JSH: Importance of knowledge
in [Math]
|
Prev: zapping a moving car with the alleged Doppler radar yet the radio wave has no Doppler Chapt 8 #140; ATOM TOTALITY
Next: [Combinatorics] Ball choosing problem
From: Joshua Cranmer on 6 Jun 2010 21:26 On 06/06/2010 04:13 PM, JSH wrote: > There are possibly national security implications with a fundamental > result in modular arithmetic that involves factoring. It raises the > issue of the big unknown. And national security people not only hate > the "big unknown" they like to be informed of such things rapidly. I am pretty sure that the secret stuff of the government does not rely on something so weak as RSA encryption; I'd be surprised if anything weaker than ECC were used. To my knowledge, the ease of factoring has no bearing on the difficulty of ECC. In any case, you missed the part where I mentioned that the more important stuff would take longer to review. If you claim to be able to crack factoring, the reviewer is going to make damn sure that your thesis supports that claim. I've not seen your submission, but extrapolating from what I've seen in the newsgroup, it's going to be one of those that takes a bit more time than the average to properly review. -- Beware of bugs in the above code; I have only proved it correct, not tried it. -- Donald E. Knuth
From: Joshua Cranmer on 6 Jun 2010 21:43 On 06/06/2010 05:34 PM, JSH wrote: > Ok, as stock markets continue to reel around the world investors can > take comfort in your opinion. You seem to think us wildly more influential than we really are. I think that the most influence I've ever had is that I may have impacted the API of a possible AST reflection API in JavaScript. Maybe; I'm not so sure the author of the patch wouldn't have come to a similar conclusion on his own. I will posit, though, that the possibility that an eavesdropper may be able to more easily listen in on secure communications is of little import, considering that hackers already routinely steal data through a combination of user idiocy, incompetency, or just plain lucky guesswork (especially when you factor in that most people would tell you their import login details at the drop of a hat anyways); that itself is of little import compared to the more immediate concerns of major rules affecting fiscal institutions, demagogues in power during a particularly anti-financial sentiment, the specter of sovereign debt crises being compounded by the possibility that growth in major countries (most notably the U.S. and China) is going to head back downwards again (the last job reports was depressing), and not to mention that North Korea is going into a feisty period and the major involved powers don't really have plans for what to do if the country truly melts down, Israel is once again fighting for the title of "most arrogant country," plus the normal level of dysfunctional politics in countries like Iran, Thailand, and the Philippines. And yes, that really was just a single sentence. > Looks like Monday will be a melt-down in the US. Correlation does not imply causation. The volatility indices have already climbed up anyways, the LIBOR isn't anything to jump up and down about right now, the Australian dollar has taken a hit, gold is soaring while other commodities retreat somewhat, and U.S. and British bond yields are once again plumbing the depths. And if you don't understand the impact of all of those, well, you have no business trying to discern anything about the stock market. -- Beware of bugs in the above code; I have only proved it correct, not tried it. -- Donald E. Knuth
From: Joshua Cranmer on 6 Jun 2010 22:33 On 06/06/2010 09:50 PM, JSH wrote: > No, the NSA says it relies on some kind of elliptic logarithmic stuff > or something. > > That's not important here though. Ah, but it is. It indicates that the government doesn't rely on factoring for communicating its secrets, like, say, the nuclear launch codes you mention later. To my knowledge, the military actually uses a separate network from the Internet, so the most secure information is probably not even hooked up to the World Wide Web. > So say, some hostile nation finds this out, while the US and Britain > sit on their hands because top mathematicians just, oh, don't feel > like mentioning the result! > > That nation gets a lot of its mathematicians together and puts them on > a fast-paced secret program to exploit the information. Which nation are you talking about? The ability to devise a new, fast factoring algorithm is probably outside the domain of those who are not quite well-educated, which pretty much means that the search is effectively narrowed to old Europe, the U.S., and China. The bright kids elsewhere probably don't stay back home long enough. > Learning how to crack RSA--maybe trivial to them by then--they figure > out ways to crack all the other systems as well. I am a bit flaky on my cryptographic systems, but the mathematics behind RSA and the ones behind ECC are vastly different beasts. It would be like expecting an advanced biochemist to announce ground-breaking results in solid-state chemistry. > Months of effort pay off as that nation hacks into computers in the US > and Britain and all over the world, downloads top secret information > from all levels, including nuclear launch codes. There is already a mind-boggling amount of hacking going on. Knowing RSA lets you decrypt encrypted conversations you eavesdrop on. For it to be useful, you have to first eavesdrop on conversations. Even then, you can only know what people send, electronically, to each other. If there is a brain in national security organizations, nuclear launch codes are not among these. Let me reiterate more clearly: knowing RSA isn't going to help you hack into a computer (unless you uncover someone's login on a remote session, but you can't connect to the secure systems remotely anyways, so that's moot). And there already exist plenty of techniques for hacking into computers that don't rely on currently-unknown techniques. It's called social engineering, and it's already been used to ruin the lives of millions. > You die in flames along with millions of people all over the world too > civilized to remember that the world is a brutal place. How does this happen? Using the stolen nuclear codes to launch a nuclear strike? Anyone who did such an action would likely find himself or herself part of a glass parking lot before long. The United States (and I think Britain and China as well) still have second-strike capabilities. You're relying a long train of hypotheticals here, and one of the middle steps is already outlandishly unbelievable. Say, I have this bridge I want to sell you... -- Beware of bugs in the above code; I have only proved it correct, not tried it. -- Donald E. Knuth
From: JSH on 7 Jun 2010 00:02 On Jun 6, 7:33 pm, Joshua Cranmer <Pidgeo...(a)verizon.invalid> wrote: > On 06/06/2010 09:50 PM, JSH wrote: > > > No, the NSA says it relies on some kind of elliptic logarithmic stuff > > or something. > > > That's not important here though. > > Ah, but it is. It indicates that the government doesn't rely on > factoring for communicating its secrets, like, say, the nuclear launch > codes you mention later. To my knowledge, the military actually uses a > separate network from the Internet, so the most secure information is > probably not even hooked up to the World Wide Web. Doesn't need to be. And your knowledge clearly isn't worth squat here. And of course nuclear launches are regulated by an entirely different multiply redundant system designed to be impervious to penetration. So breaching it means there is no protection left, as supposedly, it can't be breached. At that point, it's game over. No hope. It's called barbarian through the gates. > > So say, some hostile nation finds this out, while the US and Britain > > sit on their hands because top mathematicians just, oh, don't feel > > like mentioning the result! > > > That nation gets a lot of its mathematicians together and puts them on > > a fast-paced secret program to exploit the information. > > Which nation are you talking about? The ability to devise a new, fast > factoring algorithm is probably outside the domain of those who are not > quite well-educated, which pretty much means that the search is > effectively narrowed to old Europe, the U.S., and China. The bright kids > elsewhere probably don't stay back home long enough. Actually I'd say roughly a dozen nations should be quite capable, including I notice you completely forget Middle East nations. Algebra was invented in the Middle East. But those are meaningless hypotheticals around the drama that might occur if the result others here have claimed isn't new, is. Fun to talk out like science fictions stories are fun, but not really relevant. Your arrogance is Western stupidity run amuck. You believe you are very much advanced from the rest of the world. Makes one wonder how much it would take to remind you that there are other human beings on the planet quite intelligent outside the sphere you recognize as superior. And some of them would really like to destroy your nation, and you with it. But enough mindless fun. The real issue for Usenet posters is, is this thing new? Other than that, you're worthless. James Harris
From: Richard Henry on 7 Jun 2010 01:33
On Jun 6, 6:50 pm, JSH <jst...(a)gmail.com> wrote: > > No, the NSA says it relies on some kind of elliptic logarithmic stuff > or something. > Anything the NSA tells you they are doing, they're not doing. |