I honestly haven't followed them that closely. I agree that any DDoSing they did is stupid as it doesn't really prove anything to do with security. But as for the actual hacks, while I very much dislike the actions, and they're probably not the sorts of people I'd call friends, I think they are doing a service in telling people "hey that door is not nearly as secure as you thought it was" encouraging people to either get a better lock, or at least be more careful before putting their valuables inside.
Right now the only other way that security issues come to the attention of the media is through public hostile hacks, and who knows how many of those go unreported. LulzSec is bringing a lot more attention to the problem of security, without many of the costs of more hostile hacks, on the whole I think that's a good thing.
I've seen that study before and I think it's misinterpreted and what they're measuring is a short term argumentative effect, not a long term belief effect.
When I'm presented with some information that contradicts my views my first reaction is that my credibility and intelligence are being questioned, so I try to defend myself.
It's only later, after the argument is done, when I've have a few days to digest the information and arguments, that I may start changing my mind, or realize the arguments and evidence that were presented to me were bunk. This isn't even necessarily a mental malfunction as a defence mechanism, there are some very skilled persuaders and if I changed my mind every time I was presented with a seemingly airtight argument I would be a very confused person indeed.
I'm 95% sure you're just trolling, so if you are serious and want me to give a real reply please put in the effort of listing some examples of media manufactured scandals.
The problem with Palin is a lot of people take her very seriously despite an absurd number of warning signs to the contrary. And if the media didn't keep jumping on her scandals and missteps in the past she very well could still be considered a contender for the Republican nomination. Sometimes these media scandals are just a distraction from important issues, but other times they're a chance to say "hey, that person you've been praising as the second coming? Well here's yet another piece of evidence that they're complete nutjob, now go back and think about why you were such an idiot before".
Sure, most of her followers won't get the message, but a few will, and the world will be that much wiser for it.
I think you're being too harsh on the actual AGW proponents. In my experience the scientists actually make fairly conservative claims, "the lies and ridiculous hyperbole about what global warming is going to do" mostly come from the AGW-deniers. They either blow out of proportion some isolated statement from a scientist, or in this case heavily distort the original report to make it sound ridiculous.
As to why so many smart people believe in AGW, I think this article is a great illustration. If you just read the summary a smart person might think this was pretty damning for AGW, it's only when you read the comments when you realize how distorted the original article was. If you don't happen to read the right resources and are stuck in the information bubble where the original article came from, a smart person could very easily be mislead about many topics.
"I may not agree with what you say, but I will defend to the death your right to say it."
But that has never been Apple's policy with app store, they've actually been fairly strict with what apps they've allowed and I'm not sure everyone is clear exactly what the guidelines are.
Now I don't think for a moment that Apple actually agrees with the "Gay Cure" app, but I wouldn't blame someone for thinking that Apple accepting an app means Apple agrees with it.
Who cares, if Watson's artificial reflexes gave it a few milliseconds' advantage on the buzzer? Who even cares if it'd take it a second longer to read the clue via OCR? So what if Watson would be 5% faster or 10% slower, if conditions were slightly different? Moore's Law makes that level of difference utterly irrelevant - in 18 months time, Watson will be *100% faster* (or even today, if IBM just threw more hardware at it).
It actually is fairly important.
Watson won convincingly, but how much of that win was due to it's superior answering ability, and how much was due to it's speed?
It's clearly a better Jeopardy player than the top humans, and will become only stronger as time goes on. But Jeopardy is a game of trivia answering and reaction speed, and since the humans were often denied the opportunity to answer by Watson's speed we don't know who was better at actually answering trivia.
Clearly the trivia answering abilities will improve in the future, but not necessarily as fast as the speed, and we don't know what gap, if any, there is.
Now if I ever managed to USE the icons / links at the top of the Slashdot page (and now on the Slashdot toolbar) more than once every 3 months, it might be good to have them handy.
You mean in the middle of reading comments you don't suddenly decide to mail slashdot some feedback, submit a story, change your options, etc?
Worst part is that I remember being able to minimize the previous one, this one just sticks there. How am I supposed to read the comments page? Using arrow keys is tedious and the spacebar always means some new content will be hidden behind the bar.
So my understanding is that P is the set of problems both verifiable and solvable in polynomial time.
NP is the set of problems verifiable in polynomial time, but we can't currently solve in polynomial time
NP-Complete is a set of problems in NP where one can be reduced to another in polynomial time, so if you could solve one in poly time you could solve all in poly time
3-SAT is NP-Complete.
So assuming this code is correct (which sounds very unlikely), wouldn't it merely prove that NP-Complete is in P? Not to say this wouldn't be a major result, but that's not the same as the more general P==NP.
No chilling effects doesn't mean you can say or read whatever you want with absolutely no adverse effects, your actions can still be used as evidence of a crime, or just piss people off. Chilling effects is when the speech itself is punishable.
Your search history can be used as evidence, as can your speech history, or your library record, that's why we have courts. To issue warrants to look for evidence, and to decide if such evidence is actually evidence of guilt.
We still don't know if the suit was properly served, but if it was, and the guy just completely ignored it, than he does deserve at least a little slap for ignoring the letter and wasting even more peoples time (though obviously not an $11 million slap).
Besides, I'm sure DomainsByProxy receives more than a few notifications and has a system in place for making sure paperwork reaches the proper parties.
Myers complaint was that Kurzweil was claiming that all information necessary to create the brain was stored in the letters of the genome, and that Kurzweil was ignoring/ignorant of all the complex interactions of the environment in actually creating the brain.
Kurzweil's response was that the report was incorrect and Myers misunderstood what he was talking about.. He wasn't saying that you could reproduce the brain using only the genome, but that the initial design was contained in the genome, and that even with the information added by the environment later you could still get only so much complexity.
For instance consider a 5 line c program. To run it you need a compiler, cpu, os, hard disk, ram, etc. But the essential information isn't in the computer, it's in the 5 lines of code.
Kurzweil does have a tendency of exaggeration and is overly optimistic in assuming exponential growth curves will simply continue, and his timelines are off since I don't think we'd have the computational power to simulate brains on his timescale. But I don't think PZ's primary complaint applied since it pertained to a claim that Kurzweil never really made.
If Vinay Deolalikar is awarded the $1,000,000 Clay Millennium Prize for his proof of PNP, then I, Scott Aaronson, will personally supplement his prize by the amount of $200,000.
I’m dead serious—and I can afford it about as well as you’d think I can.
People yearn to come here to get quality higher education. Ask any international (undergraduate/graduate) student who is studying here.
Sorry, but you are making a sweeping and entirely false generalization there. From what I have seen, most of the international students in my engineering program came here because a degree from an American university was perceived as more valuable than a degree from their own country. I saw far more cheating and far less competence among the international students, even those that spoke English fluently, than I did among the American students; they were not going to school because they were seeking a better education than what they could get back home.
Though you're dealing with quite a sampling bias. It's the students from countries with poor education systems who will be most motivated to go to another country for schooling. The schools in my country are quite good, why do I want to move to the states, and pay a massively inflated tuition, when I can get the same education at home for cheaper?
Partially, according to that post java.vendor and java.vm.vendor also have the Sun ID and those were never changed, eclipse was looking for the name in another (less standard?) location which did get changed.
So if this is the case and eclipse did mess up by looking for the name from a non-standard location, but my comment is still valid as to the issues with Java.
What was described is not "an approved interface to figure this out", it's an approved interface to figure out the vendor, but it says nothing about the vendor and those properties still need to be done on an ad-hoc per vendor basis.
It's not quite that simple. If you read the bug reports it sounds like the Sun (ok Oracle) VM has a lower than usual memory limit, thus requiring the "-XX:MaxPermSize" argument (which may not always be available) to be used to increase the memory limit.
Therefore you need to use -XX:MaxPermSize for Sun, but for another VM the argument might not only be unnecessary, but break that VM since it's an unsupported argument. Eclipse, and other applications also running into memory limits who want to fix the issue automatically, need two things, 1) a way to detect the memory limit, and 2) a way to detect if -XX:MaxPermSize is supported. Since they've apparently written an isSunVM method to try to solve this issue I'm guessing 1&2 aren't available.
Yes it's an ugly hack, but it sounds like it's a hack necessitated by a shortcoming in Java.
As for Oracle it's a bit of bad luck on their part, but I am a little surprised that at no point in their QA did someone try running Eclipse, the most popular Java IDE, on their update.
", I'd think this wouldn't get very good designs since people wouldn't do that much work when they probably won't get paid." that explains why there is no open source programs~
The people that do ti for the love of doing it will continue to do it even if it's just a hobby. and they will be good. People that want to do it because they can get a regular check from come company will find another way to make work with minimal effort.
Though you can do even better paying the enthusiastic people, and anyway, it's not clear if people can actually make a living with this model.
By the way, the way to win is to bid a dollar, then split the money amongst all the players.
That only works if every player is in on the single bid, and is happy getting $20/n. What happens if an individual bids $2 hoping that the group will get scared off and they'll get the $20 for themselves?
I'm kinda curious how this will work out long term, I'd think this wouldn't get very good designs since people wouldn't do that much work when they probably won't get paid.
On the other hand if you've worked X hours already, than it's worthwhile to work X+1 to make sure you're the best design so you do get paid.
It reminds me of an economist trick of auctioning off a $20 bill. Basically the idea is just that, auction off the $20 bill, except not only does the winner have to pay, but the second place bidder as well.
So if I bid $18, and another student bids $19 than I'm still paying $18 but I won't get the $20. So I now have to bid $20 just to get the $20 bill and break even, and the other student has to bid $21 so they're only losing $1 instead of $19 and so on until one of the participants smartens up and decides the bill isn't worth the cost.
Anyone know of any other industries where such a big proportion of the work is done in the bidding process?
Not quite, I think the main difference is that a Matte screen is a lambertian surface, meaning it doesn't necessarily reflect less light, but it reflects it diffusely.
Now I haven't really tried this so it's based on theory, but I think the difference is thus.
Reflected light is bad regardless, the more reflected light, the harder it is to see the light emitted from the screen.
With the matte screen you never get a direct reflection from a light source, thus its harder to overwhelm the screen light. However, if you are in a high ambient light environment (ie outside) than the amount of ambient light may still be too much for the matte screen, and even if you're in a medium light environment you'll still be dealing with some reflected light reducing the contrast.
With the glossy screen if you avoid a direct reflection you see very little reflected light and get a better image, this makes it potentially readable outside. Inside if you can avoid reflections you'll get almost no reflection and have a higher contrast image. Of course this assumes it's easy to avoid getting a direct reflection, which may not be feasible in all cases. Particularly in an environment with a lot of bright objects (outside again) you may get a lot of visible reflections depending on the surroundings.
Thus it largely depends on your individual circumstance whether glossy or matte is better.
ps. I can't remember the last time I tried my laptop outside so please correct me if I've made any errors in my assumptions.
DNA fingerprints are not as random as many think. The markers used were not designed for a nation wide database situation. Hence collisions could be a big problem. That is two people with the same fingerprint (at least at the very small parts of DNA we look at) can in fact be very likely with a database this size.
The collision is a problem only if both are plausible suspects:
Or if the actual killer isn't even in your database.
In this case I'm assuming the scientists do understand the odds of false positives, and they did find real evidence linking the suspect to the killing. However when you are using a DNA fishing expedition to find the suspect I don't think that DNA can also be considered much in the way of evidence of guilt.
Though along those lines if, during every meeting, some marketing guy is saying "you're a dumb fraud, give me all your code so I can show you all the dumb mistakes you made".
Now this person may find the occasional small real mistake, but they don't really know how to code, and most of the time they're just misunderstanding something and wasting everyone's time.
Would you really be that motivated to go out of your way to get them all your code?
I'm not saying this is the actual state, but I suspect that many of the climatologists have this opinion which is why the data isn't released as much as it should.
Slashdot Mirror
The serfs thing is hyperbole, but the rest is known as starving the beast.
I honestly haven't followed them that closely. I agree that any DDoSing they did is stupid as it doesn't really prove anything to do with security. But as for the actual hacks, while I very much dislike the actions, and they're probably not the sorts of people I'd call friends, I think they are doing a service in telling people "hey that door is not nearly as secure as you thought it was" encouraging people to either get a better lock, or at least be more careful before putting their valuables inside.
They're not necessarily doing a disservice.
Right now the only other way that security issues come to the attention of the media is through public hostile hacks, and who knows how many of those go unreported. LulzSec is bringing a lot more attention to the problem of security, without many of the costs of more hostile hacks, on the whole I think that's a good thing.
You mean like throwing him off a boat?
Yeah, that would probably be effective.
I've seen that study before and I think it's misinterpreted and what they're measuring is a short term argumentative effect, not a long term belief effect.
When I'm presented with some information that contradicts my views my first reaction is that my credibility and intelligence are being questioned, so I try to defend myself.
It's only later, after the argument is done, when I've have a few days to digest the information and arguments, that I may start changing my mind, or realize the arguments and evidence that were presented to me were bunk. This isn't even necessarily a mental malfunction as a defence mechanism, there are some very skilled persuaders and if I changed my mind every time I was presented with a seemingly airtight argument I would be a very confused person indeed.
I'm 95% sure you're just trolling, so if you are serious and want me to give a real reply please put in the effort of listing some examples of media manufactured scandals.
The problem with Palin is a lot of people take her very seriously despite an absurd number of warning signs to the contrary. And if the media didn't keep jumping on her scandals and missteps in the past she very well could still be considered a contender for the Republican nomination. Sometimes these media scandals are just a distraction from important issues, but other times they're a chance to say "hey, that person you've been praising as the second coming? Well here's yet another piece of evidence that they're complete nutjob, now go back and think about why you were such an idiot before".
Sure, most of her followers won't get the message, but a few will, and the world will be that much wiser for it.
I think you're being too harsh on the actual AGW proponents. In my experience the scientists actually make fairly conservative claims, "the lies and ridiculous hyperbole about what global warming is going to do" mostly come from the AGW-deniers. They either blow out of proportion some isolated statement from a scientist, or in this case heavily distort the original report to make it sound ridiculous.
As to why so many smart people believe in AGW, I think this article is a great illustration. If you just read the summary a smart person might think this was pretty damning for AGW, it's only when you read the comments when you realize how distorted the original article was. If you don't happen to read the right resources and are stuck in the information bubble where the original article came from, a smart person could very easily be mislead about many topics.
How nice for them.
"I may not agree with what you say, but I will defend to the death your right to say it."
But that has never been Apple's policy with app store, they've actually been fairly strict with what apps they've allowed and I'm not sure everyone is clear exactly what the guidelines are.
Now I don't think for a moment that Apple actually agrees with the "Gay Cure" app, but I wouldn't blame someone for thinking that Apple accepting an app means Apple agrees with it.
Who cares, if Watson's artificial reflexes gave it a few milliseconds' advantage on the buzzer? Who even cares if it'd take it a second longer to read the clue via OCR? So what if Watson would be 5% faster or 10% slower, if conditions were slightly different? Moore's Law makes that level of difference utterly irrelevant - in 18 months time, Watson will be *100% faster* (or even today, if IBM just threw more hardware at it).
It actually is fairly important.
Watson won convincingly, but how much of that win was due to it's superior answering ability, and how much was due to it's speed?
It's clearly a better Jeopardy player than the top humans, and will become only stronger as time goes on. But Jeopardy is a game of trivia answering and reaction speed, and since the humans were often denied the opportunity to answer by Watson's speed we don't know who was better at actually answering trivia.
Clearly the trivia answering abilities will improve in the future, but not necessarily as fast as the speed, and we don't know what gap, if any, there is.
Now if I ever managed to USE the icons / links at the top of the Slashdot page (and now on the Slashdot toolbar) more than once every 3 months, it might be good to have them handy.
You mean in the middle of reading comments you don't suddenly decide to mail slashdot some feedback, submit a story, change your options, etc?
Worst part is that I remember being able to minimize the previous one, this one just sticks there. How am I supposed to read the comments page? Using arrow keys is tedious and the spacebar always means some new content will be hidden behind the bar.
So my understanding is that P is the set of problems both verifiable and solvable in polynomial time.
NP is the set of problems verifiable in polynomial time, but we can't currently solve in polynomial time
NP-Complete is a set of problems in NP where one can be reduced to another in polynomial time, so if you could solve one in poly time you could solve all in poly time
3-SAT is NP-Complete.
So assuming this code is correct (which sounds very unlikely), wouldn't it merely prove that NP-Complete is in P? Not to say this wouldn't be a major result, but that's not the same as the more general P==NP.
Am I misunderstanding something?
No chilling effects doesn't mean you can say or read whatever you want with absolutely no adverse effects, your actions can still be used as evidence of a crime, or just piss people off. Chilling effects is when the speech itself is punishable.
Your search history can be used as evidence, as can your speech history, or your library record, that's why we have courts. To issue warrants to look for evidence, and to decide if such evidence is actually evidence of guilt.
5 of the places listed (including the #1, Dildo) are in Newfoundland. Which isn't that surprising.
The address on file for the domain is that of DomainsByProxy, and notice was probably never delivered to the actual site owner.
It wasn't DomainsByProxy at the time of the suit
"Our Whois was public prior to this article, and they still sued the wrong company."
We still don't know if the suit was properly served, but if it was, and the guy just completely ignored it, than he does deserve at least a little slap for ignoring the letter and wasting even more peoples time (though obviously not an $11 million slap).
Besides, I'm sure DomainsByProxy receives more than a few notifications and has a system in place for making sure paperwork reaches the proper parties.
Well no.
Myers complaint was that Kurzweil was claiming that all information necessary to create the brain was stored in the letters of the genome, and that Kurzweil was ignoring/ignorant of all the complex interactions of the environment in actually creating the brain.
Kurzweil's response was that the report was incorrect and Myers misunderstood what he was talking about.. He wasn't saying that you could reproduce the brain using only the genome, but that the initial design was contained in the genome, and that even with the information added by the environment later you could still get only so much complexity.
For instance consider a 5 line c program. To run it you need a compiler, cpu, os, hard disk, ram, etc. But the essential information isn't in the computer, it's in the 5 lines of code.
Kurzweil does have a tendency of exaggeration and is overly optimistic in assuming exponential growth curves will simply continue, and his timelines are off since I don't think we'd have the computational power to simulate brains on his timescale. But I don't think PZ's primary complaint applied since it pertained to a claim that Kurzweil never really made.
Note this is from a respected MIT prof:
My hunch is he's pretty sure it's broken.
People yearn to come here to get quality higher education. Ask any international (undergraduate/graduate) student who is studying here.
Sorry, but you are making a sweeping and entirely false generalization there. From what I have seen, most of the international students in my engineering program came here because a degree from an American university was perceived as more valuable than a degree from their own country. I saw far more cheating and far less competence among the international students, even those that spoke English fluently, than I did among the American students; they were not going to school because they were seeking a better education than what they could get back home.
Though you're dealing with quite a sampling bias. It's the students from countries with poor education systems who will be most motivated to go to another country for schooling. The schools in my country are quite good, why do I want to move to the states, and pay a massively inflated tuition, when I can get the same education at home for cheaper?
Partially, according to that post java.vendor and java.vm.vendor also have the Sun ID and those were never changed, eclipse was looking for the name in another (less standard?) location which did get changed.
So if this is the case and eclipse did mess up by looking for the name from a non-standard location, but my comment is still valid as to the issues with Java.
What was described is not "an approved interface to figure this out", it's an approved interface to figure out the vendor, but it says nothing about the vendor and those properties still need to be done on an ad-hoc per vendor basis.
It's not quite that simple. If you read the bug reports it sounds like the Sun (ok Oracle) VM has a lower than usual memory limit, thus requiring the "-XX:MaxPermSize" argument (which may not always be available) to be used to increase the memory limit.
Therefore you need to use -XX:MaxPermSize for Sun, but for another VM the argument might not only be unnecessary, but break that VM since it's an unsupported argument. Eclipse, and other applications also running into memory limits who want to fix the issue automatically, need two things, 1) a way to detect the memory limit, and 2) a way to detect if -XX:MaxPermSize is supported. Since they've apparently written an isSunVM method to try to solve this issue I'm guessing 1&2 aren't available.
Yes it's an ugly hack, but it sounds like it's a hack necessitated by a shortcoming in Java.
As for Oracle it's a bit of bad luck on their part, but I am a little surprised that at no point in their QA did someone try running Eclipse, the most popular Java IDE, on their update.
", I'd think this wouldn't get very good designs since people wouldn't do that much work when they probably won't get paid."
that explains why there is no open source programs~
The people that do ti for the love of doing it will continue to do it even if it's just a hobby. and they will be good.
People that want to do it because they can get a regular check from come company will find another way to make work with minimal effort.
Though you can do even better paying the enthusiastic people, and anyway, it's not clear if people can actually make a living with this model.
By the way, the way to win is to bid a dollar, then split the money amongst all the players.
That only works if every player is in on the single bid, and is happy getting $20/n. What happens if an individual bids $2 hoping that the group will get scared off and they'll get the $20 for themselves?
I'm kinda curious how this will work out long term, I'd think this wouldn't get very good designs since people wouldn't do that much work when they probably won't get paid.
On the other hand if you've worked X hours already, than it's worthwhile to work X+1 to make sure you're the best design so you do get paid.
It reminds me of an economist trick of auctioning off a $20 bill. Basically the idea is just that, auction off the $20 bill, except not only does the winner have to pay, but the second place bidder as well.
So if I bid $18, and another student bids $19 than I'm still paying $18 but I won't get the $20. So I now have to bid $20 just to get the $20 bill and break even, and the other student has to bid $21 so they're only losing $1 instead of $19 and so on until one of the participants smartens up and decides the bill isn't worth the cost.
Anyone know of any other industries where such a big proportion of the work is done in the bidding process?
Not quite, I think the main difference is that a Matte screen is a lambertian surface, meaning it doesn't necessarily reflect less light, but it reflects it diffusely.
Now I haven't really tried this so it's based on theory, but I think the difference is thus.
Reflected light is bad regardless, the more reflected light, the harder it is to see the light emitted from the screen.
With the matte screen you never get a direct reflection from a light source, thus its harder to overwhelm the screen light. However, if you are in a high ambient light environment (ie outside) than the amount of ambient light may still be too much for the matte screen, and even if you're in a medium light environment you'll still be dealing with some reflected light reducing the contrast.
With the glossy screen if you avoid a direct reflection you see very little reflected light and get a better image, this makes it potentially readable outside. Inside if you can avoid reflections you'll get almost no reflection and have a higher contrast image. Of course this assumes it's easy to avoid getting a direct reflection, which may not be feasible in all cases. Particularly in an environment with a lot of bright objects (outside again) you may get a lot of visible reflections depending on the surroundings.
Thus it largely depends on your individual circumstance whether glossy or matte is better.
ps. I can't remember the last time I tried my laptop outside so please correct me if I've made any errors in my assumptions.
DNA fingerprints are not as random as many think. The markers used were not designed for a nation wide database situation. Hence collisions could be a big problem. That is two people with the same fingerprint (at least at the very small parts of DNA we look at) can in fact be very likely with a database this size.
The collision is a problem only if both are plausible suspects:
Or if the actual killer isn't even in your database.
In this case I'm assuming the scientists do understand the odds of false positives, and they did find real evidence linking the suspect to the killing. However when you are using a DNA fishing expedition to find the suspect I don't think that DNA can also be considered much in the way of evidence of guilt.
Though along those lines if, during every meeting, some marketing guy is saying "you're a dumb fraud, give me all your code so I can show you all the dumb mistakes you made".
Now this person may find the occasional small real mistake, but they don't really know how to code, and most of the time they're just misunderstanding something and wasting everyone's time.
Would you really be that motivated to go out of your way to get them all your code?
I'm not saying this is the actual state, but I suspect that many of the climatologists have this opinion which is why the data isn't released as much as it should.