4 ms·

That's weird. You're turning the argument on its head. YOU started supporting the process and said something like Don't people who're manufacturing explosive

That's weird. You're turning the argument on its head. YOU started supporting the process and said something like Don't people who're manufacturing explosives in secret.. Well, based on what evidence? The machine that detects chemicals here? It seems that this supports my claim, you think that machine is 'working' and somewhat reliable. Without having provided any kind of proof or source, by the way. The article, the author, proceeds to talk about potential chemicals to set off the detector (and we had more posts like this over the years, for example from people working with plants, farmers etc). Chemicals you're able to get everywhere. If that is a known issue and this machine can beep if a) you've built a bomb b) you've moved your plants from inside the house to your garden then you should reevaluate the idea of this particular test in the first place. The author writes that one participant in that theater claimed to know the exact substance that triggered the detector. Really? In that case it should be possible to - dismiss the idea, it's stupid. That stuff? Everyone can buy it - ask disguised/hidden questions to reveal if he used one of the gazillion things that might trigger this warning and leave the man alone The option - keep him, make him feel like a criminal, treat him in an inhuman (no water..? Really?) way for as long as it takes to break into his appartment is really not on the list of things you can rationally describe as necessary. That's insane. Ah, well. I come up empty handed at the end and cannot provide a source for the reliability of the detectors used in the US of A. I guess that makes my whole post moot. Obviously you don't need to provide the same thing, since "It's used at the airports, certainly it works!"?

zaroth 13y ago
> If that is a known issue and this machine can beep if a) you've built a bomb, b) you've moved your plants from inside the house to your garden, then you should reevaluate the idea of this particular test in the first place. It's more complicated than that. Anecdotal evidence suggests the trace analysis machine has a very low false positive rate. The fact that agents are trained to re-run the test repeatedly also suggests the machine has a very low false negative rate. Given a sufficiently low false positive and false negative rate, the right answer may be, if it repeatedly triggers on a passenger, to question that passenger further. Of course, with increasing reliability of the test comes increasingly severe questioning when the test repeatedly fails. In theory you want the level of the response to be commensurate with the reliability of the test. A perfect test convicts you as soon as you fail it. An imperfect test such as this requires humans to actually follow-up and investigate the failure. The tactics, ethics, and legality of the how that investigation proceeds... I think that's another matter entirely, from whether we should be employing trace analysis in airports in the first place.
samfisher83 13y ago
You can't tell between the moving plants and making bombs because materials in one can be used for the other.

zaroth 13y ago

> If that is a known issue and this machine can beep if a) you've built a bomb, b) you've moved your plants from inside the house to your garden, then you should reevaluate the idea of this particular test in the first place. It's more complicated than that. Anecdotal evidence suggests the trace analysis machine has a very low false positive rate. The fact that agents are trained to re-run the test repeatedly also suggests the machine has a very low false negative rate. Given a sufficiently low false positive and false negative rate, the right answer may be, if it repeatedly triggers on a passenger, to question that passenger further. Of course, with increasing reliability of the test comes increasingly severe questioning when the test repeatedly fails. In theory you want the level of the response to be commensurate with the reliability of the test. A perfect test convicts you as soon as you fail it. An imperfect test such as this requires humans to actually follow-up and investigate the failure. The tactics, ethics, and legality of the how that investigation proceeds... I think that's another matter entirely, from whether we should be employing trace analysis in airports in the first place.

samfisher83 13y ago

You can't tell between the moving plants and making bombs because materials in one can be used for the other.