Scientific Red Cards: a good idea or opening a hornet's nest?

From The Scientist: Flagging fraud:

A team of French life sciences grad students has launched an online repository of fraudulent scientific papers, and is calling on researchers to report studies tainted by misconduct.

The website -- called Scientific Red Cards -- is still in a beta version, but once it's fully operational it should help the scientific community police the literature even when problems slip past journal editors, the students claim.

The database might also prevent researchers from citing papers that they don't even realize are fraudulent, said Claire Ribrault, a PhD student in neurobiology at the Ecole Normale Superieure in Paris, who unveiled the new website last month at a workshop in Madrid, Spain, organized by the European Science Foundation's Research Integrity Forum.

After misconduct is detected in a published paper, "sometimes the paper is not retracted, depending on the policy of the journal, and even if the paper is retracted sometimes it's still cited after the retraction," Ribrault said in a press release.

The website color-codes misconduct into three categories: red for data-related misconduct, including fabrication, falsification, and plagiarism; blue for publication-related misconduct, such as when editorial policies are not followed; and green for research practice misconduct, including problems with consent forms.

Each problematic paper in the register includes a full bibliographic reference, a link to published accounts of the misconduct, and a discussion board for users to leave their comments. So far, only around 30 papers have been listed.

Scientific Red Cards received a cautious thumbs-up from the meetings' attendees, although some voiced concerns over legal problems and the site being used for scientific smear campaigns. Other countered that it provides a transparent way to patrol the literature.

More like this

For me, too. Drove me nuts. But if they just started the site, I guess they'll fix such glitches over time.

Hmm... I like the idea, on the face of it, since I think that peer review can only be strengthened by shining outside lights such as these onto the process. Having said that, the one thing that immediately concerns me is the nature of the comments on each post, which are anonymous. Here's an example, from Risk markers of oral cancer in clinically normal mucosa as an aid in smoking cessation counseling.:

Be careful about other publications by Sudbo : according to the report "the bulk of Jon Sudb�'s scientific publications are invalid due to the fabrication and manipulation of the underlying data material"

I'm normally fine with anonymous commenting on the web, but given the amount of damage that even an untrue accusation of widespread fraud could do to a scientist's career, I would like to see a full profile on commenters to add a layer of accountability. Given that the "Red Cards" themselves seem to be reasonably well-documented - at least, from a cursory examination - I'd be happier if the same were true of the commenters. This comment is probably fine, since it's pulling the quotation from the source material for the Red Card itself, but I smell the potential for abuse (almost along the lines of a Prisoner's Dilemma: everyone's better off if everyone is honest and only says true things, but the temptation to smear a fellow scientist to advance one's own career might lead to an orgy of mud-slinging on a site like this).

I'm not na�ve to believe that this is a novel concern which hasn't arisen in other contexts, but I'm ignorant of the general line of thought on the question...