Popular Science magazine yesterday took a bold step toward improving the civil discourse, cutting off reader comments to its online edition. At first blush, the decision looks like a move toward censorship.

popular scienceSuzanne LaBarre, the magazine's online comment editor, sees it differently. To her, the decision is to eliminate rude responses, ad hominem attacks, spewings from wackos, political cheap shots and luddite rants that cast doubt upon the science-backed reports of the magazine.

She cites research that found negative and off-the-wall comments wield enough influence to skew a reader’s perception of the best-researched story. Those responses undermine the authority of the media outlet and its sources.

Wrote LaBarre: "A politically motivated, decades-long war on expertise has eroded the popular consensus on a wide variety of scientifically validated topics. Everything, from evolution to the origins of climate change, is mistakenly up for grabs again. Scientific certainty is just another thing for two people to 'debate' on television. And because comments sections tend to be a grotesque reflection of the media culture surrounding them, the cynical work of undermining bedrock scientific doctrine is now being done beneath our own stories, within a website devoted to championing science."

The last thing Popular Science wants is to provide a forum that duplicates cable TV shouting matched between no-nothing pundits. (PS will provide other platforms for reader responses to stories.)

Other media outlets should follow LaBarre’s lead. The media, in its quest to present both sides of a story, provide space to many who either lack qualifications to respond or have an ax to grind.

The debate over global warming is the prime example. Though 97 percent of climate scientists say warming is an important issue, the media search high and low to find a source to cast overall doubt on warming.

LaBarre's move may deal a blow to the PR community, but it's a godsend for the media. Who will be next to follow in her footsteps?