Britannica lambasts Wikipedia findings

The Encyclopaedia Britannica has reopened the debate over how its accuracy compares to that of its online rival, Wikipedia.

The publisher of the venerable encyclopaedia this week released a scathing 20-page rebuttal to a December article in the journal Nature that tallied errors in both Britannica and Wikipedia and found that the Web upstart more than held its own. The experts who reviewed comparable entries found 162 factual errors, omissions or misleading statements for Wikipedia, compared with 123 for Brittanica.

The article was widely seen as a validation of Wikipedia's content and methods.

The Britannica retort criticised both the techniques Nature used in its comparison and the manner in which the assessment was presented to readers. It also demanded a public retraction. (Read a PDF of the 20-page report.)

According to Britannica, Nature sent out re-edited, rearranged and truncated versions of Britannica entries to reviewers and included samples that were not even from its encyclopaedia texts. Britannica also accuses Nature editors of failing to verify its reviewers' findings of inaccuracy, saying that in many cases the findings were scientifically or factually wrong.

For example, one reviewer in the Nature article said that Subrahmanyan Chandrasekhar's Principles of Stellar Dynamics was published in 1943, not 1942 as its Britannica entry states. Britannica disputes the charge, citing the Library of Congress as its primary source.

"This study has been cited all over the world, and it's invalid," Dale Hoiberg, Britannica's editor in chief, said in a statement. "We have never claimed that Britannica is error-free, but Nature attributed to us dozens of inaccuracies that simply were not inaccuracies at all. We practice strong scholarship, reasoned judgment, and continuous editorial review, and we publish a reliable, high-quality encyclopedia. By its flawed analysis and false accusations, Nature did us a great disservice."

The Nature article came out at a critical time for Wikipedia, whose credibility had been very publicly called into question. Most notably, it was under attack for an entry on John Siegenthaler that erroneously linked the journalist and former Washington insider to the Kennedy assassinations.

Where Britannica and other traditional encyclopedias tap subject matter experts to produce articles in a closed editorial system, Wikipedia entries can be created and edited by anyone with access to the Internet. Wikipedia founder Jimmy Wales has long maintained that the site's accuracy is ensured through self-policing by its readers and contributors.

Nature answered the charges this week in a statement indicating no remorse and reaffirming its original assessment. "We reject those accusations, and are confident our comparison was fair," the publication said.

Meanwhile, Wikipedia's burgeoning status on the Internet continues apace. Earlier this month, its publisher, the Wikimedia Foundation, reported the addition of the 1 millionth article to the English-language version of Wikipedia.