Boston Flu on A screen grab of's Boston Flu page.Boston Flu on A screen grab of (then broken)’s Boston Flu page.

I’ve been paying more attention to 10000words recently, which bills itself as the place “where journalism and technology meet.” I don’t know if it’s gotten better or it had simply fallen to the bottom of my RSS reader, but they have some great pieces, including a recent one on news orgs using polling data to drive news. The post profiles sites like RiskyRoads and Tampa Bay Mugshots</a>*, which allow readers to interact in some pretty cool ways with otherwise dry data bases.

One of the sites was, which uses some novel Twitter searches to, very non-scientifically, gauge where outbreaks are occurring. The idea is, over time, if more people start Tweeting “flu,” “food poisoning,” or “sore throat,” you can get a rough idea of what’s ailing a given city.

It’s a very similar idea to Google Flu Trends, in fact, but made without internal access to vast troves of search data. Anyone with enough programming know-how could duplicate the mash-up.

But it also highlights one of the dangers I think that are often overlooked. The first time I tried the service, all searches returned “0.” I don’t know the reason, but I suspect there were API problems: Maybe Twitter changed how they delivered results, or the SickCity pinged Twitter too often, but the point is that the 3rd party is entirely reliant on another single company for their usefulness. It wouldn’t be so bad if SickCity were able to distribute their reliance to multiple services, like is trying to provide. With Twitter the dominant micro-blogging tool, that can be a tough call to make, but the other option is to risk the service’s entire existence on the whims of a third-party notorious for unexplained changes without warning.

Further Reading:

*Not long ago, somebody took the same idea and made a profitable new PRINT product with nothing but mug shots, charging a buck a copy. Opportunities still abound for opportunists.