Thus, the damage wrought by any single untruth temporarily floating around in the vast ocean of Wikipedia isn’t likely to last long. Certainly, the Seigenthaler episode is a reminder that such information can be inaccurate, and he’s right to complain about other sites scraping up Wikipedia content and presenting it as their own. But the general trend is away from the proliferation of such rumors.
More than that, however, what’s been obscured in much of the coverage is that the Wikipedia model actually works. The English-language Wikipedia boasts over 800,000 articles, all generated for free, and freely available to anyone with Internet access. That’s a tremendous amount of information being made far more accessible to far more people than print or broadcast can reach.
Wikipedia’s content itself is, in general, quite good. Most of the entries are quick overviews synthesizing information from other sources. If you’re looking for basic factual information about, say, zebras, or coffee or even Columbia University, Wikipedia is a good place to start. Moreover, a number of experts contribute to Wikipedia (particularly when it comes to technology), making its longer articles on more esoteric topics especially useful. (See the entry on DNS, the domain name system underlying the Internet, for just one example.) And a survey published this week by the British scientific journal Nature found that Wikipedia was nearly as accurate as Encyclopedia Brittanica about scientific information.
The fact that everyone can produce and edit entries on Wikipedia is, generally speaking, a great virtue. Entries tend to be edited and refined iteratively, not unlike a breaking news story that develops and merits more articles after the initial report. The effect is a gigantic collective data download from the brains of contributors. There are always, as anyone with a Web site that includes a comments section can tell you, a couple of idiots who want to ruin things for everyone — but, all in all, those are surprisingly few and far between. And, increasingly, deletion of false information occurs swiftly at Wikipedia. If, say, Fortune magazine prints an error, that error is out there on newsstands for two full weeks before a new edition of Fortune comes along. Wikipedia’s users don’t have to wait two weeks to clean up someone else’s mess; they can do it in two minutes, once they spot it.
Then there is the question of responsibility for the content itself. While Wikipedia is published and overseen by the Wikimedia Foundation (and by Jimmy Wales, the founder of the site), it relies on unpaid, usually anonymous, contributors and editors. (Tracking down Seigenthaler’s libeler turned out to be onerous — though not impossible — because of the relative lack of data about users.) Wikipedia does make available the edit history of most entries (you can see, for example, that the zebras entry has been edited a few dozen times), but it doesn’t require verification of a user’s identity (a not-uncommon practice on the Web).
The downside of that system is obvious in episodes like the Seigenthaler case. But there’s a practical reason to do it Wikipedia’s way, as well — letting people anonymously post and edit entries generates the maximum amount of content for the least amount of overhead on the part of contributors and editors. That’s the nature of distributed enterprises like Wikipedia: the more barriers you erect to producing content, the less of it you’re likely to get. And the anonymous nature of Wikipedia’s editors also emphasizes that the project is collaborative — no one gets ownership over any single entry, which encourages people to contribute as they see fit.