At ScienceInsider, part of the news operation at AAAS’s Science magazine, reporter Sara Reardon latched on to a sure-fire different angle on the killing of Osama bin Laden  by US special forces troops. However, after following the comments appended to it from readers and editors, one imagines that she and section boss Eli Kintisch wish they had taken one more steady look at the story before putting it up.

The underlying news, which I recall the first time it broke a year or so ago, is that a student-professor team of geographers at UCLA calculated on the basis of biogeography principles the most likely sorts of places the al Qaeda leader might be hiding.  Their paper is below in Grist.

But, apparently, and again judging from comments, the first version of the ScienceInsider piece said that not only did the exercise predict a very high likelihood that the world’s foremost mass murderer who does not run a nation would be found in a large, walled compound in a sizeable town not far from the Afghan-Pakistani border, but that it gave nearly a 90 percent chance that town would be Abbottobad.

Since then, it has been corrected to say more merely that Abbottobad is within a circular area of probability of that scale, but that includes plenty of other cities. Furthermore, it assigned its highest probability to a different town, and even provided pictures of compounds in that town that look suitable for hiding a tall, bearded, most-wanted person. They look spookily a little like the property where the assassination actually occurred a few days ago. Commenters are still arguing with ScienceNow, it appears, over what probability to assign to Abbatobad’s category of towns, too.

Good for Science to amend the piece rapidly. But errors and oversights (as I well know) don’t disappear from the web even after re-edit. The original, as of this writing, is still excerpted at blogs, including this one by Jim Galloway at the Atlanta Journal Constitution. Ditto for a blogpost at the Los Angeles Times. A better job of picking up the story from ScienceInsider, but thinking it through afresh, is at a Foreign Policy blogpost by Joshua Keating. He notes that while the conclusion he’d be in a crowded urban area, and in a distinctively-build compound, was correct and perhaps even brilliant, the percentages of probability and the selection of which town was not.

Grist for the Mill: Journal article Finding bin Laden ;

(..hat tip to Sylvia Wright for the lead)

- Charlie Petit

This entry was posted on Tuesday, May 3rd, 2011 at 3:23 pm and is filed under Science Stories. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.