It's not the first time something is hacked before it's even released, but it's always funny.
What really makes this one a good example is that for once this lock used some kind of real crypto (RSA), not some security-through-obscurity stuff. And yet, of course, that defeated, by not even letting the check occured.
Because crypto scenario were Bob tries to hide something to bob, after giving Bob the key are just a bit to stupid to work.
The effects of controling the medias are subtle and effective, and every succeful control will also convince the population it is itself necessary.
Take the example of Russia: the elections were cheated (some small towns were 105% pro government...), but even perfectly fair elections would probably show that a majority (like 55%) think Putin was a good leader. But thinking that 95%, of the country agrees with the government will make you more prone to agree yourself, whereas at 55% you'll start beleiving that alternatives exist.
I could also speak about Fox in the US, and the necessity for antiterrorist laws.
They also apparently have an army of unlicensed private investigators.
It seems that their tactic was:
1)"illegally enter the hard drives of tens of thousands of private American citizens to look for music recordings stored there". That was MediaSentry's job.
2) Fill "thousands" of anonymous lawsuits, only to subpoena the ISP, and then "discover" the IPs that they already illegaly found. The lawsuit is then discarded, having served it's purpose.
3) Profit, by settling out of court, harrassing and such.
I thought I was pretty well informed on those things, and yet it's the first time I hear about that. It sheds a very new light on the fact that they often couldn't give the proofs. (What I still don't get though, is how they ended suing guys without computers.)
Well, for starters, the judge seems to be well-informed of the case. That should really help.
Ms Andersen is fortunate that Judge Anna Brown is very familiar with the history of this and other RIAA cases and will not stand for the RIAA's usual tactics of delay. Sorry, the question was addressed to a lawyer, and IANAL.
You gotta love the style the linked pdf is worderd:
1) "Killing "Dolphins" by Direct Threat and Intimidation" The funniest thing: that's actually the RIAA's own word:
As a senior RIAA spokeswoman explained: "When you fish with a net, you are going to catch a few dolphins". This will make for interesting reading, and has a good potential for quotations; I think we can hope some real prime-time awareness this time.
It's no more circular than saying that a mathematical theorem is considered true if (and as long as) it can hold against all atempts at proving it wrong.
Yes, it doesn't give you any insight on how to write correct theorem.
But it doesn't mean "testers" will necessarly disprove most valid theorems.
(I am voluntarly ignoring your circular point about your own intelligence, so as to make for an interesting reading to a genuinly interested reader)
But the point is that if they were subjected to the Turing Test, many would fail, even though they are "intelligent" in the way that the Turing Test supposedly tests for. If the testers are good, they won't label humans as robots. If some robots are in effect to intelligent to detect, the testers will label them humans, too.
If your idea of the TT labels "many" humans as robots, you would simply make a bad tester:) .
If you somehow spotted three recursion in my post, then that would certainly be pretty ironic ^^.
Then again, I'm a mathematician, so I wouldn't fear any level of recursion!
Seriously though, I think that if you started calculating the probability of error in real-life sentences, you'd find that correct sentences with three level of recursion are pretty rare. Except in maths, of course.
The prize you are referring to is just an implementation of the Turing test (probably the only formal one).
And I find very fishy that it relies on the good will of the judges (who must "respond naturally" and use "no trickery"). If it's a test, the test should be able to ask any questions, not just those he "thinks" are easy.
As for the article, I think they reffered to "people passing by the [AI operated] avatar, and maybe not noticing the difference". Which is VERY different from the Turing test, or even a simplified version of it. Heck, it's not even a test.
That's hogwash. Any number of real people I talk to could easily be simulated by some non-intelligent machine. Yes, but you're not asked to test wether they are humans or not. If that was the case, you would start to ask tricky questions, etc.
But I agree that the Turing test is a bit empty in that it gives to the examiner the responsability to define a "human". I think the best way to see it is to say: "the more you can ask from an AI is to be indistinctable from a human from the outside".
Which is already saying quite a lot (although it's philosophy, not science). Mainly, it rules out people conplaining that the AI has no soul or whatever.
Somewhere around five years of age, however, children begin to have second-order beliefs--that is, beliefs about the beliefs of others, enabling them to understand that other people can have beliefs different from their own. Now, Bringsjord's research group claims to have achieved second- and third-order beliefs in their synthetic characters. Funny how recursion is always a key for "real" abstract thoughts. You could think that adding them to the langage of the AI will bring all the problems it does in logic, but then you realize that real humans always doubt sentences with three levels of recursions (or above), and try to avoid them.
That makes this approach all the more interesting.
What sort of oversight could be put in place What sort? What about meta-moderation, like here on slashdot?
Currently, Wikipedia lacks a proper meta-moderation for things like deleting articles, banning account and everything that touches accounts with special privileges. The worst thing being that (on the english wiki), some get privileges for undetermined time, without the possibility of an "impeachment".
But it wouldn't be hard to make it right: simply add "jury duty" like meta moderation, and demote moderators when they have a low score. Important votes (like banning account or electing moderators) could stay active for a week or more before closing.
On a side note, meta-moderation wouldn't be entirely enough; you would also need a way to prove to anyone that the website respected the procedure (with hashs, and only pseudo randomness so that it can be checked); for example, on slashdot, you can't be sure that the owners of the website don't cheat with the way karma or modpoints are given.
It's strange to see a post starting with concerns of efficiency and ending by saying an oppressive regime shouldn't have a say in ICANN.
I agree that one can not enjoy the benefits of the US influence on the world without recognising them (as opposed to China's influence; I'm NOT saying there are only benefits to US actions).
BUT, checks and balances between countries could probably lead to a more neutral governance; especially, no single country would have the power to shut off an entire countries worth of domain names (like all *.fr, or all *.cx ^^). Something like a veto system sounds pretty good (the UN was TOO neutral!)
And the loss of efficiency? That's what a controlled decision system costs. Sure, a dictatorship will always be more "efficient" than a democracy, but that's not really a good point, is it?
I don't know if it was the case in the US, but in France, when cell phones became popular, they were mostly used by teenagers. And there was a lot of FUD, about how that would make them antisocial, or stupid, or unaware of the outside world, etc.
Mostly it came from people (and journalists) ranting because they found teens phoning on the street obnoxious.
Then all of a sudden, it stopped. Why? because then the amount of 60+ old people owning a cell phone had risen to 70% !
It's the same thing with video games; now they're becoming a common thing among adults, ie those who vote, buy newspaper and work as journalists. That's all.
First, we would need legal recognitions for some key tech related concepts:
a) Open Source. Open source (and CC) licences sould not have to rely on copyright to protect them; the fact that someone donates it's work to the community should be acknowledged, and given a special status (so that you can give it *more* protection than public domain). (This does realte to a "bill of right" because free speech issues).
b) Personnal informations. There needs to be a list of datas known as "personnal informations". Then you can enforce special laws about them. For example, that no organization except the government can force you to give them if they can provide the service without it (no "required field" with your name and address on a forum, for example). Maybe instaure a national service whitch lets you generate identifying numbers that the government can read (for tracability of payments, etc), but the company (the online store) has no right to read.
c) Cryptography. Crypto changes everything on the internet; the very nature of crypted or unencrypted data is different, and this must be acknowledged. For example, it must be stated somewhere that making a crypted file public does not mean you are making the file itself public. The definition of encrypted content should include a reference to the "current knowledge of the scientific community". A good idea could also be to give everyone a public key on some government hosted server.
All of these concept can be used with the law as it is, but they are left to the appreciation of judges, who do not always know the subject.
about #4 : You can't hope to make a law that defines "human readable" as opposed to legalese; but I agree stronger legislation on what is allowed or not to put in a contract should exist. Moreover, putting abusive clause (which is the case on alsmost all EULAs) should be motive to sue the company. Anyone should be able to do it (ie, competitors, not just users), justa as false advertising (maybe it's already the case?).
about #5 : IANAL, but I think this is already illegal (I received an email when ebay changed some of their EULA; then again, I'm in europe.)
about #6 : Again to vague to transcribe in legalese. Adding specific customer protection laws should do the trick, that way you can explicit those "assumptions"
about #10 : To much of a hassle to everyone to be enforced. When I'm quoting a slashdot post, or showing a picture in a video mashup, I don't have the time to contact or acknowledge every owner. That's actually the current state of IP, and it means everybody is breaching the law a little (modulo fair use, but it's pretty hard to prove it without paying lawers), so everyone is open to takedown letters.
I do agree with the idea; but laws are tricky things, and you have to make them legalese-compatible
Exaclty. People can do this in good faith, because the service could exist for free.
It's the same with the video streaming of tv shows; some channels (nbc?) make their show available for free, with ads (like, you know, on TV).
It's different of course if you access a warez website, or some other thing which is obviously illegal; but in this case, it really seems stupid to try to sue; they should just add a n intermediate page stating that the video is pay only (or better, use some kind of crypto!)
Well, the fact is, "experts" get some kind of reward for posting there (I don't think they get money, but it happens on some other sites).
I would not go as far as saying that their buisness model is flawed; rather, I'm saying that you can in good faith come across that site without paying (as I did the first time). Some sites serve obviously illegal content, other offer something which is to good to be legal (full recent games download, etc), but when it's just a video strem of a TV show, or an answer about some bash command question, you can't just blame the user saying he should have "guessed" it was illegal.
And the same applies to the "dumb" bots of google.
As always, that kind of position is missing the fact that google is technically doing the same thing.
It's not that far fetch: imagine you are googling for your favorite show, and find some url with a video stream; and it's form a respectable "nbc.com" or the like website. How do you guess it's supposed to be a paying service? Want a real life example? The other day I was looking for some bash command help, and the third google result was from http://www.experts-exchange.com./ If you access it directly, it hides the answers and asks you to pay. But from google, you get to the answers directly because of some glitch.
What I'm saying is you can't blame the user (or here, the website) if they never went through a dsiclaimer page that made them realise: "well, if I click this link, I will have done something illegal". Free equivalent services exist.
It's an important example; and in fact intelligent design would never be called the accepted theory even with a vote system.
Why? Because on Wikipedia you have to write verifiable facts; and when intelligent design claims that "there is a significant part of the scientific community which disagree with evolution", they simply don't have the proofs to support it (all articles they could link are from the same few intelligent design advocates who have a phd).
That kind of "guidelines" on wikipedia effectively serve as a constitution; it's pretty efficient, but not as much as full blown moderation and meta-moderation as we have here, on Slashdot. Personaly, I think it's a shame, and I would love to see more "jury duty" like moderation on wikipedia, or some fork of it!
Take Slashdot: it's far more democratic; even though the founders still have special powers, it's hard to imagine the oligarchy problem that resulted in the abusive ban of an IP range on Wikipedia: http://yro.slashdot.org/yro/07/12/07/1434221.shtml
I think Wikipedia CAN and SHOULD be a democracy experiment.
Slashdot Mirror
It's not the first time something is hacked before it's even released, but it's always funny.
What really makes this one a good example is that for once this lock used some kind of real crypto (RSA), not some security-through-obscurity stuff. And yet, of course, that defeated, by not even letting the check occured.
Because crypto scenario were Bob tries to hide something to bob, after giving Bob the key are just a bit to stupid to work.
The effects of controling the medias are subtle and effective, and every succeful control will also convince the population it is itself necessary.
Take the example of Russia: the elections were cheated (some small towns were 105% pro government...), but even perfectly fair elections would probably show that a majority (like 55%) think Putin was a good leader. But thinking that 95%, of the country agrees with the government will make you more prone to agree yourself, whereas at 55% you'll start beleiving that alternatives exist.
I could also speak about Fox in the US, and the necessity for antiterrorist laws.
They also apparently have an army of unlicensed private investigators.
It seems that their tactic was:
1)"illegally enter the hard drives of tens of thousands of private American citizens to look for music recordings stored there". That was MediaSentry's job.
2) Fill "thousands" of anonymous lawsuits, only to subpoena the ISP, and then "discover" the IPs that they already illegaly found. The lawsuit is then discarded, having served it's purpose.
3) Profit, by settling out of court, harrassing and such.
I thought I was pretty well informed on those things, and yet it's the first time I hear about that. It sheds a very new light on the fact that they often couldn't give the proofs. (What I still don't get though, is how they ended suing guys without computers.)
Unless the labels stop funding them, which they already considered doing since the RIAA is starting to lose money rather than earning any.
It's no more circular than saying that a mathematical theorem is considered true if (and as long as) it can hold against all atempts at proving it wrong.
Yes, it doesn't give you any insight on how to write correct theorem.
But it doesn't mean "testers" will necessarly disprove most valid theorems.
(I am voluntarly ignoring your circular point about your own intelligence, so as to make for an interesting reading to a genuinly interested reader)
Yes, but then it would not be enough to pretend that your car is street-legal, which is what they do since they use the word "Turing test".
:)
(And don't get me started, because this is my OTHER car analogy!
If your idea of the TT labels "many" humans as robots, you would simply make a bad tester
If you somehow spotted three recursion in my post, then that would certainly be pretty ironic ^^.
Then again, I'm a mathematician, so I wouldn't fear any level of recursion!
Seriously though, I think that if you started calculating the probability of error in real-life sentences, you'd find that correct sentences with three level of recursion are pretty rare. Except in maths, of course.
The prize you are referring to is just an implementation of the Turing test (probably the only formal one).
And I find very fishy that it relies on the good will of the judges (who must "respond naturally" and use "no trickery"). If it's a test, the test should be able to ask any questions, not just those he "thinks" are easy.
As for the article, I think they reffered to "people passing by the [AI operated] avatar, and maybe not noticing the difference". Which is VERY different from the Turing test, or even a simplified version of it.
Heck, it's not even a test.
But can it do THAT:
http://xkcd.com/329/ ?
But I agree that the Turing test is a bit empty in that it gives to the examiner the responsability to define a "human". I think the best way to see it is to say: "the more you can ask from an AI is to be indistinctable from a human from the outside".
Which is already saying quite a lot (although it's philosophy, not science). Mainly, it rules out people conplaining that the AI has no soul or whatever.
That makes this approach all the more interesting.
Currently, Wikipedia lacks a proper meta-moderation for things like deleting articles, banning account and everything that touches accounts with special privileges. The worst thing being that (on the english wiki), some get privileges for undetermined time, without the possibility of an "impeachment".
But it wouldn't be hard to make it right: simply add "jury duty" like meta moderation, and demote moderators when they have a low score. Important votes (like banning account or electing moderators) could stay active for a week or more before closing.
Wikipedia itself claims that it does not want to be a democracy: http://en.wikipedia.org/wiki/Wikipedia:What_Wikipedia_is_not#Wikipedia_is_not_a_democracy ,
and that it uses "consensus" to determine what action to take. When I first saw that, I thought "well, the all thing isn't very transparent, but it still works okay". But the fact is, simply being suspected of scandals like this one, or the more serious one a while ago ( http://yro.slashdot.org/yro/07/12/07/1434221.shtml ) is reason enough to become more transparent.
Wikipedia CAN and SHOULD be a democracy.
On a side note, meta-moderation wouldn't be entirely enough; you would also need a way to prove to anyone that the website respected the procedure (with hashs, and only pseudo randomness so that it can be checked); for example, on slashdot, you can't be sure that the owners of the website don't cheat with the way karma or modpoints are given.
At about 10,000$ of damages per song, 32MB doesn't seems that small!
;)
In fact, it should be "engough for everybody"
It's strange to see a post starting with concerns of efficiency and ending by saying an oppressive regime shouldn't have a say in ICANN.
I agree that one can not enjoy the benefits of the US influence on the world without recognising them (as opposed to China's influence; I'm NOT saying there are only benefits to US actions).
BUT, checks and balances between countries could probably lead to a more neutral governance; especially, no single country would have the power to shut off an entire countries worth of domain names (like all *.fr, or all *.cx ^^). Something like a veto system sounds pretty good (the UN was TOO neutral!)
And the loss of efficiency? That's what a controlled decision system costs. Sure, a dictatorship will always be more "efficient" than a democracy, but that's not really a good point, is it?
I don't know if it was the case in the US, but in France, when cell phones became popular, they were mostly used by teenagers. And there was a lot of FUD, about how that would make them antisocial, or stupid, or unaware of the outside world, etc.
Mostly it came from people (and journalists) ranting because they found teens phoning on the street obnoxious.
Then all of a sudden, it stopped. Why? because then the amount of 60+ old people owning a cell phone had risen to 70% !
It's the same thing with video games; now they're becoming a common thing among adults, ie those who vote, buy newspaper and work as journalists.
That's all.
First, we would need legal recognitions for some key tech related concepts:
a) Open Source.
Open source (and CC) licences sould not have to rely on copyright to protect them; the fact that someone donates it's work to the community should be acknowledged, and given a special status (so that you can give it *more* protection than public domain). (This does realte to a "bill of right" because free speech issues).
b) Personnal informations.
There needs to be a list of datas known as "personnal informations". Then you can enforce special laws about them. For example, that no organization except the government can force you to give them if they can provide the service without it (no "required field" with your name and address on a forum, for example). Maybe instaure a national service whitch lets you generate identifying numbers that the government can read (for tracability of payments, etc), but the company (the online store) has no right to read.
c) Cryptography.
Crypto changes everything on the internet; the very nature of crypted or unencrypted data is different, and this must be acknowledged. For example, it must be stated somewhere that making a crypted file public does not mean you are making the file itself public. The definition of encrypted content should include a reference to the "current knowledge of the scientific community".
A good idea could also be to give everyone a public key on some government hosted server.
All of these concept can be used with the law as it is, but they are left to the appreciation of judges, who do not always know the subject.
about #4 : You can't hope to make a law that defines "human readable" as opposed to legalese; but I agree stronger legislation on what is allowed or not to put in a contract should exist. Moreover, putting abusive clause (which is the case on alsmost all EULAs) should be motive to sue the company. Anyone should be able to do it (ie, competitors, not just users), justa as false advertising (maybe it's already the case?).
about #5 : IANAL, but I think this is already illegal (I received an email when ebay changed some of their EULA; then again, I'm in europe.)
about #6 : Again to vague to transcribe in legalese. Adding specific customer protection laws should do the trick, that way you can explicit those "assumptions"
about #10 : To much of a hassle to everyone to be enforced. When I'm quoting a slashdot post, or showing a picture in a video mashup, I don't have the time to contact or acknowledge every owner. That's actually the current state of IP, and it means everybody is breaching the law a little (modulo fair use, but it's pretty hard to prove it without paying lawers), so everyone is open to takedown letters.
I do agree with the idea; but laws are tricky things, and you have to make them legalese-compatible
Exaclty. People can do this in good faith, because the service could exist for free.
It's the same with the video streaming of tv shows; some channels (nbc?) make their show available for free, with ads (like, you know, on TV).
It's different of course if you access a warez website, or some other thing which is obviously illegal; but in this case, it really seems stupid to try to sue; they should just add a n intermediate page stating that the video is pay only (or better, use some kind of crypto!)
Well, the fact is, "experts" get some kind of reward for posting there (I don't think they get money, but it happens on some other sites).
I would not go as far as saying that their buisness model is flawed; rather, I'm saying that you can in good faith come across that site without paying (as I did the first time).
Some sites serve obviously illegal content, other offer something which is to good to be legal (full recent games download, etc), but when it's just a video strem of a TV show, or an answer about some bash command question, you can't just blame the user saying he should have "guessed" it was illegal.
And the same applies to the "dumb" bots of google.
As always, that kind of position is missing the fact that google is technically doing the same thing.
It's not that far fetch: imagine you are googling for your favorite show, and find some url with a video stream; and it's form a respectable "nbc.com" or the like website. How do you guess it's supposed to be a paying service?
Want a real life example? The other day I was looking for some bash command help, and the third google result was from http://www.experts-exchange.com./ If you access it directly, it hides the answers and asks you to pay. But from google, you get to the answers directly because of some glitch.
What I'm saying is you can't blame the user (or here, the website) if they never went through a dsiclaimer page that made them realise: "well, if I click this link, I will have done something illegal". Free equivalent services exist.
It's an important example; and in fact intelligent design would never be called the accepted theory even with a vote system.
Why? Because on Wikipedia you have to write verifiable facts; and when intelligent design claims that "there is a significant part of the scientific community which disagree with evolution", they simply don't have the proofs to support it (all articles they could link are from the same few intelligent design advocates who have a phd).
(it doesn't mean there is no article on inteligent design: http://en.wikipedia.org/wiki/Intelligent_design; just that there is no mention of it on the "evolution" article)
That kind of "guidelines" on wikipedia effectively serve as a constitution; it's pretty efficient, but not as much as full blown moderation and meta-moderation as we have here, on Slashdot. Personaly, I think it's a shame, and I would love to see more "jury duty" like moderation on wikipedia, or some fork of it!
Yes, and this is the problem.
Take Slashdot: it's far more democratic; even though the founders still have special powers, it's hard to imagine the oligarchy problem that resulted in the abusive ban of an IP range on Wikipedia: http://yro.slashdot.org/yro/07/12/07/1434221.shtml
I think Wikipedia CAN and SHOULD be a democracy experiment.