Friday, March 24, 2006

Britannica vs Wikipedia: Fight!

There was an article in Nature on 12/15 arguing that Wikipedia was just as accurate as Britannica. Here is an article about Britannica's 20-page retort, which includes a demand that article be retracted.

Here is an interesting tidbit from the newspaper article:

The article, which has since been updated, differed from the normal practice in that it was "an expert-led investigation carried out by Nature" rather than a paper written by scientists and submitted to the journal for peer review. It also came out at a time Wikipedia was under criticism for high-profile errors in some entries.

Nature concluded such errors appear to be the exception rather than the rule.



Now, Nature is only partly a research journal. It's also partly a magazine, and their primary goal is to sell issues rather than to produce science. I hadn't realized that this was a Nature investigation rather than a more traditional scientific study. Given that such an article would sell magazines, I think we need to see the extensive article in Social Networks or a similar reputable journal.

My own intuition is that wikipedia entries should asymptotically achieve comparable accuracy, but it's not at all clear how fast this convergence occurs.

Any thoughts?

No comments: