Unnatural acts at Nature

Scientific journal Nature, which delivered an evangelical editorial urging scientists to write for the online website Wikipedia last year, has defended the exercise.

But the journal, which usually publishes other people's research, won't show the world its own.

Last year Nature's news team ran a feature that demonstrated, staff claimed, that there wasn't much difference between Wikipedia and the Encyclopedia Britannica (in fact, by Nature's own statistics, Wikipedia's contained 31 per cent more errors). The news story became the basis for this remarkable advertorial.

Britannica has accused the journal of shoddy practices, such as including material from a kid's version, using articles from the yearbook not the Encyclopedia, sending selective extracts from articles - which were then criticised for omissions - and even stitching together related articles and passing them off as its own. This week it demanded the journal retract the article.

Nature has issued a press release defending its own news team's version of the scientific method, and says it won't be issuing a retraction.

Nature also says Encyclopedia Britannica didn't asked the right questions of it. "While we were quite willing to discuss the issues, the company failed to provide specific details of its complaints when we asked for them," says the journal.

This doesn't tally with Britannica's version of events. Britannica says Nature prevaricated and still won't release the referees' reports in full.

"We asked for the data so that anyone could replicate the results," Britannica spokesman Tom Panelas said.

"At first Nature said they couldn't release the data because they had promised their referees anonymity. We said that's fine. Then they said it's too much trouble."

Nine days after Nature's news story and rallying editorial, the magazine published a file of "supplementary information" which merely listed the errors referees had found. Britannica had to figure out what the sources were. That information still hasn't been released.

"Nature actually has a string of embarrassments behind it," writes reader Fazal Majid.

"They were the ones who published Benveniste's infamous 'Memory of water' paper some years ago, an apology for homeopathy, which could never be reproduced."

"More recently, they published a pretty bogus article purporting to detect the fashionable - among breathless bloggers - "power law" in email response time patterns."

"I would say their credibility is actually fairly low, especially when the question is on a fashionable or controversial subject," he concludes.

Still, some readers are surprised.

"Nature would have been the last place I ever would have expected such behaviour. If Carl Sagan was still alive he would probably beat Steve Ballmer's record for throwing chairs around after seeing this," writes Pete Hines.

"These two posters sum up the idea of "collective wisdom" quite well for me."

Michael Rouse thinks Britannica could have done a better job of showing which mistakes it did correct - something absent from its statement this week.

"It's quite possible that every objection they raised in their rebuttal was absolutely true - and I would be surprised if they didn't check it out very thoroughly before releasing it, because Wikipedia would have a field day otherwise - but the best lie is the truth partly told," writes Michael.

"There is a reason that you are supposed to 'tell the truth, the whole truth, and nothing but the truth' in a court of law. By omitting examples of their mistakes and the corrections they performed - or better yet, giving a comprehensive rebuttal/mea culpa for every point in the Nature article - they show they are as interested in slanting the result as they claim Nature was in the original story."

However, readers continue to find the issue deeply troubling. Reader Tim compares the faith in Wikipedia to the enthusiasm for 'Intelligent Design'.

"The only rationale I can concoct for Jim Giles' [Nature's feature editor] behavior is that it is better to be right than honest," he writes.

"This is the mentality of a mediocre student trying to remember various facts and formulas so as to pass a standardized test. Mr. Giles already knew the 'right' answer: Wikipedia is 'better' than the Brittanica because of the 'open' process of how Wikipedia entries are created is clearly superior to the 'closed' and hierarchical methods of Brittanica. It is also the behavior of a less than scrupulous researcher: Hwang Woo Suk of South Korea comes to mind.