Every deconnection case will have to be decided by a judge, according to the law. This was a last-minute arrangement, brought upon because the French equivalent to the supreme court said any other way would be tantamount to (1) guilty-until-proven-innocent and (2) the executive power taking on judicial powers too.
Maybe but the gov. cannot control all the ISPs ; note that Orange already has the highest fees, so they might be onto a losing proposition if they are the only one who play nice with Hadopi.
In this case there would be no jury, just a judge, because this is not a criminal offence, only a misdemeanor. However, defendants can appeal any decision, and if enough people do this, soon the judicial system will become completely clogged by this issue, in effect reaching the same effect.
In general I don't know if jury nullification exists in France. Popular juries are only called in for major crimes like murders, so there is little chance of nullification in this case. Civil law matters are dealt with through a judge or a professional jury.
I heard Claude Cohen-Tanoudji (one of the recipients of the 1997 Nobel) describe their method in 1992 at MIT. It was one of the coolest lecture ever, no pun intended.
I don't think you should judge what people really think of white collar criminals from some random Slashdotter's opinion. However, while there might not be admiration for such criminal deeds, there is at least envy. Also perhaps many associate such sociopathic behavior with others such as found in some well publicized people in power position (CEOs, etc), who sometime use barely legal means to get their way, and certainly don't worry too much about the welfare of their employees.
Actually, there was quite a fight on where to put it, the final contender were Japan and France. I'm not sure why France won, perhaps fewer earthquake (a significant factor over several decades).
It could make sense because the method of the proof in the limited case is in fact very interesting (the infinite descent). Yours is not a sufficient argument.
First I'm amazed that someone took on Autodesk over this. Second, this is "natural" given the overall current court perspective that basically says that EULAs are enforceable. This would be more or less OK if you could return the software if you didn't agree on the EULA. I don't know if this is the case with Autocad.
In this case, unfortunately Autocad has relatively few competitors (try CATIA), costs an arm and a leg, and have been around forever. There is not going to be a FOSS equivalent for a while yet.
Known defects I don't know, but unknown defects, for sure. Look at how many error-correcting printings the 3rd edition of Stroustrup's C++ book has gone through.
...and still has a paltry 512 mb of memory, which, when they eventually get around to implementing multitasking, means that what you're actually going to get is something on the order of windows 3.1 multitasking with a few services, not actual task switching, etc
When Windows 3.1 came out PCs had in general ONE mb of RAM or less, and a few hundred MB or HD. I wrote my 300-page dissertation in 1993 on a NeXT cube, a massive workstation that cost more than most new cars (close to US 100k$ then) and that had a whopping 32 MB of RAM plus 8 MB of video RAM and 600MB of HD and that was a monster back then. When Windows 95 came out PCs had *8* MB or less and about 1GB of HD space, I remember it well because I tried to make my NeXT programs run on the one I had then and they were swapping like hell.
You got yourself confused over three orders of magnitude here. The iPad has as much RAM and as much HD space as a well-loaded win2k-era PC. So you can have real multitasking on the iPad, there is no problem there. In fact the iPad runs the Darwin kernel perfectly OK, and is multitasking just fine. The whole GNU development toolkit works fine on it.
Things are only easy in retrospect. There is plenty of evidence, both theoretical and practical, that say that controlled nuclear fusion is feasible. We have actually already achieved it by many means. Yet current methods do not scale well, and it will not be useful or practical for the next 20-40 years, despite concerted international effort on the matter.
In the case of cloning, the obstacles were technical. In theory we have known for a long time what was required to achieve it. It turns out that the technical obstacles were less than expected, and that we could achieve it by letting Nature do *most* of the work.
In the AI case, we do not have theoretical firm ground that say that it can work, other than "we can always simulate nature, and this will require a computer at least this fast and this big".
I can't give you a definition of sentience because the term is more philosophical than scientific, however there exists a somewhat agreed-upon definition of "Strong AI" that is at least mildly useful to computer scientists for describing something that human-built computers in particular and machines in general are not capable of yet. You can look it up on wikipedia or other if you want. The definition is mildly confusing and unclear but does exist, and is at least partly testable using for instance the (controversial) Turing test.
Now all normal humans are capable of "Strong AI" because this is precisely what the definition means (except not artificial).
Now I cannot prove that there does not exist a computer or a machine that has exhibited or exhibits now "Strong AI", however even the most optimistic of futurists seem to say that we are about 20 years away from getting them on the strength of simple back-of-the-enveloppe calculation and basic assumption that current hardware improvement trend will continue. Note: they have been saying that for a long time.
If thought depend on quantum processes that cannot be well approximated classically (which is possible), duplicating them might prove difficult. At present we just don't know.
Assuming continuing exponential growth in the computer industry with a 18 month doubling constant (a.k.a Moore's law), log(10^4)/log(2) = 13.3 doubling, i.e 13.3*1.5 = 19.5 years.
Ok, interesting reply, but the parent is still mostly right. You are adding that there are cognitive issues associated with the color red.
This is correct. As we are born, we are all color-blind. Vision is learned, not simply acquired, so someone physiologically incapable of perceiving the stated visible range using the correct set of cone cells in your eyes will not associate it with the color red. Most people can and do, hence the common name for the perception.
The red afterimage you mention is definitely "false red". It is due to the fact that your green receptor saturate after a while, and when you turn them to a white background they respond less than the one that surround them, hence the perception of a very pale red, as the white background is in effect less green-saturated.
The red you perceive in your dream is remembered, not the result of a stimulus.
I'm not aware of the synesthesia of hearing a sound while seeing a color, I do know the converse exists: some people associate some notes with a given color, which is a relatively rare but not ultra-rare talent. This and the other example don't contradict the fact that the perception of "red" is learned.
The Parent tells you acupuncture is used at major hospitals around the US. This is not BS, this is because it does work. There are plenty of control trials with acupuncture. The main use is pain suppression, but this not the only one. There exists a whole journal devoted to the scientific approach to acupuncture. Here is one of their published study:
http://aim.bmj.com/content/27/4/174.abstract
It's good to be a skeptic, it is even better to research a bit.
One we have discovered what intelligence really is, for sure it will be somewhat simple by some standards. At the moment we do not even have a working definition of what it is really, so I'm not so convinced as you are, and I doubt we will have made that much more progress by 2030. My reasoning is this: even riding some kind of exponential, locally progress rate is still linear. Fundamentally we have made some but not sufficient progress in the last 30 years in computer science in general. Over the next 20 years I expect more of the same. We need more biology insight and that is going to require a truly massive effort. We do not even have yet the required probing instruments (e.g. sufficiently precise functional MRI machines for instance)
Mostly I think that in general the human population, myself included, is an incredibly stupid, short sighted, nasty, egotistic species with occasional streaks of artistic, political or scientific brilliance. Hopefully we can do better than simulate that.
Sometimes a poor artist is really not a very good one, but sometimes artists are incredibly ahead of their time and not understood, although that may have been more true around the time of Vincent van Gogh than now.
At any rate exceptional artists are by definition very rare exceptions.
Slashdot Mirror
Every deconnection case will have to be decided by a judge, according to the law. This was a last-minute arrangement, brought upon because the French equivalent to the supreme court said any other way would be tantamount to (1) guilty-until-proven-innocent and (2) the executive power taking on judicial powers too.
No they wont. They can only request ISP numbers of people who connect to torrents feed of French media companies contents.
The net effect of this is rather that no one will soon download anything made by French artists. They are shooting themselves in the foot big time.
Maybe but the gov. cannot control all the ISPs ; note that Orange already has the highest fees, so they might be onto a losing proposition if they are the only one who play nice with Hadopi.
In this case there would be no jury, just a judge, because this is not a criminal offence, only a misdemeanor. However, defendants can appeal any decision, and if enough people do this, soon the judicial system will become completely clogged by this issue, in effect reaching the same effect.
In general I don't know if jury nullification exists in France. Popular juries are only called in for major crimes like murders, so there is little chance of nullification in this case. Civil law matters are dealt with through a judge or a professional jury.
I heard Claude Cohen-Tanoudji (one of the recipients of the 1997 Nobel) describe their method in 1992 at MIT. It was one of the coolest lecture ever, no pun intended.
I don't think you should judge what people really think of white collar criminals from some random Slashdotter's opinion. However, while there might not be admiration for such criminal deeds, there is at least envy. Also perhaps many associate such sociopathic behavior with others such as found in some well publicized people in power position (CEOs, etc), who sometime use barely legal means to get their way, and certainly don't worry too much about the welfare of their employees.
Look up Solar Tres.
They say you can donate to the project on their web page. Somehow this does not inspire much confidence.
Actually, there was quite a fight on where to put it, the final contender were Japan and France. I'm not sure why France won, perhaps fewer earthquake (a significant factor over several decades).
Please mod parent up, close to only informed post on this thread.
It could make sense because the method of the proof in the limited case is in fact very interesting (the infinite descent). Yours is not a sufficient argument.
First I'm amazed that someone took on Autodesk over this. Second, this is "natural" given the overall current court perspective that basically says that EULAs are enforceable. This would be more or less OK if you could return the software if you didn't agree on the EULA. I don't know if this is the case with Autocad.
In this case, unfortunately Autocad has relatively few competitors (try CATIA), costs an arm and a leg, and have been around forever. There is not going to be a FOSS equivalent for a while yet.
Known defects I don't know, but unknown defects, for sure. Look at how many error-correcting printings the 3rd edition of Stroustrup's C++ book has gone through.
Hello,
that was a good read:
When Windows 3.1 came out PCs had in general ONE mb of RAM or less, and a few hundred MB or HD. I wrote my 300-page dissertation in 1993 on a NeXT cube, a massive workstation that cost more than most new cars (close to US 100k$ then) and that had a whopping 32 MB of RAM plus 8 MB of video RAM and 600MB of HD and that was a monster back then. When Windows 95 came out PCs had *8* MB or less and about 1GB of HD space, I remember it well because I tried to make my NeXT programs run on the one I had then and they were swapping like hell.
You got yourself confused over three orders of magnitude here. The iPad has as much RAM and as much HD space as a well-loaded win2k-era PC. So you can have real multitasking on the iPad, there is no problem there. In fact the iPad runs the Darwin kernel perfectly OK, and is multitasking just fine. The whole GNU development toolkit works fine on it.
It was to be announced at the Party Congress on Monday. As you know, the Premier loves surprises.
Things are only easy in retrospect. There is plenty of evidence, both theoretical and practical, that say that controlled nuclear fusion is feasible. We have actually already achieved it by many means. Yet current methods do not scale well, and it will not be useful or practical for the next 20-40 years, despite concerted international effort on the matter.
In the case of cloning, the obstacles were technical. In theory we have known for a long time what was required to achieve it. It turns out that the technical obstacles were less than expected, and that we could achieve it by letting Nature do *most* of the work.
In the AI case, we do not have theoretical firm ground that say that it can work, other than "we can always simulate nature, and this will require a computer at least this fast and this big".
If you are extra-careful, sometime you get 0.1 as the answer, which probably means we are about to destroy ourselves as the abomination that we are.
I can't give you a definition of sentience because the term is more philosophical than scientific, however there exists a somewhat agreed-upon definition of "Strong AI" that is at least mildly useful to computer scientists for describing something that human-built computers in particular and machines in general are not capable of yet. You can look it up on wikipedia or other if you want. The definition is mildly confusing and unclear but does exist, and is at least partly testable using for instance the (controversial) Turing test.
Now all normal humans are capable of "Strong AI" because this is precisely what the definition means (except not artificial).
Now I cannot prove that there does not exist a computer or a machine that has exhibited or exhibits now "Strong AI", however even the most optimistic of futurists seem to say that we are about 20 years away from getting them on the strength of simple back-of-the-enveloppe calculation and basic assumption that current hardware improvement trend will continue. Note: they have been saying that for a long time.
If thought depend on quantum processes that cannot be well approximated classically (which is possible), duplicating them might prove difficult. At present we just don't know.
Just to check the parent's math
Assuming continuing exponential growth in the computer industry with a 18 month doubling constant (a.k.a Moore's law), log(10^4)/log(2) = 13.3 doubling, i.e 13.3*1.5 = 19.5 years.
Ok, interesting reply, but the parent is still mostly right. You are adding that there are cognitive issues associated with the color red.
This is correct. As we are born, we are all color-blind. Vision is learned, not simply acquired, so someone physiologically incapable of perceiving the stated visible range using the correct set of cone cells in your eyes will not associate it with the color red. Most people can and do, hence the common name for the perception.
The red afterimage you mention is definitely "false red". It is due to the fact that your green receptor saturate after a while, and when you turn them to a white background they respond less than the one that surround them, hence the perception of a very pale red, as the white background is in effect less green-saturated.
The red you perceive in your dream is remembered, not the result of a stimulus.
I'm not aware of the synesthesia of hearing a sound while seeing a color, I do know the converse exists: some people associate some notes with a given color, which is a relatively rare but not ultra-rare talent. This and the other example don't contradict the fact that the perception of "red" is learned.
Hello,
The Parent tells you acupuncture is used at major hospitals around the US. This is not BS, this is because it does work. There are plenty of control trials with acupuncture. The main use is pain suppression, but this not the only one. There exists a whole journal devoted to the scientific approach to acupuncture. Here is one of their published study:
http://aim.bmj.com/content/27/4/174.abstract
It's good to be a skeptic, it is even better to research a bit.
One we have discovered what intelligence really is, for sure it will be somewhat simple by some standards. At the moment we do not even have a working definition of what it is really, so I'm not so convinced as you are, and I doubt we will have made that much more progress by 2030. My reasoning is this: even riding some kind of exponential, locally progress rate is still linear. Fundamentally we have made some but not sufficient progress in the last 30 years in computer science in general. Over the next 20 years I expect more of the same. We need more biology insight and that is going to require a truly massive effort. We do not even have yet the required probing instruments (e.g. sufficiently precise functional MRI machines for instance)
Mostly I think that in general the human population, myself included, is an incredibly stupid, short sighted, nasty, egotistic species with occasional streaks of artistic, political or scientific brilliance. Hopefully we can do better than simulate that.
Sometimes a poor artist is really not a very good one, but sometimes artists are incredibly ahead of their time and not understood, although that may have been more true around the time of Vincent van Gogh than now.
At any rate exceptional artists are by definition very rare exceptions.
Thanks for the link, excellent stuff.
However I think both Orwell and Huxley were right. Both "brave new world" and "1984" are upon us right now.