We present a prototype system that can be used to capture longitudinal socialising processes by recording people's encounters in space. We argue that such a system can usefully be deployed in prisons and other detention facilities in order help intelligence analysts assess the behaviour or terrorist and organised crime groups, and their potential relationships. Here we present the results of a longitudinal study, carried out with civilians, which demonstrates the capabilities of our system.

Frighteningly bad science.

The idea was that these two physicists wanted to help figure out a way to track prisoners who might be up to no-goodery. Ideally they'd put RFID's into every prisoner's ass and then watch and track the no-gooders. Then after some statistical modeling magic they'd be able to know that prisoner A spends a lot of time humping prisoner B.

That's all fine and dandy, but the PoC implemented about wasn't tested in prisons. It was tested on the the people who live in and around the town of Bath, UK using the resident's bluetooth devices. They have plots and models in their paper showing off some awfully frightening ideas.

So their original idea has nothing to do w/ the implemented idea except for all the spying.

--timball

[0804.3064] Intelligence gathering by capturing the social processes within prisons