I once beat a man to death with a Model M keyboard. Worked just fine before and since.
OK, I didn't really. But I'm sure I could have, those things really are invincible (also big and heavy) - fire, physical shock, water (or beer), nothing hurts them.
Everyone seems to ignore the fact the with nanotechnology any disgruntled employee can manufacture a nanoplague that will kill off the whole world's population. Imagine something like a miniature time bomb that spreads like a cold virus.
See also: Herbert's "The White Plauge" (hardly his best book, but similiar to what you're talking about here).
What you write is all nice and reasonable as long as you s/Trusted Computing/the first version of Trusted Computing/g
Trusted Computing may not (probably will not) do X (say, prevent you from running Linux) in the first releases, but the capability is there. You have hardware that can verify the software it loads with public keys or hashes. You have hardware for storing secret keys used to encrypt software and/or other keys. Between those two things, a group consisting of Microsoft/BIOS vendors/etc could indeed do all of the things you are claiming are impossible.
Will they? I kind of doubt it (I would be interested to see what kind of shitstorm it would cause if they did try it). But don't claim that anyone theorizing this stuff is a total whacko, because technologically (if not politically) it's all possible under TC.
OK, totally offtopic now.
Re:Not the source, really
on
Real Security?
·
· Score: 1
no better system than passwords has yet been devised
Depends on how you define 'better'. A smartcard that did RSA signatures would kick a passwords ass. (A simple PIN system that locks the device after N bad attempts would be a good idea for this). The problem is that you need special hardware to interface with it, and it's not going to be supported by a lot of stuff. Even if you could modify your core systems to deal with it, there are a lot of random things that are going to expect just a plain ordinary password no matter what (for example, websites).
You could say similiar things about S/Key, EKE, or half a dozen other things. The problem is that right now most applications don't have any concept that things like this might be used. PAM helps a bit, but it's not like your ASP or Java code is going to be using PAM anytime soon.
Once you've read that, treat yourself to a post where the poster attempts to achieve a maximum crackpot index score by violating all the rules in sequential order.
The best part is the people who respond, half of whom seem totally unable to realize that it's a joke, and the other half who troll along with the original post. A good laugh.
Moore's Law is not the only thing that can make a machine faster. ABIs, compiler/language improvments, and especially the ISA (obvious example: Alpha vs x86) makes a huge difference. In terms of transistor density, sure, someday it's going to have to stop doubling. But that doesn't mean computers can't keep getting faster.
From the article: "As one would expect, hackers need to operate undercover. Thus, in order to find interviewees, Turgeman had to do some detective work and, through journalists, conferences and Web sites, she managed to find hackers willing to talk to her."
Or, she mananged to find some script kiddies or, random people who felt like showing off. According to the article, 'hackers' are considered cool ("Apparently, the image that society has of hackers is generally positive"), so maybe someone thought it would be fun to 'be' one. The quotes by the interviewees are highly non-technical (for example, "When you crack a code, it gives you an amazing feeling", and rants about MS); did she ask any of them if they knew how a TCP handshake worked, or anything?
We don't need a 'new master, same as the old master' ascending to power, but some here seem to think it would be okay.
Historically, Apple has been too fucking stupid to actually become the next IBM or MS. They (sometimes) make really good stuff, and then make dumb business choices. Of course these days with OS X and the G4/G5, they're doing a little better, but they're a hell of a long way from being the new master.
Your concern about Apple people contributing to GPL/BSD stuff is interesting, but basically it would be Apple cutting their own throats. I know Apple has people working full-time on GCC, and probably Darwin/*BSD as well. If they did something like that, trying to "take back" code that someone had written for an FSF project, they would suddenly find that their employees would have a very hard time contributing stuff even when authorized by Apple. After all, Apple might just change their minds, and ask for *that* code back, too. The possible hassle of getting C&Ds from Apple's lawyers at some point later on would outweight the benefits of getting the code.
Of course we should, but are you willing to bet that nobody ever assumed that "long" was a 4 byte quantity when they were allocating memory? That everybody used ~0x4f instead of 0xffffffb0?
Sure they have. Except hopefully by now somebody has tried to use their code on Alpha, or SPARC64, or PPC64, or MIPS64, or any of the other 64-bit CPUs out there, and complained that it didn't work. Either they fixed it (no problem, then), or they didn't (in which case they're both lazy coders and bad maintainers, and so it'd be best to stay away from their stuff anyway).
Best Buy has portable CD players that they sell for 30.00 that read MP3s and there you go, the cheap solution
What model do you use? I've got one like this (about a year old), and it kind of sucks - there is a 5-15 second delay between each song, which can be really annoying.
Do you know if it is based more on the manga, or is it an independent story?
From what I've seen (first 6 episodes), it reflects the manga in style better than the movie did. Less straight-up philosophy, more ass-kicking, but still some really weird ideas coming at you ever once in a while. The fuchikoma/techikoma AI mini-tanks are back, and with the perfect voice; they sound like exited little kids, which might seem weird, but it fits them really well.
The story seems roughly based on the manga plotlines (but keep in mind I've only seen the first 6 episodes), but no closer than the movie was. However there are also what at least seem to be single stories, just one episode things that have no relation to anything else (either the manga or other episodes). I imagine pieces of them will come back in the later episodes though.
Quantum computing does have the potential to make this obsolete. All SSL -- used by banks, governments, might be breakable. PGP would be breakable.
Bzzt. A quantum computer can do a keysearch (for something like DES or AES) in the square root of the number of ops it takes a normal computer. So, for example, DES can be broken with 2^28 effort (sqrt(2^n) = 2^(n/2)). AES with a 128 bit key can be broken with 2^64 effort. And AES (or anything else) with a 256 bit key can be broken with 2^128 effort == impossible. Why do you think AES was required to support 256 bit keys, when 128 bit is unbreakable by anything BUT a quantum computer. Answer: because NIST was worried about quantum computers.
Shor's algorithm makes things possibly dicey for RSA, but nobody (except possibly the NSA, but I would be surprised if they had a machine with more than 50 qubits) knows what the constants are, and nobody will until we build something bigger and try to factor a 'real' composite. It's in P, but that doesn't tell us how hard it is - it just tells us that it doesn't get much harder as the problem gets bigger.
It seems reasonable that governments will tightly control developments in this field once they catch on to what's at stake.
Everyone knows what will happen to crypto when quantum computers are built. NIST, NSA, researchers, anyone who has taken a class in complexity theory, any idiot who reads slashdot, etc, etc. It is not a secret.
Researchers who love quantum mechanics would do better to work on quantum encryption
For the purposes you suggest, quantum crypto is useless. You can only do quantum crypto if you have a fiber optic line between you and whoever you want to talk to (though I hear recently they've got it working over line-of-sight, which means a fiber optic line OR living within 10-30 miles of who you want to talk to). Quantum crypto will not save you from quantum computers.
or on inventing more useful algorithms for quantum computers.
Useful algorithms like the Shor's factorization algorithm? I call that a pretty useful algorithm.
What good is a quantum computer for besides breaking encryption?
Not that much AFAICT. The faster searching and stuff seems interesting, but I doubt I'll be alive when the first database built on a quantum computer is built, so I don't get too into that. I basically think of it as an interesting research project, nothing more.
Would quantum computers be able to do anything with things like elliptic curve cryptography?
Probably. The best known ways of computing discrete logarithms on an elliptic curve are based on the square root of the size (while the hardness of discrete log on a group mod p is logarithmic with size, thus we have to use larger p's). It's rather likely that there are algorithms which can do it faster than square root on a quantum computer, though I haven't heard about anyone developing an algorithm (so far).
Free Trade Area of the Americas (FTAA) treaty, which is an attempt to create a single free trade agreement for the Western Hemisphere
First thing that popped into my head when I read the above sentence: "Greater American Coprosperity Sphere". Obviously comparing this with Japan's policies 1940 is not exactly fair*, but it was almost like a word association.
*: Though it may well end up having a simliar effect in the long run, who knows.
Nuclear reactors are just a new(ish) way of heating water.
Spinning turbines with steam to generate electricity is as old as your hat.
However, there is the method used in the Cassini probe, which basically converts the heat of the nuclear decay directly into electricity. Ah, which are called 'radioisotope thermoelectric generators' (according to here). I can't remember how these worked (I took a class in nuclear engineering ~6 years ago and at one point they explained it, but I'm forgetful), but I'm reasonably sure it's not something that could have been done 150 years ago (unlike nuclear reactors, which, as you say, just uses a new way of heating water).
From article: Dongarra said the cost is so low he questioned whether the college got a special discount.
At $5.2 mil for 1100 machines, I think they paid full market price; that's over $4,500 per machine, and currently Apple is selling dual 2 Ghz G5's for ~$3000. And that's with lots of extras that they wouldn't want in a cluster (ATI 9600, CDRW, etc), which hopefully they convinced Apple they didn't need... (else they've got a whole lot of Mac keyboards sitting around!)
I wonder how much of the cost was the actual machines, and how much was infrastructure and networking stuff (I can just see 1,100 Macs all powered off one extension cord and a bunch of surge protectors).
The major stumbling block today is the motherboard, which will run you a good five hundred bucks.
Yeah, but why bother when you can get an Athlon64 board for = $100? (Unless you want, and can afford, to put 2-4 of those little beauties all together, in which case the answer is obvious).
actually I don't know this... does Athlon 64 have a 64 bit x 64 bit = 128 bit result operation ?
It does (it's just like the 32-bit x86 multiply except it puts the 128-bit result into rax:rdx, instead of the 64-bit result into eax:edx). And so, yes, you can do sick fast PK crypto on it. Half the number of words => 1/4 as much work, cause lots of the underlying algorithms (like multiplication) are O(n^2), or sometimes O(n^~1.6).
The people who use Comeau's compiler are generally concerned with portability, because, as any C++ programmer knows, most compilers completely blow when it comes to standards conformance at the moment.
This makes 0 sense. If someone is worried about C++ portability, they will a) test on multiple compilers, and b) not use super-new features (though AFAIK everything except export is supported by every major compiler). Why would you want a compiler that supported such otherwise-not-supported-by-anyone features if you were worried about portability?
Slashdot Mirror
I once beat a man to death with a Model M keyboard. Worked just fine before and since.
OK, I didn't really. But I'm sure I could have, those things really are invincible (also big and heavy) - fire, physical shock, water (or beer), nothing hurts them.
Everyone seems to ignore the fact the with nanotechnology any disgruntled employee can manufacture a nanoplague that will kill off the whole world's population. Imagine something like a miniature time bomb that spreads like a cold virus.
See also: Herbert's "The White Plauge" (hardly his best book, but similiar to what you're talking about here).
It's also possible that the US Government will invade your house tomorrow and take away all of your rights.
Indeed.
What you write is all nice and reasonable as long as you s/Trusted Computing/the first version of Trusted Computing/g
Trusted Computing may not (probably will not) do X (say, prevent you from running Linux) in the first releases, but the capability is there. You have hardware that can verify the software it loads with public keys or hashes. You have hardware for storing secret keys used to encrypt software and/or other keys. Between those two things, a group consisting of Microsoft/BIOS vendors/etc could indeed do all of the things you are claiming are impossible.
Will they? I kind of doubt it (I would be interested to see what kind of shitstorm it would cause if they did try it). But don't claim that anyone theorizing this stuff is a total whacko, because technologically (if not politically) it's all possible under TC.
OK, totally offtopic now.
no better system than passwords has yet been devised
Depends on how you define 'better'. A smartcard that did RSA signatures would kick a passwords ass. (A simple PIN system that locks the device after N bad attempts would be a good idea for this). The problem is that you need special hardware to interface with it, and it's not going to be supported by a lot of stuff. Even if you could modify your core systems to deal with it, there are a lot of random things that are going to expect just a plain ordinary password no matter what (for example, websites).
You could say similiar things about S/Key, EKE, or half a dozen other things. The problem is that right now most applications don't have any concept that things like this might be used. PAM helps a bit, but it's not like your ASP or Java code is going to be using PAM anytime soon.
Once you've read that, treat yourself to a post where the poster attempts to achieve a maximum crackpot index score by violating all the rules in sequential order.
The best part is the people who respond, half of whom seem totally unable to realize that it's a joke, and the other half who troll along with the original post. A good laugh.
Moore's Law is not the only thing that can make a machine faster. ABIs, compiler/language improvments, and especially the ISA (obvious example: Alpha vs x86) makes a huge difference. In terms of transistor density, sure, someday it's going to have to stop doubling. But that doesn't mean computers can't keep getting faster.
From the article: "As one would expect, hackers need to operate undercover. Thus, in order to find interviewees, Turgeman had to do some detective work and, through journalists, conferences and Web sites, she managed to find hackers willing to talk to her."
Or, she mananged to find some script kiddies or, random people who felt like showing off. According to the article, 'hackers' are considered cool ("Apparently, the image that society has of hackers is generally positive"), so maybe someone thought it would be fun to 'be' one. The quotes by the interviewees are highly non-technical (for example, "When you crack a code, it gives you an amazing feeling", and rants about MS); did she ask any of them if they knew how a TCP handshake worked, or anything?
Sure, it is crackable, but it takes hours to do... making it impractical for eavesdropping on a conversation in real-time.
Actually, A5/1 and A5/2 (the GSM algorithms) can be cracked in real-time.
We don't need a 'new master, same as the old master' ascending to power, but some here seem to think it would be okay.
Historically, Apple has been too fucking stupid to actually become the next IBM or MS. They (sometimes) make really good stuff, and then make dumb business choices. Of course these days with OS X and the G4/G5, they're doing a little better, but they're a hell of a long way from being the new master.
Your concern about Apple people contributing to GPL/BSD stuff is interesting, but basically it would be Apple cutting their own throats. I know Apple has people working full-time on GCC, and probably Darwin/*BSD as well. If they did something like that, trying to "take back" code that someone had written for an FSF project, they would suddenly find that their employees would have a very hard time contributing stuff even when authorized by Apple. After all, Apple might just change their minds, and ask for *that* code back, too. The possible hassle of getting C&Ds from Apple's lawyers at some point later on would outweight the benefits of getting the code.
Of course we should, but are you willing to bet that nobody ever assumed that "long" was a 4 byte quantity when they were allocating memory?
That everybody used ~0x4f instead of 0xffffffb0?
Sure they have. Except hopefully by now somebody has tried to use their code on Alpha, or SPARC64, or PPC64, or MIPS64, or any of the other 64-bit CPUs out there, and complained that it didn't work. Either they fixed it (no problem, then), or they didn't (in which case they're both lazy coders and bad maintainers, and so it'd be best to stay away from their stuff anyway).
Best Buy has portable CD players that they sell for 30.00 that read MP3s and there you go, the cheap solution
What model do you use? I've got one like this (about a year old), and it kind of sucks - there is a 5-15 second delay between each song, which can be really annoying.
Do you know if it is based more on the manga, or is it an independent story?
From what I've seen (first 6 episodes), it reflects the manga in style better than the movie did. Less straight-up philosophy, more ass-kicking, but still some really weird ideas coming at you ever once in a while. The fuchikoma/techikoma AI mini-tanks are back, and with the perfect voice; they sound like exited little kids, which might seem weird, but it fits them really well.
The story seems roughly based on the manga plotlines (but keep in mind I've only seen the first 6 episodes), but no closer than the movie was. However there are also what at least seem to be single stories, just one episode things that have no relation to anything else (either the manga or other episodes). I imagine pieces of them will come back in the later episodes though.
(And the animation is just absolutely amazing)
Is there a good sci-fi book dealing with Nanotech that is *not* written on a Victorian framework? Something a bit more believable?
Gibson's San Francisco Trilogy deals with it sometimes (it's central to the overall plot, but not mentioned much).
Quantum computing does have the potential to make this obsolete. All SSL -- used by banks, governments, might be breakable. PGP would be breakable.
Bzzt. A quantum computer can do a keysearch (for something like DES or AES) in the square root of the number of ops it takes a normal computer. So, for example, DES can be broken with 2^28 effort (sqrt(2^n) = 2^(n/2)). AES with a 128 bit key can be broken with 2^64 effort. And AES (or anything else) with a 256 bit key can be broken with 2^128 effort == impossible. Why do you think AES was required to support 256 bit keys, when 128 bit is unbreakable by anything BUT a quantum computer. Answer: because NIST was worried about quantum computers.
Shor's algorithm makes things possibly dicey for RSA, but nobody (except possibly the NSA, but I would be surprised if they had a machine with more than 50 qubits) knows what the constants are, and nobody will until we build something bigger and try to factor a 'real' composite. It's in P, but that doesn't tell us how hard it is - it just tells us that it doesn't get much harder as the problem gets bigger.
It seems reasonable that governments will tightly control developments in this field once they catch on to what's at stake.
Everyone knows what will happen to crypto when quantum computers are built. NIST, NSA, researchers, anyone who has taken a class in complexity theory, any idiot who reads slashdot, etc, etc. It is not a secret.
Researchers who love quantum mechanics would do better to work on quantum encryption
For the purposes you suggest, quantum crypto is useless. You can only do quantum crypto if you have a fiber optic line between you and whoever you want to talk to (though I hear recently they've got it working over line-of-sight, which means a fiber optic line OR living within 10-30 miles of who you want to talk to). Quantum crypto will not save you from quantum computers.
or on inventing more useful algorithms for quantum computers.
Useful algorithms like the Shor's factorization algorithm? I call that a pretty useful algorithm.
What good is a quantum computer for besides breaking encryption?
Not that much AFAICT. The faster searching and stuff seems interesting, but I doubt I'll be alive when the first database built on a quantum computer is built, so I don't get too into that. I basically think of it as an interesting research project, nothing more.
Would quantum computers be able to do anything with things like elliptic curve cryptography?
Probably. The best known ways of computing discrete logarithms on an elliptic curve are based on the square root of the size (while the hardness of discrete log on a group mod p is logarithmic with size, thus we have to use larger p's). It's rather likely that there are algorithms which can do it faster than square root on a quantum computer, though I haven't heard about anyone developing an algorithm (so far).
Free Trade Area of the Americas (FTAA) treaty, which is an attempt to create a single free trade agreement for the Western Hemisphere
First thing that popped into my head when I read the above sentence: "Greater American Coprosperity Sphere". Obviously comparing this with Japan's policies 1940 is not exactly fair*, but it was almost like a word association.
*: Though it may well end up having a simliar effect in the long run, who knows.
However, there is the method used in the Cassini probe, which basically converts the heat of the nuclear decay directly into electricity. Ah, which are called 'radioisotope thermoelectric generators' (according to here). I can't remember how these worked (I took a class in nuclear engineering ~6 years ago and at one point they explained it, but I'm forgetful), but I'm reasonably sure it's not something that could have been done 150 years ago (unlike nuclear reactors, which, as you say, just uses a new way of heating water).
From article: Dongarra said the cost is so low he questioned whether the college got a special discount.
At $5.2 mil for 1100 machines, I think they paid full market price; that's over $4,500 per machine, and currently Apple is selling dual 2 Ghz G5's for ~$3000. And that's with lots of extras that they wouldn't want in a cluster (ATI 9600, CDRW, etc), which hopefully they convinced Apple they didn't need... (else they've got a whole lot of Mac keyboards sitting around!)
I wonder how much of the cost was the actual machines, and how much was infrastructure and networking stuff (I can just see 1,100 Macs all powered off one extension cord and a bunch of surge protectors).
The major stumbling block today is the motherboard, which will run you a good five hundred bucks.
Yeah, but why bother when you can get an Athlon64 board for = $100? (Unless you want, and can afford, to put 2-4 of those little beauties all together, in which case the answer is obvious).
actually I don't know this... does Athlon 64 have a 64 bit x 64 bit = 128 bit result operation ?
It does (it's just like the 32-bit x86 multiply except it puts the 128-bit result into rax:rdx, instead of the 64-bit result into eax:edx). And so, yes, you can do sick fast PK crypto on it. Half the number of words => 1/4 as much work, cause lots of the underlying algorithms (like multiplication) are O(n^2), or sometimes O(n^~1.6).
The people who use Comeau's compiler are generally concerned with portability, because, as any C++ programmer knows, most compilers completely blow when it comes to standards conformance at the moment.
This makes 0 sense. If someone is worried about C++ portability, they will a) test on multiple compilers, and b) not use super-new features (though AFAIK everything except export is supported by every major compiler). Why would you want a compiler that supported such otherwise-not-supported-by-anyone features if you were worried about portability?
Comeau uses the EDG frontend, as do some others (including Intel, I believe, though I'm not sure).
I don't think so (but I could be wrong). KAI, a compiler company that Intel bought a few years ago, did use EGD.