Not Getting Close to the Whole Story

The online magazine spiked has a story entitled “Is Wikipedia part of a new ‘global brain’?” in which a writer asks some reasonable questions but then undermines herself with — at best — incomplete reporting. She writes, in part:

Much was made of a study conducted by Nature magazine at the end of 2005, which found that Wikipedia was about as accurate in covering scientific topics as was Encyclopaedia Britannica. According to the survey, based on 42 articles reviewed, the average scientific entry in Wikipedia contained four errors or omissions, while the average entry in Encyclopaedia Britannica contained three. Of eight ‘serious errors’ the reviewers found, including misinterpretations of important concepts, four came from Wikipedia and four from Encyclopaedia Britannica.

However, soon after this report was published, Encyclopaedia Britannica published a damning response accusing Nature of misrepresenting its own evidence. Dozens of inaccuracies attributed to Encyclopaedia Britannica were, in fact, not inaccuracies at all, and a number of the articles examined were not even in Encyclopaedia Britannica. It has been reported that the study was poorly carried out and its findings were laden with errors; one publication accuses Nature of ‘cooking’ the report (6).

Yet hundreds of publications jumped on the Nature story, echoing the argument that Wikipedia (based on collective intelligence) was as good as Encyclopaedia Britannica (based on professional knowledge). Jim Wales, founder of Wikipedia, continues to cite the Nature survey in his defence when quizzed about the accuracy of information on Wikipedia.

Several issues:

First, there’s no link or even a footnote pointing to the Nature report. Nor is there a link to the Britannica response, which as the story notes disputed the findings. (The reporter’s footnoted evidence of Nature’s errors is a story by a publication that has been deeply and consistently skeptical, if not downright hostile, to Wikipedia.)

And, given her damning of the publications that “jumped on” the Nature story, it’s utterly bizarre that she didn’t point to Nature’s reply to Britannica’s objections, which included a point-by-point response.

No one says Wikipedia doesn’t have its flaws. It has plenty. And Nature’s methodology, and its original headline, did leave something to be desired. But its response was thorough, and the fundamental points it raised essentially held up.

But when the reporter fails to point to any of the relevant material, she does readers no favors. Given that more than 10 months had passed since the magazine’s full replies, the missing links undermine the entire article.

(Note: Wikipedia founder Jimmy Wales is a member of this center’s board of advisors, and I’m an investor in his separate for-profit company.)