I recently came across the story of the Significant Objects Project, an interesting experiment where Joshua Glenn and his other collaborators take cheap thrift-store finds and see if they can create something more valuable by contriving a back story to match. For example, they found a "plastic Russian doll with a big, cloth moustache that's mounted on a little piece of wood" and claimed that it was a "woodcutter named Vralkomir, who during a particularly bad blizzard, when everybody in his village in Russia were freezing to death, danced so hard on a little piece of wood that it burst into flame." Turns out, at least one person bought the story: he bought the $3 thrift-store tchotchke for $193.

Of course, the creative spin is done by marketers and PR folks all the time. But sometimes stories just don't really match up. In particular, when fabricated stories masquerade as news, people subject them to a different set of rigorous, ethical rules.

Take the New York Times' analysis of Kindle's reviews by Nick Bilton. When the story first ran, the graphs were mislabeled. I highly respect Nick and his deft hand at using infographics to tell a story, but, in this case, there was no story—or at least, the materials he used didn't corroborate his story. Seth Godin has a good analysis and iReader Review offers a more thorough one—both pointed out that you can't rely on the star-ratings alone without reading their accompanying comments to draw conclusions (apparently, many of them came from non-Kindle owners). Unfortunately, I think this was a case of "bad" infographics (simplified star-ratings) leading to more bad infographics that didn't take into consideration the value of sentiment within the written reviews.

As we increasingly scan, rather than read, we often fall prey to relying on data visualizations to get a quick summary of large, complex amounts of data. The problem, of course is we lose the fine-grained resolution of the story and, instead, only see a single, simplified, interpreted view of things.