Month: August 2013

Remember the furor a month or two ago when it came to the national attention that the United States government was conducting analysis of phone records? How the claim was that they weren’t using the contents of our phone conversations and SMS messaging, merely looking at the metadata these records could provide? One of the prevailing questions at the time was what could be done with this information? Why was this important?

And that is a difficult question to answer. What kinds of information can be derived from records of who calls or texts whom? This is a question that I was informally asked if I could shed some light on in a public venue. At the time I had to decline as my experience and training with metadata has more to do with preservation and authentication of documents.

This article¹, on the other hand, does a reasonably good job of demonstrating what can be deduced from something as simple as the membership roles of a few colonial societies that are suspected of revolutionary tendencies.

For those of you who are tl;dr, the short of it is, by looking at membership in just seven societies, it becomes clear based solely on this information that Paul Revere is a central player in revolutionary communications.

Now to look the implications of this fact.

If you were Tory-oriented, you would definitely be wanting to keep an eye on his comings and goings. If the British had been doing that, the Ride of Paul Revere would have been more dark limerick and less epic poem…and the Battle of Concord may have not gone so swimmingly. Lacking the buoyancy of this success, the Massachusetts push for revolution may have sputtered. Etc.

It doesn’t take a lot of extrapolation to see how “persons of interest” can be identified just by looking at the amalgamation of our social graphs (as represented by our phone records). Persons of interest need not be doing anything in order to be interesting to watch. Their movements and behaviors are markers for the groups they connect. By closely watching these individuals, various security apparatus can keep their finger on the pulse of the general population without the resource expenditure required to monitor every single person.

Getting darker, this perfunctory analysis sheds some light on how the federal government identifies targets for drone strikes. The target (person or location) can be “interesting” without knowing a single thing about what the target is actually doing. By virtue of being identified as playing a central role in a social graph, the target demonstrates the value of its elimination–the disruption of a communication network that has been identified as “problematic.”

Not to fear monger–it’s really not my thing–but I came across this article and wanted to share since it does a decent job of translating the math into concepts graspable by regular old folks. The example and POV of the writing help demonstrate the gravity of what this kind of analysis can illuminate.

¹ http://kieranhealy.org/blog/archives/2013/06/09/using-metadata-to-find-paul-revere/