By Frank Ahrens
Sunday, August 6, 2006
Last week provided another fascinating chapter in the evolving story of Wikipedia, the almost-anybody-can-edit-it online encyclopedia.
On Monday's night's "Colbert Report" on the Comedy Central cable channel, host Stephen Colbert praised Wikipedia for its fungible factuality. Colbert's character is a parody of television blowhards whom Colbert (the real person, not the character) thinks value ideology over fact.
On the show, Colbert praised what he called "Wikiality," the idea that if you claim something to be true and enough people agree with you, it becomes true. It was Colbert's jibe at the online encyclopedia, which does not rely on formal peer review to proof its entries. Wikipedia's peer review comes from a coterie of interested parties -- citizen editors -- on particular topics, and each person can have conflicting interpretations of fact. Such problems have led to Wikipedia recently barring new or anonymous users from editing the entry on Israel, for instance.
There is, however, a critical difference between longtime and dedicated Wikipedia editors and newbie flamers who just want to blow up an entry or use it for their personal agenda. It's not entirely a free-for-all.
Colbert stepped farther through the looking glass by editing Wikipedia's "Stephen Colbert" entry during his show. He railed against the Encyclopedia Britannica's assertion that George Washington owned slaves.
"If I want to say he didn't, that's my right," Colbert said. On Wikipedia's "George Washington" entry, the following phrase appeared at the end: "In conclusion, George Washington did not own slaves."
Colbert then urged viewers to take part in rewriting history and fact. To jibe environmental activists, Colbert told viewers to log on to Wikipedia and edit the entry on elephants to say that the African population of elephants had tripled in the past six months.
Naturally, enough people obeyed Colbert to crash Wikipedia's servers.
That's sort of interesting, if not surprising -- the crossover audience between "Colbert Report" and Wikipedia is probably pretty substantial. What happened after the servers came back up is what's notable.
Wikipedia's truth-squadders swung into action. They locked down 20 entries on elephants to all but longtime users. They did the same to Colbert's entry, and they barred the screen name StephenColbert from making further changes. The last move is more symbolic than practical; there's no way of knowing whether StephenColbert is the Stephen Colbert -- the real one or the character -- from the show.
Then Wikipedia took the smart step of posting the pre-Colbert entries alongside the many, many post-Colbert ones to show exactly what was changed and when it was changed by subsequent editors.
The whole process has a mind-boggling, recursive-loop feel to it, as one Wiki-editor edits an entry and seconds later, another re-edits it. At one point, a post-Colbert entry took on a "yes, you did/no, you didn't" tone. Gaah!
But if Wikipedia is going to exist as an open-source resource and is going to resist single-peer review for its entries, then it needs to be transparent, as it has been in l'affaire Colbert. If Wikipedia's DNA prevents it from hosting a single standard for truth -- or truthiness -- then its sources of information need to be evident and their tracks easily seen so readers can have as many facts as possible to determine their accuracy. Not, of course, that anyone would or should use Wikipedia -- or really, anything else besides this column -- as a single and authoritative source on any topic.
As a coda, Colbert's assault happened right before Wikimania 2006, the encyclopedia's annual event held over the weekend at Harvard University, no doubt giving the Wiki-heads plenty to talk about. And revise.