President Obama has made it a top priority for his administration to make government data transparent and available to all agencies, government employees and public citizens. In a recent article in FCW, author Adam Mazmanian gives a detailed background on the open data initiative and poses the question of whether open data can change the culture of government. A key example of the positives of open data likely rests on many reader’s car dashboards – the GPS device that was made possible by the government unlocking the global positioning system for commercial use. Aside from the economic benefits of the jobs GPS devices created (manufacturing, marketing, retail etc.), we all now benefit by hearing a comforting and all-knowing voice say “recalculating” when we miss the exit on the highway. That said, I spot a potential wrinkle when Mazmanian writes: “Agencies are charged with making a detailed inventory of their data holdings – even data that is not going to be released because it is sensitive or contains personally identifiable information.”
Here’s my concern; undertaking such a task is monumental and privacy concerns should be raised on how classification is done to determine what is personally identifiable information (PII) versus what is safe for release. A recent article in CSO Online provides what I find to be an eye-opening scenario raised by David Navetta of the Information Law Group about an individual who buys a deep fryer and consents to sharing their information for marketing purposes only to have their insurance rates affected as a result of the insurance company’s access to that information. Now the US government isn’t in the deep fryer business (not to my knowledge anyway), but it raises the possibility of what may seem an innocuous sharing of information in one facet of government being used by another or even a private citizen or corporation’s access to that data once made open. The reality is that once the open data door is opened and the information released, it’s out there and can’t feasibly be contained at a later date. So, one would hope that all classification of the data being released is by no means rushed, done properly and takes all privacy concerns into careful consideration.