> If there are 1,000 cases of such fraud a year, then I don't think it's worthwhile. Well, it's probably not really for safety or security, and they clearly do

> If there are 1,000 cases of such fraud a year, then I don't think it's worthwhile. Well, it's probably not really for safety or security, and they clearly don't give a hoot if you're inconvenienced or not. Consider instead: - Government contracts routed to political constituents (follow the money -- we know the millimeter scanners were corruptly funded) - Tons of ground-truth data for further development and deployment. It'd be interesting to find out if they correlate against images captured by vanilla airport security cameras, as well as the cameras on the roadways or public transportation to the airport, and in rental car lots. This helps answer the question of "do we have to upgrade all the crappy domestic cameras to get ubiquitous surveillance, or can we get away with what's out there now?" A billion or two spent to catch one or two terrorists isn't the point; actual bad guys are utterly optional, and maybe even a distraction.