This is the old Luddite argument: without technology a lot more effort is required to get things done -- so more people get work. It follows that technology is bad.
In fact, the situation is exactly the opposite: if a machine can drive a car, then having a person drive the car is a waste of the person's time. They can instead do something else with their time, so society get both that and the driving done. In the 19th century, more than 80% of US population directly worked in agriculture. Today, the propotion is 2-3% -- yet we have a lot more food, and many other things to boot.
It's true that in the short term, there is a loss when the specialized skills (say driving) of the people displaced become less valuable, and those people lose their jobs. But this is a transient effect. Some skills were standard 30 years ago, yet rare today.
The more important issue is that technology more easily replaces low-skilled workers. Computers have reduced the demand for secretarial work; robots and other industrial automation reduce the demand for factory workers, and so on. This increases the returns to IQ and education, and reduces the number of well-paying jobs available to less-educated workers. But this seems inevitable, and needs to be solved by changing the attitudes of society toward education rather than by hamstringing technological progress.
Wake up and smell the roses. And then try to fix elementary school. It's practically impossible to repair this damage by the time students reach college (and even highschool is too late).
This is definitely not a scam. This company built a device which uses quantum-mechanical effects to quickly solve simulated annealing problems. They get a huge speedup in solving quatum annealing problems — which is what the customers are paying for. The customers understand exactly what they are buying -- no shenanigans here.
However, D-Wave's publicity is rather dishonest. They call their device a "quantum computer" and issue press releases with that term, despite the fact that their device is definitely not a quantum computer in the sense that theoretical computer scientists use the word. It may be that we need to redefine what "quantum computer" means, especially since D-Wave are the only ones with a product on the market that uses quantum mechanics in a computation, but so far this hasn't changed.
So the right to free speech doesn't include the right to speak together with other people, unless you grant me that right? That's downright offensive. I have a right to free speech — including the right to band together with other people and speak jointly.
Specifically, do you think the ACLU and the NAACP (both of which are corporations) have a right to free speech? What about the AFL-CIO? Of do you think that it's OK for the governmet to limit what the ACLU can say on the theory that it's a corporation, not a person?
I think you miss the point of arguing by analogy here, which is to establish a moral or legal reference point (depending on the discussion). Most of us have a personal idea of the moral weight of (relatively) common actions like robbing a bank, stealing a car for a joyride (you asked for it!) and helping an old woman cross the street.
When we are faced with a new phenomenon (abusing the fact that users run your code to suborn their computing power for personal gain), we need to decide what moral weight to give it. The natural approach is not to start from first principles, but rather to compare it with our existing framework -- in other words to argue by analogy. We say "this was not nearly as serious as bank robbery" or "this is certainly more serious than selling crappy software".
The situation is very similar when we address the legal question ("considering our existing set of legal rules, what should the punishment be?"). To me such thinking is very important, or you end up with the current US regime where criminal hacking into a computer can lead to more jail time than raping the sysadmin.
Faster computation of hash functions is very important, especially to low-power devices. In other words, even if the improvements in cryptographic strength are irrelevant I'd expect the new standard to be adopted quickly.
Back in the day, it was possible to configure Cisco routers by attaching a VT terminal to their RS-232 interface. Perhaps users not wishing to use the "cloud" interface can configure this way instead.
You seem to be repeating the common misunderstanding that "freedom of the press" is about securing rights to the "news media".
In fact, the word "press" has only acquired this meaning much later. The "press" mentioned in the First Amendment is actually a machine, the printing press. In context, what the amendment is saying is that everyone has the right to communicate with the general public both orally ("freedom of speech") and in writing ("freedom of the press").
Even so, these changes are wholly dependent on these machines - remove the wind fanss and the weather will revert to its previous state. This has nothing to do with climate change, which is a change to the underlying system.
Teller is NOT suing over the technical aspects of the trick (how the rose petals are actually made to fall). They are not copyrightable, and indeed are not described in his copyright registration. Rather, he is suing over the dramatic presentation; essentially he is claiming coypright in a short (2-minute) theatrical performance.
I don't think Teler would have protested his competitor adopting the trick itself (magician interacts with shadow on screen but affecting the physical object casintg the shadow) -- he'd have expected credit ("This trick was invented by Teller") but wouldn't have claimed legal ownership. But Teller should be able own the theatre he creates.
Here's a problem: you have a data structure intended to be kept in an array and would like to extend it (say by adding a field) in a backward-compatible way.
Here's a standard solution going back to the first days of computers: chain successive entries in the array, with the first entry being kept roughly the same as it was before while the succeeding ones are made "invalid" in some way so that the legacy programmes ignore them while the new programmes know how to deal with the extra data. A twist on this is putting a "magic cookie" in one of the fields for the "legacy entry" making it easier for the "new programmes" to know to look at the succeeding entries.
What Linus is saying is: hey -- your problem is of this type (the data structure is the directory entry and the array is the directory file), and the standard solution applies to it. I don't think Linus is showing extraordinary skill here.
Of course for this patent Linus's discussion serves are prior art. But I want to draw a more general lesson. This patent is for the natural solution to a trivial problem. It should be invalid on obviousness grounds even if no-one has documented the solution. In fact, the current system fails in that the more trivial the problem the less likely it is that someone had bothered to publish a solution, hence the harder it is to locate prior art. Obviousness is supposed to fill this gap but it doesn't because the PHOSITA used by the court is too stupid relative to the actual skill of actual programmers.
Microsoft may have gotten the idea from the comment, but in principle a patent should require actually putting the idea to practice -- to actually make it work. Patenting the implementation (how do the connectors work? how do you make parts fit together snugly?) is not so absurd.
If the patent really is for the idea "mobile phone with interchangable parts" then it's wrong, but if it's merely for "this particular design of a mobile phone with these particular interchangable parts, with this particular method of fitting them together and connecting them" then there's nothing wrong with it. In fact, the Engadget commenter could still also patent his idea as long as his method for connecting the pieces and so on was different from Microsoft's.
A major problem with the way courts have analyzed software patents is their low-balling the skill level of a "Person having ordinary skill in the art". This is significant since anything such a person can do given the prior art is considered "obvious" and non-patentable. Basically, judges don't understand software well enough to distinguish true invention from routine solutions to problems.
In the case of software patents, many of them (the long-filename patent, the BCD patent [invalidated by the Supreme Court on other grounds] and so on) are for solutions that a typical developer will propose given the problem. But, because PHOSITA is basically taken to be an idiot the "obviousness" limitation on patentability has no effect unless by some chance an actual person bothered to write down this solution in the past.
As far as I know, Google is a single company. That it offers "mail", "calendar", "youtube" and other services doesn't mean it is different companies.
Can EU customers of department stores insist that the department not combine information from their shopping at the furniture department and clothing department but treat it separately? If that's what you want you should simply get several loyalty cards and use them separately -- which Google already supports.
In fact, if you don't want to be tracked by Google then simply don't sign in to a google account when you search. Alternatively, have several google accounts -- one for each service.
Of course, the control system required is far more complicated here. I wonder how much energy is consumed in producing and maintaining the new lampposts, controls, communication network, etc.
"NP" is a class of problems; roughly it's the class of problems for which you can quickly verify whether possible solutions are correct. For example, factoring a large integer may be hard, but checking that a given factorization is correct simply requires doing some multiplications. Making a weekly schedule for a school subject to constraints (every teacher can only teach one class at a time, every classroom can only be used for one class at a time, Dorothy gets Wednesdays off...) is hard, but checking whether a given schedule satisfies the constraints is easy.
Now some problems are easier than others. How do we make this precise? We say that a problem A is easier than B if, given a method for solving B efficiently we can use it to solve A efficiently.
Finally, a problem is "NP-hard" if it's "harder than any problem in NP" in the sense above. Now NP problems are believed to be generally difficult to solve [for example, the scheduling problem is believed to be difficult]. So a problem harder than all NP problems ought to be hard.
What the paper shows is that if you had a method for efficiently reconstructing the laws of physics from experimental data, then your method also be used to solve scheduling problems, factoring, and all other NP problems.
Part of the problem is the traditional large subsidy that agricultural water gets (both via the infrastructure costs and in direct pricing). Farming would make better use of water if it had to pay the price.
PS: "Olympic-size swimming pools per year" is a strange way to measure water usage. "about 6.8 cubic metres per day" is a much clearer way to express this number. In particular, this makes it clear that low-flow toilets have a negligible effect on water use compared to dishwashing, showers, etc.
I got the email too, and it used the unique email address I gave to the NY Times, so either they were breached or some company they gave my data to was breached.
Indeed, this will probably force the NYT to shed light on who they share their subscribers' contact information with.
To be clear regarding "flat surface". Apple might mean by this a touchscreen-only phone without buttons, which is indeed something they introduced. But in that case the argument is again about distinctiveness of the design rather than any "invention".
What's notable about this list is that nearly all items are either industry-wide practices (rectangular phones with flat surfaces) or obvious design choices (a thin rim around the front maximize screen area compared with a thick rim). In particular Apple opted for choices anyone facing the design problem would make, but is now trying to prevent others from making the choices.
Even worse is that the remaining items reflect aesthetic choices on the part of Apple (no adornment, for example). Such choices should indeed be protected, but they are not inventions which deserve patent protection. Instead they are identifying marks which should be protected under trademark law.
I think it's important to realize why the four directions/bin,/sbin,/usr/bin,/usr/sbin exist (and similarly why/lib is separate from/usr/lib).
The reason is that once upon a time discs were small, so that/usr would be mounted separately from the root partition. So/bin and/lib are small directories containing as much of the operating system as you need to get going before you mount/usr and get everything else. In particular, this means the utilities needed to mount those other filesystems and to fix errors in them (e.g. fsck).
The separation between/usr/bin and/usr/sbin means that ordinary users don't have system programs (those from/sbin) in their search path.
Today most installations have the whole system (/ and/usr) on the same partition and it seems that many users use a GUI rather than a terminal. This means that the separation is not needed.
Note that this change is not about multiple-architecture situations like/usr/lib and/usr/lib64. It's about the separation between/lib and/usr/lib (or/lib64 and/usr/lib64).
/usr/local/bin is not supposed to contain any system executables, only locally installed ones. Fedora does not install anything there already. In any case,/usr/local/ seems to have been replaced by/opt/ in most UNIX installations today.
Slashdot Mirror
This is the old Luddite argument: without technology a lot more effort is required to get things done -- so more people get work. It follows that technology is bad.
In fact, the situation is exactly the opposite: if a machine can drive a car, then having a person drive the car is a waste of the person's time. They can instead do something else with their time, so society get both that and the driving done. In the 19th century, more than 80% of US population directly worked in agriculture. Today, the propotion is 2-3% -- yet we have a lot more food, and many other things to boot.
It's true that in the short term, there is a loss when the specialized skills (say driving) of the people displaced become less valuable, and those people lose their jobs. But this is a transient effect. Some skills were standard 30 years ago, yet rare today.
The more important issue is that technology more easily replaces low-skilled workers. Computers have reduced the demand for secretarial work; robots and other industrial automation reduce the demand for factory workers, and so on. This increases the returns to IQ and education, and reduces the number of well-paying jobs available to less-educated workers. But this seems inevitable, and needs to be solved by changing the attitudes of society toward education rather than by hamstringing technological progress.
Wake up and smell the roses. And then try to fix elementary school. It's practically impossible to repair this damage by the time students reach college (and even highschool is too late).
This is definitely not a scam. This company built a device which uses quantum-mechanical effects to quickly solve simulated annealing problems. They get a huge speedup in solving quatum annealing problems — which is what the customers are paying for. The customers understand exactly what they are buying -- no shenanigans here.
However, D-Wave's publicity is rather dishonest. They call their device a "quantum computer" and issue press releases with that term, despite the fact that their device is definitely not a quantum computer in the sense that theoretical computer scientists use the word. It may be that we need to redefine what "quantum computer" means, especially since D-Wave are the only ones with a product on the market that uses quantum mechanics in a computation, but so far this hasn't changed.
So the right to free speech doesn't include the right to speak together with other people, unless you grant me that right? That's downright offensive. I have a right to free speech — including the right to band together with other people and speak jointly.
Specifically, do you think the ACLU and the NAACP (both of which are corporations) have a right to free speech? What about the AFL-CIO? Of do you think that it's OK for the governmet to limit what the ACLU can say on the theory that it's a corporation, not a person?
I think you miss the point of arguing by analogy here, which is to establish a moral or legal reference point (depending on the discussion). Most of us have a personal idea of the moral weight of (relatively) common actions like robbing a bank, stealing a car for a joyride (you asked for it!) and helping an old woman cross the street. When we are faced with a new phenomenon (abusing the fact that users run your code to suborn their computing power for personal gain), we need to decide what moral weight to give it. The natural approach is not to start from first principles, but rather to compare it with our existing framework -- in other words to argue by analogy. We say "this was not nearly as serious as bank robbery" or "this is certainly more serious than selling crappy software". The situation is very similar when we address the legal question ("considering our existing set of legal rules, what should the punishment be?"). To me such thinking is very important, or you end up with the current US regime where criminal hacking into a computer can lead to more jail time than raping the sysadmin.
Could you state the final decision regarding elements?
Faster computation of hash functions is very important, especially to low-power devices. In other words, even if the improvements in cryptographic strength are irrelevant I'd expect the new standard to be adopted quickly.
Back in the day, it was possible to configure Cisco routers by attaching a VT terminal to their RS-232 interface. Perhaps users not wishing to use the "cloud" interface can configure this way instead.
You seem to be repeating the common misunderstanding that "freedom of the press" is about securing rights to the "news media".
In fact, the word "press" has only acquired this meaning much later. The "press" mentioned in the First Amendment is actually a machine, the printing press. In context, what the amendment is saying is that everyone has the right to communicate with the general public both orally ("freedom of speech") and in writing ("freedom of the press").
Even so, these changes are wholly dependent on these machines - remove the wind fanss and the weather will revert to its previous state. This has nothing to do with climate change, which is a change to the underlying system.
Before worrying about security of the software, how about worrying about the correctness and fault-tolerance of the software and hardware?
Most famous is the Therac-25 incident, but it's not the only one.
Teller is NOT suing over the technical aspects of the trick (how the rose petals are actually made to fall). They are not copyrightable, and indeed are not described in his copyright registration. Rather, he is suing over the dramatic presentation; essentially he is claiming coypright in a short (2-minute) theatrical performance.
I don't think Teler would have protested his competitor adopting the trick itself (magician interacts with shadow on screen but affecting the physical object casintg the shadow) -- he'd have expected credit ("This trick was invented by Teller") but wouldn't have claimed legal ownership. But Teller should be able own the theatre he creates.
Here's a problem: you have a data structure intended to be kept in an array and would like to extend it (say by adding a field) in a backward-compatible way.
Here's a standard solution going back to the first days of computers: chain successive entries in the array, with the first entry being kept roughly the same as it was before while the succeeding ones are made "invalid" in some way so that the legacy programmes ignore them while the new programmes know how to deal with the extra data. A twist on this is putting a "magic cookie" in one of the fields for the "legacy entry" making it easier for the "new programmes" to know to look at the succeeding entries.
What Linus is saying is: hey -- your problem is of this type (the data structure is the directory entry and the array is the directory file), and the standard solution applies to it. I don't think Linus is showing extraordinary skill here.
Of course for this patent Linus's discussion serves are prior art. But I want to draw a more general lesson. This patent is for the natural solution to a trivial problem. It should be invalid on obviousness grounds even if no-one has documented the solution. In fact, the current system fails in that the more trivial the problem the less likely it is that someone had bothered to publish a solution, hence the harder it is to locate prior art. Obviousness is supposed to fill this gap but it doesn't because the PHOSITA used by the court is too stupid relative to the actual skill of actual programmers.
Microsoft may have gotten the idea from the comment, but in principle a patent should require actually putting the idea to practice -- to actually make it work. Patenting the implementation (how do the connectors work? how do you make parts fit together snugly?) is not so absurd.
If the patent really is for the idea "mobile phone with interchangable parts" then it's wrong, but if it's merely for "this particular design of a mobile phone with these particular interchangable parts, with this particular method of fitting them together and connecting them" then there's nothing wrong with it. In fact, the Engadget commenter could still also patent his idea as long as his method for connecting the pieces and so on was different from Microsoft's.
A major problem with the way courts have analyzed software patents is their low-balling the skill level of a "Person having ordinary skill in the art". This is significant since anything such a person can do given the prior art is considered "obvious" and non-patentable. Basically, judges don't understand software well enough to distinguish true invention from routine solutions to problems.
In the case of software patents, many of them (the long-filename patent, the BCD patent [invalidated by the Supreme Court on other grounds] and so on) are for solutions that a typical developer will propose given the problem. But, because PHOSITA is basically taken to be an idiot the "obviousness" limitation on patentability has no effect unless by some chance an actual person bothered to write down this solution in the past.
As far as I know, Google is a single company. That it offers "mail", "calendar", "youtube" and other services doesn't mean it is different companies.
Can EU customers of department stores insist that the department not combine information from their shopping at the furniture department and clothing department but treat it separately? If that's what you want you should simply get several loyalty cards and use them separately -- which Google already supports.
In fact, if you don't want to be tracked by Google then simply don't sign in to a google account when you search. Alternatively, have several google accounts -- one for each service.
Of course, the control system required is far more complicated here. I wonder how much energy is consumed in producing and maintaining the new lampposts, controls, communication network, etc.
"NP" is a class of problems; roughly it's the class of problems for which you can quickly verify whether possible solutions are correct. For example, factoring a large integer may be hard, but checking that a given factorization is correct simply requires doing some multiplications. Making a weekly schedule for a school subject to constraints (every teacher can only teach one class at a time, every classroom can only be used for one class at a time, Dorothy gets Wednesdays off ...) is hard, but checking whether a given schedule satisfies the constraints is easy.
Now some problems are easier than others. How do we make this precise? We say that a problem A is easier than B if, given a method for solving B efficiently we can use it to solve A efficiently.
Finally, a problem is "NP-hard" if it's "harder than any problem in NP" in the sense above. Now NP problems are believed to be generally difficult to solve [for example, the scheduling problem is believed to be difficult]. So a problem harder than all NP problems ought to be hard.
What the paper shows is that if you had a method for efficiently reconstructing the laws of physics from experimental data, then your method also be used to solve scheduling problems, factoring, and all other NP problems.
Part of the problem is the traditional large subsidy that agricultural water gets (both via the infrastructure costs and in direct pricing). Farming would make better use of water if it had to pay the price.
PS: "Olympic-size swimming pools per year" is a strange way to measure water usage. "about 6.8 cubic metres per day" is a much clearer way to express this number. In particular, this makes it clear that low-flow toilets have a negligible effect on water use compared to dishwashing, showers, etc.
Indeed, this will probably force the NYT to shed light on who they share their subscribers' contact information with.
To be clear regarding "flat surface". Apple might mean by this a touchscreen-only phone without buttons, which is indeed something they introduced. But in that case the argument is again about distinctiveness of the design rather than any "invention".
What's notable about this list is that nearly all items are either industry-wide practices (rectangular phones with flat surfaces) or obvious design choices (a thin rim around the front maximize screen area compared with a thick rim). In particular Apple opted for choices anyone facing the design problem would make, but is now trying to prevent others from making the choices.
Even worse is that the remaining items reflect aesthetic choices on the part of Apple (no adornment, for example). Such choices should indeed be protected, but they are not inventions which deserve patent protection. Instead they are identifying marks which should be protected under trademark law.
I think it's important to realize why the four directions /bin, /sbin, /usr/bin, /usr/sbin exist (and similarly why /lib is separate from /usr/lib).
The reason is that once upon a time discs were small, so that /usr would be mounted separately from the root partition. So /bin and /lib are small directories containing as much of the operating system as you need to get going before you mount /usr and get everything else. In particular, this means the utilities needed to mount those other filesystems and to fix errors in them (e.g. fsck).
The separation between /usr/bin and /usr/sbin means that ordinary users don't have system programs (those from /sbin) in their search path.
Today most installations have the whole system (/ and /usr) on the same partition and it seems that many users use a GUI rather than a terminal. This means that the separation is not needed.
Note that this change is not about multiple-architecture situations like /usr/lib and /usr/lib64. It's about the separation between /lib and /usr/lib (or /lib64 and /usr/lib64).
/usr/local/bin is not supposed to contain any system executables, only locally installed ones. Fedora does not install anything there already. In any case, /usr/local/ seems to have been replaced by /opt/ in most UNIX installations today.
It's not precisely a household term ...