True Random Number Generator
in [Cryptography]
|
From: bmearns on 9 Feb 2010 22:55 On Feb 9, 4:11 pm, Mok-Kong Shen <mok-kong.s...(a)t-online.de> wrote: > Am 09.02.2010 21:42, schrieb bmearns: > > > > > On Feb 9, 3:20 pm, Mok-Kong Shen<mok-kong.s...(a)t-online.de> wrote: > >> Spinner wrote: > >>> Silly statement. The fact that a totally provably random OTP has never > >>> been created is going to be a helluva suprise to a lot of people who > >>> have been using them for many manyt years. Not to mentioni there are > >>> websites where you can go get a list of quantum-generated (radioactive > >>> decay) random numbers. > > >> The problem is "proving" that what one has at hand "is" a OTP. I can't > >> conceive how one could deliver a proof in the sense of proofs in math. > >> This is analogous to the case of a point in elementary geometry. > >> One simply can't have in the real world a point that has no "size", > >> though one could have arbitrarily good approximation of that. In matter > >> of OTP in "practice", see also the story about Venona. > > This in particular is far from my realm of expertise, but as I > > understand it, it is the generator itself which can be proven to be > > random. It's true, you can't look at the OTP itself and prove that it > > is truly random, but if the values come from a truly random process, > > like a quantum effect, then the generator can be proven random, and > > therefore the data it produces is as well. > > I am afraid that there is in reality a logical problem here, i.e. a > circularity in argumentation. How does one "know" that employing > quantum effect gives "perfect" randomness? Certainly one could "define" > it that way, but that I would consider to be problematic. Or consider > a "fair coin". As a "theoretical" concept it is excellent and very > useful. But can one really make such a coin "practically", depite all > the advances e.g. in nano-technology today and in future? > > M. K. Shen Well my background in quantum mechanics is even more limited than my background in information theory, but as far as I know, quantum processes are generally accepted to be strictly random; based, I'm sure, on complex mathematics that are far above my head. I'll be back for the rest of your posts tomorrow. =) -Brian
From: Mok-Kong Shen on 10 Feb 2010 03:49 bmearns wrote: > Well my background in quantum mechanics is even more limited than my > background in information theory, but as far as I know, quantum > processes are generally accepted to be strictly random; based, I'm > sure, on complex mathematics that are far above my head. As I illuded to, the present argumentation is "theoretical", or one might say, pedantic, and has no practical significance. Invariably, one accepts in practical life, depending on individuals, something as "ideal", even though that may in fact not be so. For instance, I would have considered myself a rich person, if I possessed one million dollars, but to a multi-milliardair that might be a tiny little sum. For randomness, I would accept the outcomes of dice that I buy from a game shop to be entirely satisfactory, while others would claim that there are inevitably bias in manufacture and therefore no good. In other words, there is no boundary to "perfectness", just like "infinity" has no boundary, and hence both are never "exactly" achievable but can be better approached further and further without limit, if one works hard, spends enough resources (money, time, etc.) and as the science and technology advance with time. On the other hand, one needs concepts that are "ideal" that represent the ultimate (even though consciously not exactly attainable) goals that one "should" acieve, for it is only with the existence of them that one obtains "directions" in which one can sensibly proceed in life (instead of doing some "random walk"). M. K. Shen
From: Mok-Kong Shen on 10 Feb 2010 03:54 bmearns wrote: > You're assuming causation where none is required. AES also generates > binary ciphertext, that doesn't mean that this is the reason it > doesn't have perfect security. Likewise, just because it doesn't have > maximum entropy doesn't mean this is necessarily the reason it isn't > perfectly secure. I don't think that your second sentence ever touches what I argumented. As to the issue of perfect security, I suppose that what is in any textbook about OTP contradicts your third sentence. M. K. Shen
From: Spinner on 10 Feb 2010 09:11 Mok-Kong Shen <mok-kong.shen(a)t-online.de> wrote: >bmearns wrote: > >> Well my background in quantum mechanics is even more limited than my >> background in information theory, but as far as I know, quantum >> processes are generally accepted to be strictly random; based, I'm >> sure, on complex mathematics that are far above my head. > >As I illuded to, the present argumentation is "theoretical", or one >might say, pedantic, and has no practical significance. Invariably, >one accepts in practical life, depending on individuals, something >as "ideal", even though that may in fact not be so. For instance, I >would have considered myself a rich person, if I possessed one million >dollars, but to a multi-milliardair that might be a tiny little sum. >For randomness, I would accept the outcomes of dice that I buy from >a game shop to be entirely satisfactory, while others would claim that >there are inevitably bias in manufacture and therefore no good. In >other words, there is no boundary to "perfectness", just like "infinity" >has no boundary, and hence both are never "exactly" achievable but can >be better approached further and further without limit, if one works >hard, spends enough resources (money, time, etc.) and as the science >and technology advance with time. On the other hand, one needs concepts >that are "ideal" that represent the ultimate (even though consciously >not exactly attainable) goals that one "should" acieve, for it is only >with the existence of them that one obtains "directions" in which one >can sensibly proceed in life (instead of doing some "random walk"). > >M. K. Shen spontaneous nuclear decay is provem to be as quantum effect and such effects are proven to be random. This is known because a WHOLE lot of tech depends on the math related to nuclear physics. To prove its not random is to prove general relativity wrong. Many holes have been blown in the desert and the pacific ocean proving that nuclear physics has a sound theoretical basis. Hence, decay is provably random and spontaneous MATHEMATICALLY and practically. The ability to detect decay products is proven. A number based on detection of a random process is provably random. If you are patient, build a cosmic ray coincidence detector, a random event. You generally never give up on your pedantic side-trips throwing in a lot of verbal noise when the alternative wouild be to admit you actually got something wrong. Arguing that we cant 'prove' we exist to a metaphysical certainty is just noise and smoke, turning a logical discussion to philosophical and vice versa to suit your purposes. It is stupid but I have to admit it's entertaining. -- 2+2!=5 even for extremely large values of 2
From: Mok-Kong Shen on 10 Feb 2010 09:47
Spinner wrote: > spontaneous nuclear decay is provem to be as quantum effect and such > effects are proven to be random. > > This is known because a WHOLE lot of tech depends on the math related > to nuclear physics. To prove its not random is to prove general > relativity wrong. Many holes have been blown in the desert and the > pacific ocean proving that nuclear physics has a sound theoretical > basis. > > Hence, decay is provably random and spontaneous MATHEMATICALLY and > practically. The ability to detect decay products is proven. A > number based on detection of a random process is provably random. > > If you are patient, build a cosmic ray coincidence detector, a random > event. > > You generally never give up on your pedantic side-trips throwing in a > lot of verbal noise when the alternative wouild be to admit you > actually got something wrong. Arguing that we cant 'prove' we exist > to a metaphysical certainty is just noise and smoke, turning a logical > discussion to philosophical and vice versa to suit your purposes. > > It is stupid but I have to admit it's entertaining. > Are you aware that all thechnical appartus, no matter how fine they are manufactured, have imperfectness? That would exclude any "perfectly" randomness from being obtained practically, even if a perfectly random sourcem that can be tapped, could exist. When I said of a proof, I mean proof in the sense of an exact proof, not a proof that the probability of deviation from a uniform distribution, etc. etc. is smaller than epsilon. Evidently such can't exist. If you still think it is otherwise, please give a reference in a respectable scientific journal to support that. As I said previously, one could "define" something (a process) to be "random", but that's akin in politics of some regimes "claiming" itself to be democratic. M. K. Shen |