Last December, Nature published a News story about the accuracy of two online references sources. We compared the website of an established publication, Encyclopaedia Britannica, with that of Wikipedia, a new kind of online encyclopaedia that anyone can edit and update, regardless of expertise.

The result (see Nature 438, 900–901; 2005) surprised us, and many others. Forty-two expert reviewers carried out the comparison. After we had tallied their results, we saw that they had picked up errors (the great majority of them minor) at a rate of about three per online Britannica item and about four per Wikipedia item.

Last week, Encyclopaedia Britannica issued a statement (, and this week published a half-page advertisement in the London Times criticizing our study and demanding that we retract our story.

Entries were blinded — reviewers did not know which entry came from Wikipedia and which from Britannica.

Britannica complains that we did not check the errors that our reviewers identified, and that some of them are not errors at all. We disagree with their claims in some of the cases (others are too specialized for an immediate response), but there is a more important point to make. Our reviewers may have made some mistakes — we have been open about our methodology and never claimed otherwise — but the entries they reviewed were blinded: they did not know which entry came from Wikipedia and which from Britannica. We see no reason to believe that any misidentifications of errors would adversely affect one publication more than the other. And of the 123 purported errors in question, Britannica takes issue with fewer than half.

Another Britannica criticism concerns the fact that we provided material from other Britannica publications, such as the Britannica Book of the Year. This was deliberate: the aim of our story, as we made clear, was to compare the online material available from Britannica and Wikipedia. When users search Britannica online, they get results from several Britannica publications. They have no reason to think that any one is less reliable than the others. In the case of some year-book entries, Britannica itself asks readers to reference the articles as coming from “Encyclopaedia Britannica Online” — exactly the source we set out to compare.

Other objections are simply incorrect. The company has, for example, claimed that in one case we sent a reviewer material that did not come from any Britannica publication. When the company made this point to us in private we asked for details, but it provided none. Now Britannica has identified the review in question as being on ethanol. We have checked the original e-mail that we sent to the reviewer who looked at the Britannica article on ethanol, and it is clear to us that all the reviewer's comments refer to specific paragraphs from Britannica.

Our responses to the points raised by Britannica in its original online posting and in its subsequent advertisement can be found at Our comparison was unbiased, and we reject Britannica's allegation that we have acted in a dishonest manner. We stand by the story.