Back in early 2009, long before cries of “fake news” were commonplace, I created Fairspin with my friend Dave Baggeroer. We were concerned by how our friends and family (and, if we were honest, we ourselves) routinely consumed news from preferred sources that demonstrated biases mirroring our own. We were living in echo chambers, and we wondered if technology could help break down those walls.
Fairspin was a web site that harvested political news from the (still excellent) aggregator memeorandum. It indexed each story by reporter/writer and publication. By asking readers to rate the bias (or lack thereof) of individual stories, Fairspin over time built up a crowdsourced model that could be used to predict both the nature and extent of a newly-published story’s leanings. It could also be used to filter the day’s news, slicing it by bias or source.
Fairspin was ultimately a failed experiment but I’m proud of the work we did and I continue to worry about this problem and its effect on society. I’d like to revisit it someday, although given the present political climate I’d probably need to use a pseudonum before jumping into those waters again. Because... yikes.
TechCrunch: Fairspin Teases Out The Bias In Political News
VentureBeat: FairSpin looks for media bias on Twitter