Wikipedia:Wikipedia Signpost/2006-03-27/Britannica
Britannica responds to Nature
Three months after the prestigious science journal Nature published a comparison of the accuracy of 42 science articles in the online version of Encyclopædia Britannica and Wikipedia (see archived story), editors of Encyclopædia Britannica Inc (EBI) have published a response to the study. The open letter, titled "Fatally Flawed", was published in PDF format and linked from the "EB News" box at http://www.britannica.com/ on March 22. (An HTML version is also available.)
According to the Associated Press, an email from Patricia A. Ginnis (Senior Vice President at EBI) was sent to 5000 customers pointing towards the PDF file. The Wall Street Journal said that EBI will publish half-page-advertisements on Monday 27 March in a group of English newspapers demanding the retraction of the Nature article. The AP story was widely covered in the mass media (see further press coverage).
The document denounced the Nature study, stating that "almost everything about the journal’s investigation, from the criteria for identifying inaccuracies to the discrepancy between the article text and its headline, was wrong and misleading," and called for the journal to make a public retraction of the article.
Encyclopædia Britannica Inc. disputed about half of the errors found by Nature, and said that they considered the very structure of the investigation sloppy and unscientific.
However, they indirectly acknowledged that the other half of the errors were correctly identified, and Tom Panelas, EBI spokesman, was quoted for stating that some of these errors were already known but so far they were not corrected. Both Nature and EBI failed to identify how long these errors had been known at Britannica.
Nature's response
On March 23, the staff of Nature published a similar PDF document, responding to EBI's accusations of "misrepresentation, sloppiness and indifference to scholarly standards". Nature firmly rejected those accusations, and stood by their belief that the comparisons were fairly made; they do not plan to make any retractions.
Nature says that EBI objected privately to the article when it was first published, but after they (and Wikipedia) were given access to the reviewer's comments a few weeks after publication, the journal "did not receive any further correspondence until the publication of its open letter", and says it regrets the public and acrimonious nature of this exchange.
One of EBI's most vigorous objections is that the reviewers were given excerpts of Britannica articles, or versions taken from their Student Encyclopedia or from their Book of the Year (which, by design, includes more personal opinion and theory than the standard EB article); Nature countered that on each website, they used whatever material was presented to them upon searching for the scientific term in question.
Britannica is correct that the study undertaken by Nature was not one of their usual, rigourously peer-reviewed scientific articles; it was a more informal survey made by journalists on their news staff, and published in their news section, separate from the articles at the heart of the journal. However, Nature says that while some editorial judgement was involved in turning reviewers' comments into numerical scores, it was applied "diligently and fairly", and that "because the reviewers were blind to the source of the material they were evaluating, and material from both sources was treated the same way, there is absolutely no reason to think that any errors they made would have systematically altered the results of our inquiry."
Wikipedia's response
Neither founder Jimmy Wales nor the collective staff of Wikipedia has made a public response to EBI's accusations. They have, however, gathered some reference information for members of the Wikimedia Foundation's press representatives to use in answering media questions. That document says, in part, "Wikipedia, and all Wikimedia Foundation projects, are not in competition to EBI or other companies in the business of reference works. Our goals differ significantly from other reference publishers, and only overlap in that we are all striving to create accurate and useful knowledge tools."
It also notes a project to correct all Wikipedia errors noted in the Nature study, created the day Nature released the data. Thirty-four days later, all errors were reported corrected (see archived story).