikipedians are fundamentally suspicious of experts and unjustly confident of their own opinions. He left Wikipedia in March, 2002, after Wales ran out of money to support the site during the dot-com bust. Sanger concluded that he had become a symbol of authority in an anti-authoritarian community. "Wikipedia has gone from a nearly perfect anarchy to an anarchy with gang rule," he told me. (Sanger is now the director of collaborative projects at the online foundation Digital Universe, where he is helping to develop a Web-based encyclopedia, a hybrid between a wiki and a traditional reference work. He promises that it will have "the lowest error rate in history.") Even Eric Raymond, the open-source pioneer whose work inspired Wales, argues that "'disaster' is not too strong a word" for Wikipedia. In his view, the site is "infested with moonbats." (Think hobgoblins of little minds, varsity division.) He has found his corrections to entries on science fiction dismantled by users who evidently felt that he was trespassing on their terrain. "The more you look at what some of the Wikipedia contributors have done, the better Britannica looks," Raymond said. He believes that the open-source model is simply inapplicable to an encyclopedia. For software, there is an objective standard: either it works or it doesn't. There is no such test for truth.

The territoriality of Wikipedians is a huge problem. And to be an insider, you have to have lots of time on your hands to participate in Wikipedia editing.

Of course territoriality isn't limited to Wikipedia. It pops up all the time in human groupings. There's a seemingly desperate need to distinguish insiders from outsiders. This even happens, absurdly on Internet discussion boards where your identity can be completely hidden and the rewards of being an insider are almost nil - it's not like a country club where you get access to the best golf times and tennis courts and business networking through your insider status.

The best part of observing group dynamics on Internet discussions is that the insider/outsider games are played in black and white print, and you can easily watch and analyze it, and track it through time.