This article seems to imply that Apple's primary security model is to first verify the apps and then give them at runtime unlimited access to the system, trusting them to only do the things they promised. This seems odd, especially compared with Android, where apps are limited at runtime to whatever capabilities they were granted by the user.
This issue could be trivially solved by enforcing permissions at runtime.
Sexism, both in the past and today, is a terrible thing. We are talking here about a trailblazing woman who was a victim of sexism. There is no need to rewrite quotes to make the point that she was a victim.
Because the computers were all female. There was rampant sexism at the time – in particular in that women could be computers but not research staff (with Ms. Johnson an apparent exception). But there are better ways of highlighting this sexism (of which Ms. Johnson was a victim) than by unreasonably rewriting quotes from the article
The article says "The practice in 1960 would have been not to list the female Computers as formal co-authors". The blurb above replaces "Computers" with "contributors", painting a false and offensive picture.
Today in many fields it is common to only include as authors of a paper those who have had creative scientific input. A common example is research assistants who collate data, or technical staff who build lab equipment, but the example of someone who did a numerical computation for the author is not uncommon. Most "computers" simply did the computations, which was certainly an important contribution to the research, but not necessarily the kind of contribution that makes one an author of a paper.
Analog computers are computers, of course -- they just aren't Turing machines. In TCS, "computer" is not the name of any model of computation -- unlike "Turing machine", "register machine", "pushdown automaton" and "quantum computer". So the word "computer" retains its everyday meaning.
On the other hand, you wouldn't call a device with a stack a "pushdown automaton" unless it actually was a pushdown automaton, right? Similarly, I wouldn't call something a "quantum computer" unless it really corresponds to that model of computation. I agree that this creates a void in the terminology, but this is where things stand now.
So is this finally proof that D-Wave has actually produced a real working quantum processor and isn't just pulling the wool over everyone's eyes?
This finally proves that, in some applications, D-Wave's machine offers considerable speedup over alternatives. It also confirms that D-Wave's machine uses quantum effects to speed up computation, but this point was never in dispute.
However, the term "quantum computer" has a very specific meaning (just like "Turing machine" has a specific meaning), and D-Wave's machine isn't a quantum computer. They use that label, pretending that they mean the literal reading but hoping you get confused and think of the technical one.
Despite being a computing device that relies on quantum effects, D-Wave's machine is not a "quantum computer" as that term is defined by computer scientists.
Commendably, Google's blog post calls the device a "quantum annealer", rejecting D-Wave's self-label of "quantum computer" which is a misleading marketing ploy. Perhaps if D-Wave's device had come before theoretical CS researcher defined their computational model, the term "quantum computer" would have taken a different meaning, but as things stand the meaning of "quantum computer" was fixed well before D-Wave was founded.
Of course with the developer tools built into browsers these days, it only takes a few clicks to delete the nag layer and get to the underlying content. I wonder how they count me in their statistics?
It used to be easy to read the content off the html – no developer tools needed! Today, many websites are constructed to not serve the underlying content until the you've been served the ad.
By the way, I don't think there's anything wrong with what Springer is doing. Readers can pay cash, or pay by viewing ads. They can also choose not to read.
Climate change requires coordinated action from all (or most) major countries. If the US went ahead with Keystone, then politicians or bureaucrats in other countries would say, why should we stick out our necks on this. The USA isn't making any sacrifices and they're the worst (or in top 2) polluter.
Well, your argument assumes disapproving Keystone XL has both negative climate effect and positive economic benefits, so that disapproving it would have been a "sacrifice". In fact, Obama is arguing that it would have had no economic upside, and that the main gain is the "leadership", disclaiming any reliance on direct environmental benefits. I agree that this jives with Obama's idea of what "sacrifice" and "leadership" mean, but it's not how the rest of the world uses those terms.
It's notable that Obama is making a political calculation (wanting to retain "leadership" relating to climate change, the pipeline not increasing "energy security") rather than an economic or environmental one.
Reading his statement on the matter, his economic justifications are irrelevant ("the pipeline wouldn't create jobs or lower gas prices for Americans"): since it's not proposed that the US government pay for the pipeline, these issues are only relevant against costs -- and he doesn't discuss any costs! He isn't citing the direct environmental damage of digging the pipeline and creating associated infrastructure (roads, power cables, pumping stations etc). He isn't citing the risk of leaks.
I was wondering if Obama would claim climate risks since that would have required him to quantify his estimate of the accuracy of the models used to predict the climate effects of the pipeline. But naturally he didn't claim risks to the climate -- only risks to US leadership on climate issues. That's a fair reason to make national-level decisions, but is not a win for the environment.
I'm not sure what Nintendo's exact legal claim is (of course they'd rather not specify it!), but to this armchair lawyer is seems odd.
Since these videos are derivative works of the games, they are probably legal because they are fair use of the games (they display graphics from the game but aren't a substitute for the game etc). I guess Nintendo is claiming that "fair use" doesn't apply if your source is an infringing copy of the work. This is not impossible, though I don't see why it should be relevant. More importantly, I think that the custom ROMs involve fair use of Nintendo's ROMs, especially since Nintendo isn't offering new ROMs for sale.
In general relativity (our theory of classical gravity, without quantum effects), there are several "no hair" theorems, saying that several types of black holes are completely determined by a few overall parameters (say mass, charge and angular momentum) and without regard to their history.
We don't yet have a theory of quantum gravity, so we don't know if the quantum state of a black hole does retain information. It probably has to, but this is not understood. By the way, in any case classical GR would be an excellent approximation except in the case of very small black holes, so any information retained will not be actually accessible.
The "Intel Active Management" (a governor that runs on a secondary CPU independent of the primary one, with cryptographically signed firmware and autonomous access to LAN, WiFi, Memory etc) is also quite disconcerting, but in fact only inclued on certain chipsets (see the tables for Broadwell and Skylake). Unless you are a large institution you probably don't want remote management capabilities.
It's hard to find which chipsets will feature this DSP but quite possibly some won't. Pay attention when you buy your motherboard and all will be well.
Certainly Xerox can manufacture whatever products they like. We have the right not to buy them (and, say, buy from the competition). Two remarks anyway:
1. Doing this in secret is underhanded, and they should be upfront, Despite the negative reaction by some members of the public ("it's unfair that I'm paying more than X"), there is nothing wrong with a company trying for market segmentation. They should tell the complainers to grow up
2. Everyone should own whatever they own. So, if I own a printer or a toner cartridge, I should have the right to modify and reprogram them however I like (say, to report a different zone or to ignore zonal coding). Courts have rebuffed Xerox and Lexmark as they attempted to use the DMCA to protect their business strategies, but the DMCA (US), Bill C-11 (Canada) and their worldwide clones still apply to DVD-players, for example. That should stop.
I actually think you should start with the ownership.
I believe that, like the research results themselves, all the underlying data of publicly funded research belong to the general public. The researchers must have a right to keep their data secret for a while so they get first dibs to produce results from it, but eventually everybody else should get to try their hand at the data.
Several different agents should work on this. The granting agencies should insist all research data is properly curated and hosted for posterity – as a benefit to the public who paid. This should include the raw data, intermediate products like scripts and code, and the final processed data. The journals should insist on the same thing, this time for the benefit of science (allowing others to verify the results). In all cases there should be an appropriate embargo period, depending on the field.
By the way, I think of your code as falling under "research results" (you developed a method of handling the data) more than "research data".
I am certainly aware of the rules, and event quoted this very rule in my comment. I was arguing that some ways of implementing them are wrong. Deal with it.
The issue is a general one with research. Who owns the research project and the grants – the institution (UCSD) or the researchers (Prof. Aisen and his team). [disclosure: I'm a university professor myself]
To me it seems clear that Prof. Aisen's research is his, and if he moves universities he takes his project with him (especially the data). It's true that formally the university administers the grant (the granting agency write them a check, equipment bought with the money is university property etc). But the project itself is an intangible concept, which runs with the people and not with the university.
Since grants are formally made to institutions, of course approval of the granting agency is needed to move the grant, but this should generally be routine. It's not like USC lacks the ability to administer this research. In particular, I'm quite troubled by the idea that "the original grantee institution [may] not wish to relinquish the grant". Grant-making decisions are primarily based on scientific criteria -- the potential contributions of the researchers -- not on the identity of their home institution, so this rule seems preposterous to me. "UCSD wants to resume its management of the study" -- but I doubt anyone from UCSD other than the research team actually managed the study – by definition the PIs manage the study. UCSD provided administrative services (financial oversight, for example) and facilities (for which the grant is charged overhead), but this is not a unique contribution of UCSD.
Some grants are political (Congressional earmarks) and then it may make sense not to move them if the researcher moves, but NIH grants shouldn't work like that (and in any case these earmarks are illegitimate).
It could, of course, be that they reviewed the benefits and risks and drew their own conclusions which sometimes match the consensus and sometimes don't.
That's exactly what they're doing. But this shows that they don't really believe that scientific consensus is by itself a reason to select a course of action.
I'm always amused by the way science is suborned to political expediency.
Some people strongly tout the consensus regarding global warming/climate change. They commonly disparage and dismiss those who don't fully subscribe as politically-motivated ignoramuses who are anti-science. The doubters view themselves simply as more cautious, unwilling to risk large costs when it is not clear that science can clearly predict there will be benefits.
Other people strongly tout the consensus regarding the safety of GM foods. The opposition claims to be simply cautious, unwilling to risk any unknown dangers of these foods despite the enormous benefits they could provide.
Interestingly enough, very often it's the same people who support massive reductions in CO2 emissions based on a scientific consensus and despite the economic costs and the uncertain climate benefits, and yet would prefer to avoid the benefits of GM foods due to fear of unknown bad results, despite the scientific consensus.
So what's your stance on RSA, one of the early software patents, which is still used everywhere?
I didn't try to articulate every problem with software patents, merely those illustrated by the just-overturned patent covering SSL using RC4. Note that RC4 itself is about 30 years old, and was developed by RSA security.
In any case, regarding the RSA cryptosystem itself, it was developed by several academics (independent of its previous, secret, invention GCHQ), and clearly it would have been developed and published even without the extra bonus of patent protection. It's important to remember that patents are a means to an end ("promot[ion] of Progress of Science and useful Arts") -- which is not to make money for inventors but to provide them an incentive to invent for the public good. In other words, a Patent is a way for the public to give up something (the natural possibility of making use of an invention you hear about) in return for a different advantage (getting the invention made in the first place). If inventors would invent even without the extra incentive, there is not need for the incentive.
Since practically all the value of inventiveness in the software business can be captured simply by writing the software (and, in cryptography especially, by ordinary academic incentives such as promotion, tenure and professional recognition), software patents don't help. Instead they hinder.
For a salient example consider the LZW patent. The algorithm was designed by two academics (Lempel and Ziv of the Technion). The main effect of the patent was to end the widespread use of.gif files (the GIF standard specified LZW compression), which dominated the early internet. Rather than knuckle under and pay licensing fees (and end free-software projects like Mozilla), the internet community developed PNG, an equivalent but patent-unencumbered compressed raster image format. Practically the whole internet switched to PNG -- showing that useful technology can be invented without the patent incentive, and that when there is no patent it is much more widely used for everyone's enjoyment.
PS: It is likely that the LZW patent was invalid (patenting an abstract algorithm), but nobody wanted to take the legal risk of going to course to invalidate it. This obnoxious patent has since expired.
It's worth noting that there are knownattacks against RC4 (especially SSL using RC4). While these aren't quite practical yet, it is clear that RC4 is obsolete, and that current programmers should choose other stream cyphers (AES). Even supposing the patent was legitimate, the technology it covers has become obsolete well within its lifetime.
This illustrates one of the key reasons software (that is, algorithms) shouldn't be patentable: the field moves so fast that 20-year patent protection isn't useful. Even supposing the authors of software need patent protection to recoup their "investment" in inventing the algorithm, 20-year protection is effectively an infinite term, since by the time the protection ends, the technology is obsolete.
As an aside, note that patenting a protocol (such as RC4) automatically ends its usefulness. Protocols are only useful if the other party to the communication can participate, and interoperability is very important in software. Patents are ill-suited for this. Copyright, on the other hand, works well: the code you write is protected, but anyone else can write their own code to implement the protocol and communicate with you.
Accessing an open WiFi connection using a repeater would not violate the CFAA -- the connection is open and your device would log on to it. You'd be using it the way it was intended. Of course, The DOJ claims that simply violating terms of service can make you a federal felon, but that's wrong. Read Prof. Orin Kerr's work for more on this
On the other hand, the FCC allows anyone to use the 900MHz band but tightly regulates what can be done there (for example, no "retransmission of.. signals emanating from... radio station other than an amateur radio station", which likely does make this idea illegal. See 47 CFR Part 97.
Slashdot Mirror
This article seems to imply that Apple's primary security model is to first verify the apps and then give them at runtime unlimited access to the system, trusting them to only do the things they promised. This seems odd, especially compared with Android, where apps are limited at runtime to whatever capabilities they were granted by the user.
This issue could be trivially solved by enforcing permissions at runtime.
Sexism, both in the past and today, is a terrible thing. We are talking here about a trailblazing woman who was a victim of sexism. There is no need to rewrite quotes to make the point that she was a victim.
Because the computers were all female. There was rampant sexism at the time – in particular in that women could be computers but not research staff (with Ms. Johnson an apparent exception). But there are better ways of highlighting this sexism (of which Ms. Johnson was a victim) than by unreasonably rewriting quotes from the article
The article says "The practice in 1960 would have been not to list the female Computers as formal co-authors". The blurb above replaces "Computers" with "contributors", painting a false and offensive picture.
Today in many fields it is common to only include as authors of a paper those who have had creative scientific input. A common example is research assistants who collate data, or technical staff who build lab equipment, but the example of someone who did a numerical computation for the author is not uncommon. Most "computers" simply did the computations, which was certainly an important contribution to the research, but not necessarily the kind of contribution that makes one an author of a paper.
Alan Rickman was also noted for his long-lasting partnership away from the limelight.
The Indiana Legistlature has already tried to legislate on mathematics.
Analog computers are computers, of course -- they just aren't Turing machines. In TCS, "computer" is not the name of any model of computation -- unlike "Turing machine", "register machine", "pushdown automaton" and "quantum computer". So the word "computer" retains its everyday meaning.
On the other hand, you wouldn't call a device with a stack a "pushdown automaton" unless it actually was a pushdown automaton, right? Similarly, I wouldn't call something a "quantum computer" unless it really corresponds to that model of computation. I agree that this creates a void in the terminology, but this is where things stand now.
This finally proves that, in some applications, D-Wave's machine offers considerable speedup over alternatives. It also confirms that D-Wave's machine uses quantum effects to speed up computation, but this point was never in dispute.
However, the term "quantum computer" has a very specific meaning (just like "Turing machine" has a specific meaning), and D-Wave's machine isn't a quantum computer. They use that label, pretending that they mean the literal reading but hoping you get confused and think of the technical one.
Despite being a computing device that relies on quantum effects, D-Wave's machine is not a "quantum computer" as that term is defined by computer scientists.
Commendably, Google's blog post calls the device a "quantum annealer", rejecting D-Wave's self-label of "quantum computer" which is a misleading marketing ploy. Perhaps if D-Wave's device had come before theoretical CS researcher defined their computational model, the term "quantum computer" would have taken a different meaning, but as things stand the meaning of "quantum computer" was fixed well before D-Wave was founded.
It used to be easy to read the content off the html – no developer tools needed! Today, many websites are constructed to not serve the underlying content until the you've been served the ad.
By the way, I don't think there's anything wrong with what Springer is doing. Readers can pay cash, or pay by viewing ads. They can also choose not to read.
Well, your argument assumes disapproving Keystone XL has both negative climate effect and positive economic benefits, so that disapproving it would have been a "sacrifice". In fact, Obama is arguing that it would have had no economic upside, and that the main gain is the "leadership", disclaiming any reliance on direct environmental benefits. I agree that this jives with Obama's idea of what "sacrifice" and "leadership" mean, but it's not how the rest of the world uses those terms.
It's notable that Obama is making a political calculation (wanting to retain "leadership" relating to climate change, the pipeline not increasing "energy security") rather than an economic or environmental one.
Reading his statement on the matter, his economic justifications are irrelevant ("the pipeline wouldn't create jobs or lower gas prices for Americans"): since it's not proposed that the US government pay for the pipeline, these issues are only relevant against costs -- and he doesn't discuss any costs! He isn't citing the direct environmental damage of digging the pipeline and creating associated infrastructure (roads, power cables, pumping stations etc). He isn't citing the risk of leaks.
I was wondering if Obama would claim climate risks since that would have required him to quantify his estimate of the accuracy of the models used to predict the climate effects of the pipeline. But naturally he didn't claim risks to the climate -- only risks to US leadership on climate issues. That's a fair reason to make national-level decisions, but is not a win for the environment.
Thank you for a highly informative reply.
I'm not sure what Nintendo's exact legal claim is (of course they'd rather not specify it!), but to this armchair lawyer is seems odd.
Since these videos are derivative works of the games, they are probably legal because they are fair use of the games (they display graphics from the game but aren't a substitute for the game etc). I guess Nintendo is claiming that "fair use" doesn't apply if your source is an infringing copy of the work. This is not impossible, though I don't see why it should be relevant. More importantly, I think that the custom ROMs involve fair use of Nintendo's ROMs, especially since Nintendo isn't offering new ROMs for sale.
In general relativity (our theory of classical gravity, without quantum effects), there are several "no hair" theorems, saying that several types of black holes are completely determined by a few overall parameters (say mass, charge and angular momentum) and without regard to their history.
We don't yet have a theory of quantum gravity, so we don't know if the quantum state of a black hole does retain information. It probably has to, but this is not understood. By the way, in any case classical GR would be an excellent approximation except in the case of very small black holes, so any information retained will not be actually accessible.
The "Intel Active Management" (a governor that runs on a secondary CPU independent of the primary one, with cryptographically signed firmware and autonomous access to LAN, WiFi, Memory etc) is also quite disconcerting, but in fact only inclued on certain chipsets (see the tables for Broadwell and Skylake). Unless you are a large institution you probably don't want remote management capabilities.
It's hard to find which chipsets will feature this DSP but quite possibly some won't. Pay attention when you buy your motherboard and all will be well.
Certainly Xerox can manufacture whatever products they like. We have the right not to buy them (and, say, buy from the competition). Two remarks anyway:
1. Doing this in secret is underhanded, and they should be upfront, Despite the negative reaction by some members of the public ("it's unfair that I'm paying more than X"), there is nothing wrong with a company trying for market segmentation. They should tell the complainers to grow up
2. Everyone should own whatever they own. So, if I own a printer or a toner cartridge, I should have the right to modify and reprogram them however I like (say, to report a different zone or to ignore zonal coding). Courts have rebuffed Xerox and Lexmark as they attempted to use the DMCA to protect their business strategies, but the DMCA (US), Bill C-11 (Canada) and their worldwide clones still apply to DVD-players, for example. That should stop.
I actually think you should start with the ownership.
I believe that, like the research results themselves, all the underlying data of publicly funded research belong to the general public. The researchers must have a right to keep their data secret for a while so they get first dibs to produce results from it, but eventually everybody else should get to try their hand at the data.
Several different agents should work on this. The granting agencies should insist all research data is properly curated and hosted for posterity – as a benefit to the public who paid. This should include the raw data, intermediate products like scripts and code, and the final processed data. The journals should insist on the same thing, this time for the benefit of science (allowing others to verify the results). In all cases there should be an appropriate embargo period, depending on the field.
By the way, I think of your code as falling under "research results" (you developed a method of handling the data) more than "research data".
I am certainly aware of the rules, and event quoted this very rule in my comment. I was arguing that some ways of implementing them are wrong. Deal with it.
The issue is a general one with research. Who owns the research project and the grants – the institution (UCSD) or the researchers (Prof. Aisen and his team). [disclosure: I'm a university professor myself]
To me it seems clear that Prof. Aisen's research is his, and if he moves universities he takes his project with him (especially the data). It's true that formally the university administers the grant (the granting agency write them a check, equipment bought with the money is university property etc). But the project itself is an intangible concept, which runs with the people and not with the university.
Since grants are formally made to institutions, of course approval of the granting agency is needed to move the grant, but this should generally be routine. It's not like USC lacks the ability to administer this research. In particular, I'm quite troubled by the idea that "the original grantee institution [may] not wish to relinquish the grant". Grant-making decisions are primarily based on scientific criteria -- the potential contributions of the researchers -- not on the identity of their home institution, so this rule seems preposterous to me. "UCSD wants to resume its management of the study" -- but I doubt anyone from UCSD other than the research team actually managed the study – by definition the PIs manage the study. UCSD provided administrative services (financial oversight, for example) and facilities (for which the grant is charged overhead), but this is not a unique contribution of UCSD.
Some grants are political (Congressional earmarks) and then it may make sense not to move them if the researcher moves, but NIH grants shouldn't work like that (and in any case these earmarks are illegitimate).
That's exactly what they're doing. But this shows that they don't really believe that scientific consensus is by itself a reason to select a course of action.
I'm always amused by the way science is suborned to political expediency.
Some people strongly tout the consensus regarding global warming/climate change. They commonly disparage and dismiss those who don't fully subscribe as politically-motivated ignoramuses who are anti-science. The doubters view themselves simply as more cautious, unwilling to risk large costs when it is not clear that science can clearly predict there will be benefits.
Other people strongly tout the consensus regarding the safety of GM foods. The opposition claims to be simply cautious, unwilling to risk any unknown dangers of these foods despite the enormous benefits they could provide.
Interestingly enough, very often it's the same people who support massive reductions in CO2 emissions based on a scientific consensus and despite the economic costs and the uncertain climate benefits, and yet would prefer to avoid the benefits of GM foods due to fear of unknown bad results, despite the scientific consensus.
I didn't try to articulate every problem with software patents, merely those illustrated by the just-overturned patent covering SSL using RC4. Note that RC4 itself is about 30 years old, and was developed by RSA security.
In any case, regarding the RSA cryptosystem itself, it was developed by several academics (independent of its previous, secret, invention GCHQ), and clearly it would have been developed and published even without the extra bonus of patent protection. It's important to remember that patents are a means to an end ("promot[ion] of Progress of Science and useful Arts") -- which is not to make money for inventors but to provide them an incentive to invent for the public good. In other words, a Patent is a way for the public to give up something (the natural possibility of making use of an invention you hear about) in return for a different advantage (getting the invention made in the first place). If inventors would invent even without the extra incentive, there is not need for the incentive.
Since practically all the value of inventiveness in the software business can be captured simply by writing the software (and, in cryptography especially, by ordinary academic incentives such as promotion, tenure and professional recognition), software patents don't help. Instead they hinder.
For a salient example consider the LZW patent. The algorithm was designed by two academics (Lempel and Ziv of the Technion). The main effect of the patent was to end the widespread use of .gif files (the GIF standard specified LZW compression), which dominated the early internet. Rather than knuckle under and pay licensing fees (and end free-software projects like Mozilla), the internet community developed PNG, an equivalent but patent-unencumbered compressed raster image format. Practically the whole internet switched to PNG -- showing that useful technology can be invented without the patent incentive, and that when there is no patent it is much more widely used for everyone's enjoyment.
PS: It is likely that the LZW patent was invalid (patenting an abstract algorithm), but nobody wanted to take the legal risk of going to course to invalidate it. This obnoxious patent has since expired.
It's worth noting that there are known attacks against RC4 (especially SSL using RC4). While these aren't quite practical yet, it is clear that RC4 is obsolete, and that current programmers should choose other stream cyphers (AES). Even supposing the patent was legitimate, the technology it covers has become obsolete well within its lifetime.
This illustrates one of the key reasons software (that is, algorithms) shouldn't be patentable: the field moves so fast that 20-year patent protection isn't useful. Even supposing the authors of software need patent protection to recoup their "investment" in inventing the algorithm, 20-year protection is effectively an infinite term, since by the time the protection ends, the technology is obsolete.
As an aside, note that patenting a protocol (such as RC4) automatically ends its usefulness. Protocols are only useful if the other party to the communication can participate, and interoperability is very important in software. Patents are ill-suited for this. Copyright, on the other hand, works well: the code you write is protected, but anyone else can write their own code to implement the protocol and communicate with you.
Accessing an open WiFi connection using a repeater would not violate the CFAA -- the connection is open and your device would log on to it. You'd be using it the way it was intended. Of course, The DOJ claims that simply violating terms of service can make you a federal felon, but that's wrong. Read Prof. Orin Kerr's work for more on this
On the other hand, the FCC allows anyone to use the 900MHz band but tightly regulates what can be done there (for example, no "retransmission of .. signals emanating from ... radio station other than an amateur radio station", which likely does make this idea illegal. See 47 CFR Part 97.