He just said that science teachers should be free to discuss creationism, i.e. why it isn't science.
The creationists couldn't have asked for a better outcome. They have always claimed that scientists "believe" in evolution as if it is was a religious faith, and that they won't tolerate criticism of their ideas or other beliefs. Scientists always said that was nonsense... and yet, now we have an example of a man who has been persecuted for speaking out of turn.
The greatest threat to science isn't the creationists, but the armies of Dawkinsbots who defend "science" with fundamentalist fervour. If you're going to fight creationism, you have to stick to the facts, otherwise you're as irrational as they are.
Processes help in another way; memory sharing is explicit rather than implicit, so it is easier to maintain cache coherence on a multi-core CPU. You always know that cache coherence is not necessary between two processes, except for the areas of memory that are explicitly shared, because the processes exist in their own memory spaces. Whereas two threads always share the same memory space.
This is not useful in today's SMP systems but it will be needed in the near future since cache coherence algorithms do not scale well.
Trusting banks is unavoidable if you ever use money for anything. $1000 in a bank account is the same as $1000 in $1 bills: if you can't trust the bank, neither has any value.
The point here is that DRM authentication should be moved away from publishers to a third party, able to act on behalf of customers (who have licensed content) and on behalf of publishers (who provide the content).
I understand your point, but I don't think that selling IP should be a broken business model. I don't buy the idea that all games should be online services, because many types of game don't fit that model. More generally, most types of media don't fit that model.
If DRM is unavoidable, and I believe that it is, then it is in our interests to ensure that DRM works for us as far as possible. This is something that SecuROM simply does not do; it's an ad-hoc solution that limits installations and only works on today's platforms. We need DRM that will look after our rights as well as those of the content producers, and that can only be done through standardisation and the use of a trusted third party for authentication.
Some Steam games have both the Steam DRM and something else. Bioshock, for example, came with SecuROM regardless of whether you got it from Steam or a physical shop. (This is why the demo didn't work on my PC, and this is why I didn't buy it.)
Like the DRM used in games consoles, the Steam DRM is tolerable because it works properly, and the rules that it imposes are consistent across nearly all of the games (Bioshock being an exception). We do not see this "flash mob" rating all the XBox games as 1: why not? Because the DRM in that case doesn't get in the way.
Many of the problems with DRM can be solved by standardisation, but the standard must not only involve a single DRM platform for all software, but also a single online service for authentication. This would be a trusted third party - like a bank. It would assure us that purchases will continue to function after the publisher goes out of business. Steam does both of these things quite well, although we are all assuming that Valve won't go bankrupt and sell its IP to a company with less of a clue.
Until I reached the end of that paragraph, I assumed I was reading an excerpt from the awful book... after all, it does sound like an awful blog entry of the sort that would quite justifiably be "lost".
The Idle conspiracy theory. Do you Slashdot guys draw straws for who has to write the "idle" articles? Is there a prize for the worst "idle" article? Poll idea: let us vote for the worst "idle" article after a few more weeks of this. No way is this the bottom of the barrel - you can do worse!
I suppose you could just add an exception for the site you want to access. (Four clicks?) Or your corporate IT people could add their signing certificate to the version of Firefox they distribute.
I don't understand the "antifeature" accusation at all.
Yeah, that's fair. I should have said "fusion". I was trying to reference the old joke about fusion, "the energy source of the future and always will be", but recent news about a company called D-Wave made me think about Pons and Fleischmann.
Re:no encryption that YOU didn't write is safe
on
Is Hushmail Still Safe?
·
· Score: 4, Insightful
We got past this in the 90s; initially they said that all encryption would have to be weak (e.g. 40 bit) or go through their chips (Clipper, etc.). But they found that this didn't stand up to the reality of WWW era. What worked in the 80s for the few users of encryption at that time simply couldn't scale up for web commerce. Strong encryption was a commercial necessity, so the attempts to control the industry had to be dropped. The export restrictions disappeared, and because DES was now too weak to be useful, the new AES standard was introduced.
Is AES full of back doors for the NSA? Almost certainly not, since these could also be used by any resourceful group of cryptographers, including the Chinese version of the NSA.
Is quantum computing already being used to crack AES? No. Quantum computing is the cold fusion of our industry.
It is most annoying to see people trying to make money from their theses, especially as almost nobody actually succeeds in doing so. In many cases a thesis is written using tax money, and in all cases the purpose of the thesis is to advance scientific knowledge. So why try to stop people reading it?
The only reason I can see is that a thesis is often not very good, and if you publish it on the web, this will be obvious to anyone who cares to look. Whereas if someone has to go to the British Library or buy it for $$$ in order to read it, you're not going to get as many critics.
Fine, testing is one reason for two code paths; you want to make sure the generated code works on all x86 CPUs, but you don't want to thoroughly test on all of them. But that means that icc isn't suitable for compiling benchmarks. because different code is being run depending on the CPUID. The comparison is not a fair test; you have two variables (the code and the CPU) instead of one (the CPU).
Two questions:
- Why didn't Intel tell benchmark writers not to use icc? Obviously the results will be unfair if an Intel CPU is being compared against a non-Intel CPU, and Intel will be accused of cheating.
- Why not call the memcpy in the C library instead of doing a one-byte-at-a-time memcpy? Amateur coding? Or professional evilness masquerading as incompetence?
Is that the whole security premise of "trusted bootchain" is wrong... Someone able to reverse engineer the checksumming code can simply modify the checksummer so that the bootchain always passes validation.
No, this won't work, because the checksum is computed by a hardware device. The whole idea is that if you modify the kernel or the boot loader, you invalidate the signature. The TPM can be configured to refuse to release keys if the signature is invalid.
This is very useful. For example, on my own machine I use full-disk encryption. But the boot loader and kernel have to be unencrypted so the machine can boot. If my machine had a TPM (it doesn't, it's too old), then TCPA would allow me to (a) store part of the disk encryption key in the TPM, and (b) program the TPM to release this key if and only if the boot loader and kernel have been signed by me. This would improve the security of my machine, because an attacker might add a trojan to the boot loader or kernel in order to steal my keys.
who's to say you aren't running Windows in a virtual machine?
The TPM! However it must be said that TCPA is not intended to secure a machine against its owner; even if the TPM is integrated into the CPU you can probably still use differential power analysis to get the keys out of it. In the Linux world, TCPA is entirely a good thing because it will always be under the control of the user. Microsoft might have other ideas, but who is to say that the market will accept them.
There was a story on Digg about 200 pounds of marijuana that went to the wrong address, and the recipient immediately called the police. Everyone there was incredulous. Someone remarked that the only phone call they'd make would be to Pizza Hut.
Everyone there is an idiot. Don't you think the sender/receiver is likely to want it back? And that they might well use deadly force to get it?
200lbs = 90kg. If you calculate that using the police's idea of "street value", it's worth at least $100bn. Unless you are a gangsta with your own private army of thugs, the police are your only option.
These admissions are usually followed by "in my youth, it was a mistake, no we're not going to legalise it", which seems to me the worst sort of hypocrisy:
They rationalise statements of this sort by adding "but it wasn't as strong, back then". The usual claim is that new "genetically modified" strains (more accurately, selectively bred) have a much higher proportion of the active ingredient, and are therefore more dangerous and more likely to make you go mad. (They've mostly given up saying that pot use will turn you into a heroin addict, since half the country has tried it and not ended up on smack. The big money is in schizophrenia research now; "reefer madness" is back.)
I doubt it. How will the middle class copyright infringers defend their actions in court?
"Sure, I did download all those movies, but information wants to be free!" "Yeah, I copied that program, but I don't believe in imaginary property!"
I am, of course, not a lawyer, but I don't think those defenses will do any good. The prosecutor will say the defendant is a thief, and 99% of the other middle class people will agree. "Tough luck," they will say. "You shouldn't have stolen those films. I don't steal movies and software, I pay for them. What's wrong with you, that you think you can take things for free and get away with it?" Even if the pirates are somehow right and information truly isn't property in some way, they still lose, because they look like thieves in front of the majority of people who still pay for things.
How many people are in prison for nonviolent drugs offences, "crimes" that wouldn't even be illegal in a free country? You can criminalise millions of people as long as the majority has a reason to look down on them; you can prohibit anything that the majority doesn't do (or won't admit to doing).
We can expect the War on Pirates to be the same runaway success as the War on Drugs. I think they will probably eventually succeed in forcing piracy off the public Internet, just as they forced drug dealers into the back streets. The pirating will, of course, continue by sneakernet.
Right, so you split your computation up into small units that can be efficiently allocated to the many core array. This allows you to express the parallelism in the program properly, because you're not constrained by the coarse granularity of a thread model. Cool.
But the problem here is how you write the code itself. Purely functional code maps really well onto this model, but nobody wants to retrain all their programmers to use Haskell. We're going to end up with a hybrid C-based language: but what restrictions should exist in it? This depends on what is easy to implement in hardware - because if we wanted to stick with what was easy to implement in software, we'd carry on trying to squeeze a few extra instructions per second out of a conventional CPU architecture.
The biggest restriction turns out to be the "R" in RAM. Most of our programs use memory in an unpredictable way, pulling data from all over the memory space, and this doesn't map well to a many core architecture. You can put caches at every core, but the cache miss penalty is astronomical, not to mention the problems of keeping all the caches coherent. Random access won't scale; we will need something else, and it will break lots of programs.
This is going to lead to some really shitty science, because:
Many core architectures will only be good for running certain types of program: not just programs that can be split into tiny units of computation, but programs that access RAM in a predictable way.
The many core architects will pick the programs that work best on their system; these may or may not have anything to do with real applications for many core systems (And what is an application for a many core system anyway? Don't say graphics...)
It will be hard to quantitatively compare one many core architecture with another because of the different assumptions about what programs are able to do in each case. There are too many variables; there is no "control variable".
I think that the eventual winning architecture will be the one that is easiest to write programs for. But it will have to be so much better at running those programs that it is worth the effort of porting them. So it will have to be a huge improvement, or easy to port to, or some combination of the two. However, those are qualitative properties. Anyone could argue that their architecture is better than another - and they will.
Why is this modded down? He makes a good point. If you're downloading shady EXEs from P2P, then it's just a matter of time before you get pwned, because you are a sitting duck.
The filename says "RELOADED", and you think you can trust them, but is it really from them? If I wanted to get a keylogger on a bunch of computers, I could easily add it to a "no cd crack" for a popular game and put it on P2P. I'd put "RELOADED" in the filename and fake the text files. I'd make sure that the keylogger activated itself after a couple of weeks to maximise my infection rate. If I did this, how would you tell?
Yeah, I'm still laughing at the irony of Ubisoft being accused of "stealing" from the guys who pirated their game.
Those poor pirates! Ubisoft just ripped off their work, downloaded and distributed it for free, and never gave them anything in return. Ubisoft is destroying the pirate industry!
Christ. I suggest (based on your signature) that you restart smoking. You might have a shorter life, but you'll enjoy it more. This sort of anger is not normal; liberal attitudes have caused problems but liberalism is not a "disease" any more than "torture" is a solution.
Slashdot Mirror
But he didn't do that!
He just said that science teachers should be free to discuss creationism, i.e. why it isn't science.
The creationists couldn't have asked for a better outcome. They have always claimed that scientists "believe" in evolution as if it is was a religious faith, and that they won't tolerate criticism of their ideas or other beliefs. Scientists always said that was nonsense... and yet, now we have an example of a man who has been persecuted for speaking out of turn.
The greatest threat to science isn't the creationists, but the armies of Dawkinsbots who defend "science" with fundamentalist fervour. If you're going to fight creationism, you have to stick to the facts, otherwise you're as irrational as they are.
Hey at least that frood knows where his towel is.
Processes help in another way; memory sharing is explicit rather than implicit, so it is easier to maintain cache coherence on a multi-core CPU. You always know that cache coherence is not necessary between two processes, except for the areas of memory that are explicitly shared, because the processes exist in their own memory spaces. Whereas two threads always share the same memory space.
This is not useful in today's SMP systems but it will be needed in the near future since cache coherence algorithms do not scale well.
Trusting banks is unavoidable if you ever use money for anything. $1000 in a bank account is the same as $1000 in $1 bills: if you can't trust the bank, neither has any value.
The point here is that DRM authentication should be moved away from publishers to a third party, able to act on behalf of customers (who have licensed content) and on behalf of publishers (who provide the content).
The same single point of failure exists (and is tolerated) in every games console.
I understand your point, but I don't think that selling IP should be a broken business model. I don't buy the idea that all games should be online services, because many types of game don't fit that model. More generally, most types of media don't fit that model.
If DRM is unavoidable, and I believe that it is, then it is in our interests to ensure that DRM works for us as far as possible. This is something that SecuROM simply does not do; it's an ad-hoc solution that limits installations and only works on today's platforms. We need DRM that will look after our rights as well as those of the content producers, and that can only be done through standardisation and the use of a trusted third party for authentication.
Some Steam games have both the Steam DRM and something else. Bioshock, for example, came with SecuROM regardless of whether you got it from Steam or a physical shop. (This is why the demo didn't work on my PC, and this is why I didn't buy it.)
Like the DRM used in games consoles, the Steam DRM is tolerable because it works properly, and the rules that it imposes are consistent across nearly all of the games (Bioshock being an exception). We do not see this "flash mob" rating all the XBox games as 1: why not? Because the DRM in that case doesn't get in the way.
Many of the problems with DRM can be solved by standardisation, but the standard must not only involve a single DRM platform for all software, but also a single online service for authentication. This would be a trusted third party - like a bank. It would assure us that purchases will continue to function after the publisher goes out of business. Steam does both of these things quite well, although we are all assuming that Valve won't go bankrupt and sell its IP to a company with less of a clue.
Until I reached the end of that paragraph, I assumed I was reading an excerpt from the awful book... after all, it does sound like an awful blog entry of the sort that would quite justifiably be "lost".
The Idle conspiracy theory. Do you Slashdot guys draw straws for who has to write the "idle" articles? Is there a prize for the worst "idle" article? Poll idea: let us vote for the worst "idle" article after a few more weeks of this. No way is this the bottom of the barrel - you can do worse!
But no one knows what song out of my thousands I'm using,
Maybe they would look at the access times to see what files you'd opened recently?
I suppose you could just add an exception for the site you want to access. (Four clicks?) Or your corporate IT people could add their signing certificate to the version of Firefox they distribute.
I don't understand the "antifeature" accusation at all.
Yeah, that's fair. I should have said "fusion". I was trying to reference the old joke about fusion, "the energy source of the future and always will be", but recent news about a company called D-Wave made me think about Pons and Fleischmann.
We got past this in the 90s; initially they said that all encryption would have to be weak (e.g. 40 bit) or go through their chips (Clipper, etc.). But they found that this didn't stand up to the reality of WWW era. What worked in the 80s for the few users of encryption at that time simply couldn't scale up for web commerce. Strong encryption was a commercial necessity, so the attempts to control the industry had to be dropped. The export restrictions disappeared, and because DES was now too weak to be useful, the new AES standard was introduced.
Is AES full of back doors for the NSA? Almost certainly not, since these could also be used by any resourceful group of cryptographers, including the Chinese version of the NSA.
Is quantum computing already being used to crack AES? No. Quantum computing is the cold fusion of our industry.
It is most annoying to see people trying to make money from their theses, especially as almost nobody actually succeeds in doing so. In many cases a thesis is written using tax money, and in all cases the purpose of the thesis is to advance scientific knowledge. So why try to stop people reading it?
The only reason I can see is that a thesis is often not very good, and if you publish it on the web, this will be obvious to anyone who cares to look. Whereas if someone has to go to the British Library or buy it for $$$ in order to read it, you're not going to get as many critics.
Fine, testing is one reason for two code paths; you want to make sure the generated code works on all x86 CPUs, but you don't want to thoroughly test on all of them. But that means that icc isn't suitable for compiling benchmarks. because different code is being run depending on the CPUID. The comparison is not a fair test; you have two variables (the code and the CPU) instead of one (the CPU).
Two questions:
- Why didn't Intel tell benchmark writers not to use icc? Obviously the results will be unfair if an Intel CPU is being compared against a non-Intel CPU, and Intel will be accused of cheating.
- Why not call the memcpy in the C library instead of doing a one-byte-at-a-time memcpy? Amateur coding? Or professional evilness masquerading as incompetence?
Is that the whole security premise of "trusted bootchain" is wrong... Someone able to reverse engineer the checksumming code can simply modify the checksummer so that the bootchain always passes validation.
No, this won't work, because the checksum is computed by a hardware device. The whole idea is that if you modify the kernel or the boot loader, you invalidate the signature. The TPM can be configured to refuse to release keys if the signature is invalid.
This is very useful. For example, on my own machine I use full-disk encryption. But the boot loader and kernel have to be unencrypted so the machine can boot. If my machine had a TPM (it doesn't, it's too old), then TCPA would allow me to (a) store part of the disk encryption key in the TPM, and (b) program the TPM to release this key if and only if the boot loader and kernel have been signed by me. This would improve the security of my machine, because an attacker might add a trojan to the boot loader or kernel in order to steal my keys.
who's to say you aren't running Windows in a virtual machine?
The TPM! However it must be said that TCPA is not intended to secure a machine against its owner; even if the TPM is integrated into the CPU you can probably still use differential power analysis to get the keys out of it. In the Linux world, TCPA is entirely a good thing because it will always be under the control of the user. Microsoft might have other ideas, but who is to say that the market will accept them.
There was a story on Digg about 200 pounds of marijuana that went to the wrong address, and the recipient immediately called the police. Everyone there was incredulous. Someone remarked that the only phone call they'd make would be to Pizza Hut.
Everyone there is an idiot. Don't you think the sender/receiver is likely to want it back? And that they might well use deadly force to get it?
200lbs = 90kg. If you calculate that using the police's idea of "street value", it's worth at least $100bn. Unless you are a gangsta with your own private army of thugs, the police are your only option.
These admissions are usually followed by "in my youth, it was a mistake, no we're not going to legalise it", which seems to me the worst sort of hypocrisy:
They rationalise statements of this sort by adding "but it wasn't as strong, back then". The usual claim is that new "genetically modified" strains (more accurately, selectively bred) have a much higher proportion of the active ingredient, and are therefore more dangerous and more likely to make you go mad. (They've mostly given up saying that pot use will turn you into a heroin addict, since half the country has tried it and not ended up on smack. The big money is in schizophrenia research now; "reefer madness" is back.)
I doubt it. How will the middle class copyright infringers defend their actions in court?
"Sure, I did download all those movies, but information wants to be free!" "Yeah, I copied that program, but I don't believe in imaginary property!"
I am, of course, not a lawyer, but I don't think those defenses will do any good. The prosecutor will say the defendant is a thief, and 99% of the other middle class people will agree. "Tough luck," they will say. "You shouldn't have stolen those films. I don't steal movies and software, I pay for them. What's wrong with you, that you think you can take things for free and get away with it?" Even if the pirates are somehow right and information truly isn't property in some way, they still lose, because they look like thieves in front of the majority of people who still pay for things.
How many people are in prison for nonviolent drugs offences, "crimes" that wouldn't even be illegal in a free country? You can criminalise millions of people as long as the majority has a reason to look down on them; you can prohibit anything that the majority doesn't do (or won't admit to doing).
We can expect the War on Pirates to be the same runaway success as the War on Drugs. I think they will probably eventually succeed in forcing piracy off the public Internet, just as they forced drug dealers into the back streets. The pirating will, of course, continue by sneakernet.
Right, so you split your computation up into small units that can be efficiently allocated to the many core array. This allows you to express the parallelism in the program properly, because you're not constrained by the coarse granularity of a thread model. Cool.
But the problem here is how you write the code itself. Purely functional code maps really well onto this model, but nobody wants to retrain all their programmers to use Haskell. We're going to end up with a hybrid C-based language: but what restrictions should exist in it? This depends on what is easy to implement in hardware - because if we wanted to stick with what was easy to implement in software, we'd carry on trying to squeeze a few extra instructions per second out of a conventional CPU architecture.
The biggest restriction turns out to be the "R" in RAM. Most of our programs use memory in an unpredictable way, pulling data from all over the memory space, and this doesn't map well to a many core architecture. You can put caches at every core, but the cache miss penalty is astronomical, not to mention the problems of keeping all the caches coherent. Random access won't scale; we will need something else, and it will break lots of programs.
This is going to lead to some really shitty science, because:
I think that the eventual winning architecture will be the one that is easiest to write programs for. But it will have to be so much better at running those programs that it is worth the effort of porting them. So it will have to be a huge improvement, or easy to port to, or some combination of the two. However, those are qualitative properties. Anyone could argue that their architecture is better than another - and they will.
There really isn't any reason why one that is drunk or under the influence of drugs, should be sitting at the wheel.
You should watch North by Northwest for an example. Granted, it is slightly far fetched...
Pirates who complain about people stealing their work are also hypocrites.
However, my post was not intended to be taken seriously!
Why is this modded down? He makes a good point. If you're downloading shady EXEs from P2P, then it's just a matter of time before you get pwned, because you are a sitting duck.
The filename says "RELOADED", and you think you can trust them, but is it really from them? If I wanted to get a keylogger on a bunch of computers, I could easily add it to a "no cd crack" for a popular game and put it on P2P. I'd put "RELOADED" in the filename and fake the text files. I'd make sure that the keylogger activated itself after a couple of weeks to maximise my infection rate. If I did this, how would you tell?
Yeah, I'm still laughing at the irony of Ubisoft being accused of "stealing" from the guys who pirated their game.
Those poor pirates! Ubisoft just ripped off their work, downloaded and distributed it for free, and never gave them anything in return. Ubisoft is destroying the pirate industry!
Christ. I suggest (based on your signature) that you restart smoking. You might have a shorter life, but you'll enjoy it more. This sort of anger is not normal; liberal attitudes have caused problems but liberalism is not a "disease" any more than "torture" is a solution.