I had an interesting experience this week. It all began when I read a post on the New York Times Caucus blog that reported that more than six million questions had been submitted online for Tuesday night's presidential "town hall" in Nashville. Hmm, I said to myself, that sounds fishy. I knew enough to know that MyDebates.org was the official online partner of the Commission on Presidential Debates (CPD) and that the public's ability to send in questions for the debate through MyDebates had barely been publicized. So the idea that six million questions had somehow been submitted rang false. But that New York Times post had no external links and no named sources for that nugget of information.

So I did some digging, and it turned out that indeed the number was bunk. The six million was a casual reference at a social lunch event in Nashville that mistakenly ended up in in a Gray Lady blog post. Frank Fahrenkopf, the co-chair of the CPD, told me that he had been the source for that six million figure, saying that he picked it up from Mark Whitaker, the NBC News Washington bureau chief (who replaced Tim Russert after Russert's death). Whitaker in turn told me that the six million actually referred to every bit of communication the network got in reference to Tuesday night's debate -- not by any means, the number of actual questions that came in online, which actually turned out to number just over 25,000, about 1/240th of what was reported. Whitaker called the figure "misinterpreted and misused."

I posted on the "six million questions question" on techPresident on Wednesday afternoon, and then updated it on Thursday and Friday as I learned more. (In retrospect, it would have been good practice for me to date and timestamp those updates.) The New York Times ended up posting a "clarification."

Meanwhile, though, this bogus nugget of information was busy spreading across the world. Many publications cited "six million questions were asked online" as fact, all, it seems, because it was in a New York Times blog post. The imprimatur of the Times goes a long way, and there's the presumption that if it runs under the Times' masthead, it's true. In the pieces that picked up the six million reference, often no source was given and rarely was it linked to the original source. That wildly incorrect figure ended up being quoted everywhere from the LA Times to the Boston Globe to Australia's the Age to Grist to the Winnipeg Free Press, just to pick on just a few news outlets. Some places have since corrected it. Many others haven't.

Bloggers, of course, link stuff. It's what we do, and it's looking like an awfully good practice right now. Every reporter and writer gets things wrong, of course, but linking to sources can help minimize the damage. With no links back to the original source in those derivative pieces, the Times' "clarification" didn't help even the most careful reader all that much.

For more on the six million question questions, the Columbia Journalism Review's Megan McGinley and Armin Rosen covered the story on Friday afternoon.

UPDATE: The New York Times' Brian Stelter suggests that linking is going mainstream, writing that "[e]mbracing the hyperlink ethos of the Web to a degree not seen before, news organizations are becoming more comfortable linking to competitors." What Brian describes are mostly content-sharing arrangement and aggregator projects, but they're nonetheless a step forward. "For bloggers," he writes, "linking to original reporting, primary sources and discussions about stories is a form of etiquette, assigning credit to others who have written about a topic." Indeed.