At the risk of being gauche by following up to my own post, I'd like to point out one more thing:

People complain about the putative false alarm rate of a system like TIA. But this is arguing about the relative merits of two completely unknown merits: the FA rate of some oracular intelligence system (TIA) which hasn't been built yet, and the FA rate of current crime-fighting techniques. Neither of these have ever been measured.

I would submit to you that the FA rate of traditional techniques is extremely high, given the debate about innocent people on death row! (this implies a false alarm that passed all the way through the investigative process to the courts and led to conviction -- the worst kind of error!). If we're going to make claims, it would be helpful to have all the data.