I'm surprised more ISPs (particularly foreign ones where bandwidth is pricey) haven't looked at ways to bias traffic to share internally.
Perhaps part of the reason is that last mile bandwidth is scarcer than backbone bandwidth, so an ISP doesn't save as much by encouraging its customers to share with each other (backbone bandwidth saved, last mile bandwidth remains the same) as it does by discouraging them from sharing at all (backbone and last mile bandwidth saved)?
And just like the holy grail, it doesn't exist. Ever heard the expression "garbage in, garbage out"? Now imagine how that applies to a huge database of information compiled from diverse sources (including unverified, anonymous tips), where nothing is ever thrown away, and where nobody's quite sure what they're looking for.
The human brain is amazingly good at finding patterns - so good that it often finds patterns that aren't really there. Even with years of experience, training, and peer review, professional scientists are pretty bad at handling problems like confirmation bias, post hoc reasoning and the file-drawer effect - how are law enforcement agents likely to fare, with no statistical training and no effective oversight?
The people constructing these databases are falling into the trap of believing that more data means better data. That's an understandable mistake for people who are usually "data-poor", such as archaeologists, historians and detectives. But anyone from the "data-rich" sciences will tell you that once you have the data you face a whole new set of problems, and I very much doubt that counterterrorism officials, working in conditions of secrecy and under pressure to justify their jobs, are going to handle those problems in a rigorous way.
Please note that I'm not trying to say "police officers are too stupid to understand statistics" - scientists make these mistakes all the time, but they operate in an atmosphere of relative transparency and competition, where it's usually in some other scientist's interest to bring errors to light. The same conditions don't apply to government officials.
What does this mean? It means that false positives will lead to innocent people being monitored, blacklisted and imprisoned without trial, while false negatives will mean genuine threats go undetected. We urgently need to make our governments understand that more data doesn't necessarily lead to better decisions.
And really, that's the whole point of this [Ars Technica] 'news' story - not to tell the news, but to slant it and spin it until it is no longer recognizable and then to attach editorial comments unrelated to main story.
"Unrelated?" Do you even understand why this is news? Politicians are at risk of intimidation, harrassment or even assassination because of the routine collection and sharing of personal information, and the lax controls over access to that information, that have become normal in our society. This could end up distorting the democratic process. That is the main story.
About the spying, tapping calls going out of or into from this country to know terrorist locations is not the same as spying on Tom calling Jerry across the street.
Without judicial oversight, how do you know which is actually being done?
Does it present it truthfully or in tinfoil hat methods?
Nice false dichotomy you've got there. Let's be honest, you have no more idea who's really being spied on than I do, and that's the whole point: if the spying were well-regulated and limited in scope, it could be overseen by trusted senior judges. There's already a secret court for that specific purpose. If the administration refuses to submit the spying to judicial oversight and is willing to take major political risks to keep the whole matter out of the courts, what does that imply?
The paper's online edition is here, but unfortunately the column only seems to be available in the print edition. A timeline of the rise of the Nazis in March 1933 is here.
It was the people that made it possible not the Nazi government.
In my opinion both were necessary. The Nazis expertly manipulated public opinion, so while it's true that many people supported them, I don't believe it follows that those people would have acted the same way if the Nazi party had not existed, and likewise the Nazis could not have acted the way they did without a certain amount of public support.
The US didn't need a police state or a Nazi government in power to do what we did to the native Americans or the native Africans.
Good point, but I didn't claim a police state was necessary for genocide, just that it would make repression and genocide easier to carry out.
In answer to your first point, a society that is structured to make policing easier is a police state. There may be other, better reasons for collecting huge amounts of information about people, but the reasons being put forward in this article were related to policing. Specifically, it was suggested that people should give up their individual dignity to make policing easier.
To answer your second point, I never claimed that a police state would lead to racism, oppression or genocide, only that it would make them easier to carry out.
Your third point is well taken, I should try to keep a cool head when talking about these things.:-) But I'm not entirely convinced that complaining about the construction of a police state plays into the hands of fascists. We shouldn't get hysterical, but we shouldn't be completely quiescent either.
The worst part of nazism wasn't the "papers please" aspect of the Hitler regime, rather the racism, the oppression (not quite the same as surveillance), and the eventual genocide.
Right, but it was the construction of a police state that made the racism, oppression and genocide possible. I don't believe the current UK or US governments plan to start imprisoning their opponents or murdering people en masse, but they're building infrastructure that will make that kind of thing a lot easier for future governments.
There's a column in the International Herald Tribune that reprints the news from 100, 75 and 50 years ago. Right now the 75-year section of the column is charting Hitler's methodical replacement of the German Republic with a Fascist state. It's a horribly fascinating process to watch, for two reasons: first, we know how it ends, and second, we can see many of the same moves being attempted today.
Please stop bleating like a sheep and start using your brain. The "cost" of distributing "n" copies of music is now almost ZERO. Why do you insist people still have to pay $15-$20 for a "CD" or $.99 for a "song"?
I can't believe this was modded insightful. Did you actually read the post you're responding to, or did the red mist come down after you saw the word "copyright"? The GP wasn't "bleating" or "insisting people still have to pay $15-$20", he/she was just pointing out that regardless of what the law should be, file sharing currently is illegal in most places, so Wired is correct to describe it as such.
Giving someone a copy of a book is not like giving 3,000,000 other people a copy of a book.
3 million to 1 is a pretty impressive share ratio - most "pirates" upload roughly as much as they download, meaning they give away one copy in return for the copy they receive.
Your main point is correct - file sharing is illegal in many jurisdictions - but please don't try to dress it up with dramatic numbers.
Of course, now that we know they were wrong, they should given the resources to do a better job next time, preferably a better budget more power to operate without the ACLU breathing down their necks
You're trying to blame the ACLU for the WMD fiasco? Seriously? That makes about as much sense as blaming lesbians for 9/11.
Ping time is probably a good enough approximation of network distance, and it also helps you to avoid overloaded peers. If you use application-layer pings you don't even need raw sockets (unlike traceroute for example).
If only SSL had been designed to make man-in-the-middle attacks impossible. Oh wait, it was! Your browser contains root certificates that are used to verify that your bank's certificate hasn't been replaced or modified by an attacker. MITM attack against SSH? Maybe, if you don't check the key fingerprint (and I doubt anyone does). MITM attack against SSL? No chance, unless the server has a self-signed certificate, something no bank would consider.
They were seditionists but I wouldn't call them terrorists - they started a militia war against an occupying army but AFAIK they didn't target civilians.
Aha! So it's not just the Ploxmire awards, it's a review of the specious "research" the whalers conduct in order to continue hunting whales.
You mean they're planning to cross-breed whales with cows so they can continue hunting them on land? Chasing them across the rolling hills of Kyushu in wheeled whaling boats, sending the Laser Tank Division to keep them from attacking Tokyo after they accidentally get irradiated, harpooning them on the slopes of Mount Fuji as they moo forlornly through their blowholes... this is the kind of research we need more of!
A group of activist hackers called The Institute for Applied Autonomy had some interesting experiences with their graffiti-writing robot: when they took it to public places and started spraying messages on the pavement, members of the public would be perfectly happy to join in and have the robot write their own messages. Of course if they'd tried the same thing without a robot, nobody would have joined in and they probably would have been arrested. Somehow the robot made their actions appear legitimate, and I think the same thing is happening with this guy. If he walked around spraying people he didn't like with a water pistol and telling them to move on, he'd probably be charged with harrassment and assault. He might even be considered mentally ill. But because he uses a robot it's a "patrol" - he seems to be on the side of authority because he's using technology more commonly used by the authorities.
Furthermore, at least google has its images of public space open for people to view at all times. If you wanted to look through a government owned public camera do you know where to go, who to ask?
That's a good point, maybe Google Earth is more egalitarian than government-controlled cameras, but I don't think it creates a completely level playing field. Governments can get sensitive sites removed from Google Earth, but can you get your house removed? Instead of making a false choice between government-controlled cameras and Google-controlled cameras, how about rejecting both?
You can find out more about the repeaters here. Problem is, mesh networking protocols aren't well suited for covering an entire city. The protocol used by OLPC is based on AODV, which uses flooding for route discovery. That's efficient if the mesh is sparse and there are a few long-lived connections, but inefficient if the mesh is dense or there's a lot of traffic. You could almost say that if mesh networking is a solution in search of a problem, the problem it's been searching for is two kids under a tree in the middle of the desert.;-)
Maybe you missed the story I liked were people were stabbed?
No, I didn't miss it, but I try to base my judgements on statistical likelihoods rather than spectacular one-offs. It's possible that someone will break into my house tonight and stab me to death for no reason, but it's extremely unlikely. I'm far more likely to die by misreading the instructions on a bottle of aspirin, so if I have to choose between buying a gun and buying a pair of reading glasses I'll choose the glasses.
I don't understand this thinking where it's ok for someone to violate your home, one place you should be able to be safe from others.
When did I say it was OK for someone to violate my home? Do you really think there are only two options: everyone gets robbed all the time, or everyone owns guns? Because I have to tell you there are other possibilities, and I live in one of them: a society with very little violent crime and very few guns.
There's no such "cycle" in the US.
The US has the highest per capita murder rate of any first world country. I can't prove that a vicious cycle of gun ownership by criminals and gun ownership by non-criminals contributes to that problem - all I can say is that the theory makes sense in my mind. As I said before, I'm not blaming non-criminals for owning guns, and I'm not claiming that simply banning guns will solve the problem, but I do think a problem exists. Don't you?
Its the areas that guns are outlawed where you have higher rates of violent crime.
Within the US that may be true - I haven't seen the statistics - but it's certainly not true when you compare the US to other first world countries.
Please tell me where you live that is free from crime and violence?
My country isn't free from crime and violence, but very little of that crime and violence involves guns. Example: a couple of years back my house was burgled during the night. The burglar didn't have a gun, probably because he didn't expect us to have one. We didn't have a gun because we didn't expect him to have one. We slept through the burglary, but the guy was caught trying to rob another house in our street the same night - the occupants woke up and wrestled him to the ground. Nobody was shot.
The US seems to be caught in a vicious circle where criminals have guns because they expect victims to have guns, and potential victims have guns because they expect criminals to have guns. I'm not trying to point the finger at anyone: in a situation like that it's rational to arm yourself. I'm just trying to point out that not all countries are caught in the same cycle.
Slashdot Mirror
Perhaps part of the reason is that last mile bandwidth is scarcer than backbone bandwidth, so an ISP doesn't save as much by encouraging its customers to share with each other (backbone bandwidth saved, last mile bandwidth remains the same) as it does by discouraging them from sharing at all (backbone and last mile bandwidth saved)?
And just like the holy grail, it doesn't exist. Ever heard the expression "garbage in, garbage out"? Now imagine how that applies to a huge database of information compiled from diverse sources (including unverified, anonymous tips), where nothing is ever thrown away, and where nobody's quite sure what they're looking for.
The human brain is amazingly good at finding patterns - so good that it often finds patterns that aren't really there. Even with years of experience, training, and peer review, professional scientists are pretty bad at handling problems like confirmation bias, post hoc reasoning and the file-drawer effect - how are law enforcement agents likely to fare, with no statistical training and no effective oversight?
The people constructing these databases are falling into the trap of believing that more data means better data. That's an understandable mistake for people who are usually "data-poor", such as archaeologists, historians and detectives. But anyone from the "data-rich" sciences will tell you that once you have the data you face a whole new set of problems, and I very much doubt that counterterrorism officials, working in conditions of secrecy and under pressure to justify their jobs, are going to handle those problems in a rigorous way.
Please note that I'm not trying to say "police officers are too stupid to understand statistics" - scientists make these mistakes all the time, but they operate in an atmosphere of relative transparency and competition, where it's usually in some other scientist's interest to bring errors to light. The same conditions don't apply to government officials.
What does this mean? It means that false positives will lead to innocent people being monitored, blacklisted and imprisoned without trial, while false negatives will mean genuine threats go undetected. We urgently need to make our governments understand that more data doesn't necessarily lead to better decisions.
"Unrelated?" Do you even understand why this is news? Politicians are at risk of intimidation, harrassment or even assassination because of the routine collection and sharing of personal information, and the lax controls over access to that information, that have become normal in our society. This could end up distorting the democratic process. That is the main story.
"Prosecute?" What an old-fashioned notion! US citizens can be declared enemy combatants and held without trial if they provide "material support" to an undefined enemy in an undeclared, neverending war (such as by linking to a jihadist website). Who needs prosecutions?
Without judicial oversight, how do you know which is actually being done?
Nice false dichotomy you've got there. Let's be honest, you have no more idea who's really being spied on than I do, and that's the whole point: if the spying were well-regulated and limited in scope, it could be overseen by trusted senior judges. There's already a secret court for that specific purpose. If the administration refuses to submit the spying to judicial oversight and is willing to take major political risks to keep the whole matter out of the courts, what does that imply?
The paper's online edition is here, but unfortunately the column only seems to be available in the print edition. A timeline of the rise of the Nazis in March 1933 is here.
In my opinion both were necessary. The Nazis expertly manipulated public opinion, so while it's true that many people supported them, I don't believe it follows that those people would have acted the same way if the Nazi party had not existed, and likewise the Nazis could not have acted the way they did without a certain amount of public support.
Good point, but I didn't claim a police state was necessary for genocide, just that it would make repression and genocide easier to carry out.
To answer your second point, I never claimed that a police state would lead to racism, oppression or genocide, only that it would make them easier to carry out.
Your third point is well taken, I should try to keep a cool head when talking about these things. :-) But I'm not entirely convinced that complaining about the construction of a police state plays into the hands of fascists. We shouldn't get hysterical, but we shouldn't be completely quiescent either.
Right, but it was the construction of a police state that made the racism, oppression and genocide possible. I don't believe the current UK or US governments plan to start imprisoning their opponents or murdering people en masse, but they're building infrastructure that will make that kind of thing a lot easier for future governments.
There's a column in the International Herald Tribune that reprints the news from 100, 75 and 50 years ago. Right now the 75-year section of the column is charting Hitler's methodical replacement of the German Republic with a Fascist state. It's a horribly fascinating process to watch, for two reasons: first, we know how it ends, and second, we can see many of the same moves being attempted today.
I can't believe this was modded insightful. Did you actually read the post you're responding to, or did the red mist come down after you saw the word "copyright"? The GP wasn't "bleating" or "insisting people still have to pay $15-$20", he/she was just pointing out that regardless of what the law should be, file sharing currently is illegal in most places, so Wired is correct to describe it as such.
Okay, "piracy" is a stupid term, but not as stupid as failing to recognise the difference between is and ought.
Kids these days...
3 million to 1 is a pretty impressive share ratio - most "pirates" upload roughly as much as they download, meaning they give away one copy in return for the copy they receive.
Your main point is correct - file sharing is illegal in many jurisdictions - but please don't try to dress it up with dramatic numbers.
You're not a real group until you get tax breaks.
I have no idea why you got modded down for that, it's a good point. The same applied to self-signed certificates if you save them the first time.
You're trying to blame the ACLU for the WMD fiasco? Seriously? That makes about as much sense as blaming lesbians for 9/11.
Verizon's up to P4P already but Comcast is still working on P0P - "peer, no peer".
Ping time is probably a good enough approximation of network distance, and it also helps you to avoid overloaded peers. If you use application-layer pings you don't even need raw sockets (unlike traceroute for example).
If only SSL had been designed to make man-in-the-middle attacks impossible. Oh wait, it was! Your browser contains root certificates that are used to verify that your bank's certificate hasn't been replaced or modified by an attacker. MITM attack against SSH? Maybe, if you don't check the key fingerprint (and I doubt anyone does). MITM attack against SSL? No chance, unless the server has a self-signed certificate, something no bank would consider.
It can only be a matter of time before Nintendo comes up with the same idea. Super Mario Whale!
They were seditionists but I wouldn't call them terrorists - they started a militia war against an occupying army but AFAIK they didn't target civilians.
You mean they're planning to cross-breed whales with cows so they can continue hunting them on land? Chasing them across the rolling hills of Kyushu in wheeled whaling boats, sending the Laser Tank Division to keep them from attacking Tokyo after they accidentally get irradiated, harpooning them on the slopes of Mount Fuji as they moo forlornly through their blowholes... this is the kind of research we need more of!
A group of activist hackers called The Institute for Applied Autonomy had some interesting experiences with their graffiti-writing robot: when they took it to public places and started spraying messages on the pavement, members of the public would be perfectly happy to join in and have the robot write their own messages. Of course if they'd tried the same thing without a robot, nobody would have joined in and they probably would have been arrested. Somehow the robot made their actions appear legitimate, and I think the same thing is happening with this guy. If he walked around spraying people he didn't like with a water pistol and telling them to move on, he'd probably be charged with harrassment and assault. He might even be considered mentally ill. But because he uses a robot it's a "patrol" - he seems to be on the side of authority because he's using technology more commonly used by the authorities.
That's a good point, maybe Google Earth is more egalitarian than government-controlled cameras, but I don't think it creates a completely level playing field. Governments can get sensitive sites removed from Google Earth, but can you get your house removed? Instead of making a false choice between government-controlled cameras and Google-controlled cameras, how about rejecting both?
You can find out more about the repeaters here. Problem is, mesh networking protocols aren't well suited for covering an entire city. The protocol used by OLPC is based on AODV, which uses flooding for route discovery. That's efficient if the mesh is sparse and there are a few long-lived connections, but inefficient if the mesh is dense or there's a lot of traffic. You could almost say that if mesh networking is a solution in search of a problem, the problem it's been searching for is two kids under a tree in the middle of the desert. ;-)
No, I didn't miss it, but I try to base my judgements on statistical likelihoods rather than spectacular one-offs. It's possible that someone will break into my house tonight and stab me to death for no reason, but it's extremely unlikely. I'm far more likely to die by misreading the instructions on a bottle of aspirin, so if I have to choose between buying a gun and buying a pair of reading glasses I'll choose the glasses.
When did I say it was OK for someone to violate my home? Do you really think there are only two options: everyone gets robbed all the time, or everyone owns guns? Because I have to tell you there are other possibilities, and I live in one of them: a society with very little violent crime and very few guns.
The US has the highest per capita murder rate of any first world country. I can't prove that a vicious cycle of gun ownership by criminals and gun ownership by non-criminals contributes to that problem - all I can say is that the theory makes sense in my mind. As I said before, I'm not blaming non-criminals for owning guns, and I'm not claiming that simply banning guns will solve the problem, but I do think a problem exists. Don't you?
Within the US that may be true - I haven't seen the statistics - but it's certainly not true when you compare the US to other first world countries.
My country isn't free from crime and violence, but very little of that crime and violence involves guns. Example: a couple of years back my house was burgled during the night. The burglar didn't have a gun, probably because he didn't expect us to have one. We didn't have a gun because we didn't expect him to have one. We slept through the burglary, but the guy was caught trying to rob another house in our street the same night - the occupants woke up and wrestled him to the ground. Nobody was shot.
The US seems to be caught in a vicious circle where criminals have guns because they expect victims to have guns, and potential victims have guns because they expect criminals to have guns. I'm not trying to point the finger at anyone: in a situation like that it's rational to arm yourself. I'm just trying to point out that not all countries are caught in the same cycle.