Umm, the Chinese did not cut off the supply of Opium. They cut off the demand for opium. The British were illegally smuggling opium from India into China, then the Chinese enforced their laws, leading to war.
Maybe I'm missing something, but as someone who has working with the techniques referenced in the parent post - I'm not sure where the funny mod came from. Both clustering packet attributes and nonnegative matrix factorizations could be used for anomaly detection. And as someone who has also worked on CUDA a good bit, I think both of those problems have solutions that fit CUDA's concurrency model.
I get the impression that the mods saw big words and assumed this was a joke about buzzwords, but in fact that's a reasonable approach to this problem.
Unlike the flying car, this one is actually completely feasible.
The flying car won't happen because of problems with physics. The vehicle has to produce sufficient upward thrust to lift the weight of a car + possibly 6 humans. And the amount of force needed will only go down if we develop lighter engines or better materials.
Simulating a brain is a purely computational problem. What's more, if we simulate at the cellular level as the article suggests, it's an embarrassingly parallel problem. This means that even if further iterations of Moore's Law keep us stuck at ~3-4GHz and only expand parallelism, it would still get the full benefit. This is one situation in which simply throwing more computational power at it will eventually succeed in producing results (no guarantee of positive results obviously, but results).
Maybe I'm missing something here. I've always wondered why there was a rush to block images of child abuse like this. As long as these sites are up, there is still a possibility for authorities to identify the guilty parties through the websites.
If every ISP blocks 100%, then not even cops can get an unfiltered connection. That means that they have stopped trying to catch the child pornographers, they just want to pretend they don't exist.
These are real children being abused. Their abusers are handing the police evidence. Why the rush to ignore it? Why not just monitor them? Keep track of who visits www.kiddieporn.com or whatever.
It wouldn't be impossible to give private sandboxes to "legacy" apps that don't use the new secure APIs.
Indeed, and some previous versions of windows did exactly the same thing - they had a mode for running DOS executables that needed access to the whole system. The same will happen as happened then - trojans that want to access the old API's... still can. All we've achieved is forcing much code rewriting while preserving our vulnerability to malicious code 100%.
No, Microsoft sits down for a bit and thinks about how they can best support a particular kind of hardware. They sometimes call up the relevant hardware manufacturers and say "hey, helps up define the API for this."
Man I really wish there was a '-1 Just Plain Factually Incorrect'.
They're not putting a $185K price tag on domain names, they're putting it on the application for a new generic top level domain. That is, in www.google.com, the $185K isn't to apply for the "google" pieces, it's to apply for the "com" piece.
As an individual, why would you care? And what would a "domain-name-pimp" get out of the deal? Nothing as far as I can see. This looks more like it's being marketed to governments.
This is pretty simple if an actual non-profit is using your open-source product. Just set up a web page selling *discs* of your software. Then, donate the software, on discs, to your local charity.
This way, you are donating an actual good to an actual charity. Further, as long as you keep the price somewhat in line with competing proprietary products, you decide market value. And to sweeten the deal further, you can donate more discs at each major update (as long as you don't go too crazy).
Of course, you can still distribute the software for free over the internet - after all you aren't selling the software, you're selling the disc.
The big difference here is that you aren't donating your time - the time is all on you. However, you do have a disc which has a market value that you can donate.
Umm, you are aware that "99.99999% more" roughly means "slightly less than twice as much", right? As in, he reads 199.99999% as much XKCD as he does NYT.
to "fix" the key when you were that age? I'd probably destroy the first key trying, but I bet I'd get there on the second one. That's not a prevention mechanism, that's a challenge:-)
I would disagree. When I was a small child, the parents would spank me when I did wrong, but the only thing I ever "learned" from it was that my parents had anger management issues. They didn't really, of course, but that's how my mind spinned the events.
Then one day something happened - I got to be stronger than both my mom (who was the discipliner of the family). You can't effectively spank a kid who can catch your hand. Then she started disciplining me in other ways - like groundings, or (the worst) talking to me about what I had done. And you know what happened? I learned to respect her. Not fear her, not hate her, respect her.
It has nothing to do with children. When the bounds on an interpersonal relationship are enforced with violence, there cannot be respect. There can be fear, which is what most parents won't admit to wanting, but not real respect.
Right, but the point is that this is not (or at least should not) be to replace to pairs of eyes looking at the results. The opinions of two human radiologists is still the best. The proposal is that adding a computer to the one pair of eyes looking at these pictures currently will improve the accuracy. That 199th woman would get missed anyways, because she wouldn't have two doctors looking at it, she'd have one doctor looking at it.
Of course, there is a greater issue of trust. If you accept chips made by unknown fabricators, do you know what microcode has been implemented? If you cannot examine the "source code" of the chips being used how can you actually trust that these chips are not doing things behind your back to reveal your identity and files?
Haha, seriously? And where, pray tell, in an instruction set is one going to hide operations that complex? Is it in the microcode for XOR that they will embed a driver for a network card and the code necessary to compromise your files? Or perhaps they will hide it in the "branch-on-equal" instruction?
Nothing is going to hide in the microcode. Not because the govenment/Microsoft/the Illuminati/your mother in law wouldn't do that, but because they can't. Any instruction that took that long and had those side-effects would have immediately noticeable effects on system performance.
The closest you could get is to embed an undocumented instruction that allowed other code to escalate to ring-0, but that still require malicious software to exploit it.
Agreed. Every voting system has a point of failure somewhere. Those who actually count the votes must trust that those who bring votes to them are bringing actual ballots. Once each polling place tolls its own votes, these are added up elsewhere, where they are assuming they are getting honest data.
The point isn't that it's bulletproof, the point is that its better. The more humans we can involve openly before the election and the fewer humans we can involve unsupervised during the election, the better.
Electronic voting is such a horrible, horrible idea.
Not really true. Electronic voting is a great idea implemented very badly. Let's consider how it compares to paper ballots:
If everyone is honest.
Paper ballots produce the correct output.
Electronic ballots produce the correct output.
If those counting the votes are not honest.
Paper votes get recounted if and only if someone powerful enough convinces a judge to order a recount.
The incorrect outcomes are hopefully detected beforehand in the massive open nationwide audit (because that should exist).
The problem isn't that electronic voting is a bad idea, the problem is that as a citizen I can't audit the code. Remember, if you put an honest algorithm into a computer, you get an honest answer.
Is this true? If so, I'd like to take a look. I'd probably still use LaTeX, as I've already invested the learning curve, but it could be interesting.
But then, remember that beautiful maths are just a feature of LaTeX, not the main focus. It's a document processor, which is nice in that someone like me with little aesthetic taste can simply specify the *content* of my paper and let LaTeX worry about the layout. And maybe you can write a paper in word that looks as good, but I know I've never been able to.
You put your money in a bank and trust that they're gonna give it back when you ask for it.
I trust that they'll give me my money back because they're a heavily regulated industry with a lot to lose if they don't. A "cloud computing" business has nothing to lose except your business.
You pay a cable company to provide you internet, and trust that none of their techs are reading your email.
Really? I have an idea how easy it would be for them, so no - I don't trust that my ISP is not reading my email. My internet comes from a phone company, not a cable company, and most of them have already wiretapped phones (and got away with it) so why wouldn't they read my email?
You use your credit/debit card at countless businesses, and trust a whole chain of people not to swipe your card number.
But if any of these people misuse my card, I have some legal recourse - and my liability is limited to $50.
On the other hand, if I keep my businesses customer communications in my GMail, and then Google stops offering it, I could potential lose thousands of dollars and, here's the biggie, have no recourse.
This is why Comcast doesn't want to say anything because rather than admit they are overselling the service, they are putting in amorphous 'caps' of service in which customers have no idea what the limit is or how to know if they are close to it.
Umm, since when is 250GB "amorphous"? I mean, I guess we could quible over the prefix mega, but we both know that's splitting hairs. Did you RTFS?
All I'm saying is that this is a turn for the better. Comcast used to do shifty things with their customers traffic to avoid capping. Now they are publicly announcing a cap.
They want to pretend or make it sound like they offer unlimited service while at the same time cutting off people for going over some unstated, unannounced limit
Again, a good scare monger, but it runs directly contrary to the facts of the situation. They just stated and announced the cap - 250GB
I could make all the calls I want to anyone, anywhere within the service area, for as long as I want to call. Or make no calls at all, only receive them. Or not use the phone at all. In any case, I'd pay the same rate.
Okay, but the comparison isn't really relevant. After all, how would you feel if your ISP was tracking which webpages you visited (who you called), how long you viewed (how long you talked), and what you downloaded (gov't wiretaps)?
They can also transmit phone via lossy compression, which I don't think you'd like them doing to your data. Telephones != the internet.
Don't get me wrong, I hate Comcast as much as the next guy, and I think they should provide a meter, but this is a step in the direction of honesty.
Seriously? Let's take a second to (there goes my karma) see this from Comcast's point of view. They want to run a profitable business. We want net neutrality - the ability to use our bits as we see fit.
Comcast sold unlimited internet, and found that they were losing money on too many accounts - so they switched to selling limited internet. This is a step *forward*, not back. Comcast is now being honest about its caps.
Ultimately, we need them more than they need us. There's two ways to handle the increased bandwidth usage that bittorrent and streaming HD have created - we can let Comcast DPI, throttle, and otherwise mess with our bits, or we can accept that the cost of a commodity is fixed to how much we use it (like every other commodity we use).
I don't see why its so bad that Comcast should enforce bandwidth caps. And we get all excited about the cutoff, but think how mad you would be if they went the other route - charging you a higher rate per Mb when you went over.
I quite agree. This begs the significant question, how much more would a VIA chip cost of VIA payed the patent fees? It seems that we should pay the same amount as consumers, because if VIA pays the patents, then (correct me if I'm wrong) all legal problems regarding open drivers vanish. I'd pay a bit more for a card with open drivers.
* There are 9 planets orbiting the sun. Turns out Pluto isn't even a planet.
You do realize that Pluto didn't shrink or turn out to be significantly smaller than we thought. We just decided to stop calling it a planet any more. Nothing happened there except that we changed an arbitrary classification.
Jeez, can we please stop pretending that the word "art" has an objective meaning and that impartial, objective, rational decisions can be made regarding what does and doesn't fall under the category of "art".
"Art" is, and always has been, given a different definition by each observer.
Slashdot Mirror
Umm, the Chinese did not cut off the supply of Opium. They cut off the demand for opium. The British were illegally smuggling opium from India into China, then the Chinese enforced their laws, leading to war.
Maybe I'm missing something, but as someone who has working with the techniques referenced in the parent post - I'm not sure where the funny mod came from. Both clustering packet attributes and nonnegative matrix factorizations could be used for anomaly detection. And as someone who has also worked on CUDA a good bit, I think both of those problems have solutions that fit CUDA's concurrency model.
I get the impression that the mods saw big words and assumed this was a joke about buzzwords, but in fact that's a reasonable approach to this problem.
Apparently the lousy moderators have won that game--and I expect the moderation of this post to prove my point (yet again).
Anyone ever notice how many posts with quotes like "Go ahead, mod me down, but..." or "Hey, I've got karma to burn..." seem to get highly moderated?
Unlike the flying car, this one is actually completely feasible.
The flying car won't happen because of problems with physics. The vehicle has to produce sufficient upward thrust to lift the weight of a car + possibly 6 humans. And the amount of force needed will only go down if we develop lighter engines or better materials.
Simulating a brain is a purely computational problem. What's more, if we simulate at the cellular level as the article suggests, it's an embarrassingly parallel problem. This means that even if further iterations of Moore's Law keep us stuck at ~3-4GHz and only expand parallelism, it would still get the full benefit. This is one situation in which simply throwing more computational power at it will eventually succeed in producing results (no guarantee of positive results obviously, but results).
Maybe I'm missing something here. I've always wondered why there was a rush to block images of child abuse like this. As long as these sites are up, there is still a possibility for authorities to identify the guilty parties through the websites.
If every ISP blocks 100%, then not even cops can get an unfiltered connection. That means that they have stopped trying to catch the child pornographers, they just want to pretend they don't exist.
These are real children being abused. Their abusers are handing the police evidence. Why the rush to ignore it? Why not just monitor them? Keep track of who visits www.kiddieporn.com or whatever.
It wouldn't be impossible to give private sandboxes to "legacy" apps that don't use the new secure APIs.
Indeed, and some previous versions of windows did exactly the same thing - they had a mode for running DOS executables that needed access to the whole system. The same will happen as happened then - trojans that want to access the old API's... still can. All we've achieved is forcing much code rewriting while preserving our vulnerability to malicious code 100%.
No, Microsoft sits down for a bit and thinks about how they can best support a particular kind of hardware. They sometimes call up the relevant hardware manufacturers and say "hey, helps up define the API for this."
Man I really wish there was a '-1 Just Plain Factually Incorrect'.
They're not putting a $185K price tag on domain names, they're putting it on the application for a new generic top level domain. That is, in www.google.com, the $185K isn't to apply for the "google" pieces, it's to apply for the "com" piece.
As an individual, why would you care? And what would a "domain-name-pimp" get out of the deal? Nothing as far as I can see. This looks more like it's being marketed to governments.
This is pretty simple if an actual non-profit is using your open-source product. Just set up a web page selling *discs* of your software. Then, donate the software, on discs, to your local charity.
This way, you are donating an actual good to an actual charity. Further, as long as you keep the price somewhat in line with competing proprietary products, you decide market value. And to sweeten the deal further, you can donate more discs at each major update (as long as you don't go too crazy).
Of course, you can still distribute the software for free over the internet - after all you aren't selling the software, you're selling the disc.
The big difference here is that you aren't donating your time - the time is all on you. However, you do have a disc which has a market value that you can donate.
Umm, you are aware that "99.99999% more" roughly means "slightly less than twice as much", right? As in, he reads 199.99999% as much XKCD as he does NYT.
to "fix" the key when you were that age? I'd probably destroy the first key trying, but I bet I'd get there on the second one. That's not a prevention mechanism, that's a challenge :-)
I would disagree. When I was a small child, the parents would spank me when I did wrong, but the only thing I ever "learned" from it was that my parents had anger management issues. They didn't really, of course, but that's how my mind spinned the events.
Then one day something happened - I got to be stronger than both my mom (who was the discipliner of the family). You can't effectively spank a kid who can catch your hand. Then she started disciplining me in other ways - like groundings, or (the worst) talking to me about what I had done. And you know what happened? I learned to respect her. Not fear her, not hate her, respect her.
It has nothing to do with children. When the bounds on an interpersonal relationship are enforced with violence, there cannot be respect. There can be fear, which is what most parents won't admit to wanting, but not real respect.
Right, but the point is that this is not (or at least should not) be to replace to pairs of eyes looking at the results. The opinions of two human radiologists is still the best. The proposal is that adding a computer to the one pair of eyes looking at these pictures currently will improve the accuracy. That 199th woman would get missed anyways, because she wouldn't have two doctors looking at it, she'd have one doctor looking at it.
Of course, there is a greater issue of trust. If you accept chips made by unknown fabricators, do you know what microcode has been implemented? If you cannot examine the "source code" of the chips being used how can you actually trust that these chips are not doing things behind your back to reveal your identity and files?
Haha, seriously? And where, pray tell, in an instruction set is one going to hide operations that complex? Is it in the microcode for XOR that they will embed a driver for a network card and the code necessary to compromise your files? Or perhaps they will hide it in the "branch-on-equal" instruction?
Nothing is going to hide in the microcode. Not because the govenment/Microsoft/the Illuminati/your mother in law wouldn't do that, but because they can't. Any instruction that took that long and had those side-effects would have immediately noticeable effects on system performance.
The closest you could get is to embed an undocumented instruction that allowed other code to escalate to ring-0, but that still require malicious software to exploit it.
If those counting the votes are not honest.
This assumes that those counting the votes did not substitute their own votes for the ones actually cast.
Emphasis (and original quote) mine.
Agreed. Every voting system has a point of failure somewhere. Those who actually count the votes must trust that those who bring votes to them are bringing actual ballots. Once each polling place tolls its own votes, these are added up elsewhere, where they are assuming they are getting honest data.
The point isn't that it's bulletproof, the point is that its better. The more humans we can involve openly before the election and the fewer humans we can involve unsupervised during the election, the better.
Electronic voting is such a horrible, horrible idea.
Not really true. Electronic voting is a great idea implemented very badly. Let's consider how it compares to paper ballots:
The problem isn't that electronic voting is a bad idea, the problem is that as a citizen I can't audit the code. Remember, if you put an honest algorithm into a computer, you get an honest answer.
Is this true? If so, I'd like to take a look. I'd probably still use LaTeX, as I've already invested the learning curve, but it could be interesting.
But then, remember that beautiful maths are just a feature of LaTeX, not the main focus. It's a document processor, which is nice in that someone like me with little aesthetic taste can simply specify the *content* of my paper and let LaTeX worry about the layout. And maybe you can write a paper in word that looks as good, but I know I've never been able to.
You put your money in a bank and trust that they're gonna give it back when you ask for it.
I trust that they'll give me my money back because they're a heavily regulated industry with a lot to lose if they don't. A "cloud computing" business has nothing to lose except your business.
You pay a cable company to provide you internet, and trust that none of their techs are reading your email.
Really? I have an idea how easy it would be for them, so no - I don't trust that my ISP is not reading my email. My internet comes from a phone company, not a cable company, and most of them have already wiretapped phones (and got away with it) so why wouldn't they read my email?
You use your credit/debit card at countless businesses, and trust a whole chain of people not to swipe your card number.
But if any of these people misuse my card, I have some legal recourse - and my liability is limited to $50. On the other hand, if I keep my businesses customer communications in my GMail, and then Google stops offering it, I could potential lose thousands of dollars and, here's the biggie, have no recourse.
This is why Comcast doesn't want to say anything because rather than admit they are overselling the service, they are putting in amorphous 'caps' of service in which customers have no idea what the limit is or how to know if they are close to it.
Umm, since when is 250GB "amorphous"? I mean, I guess we could quible over the prefix mega, but we both know that's splitting hairs. Did you RTFS?
All I'm saying is that this is a turn for the better. Comcast used to do shifty things with their customers traffic to avoid capping. Now they are publicly announcing a cap.
They want to pretend or make it sound like they offer unlimited service while at the same time cutting off people for going over some unstated, unannounced limit
Again, a good scare monger, but it runs directly contrary to the facts of the situation. They just stated and announced the cap - 250GB
I could make all the calls I want to anyone, anywhere within the service area, for as long as I want to call. Or make no calls at all, only receive them. Or not use the phone at all. In any case, I'd pay the same rate.
Okay, but the comparison isn't really relevant. After all, how would you feel if your ISP was tracking which webpages you visited (who you called), how long you viewed (how long you talked), and what you downloaded (gov't wiretaps)? They can also transmit phone via lossy compression, which I don't think you'd like them doing to your data. Telephones != the internet.
Don't get me wrong, I hate Comcast as much as the next guy, and I think they should provide a meter, but this is a step in the direction of honesty.
Seriously? Let's take a second to (there goes my karma) see this from Comcast's point of view. They want to run a profitable business. We want net neutrality - the ability to use our bits as we see fit. Comcast sold unlimited internet, and found that they were losing money on too many accounts - so they switched to selling limited internet. This is a step *forward*, not back. Comcast is now being honest about its caps. Ultimately, we need them more than they need us. There's two ways to handle the increased bandwidth usage that bittorrent and streaming HD have created - we can let Comcast DPI, throttle, and otherwise mess with our bits, or we can accept that the cost of a commodity is fixed to how much we use it (like every other commodity we use). I don't see why its so bad that Comcast should enforce bandwidth caps. And we get all excited about the cutoff, but think how mad you would be if they went the other route - charging you a higher rate per Mb when you went over.
I quite agree. This begs the significant question, how much more would a VIA chip cost of VIA payed the patent fees? It seems that we should pay the same amount as consumers, because if VIA pays the patents, then (correct me if I'm wrong) all legal problems regarding open drivers vanish. I'd pay a bit more for a card with open drivers.
I wonder how many other people would even get that reference?
At least one. And if this patent gets approved, it will be a typical case of American blind justice. ;-)
* There are 9 planets orbiting the sun. Turns out Pluto isn't even a planet.
You do realize that Pluto didn't shrink or turn out to be significantly smaller than we thought. We just decided to stop calling it a planet any more. Nothing happened there except that we changed an arbitrary classification.
Jeez, can we please stop pretending that the word "art" has an objective meaning and that impartial, objective, rational decisions can be made regarding what does and doesn't fall under the category of "art". "Art" is, and always has been, given a different definition by each observer.