A Modern Reappraisal of the One-Time Pad Cipher.
in [Cryptography]
|
Prev: A Randomness Hypothesis.
Next: How cool is this?
From: J.D. on 5 May 2010 09:14 On May 5, 1:33 am, David Eather <eat...(a)tpg.com.au> wrote: > > or it is decided that the case may involve matters of national security > - in which case anything from a star chamber to a military court might > be the process (what ever happened in gitmo?) . The military commissions for trying Gitmo detainees have so far done nothing to earn the moniker 'star chamber' or 'kangaroo court' or whatever other paranoid term one might want to apply. So far they have done basically nothing except dismiss charges on technical legal grounds and sit around waiting for jurisdictional matters to be resolved. The judges of the military commissions are far and away not the scary part of Gitmo -- the scary part of Gitmo is that if you are brown-skinned and Muslim and had the misfortune to be accused of witchcra...I mean terrorism, then you could be locked away and tortured for years on end without ever seeing a judge. Judges are not scary -- the scary things are powerful politicians who cannot ever set you free even though they know you are innocent because to do so would admit their incompetence and criminal culpability to the world. > Or you waive the right > to a jury either of your own free will or by coercion, Out of curiosity, what specific case do you know of where a person waived his or her right to a jury trial because of coercion?* Preferably a case where the judge was then less fair than a jury would have been... * I suppose it is coercive to say "we'll let you have a jury trial, but if you lose you have to pay all the additional costs that entails", but for criminal matters that is only (to my knowledge) the case for trivial offenses (like traffic tickets) where States are not required to give you a jury trial at all. > or if there is a > suspicion (real or imagined) of jury tampering you may be heard without > a jury. When did this happen? That would be Unconstitutional. > In any case, what the judge thinks is relevant as he may decide > what evidence is in or out and he may instruct the jury what weight to > give a piece of evidence or how to interpret it or even throw out a > verdict as having no basis in law (and all of these have happened in US > courts). Judges can only throw out guilty verdicts. They cannot set aside acquittals on any grounds -- to do so would violate double jeopardy. And yes, the judge can decide what evidence is permissible and what jury instructions are to be issued -- that's his/her job. But if you believe that an instruction was incorrect, or some bit of evidence should or should not have been allowed, then you can appeal on that basis to a higher court. Of course, if you believe all judges (including Appellate judges) are Evil then that will be cold comfort to you, I suppose.
From: Andrew Haley on 5 May 2010 10:23 Simon Johnson <simon.johnson(a)gmail.com> wrote: >> It depends. I send to a customer quite a lot of encrypted emails, but >> altogether they make maybe one GB per year. I could have gone there >> five years ago and have personally brought them a DVD, and we could be >> using OTP for the whole time. > > Tom attacked this on the grounds of the key distribution problem but > there is another problem of equal magnitude, in my view. > > How do you know that your random number generator (RNG) gave you pad > bits are actually random? > > Sure, we have a series of engineering techniques that can produce a > large series of random bits in *theory*. But that's quite a different > than building an actual device that captures those bits accurately, > without introducing bias. [1] But that problem was solved by von Neumann in 1951, for any source that generates independent bits. Andrew.
From: MrD on 5 May 2010 12:31 Andrew Haley wrote: > Simon Johnson <simon.johnson(a)gmail.com> wrote: >>> It depends. I send to a customer quite a lot of encrypted emails, >>> but altogether they make maybe one GB per year. I could have gone >>> there five years ago and have personally brought them a DVD, and >>> we could be using OTP for the whole time. >> Tom attacked this on the grounds of the key distribution problem >> but there is another problem of equal magnitude, in my view. >> >> How do you know that your random number generator (RNG) gave you >> pad bits are actually random? >> >> Sure, we have a series of engineering techniques that can produce a >> large series of random bits in *theory*. But that's quite a >> different than building an actual device that captures those bits >> accurately, without introducing bias. [1] > > But that problem was solved by von Neumann in 1951, for any source > that generates independent bits. As far as I can tell, it's neither easy nor necessary to get a stream of unbiased random bits from a signal. Von Neumann's method reduces the bias in a stream of bits (and through repetition, can reduce it to arbitrarily low levels); it doesn't capture the bits, nor does it assure their randomness. And it's lossy, in that the debiased stream contains a lot less bits than the original. But it's surely better than hashing, which just obfuscates the bias. For the sake of clarity (and so also for the sake of confidence), one would prefer to have a bitstream that is as close as possible to the original source of random (i.e. quantum) events; the more processes the information passes through, the more likely it is that the source randomness has been degraded. It does seem to be surprisingly difficult to accurately extract bits from a stream of analogue noise. The analogue signal has to be sampled, sooner or later; doing that properly requires a good understanding of the nature of analogue noise, which can be rather problematic. For example, for any frequency you choose, white noise contains a sine wave at that frequency. So any periodic sampling procedure is potentially problematic. -- MrD.
From: MrD on 5 May 2010 12:40 J.D. wrote: >> Or you waive the right to a jury either of your own free will or >> by coercion, > > Out of curiosity, what specific case do you know of where a person > waived his or her right to a jury trial because of coercion?* > Preferably a case where the judge was then less fair than a jury > would have been... In the UK it is quite routine for suspects to opt for summary trial (by a magistrate without a jury), because conviction in a Crown Court is likely to carry a heavier penalty. I think that's a form of coercion. > > * I suppose it is coercive to say "we'll let you have a jury trial, > but if you lose you have to pay all the additional costs that > entails", but for criminal matters that is only (to my knowledge) the > case for trivial offenses (like traffic tickets) where States are > not required to give you a jury trial at all. > >> or if there is a suspicion (real or imagined) of jury tampering you >> may be heard without a jury. > > When did this happen? That would be Unconstitutional. The UK has no proper constitution; our first such case was heard last year. http://en.wikipedia.org/wiki/Jury_tampering > >> In any case, what the judge thinks is relevant as he may decide >> what evidence is in or out and he may instruct the jury what weight >> to give a piece of evidence or how to interpret it or even throw >> out a verdict as having no basis in law (and all of these have >> happened in US courts). > > Judges can only throw out guilty verdicts. They cannot set aside > acquittals on any grounds -- to do so would violate double jeopardy. On 11 September 2006, William Dunlop became the first person [in the UK] to be convicted of murder after previously being acquitted. http://en.wikipedia.org/wiki/Double_jeopardy As someone said, the courts are far from simple! -- MrD.
From: Kristian Gj�steen on 5 May 2010 12:59
nemo_outis <abc(a)xyz.com> wrote: > What I'm talking about is a backup plan for when the >outer walls have been breached and the citadel is in danger of >falling. Pretending to be stupid might work for some. I'd dial back the paranoia and rhetoric, then focus on some practical scheme instead. Identify the threat you are trying to protect against. This is a moderately interested forensic investigator. You want to create a plausible alternate scenario to explain a file containing ciphertext, a file that will be regularily updated. One possibility: an abandoned project (one of many!) that as a by-product produces moderately large, random-looking files. By the way. Don't discuss this topic in public. -- Kristian Gj�steen |