3 ms·
How do you differentiate someone who is LYING from someone who is the wrong person, or who doesn't know? e.g.: Bob hates you, and leaves a thumb drive labeled
by gknoy 13y ago
How do you differentiate someone who is LYING from someone who is the wrong person, or who doesn't know?
e.g.: Bob hates you, and leaves a thumb drive labeled "CP" behind your computer, filled with random astronomy data. He then leaves bits and pieces that make it appear to be an encrypted drive of illegal porn. (Or, any other nefarious Jack-Bauer-worthy terrorist movie plot.)
When you get interrogated by wrench-wielding thugs, it will never be possible for them to tell whether you are lying about whether it's yours, whether you know what's on it, or whether you can decrypt it.
Torture can only possibly get useful information if the person you're torturing HAS that information. Of course that can work -- no one doubts that. Think from the perspective of the wrongly-accused person, though, and you'll see that it very quickly becomes a game of "what can I say to make them stop hurting me". This is why so many people in medieval times confessed to being witches, or part of treasonous plots: their choice was death by a relatively quick means versus death by torture.
edit: My point is, "reliable" means more than "we can verify that it's true", but also means that you can avoid false positives.
- vorg 13y ago> The "hit him with this $5 wrench until he tells us his password" school of torture produces highly reliable information, because it's trivial to check whether what the guy said was correct. Sounds like the difference between P and NP problems in Computer Science, i.e. P (polynomial-time) problems can be solved in polynomial time, whereas NP (non-deterministic polynomial-time) problems can't be solved in polynomial time but their solutions can be checked in polynomial time. > How do you differentiate someone who is LYING from someone who is the wrong person, or who doesn't know? Sounds like the Halting Problem in Computer Science, i.e. checking whether an algorithm halts for all inputs in uncomputable. It seems Computational Complexity and Computation Theory have something to say about whether torture is an effective means of extracting information.
- thaumasiotes 13y ago> Torture can only possibly get useful information if the person you're torturing HAS that information. Of course that can work -- no one doubts that. You are wrong. For many people it is an item of faith that torture cannot produce useful information. In my opinion, these people are generally overattached to the principle that good results come from (morally) good acts, and evil acts necessarily produce bad results.
- DanBC 13y agoYou quote a sentence that says "Torture can get useful information (if the person being tortured has that information) - no one doubts that", and then say that is wrong, and then say that torture can get useful information. Yes, if you torture the right people you will get useful information. People who are tortured will give up the information they have at some point. The problem is that we often torture people who have no information, and so they tell us what we think we want to hear.
- thaumasiotes 13y agoYes, my "you are wrong" was not denying that torture can get useful information. It referred to the comment "no one doubts that", which is wrong. That's why I followed it by saying "for many people it is an item of faith that torture cannot produce useful information".