There is nothing new-fangled about dynamic languages. Some of the oldest and most proven programming languages, Smalltalk, Lisp, and APL, are dynamically typed. For practical purposes, Java is dynamically typed as well; Java's static type system is so badly designed that it might as well not be there.
Furthermore, dynamic typing's strength isn't in quick hacks, it's in building large, long-running systems. You cannot build those systems without reflection, introspection, duck typing, and other dynamic features. When people try to build them in strictly statically typed languages, they implement their own dynamic type systems, and they usually do a worse job than if it were built in.
Perl's problem is not that it is dynamically typed, it is that it is weakly typed (in addition to being generally poorly designed).
I'm not just talking about a mismatch; Fortran 77 and C are not that different as languages. Fortran 2008 has a ton of features for HPC, optimization, genericity, and parallelism that Java is completely lacking. Translating Fortran 2008 into Java is like translating it into Fortran 66, and the code still won't run as fast.
Benjymouse explained it very well. I was referring to multidimensional arrays, which are essential to a lot of scientific work. There is a long list of performance-relevant functionality in which the JVM is deficient relative to other languages.
Dynamic languages may have been slow in the past (personally I hate them, you can't write huge codebases in a dynamic language and get a team to understand it),
Whether a large codebase is easy to understand or not depends on documentation, architecture, and unit testing, not on static type checking.
but *Java* was always pretty damn fast:
That's because the JVM has a lot of restrictions and people only write benchmarks that conform to those restrictions. Try translating an idiomatically written modern Fortran 2008 (or C++) HPC program into Java: it's extremely hard (because Java lacks a lot of functionality), and getting anything like Fortran/C++ performance out of the Java is even harder.
The JVM may be faster than the CLR on microbenchmarks, but that doesn't mean it's faster on real code, since the JVM has less performance-relevant functionality than the CLR (e.g., generics, arrays, pointers, etc.).
Imagine, in the limit, you could come up with a really simple VM with just two or three instructions that performed spectacularly well on a very restricted set of microbenchmarks, but it wouldn't be any good for writing large programs in.
Your view of Milner is that he is exploited by his company, alienated from his work, and unable to change jobs because of a labor surplus. That is, you hold the classical Marxist view of labor; that's just an objective fact. If you didn't know that was Marxist, start doing your homework.
The capitalist view of labor is that it operates in a free market, so workers can negotiate fair wages, prefer their current work to the alternatives, and can change jobs if another job offers them a better overall value.
One can argue about whether the Marxist view of labor applies to unskilled factory workers (and that observation alone wouldn't amount to a full-scale endorsement of Marxist ideology), but it certainly does not apply to Google engineers. Milner had a choice whether to collect this data or not, and he wouldn't have starved if he had chosen not to.
So it looks like you are now adding fabricated outrage to fabricated facts.
But placing blame entirely on people who fail to encrypt their wireless is going too far in Google's favor. If I don't lock the door to my house, yeah it's my fault if I get robbed. But that doesn't make the robbery legal.
But we have a long-standing principle that if you use the public airwaves, people have a right to listen. Why do you want to change that principle and suddenly criminalize behavior that's been legal for as long as we have had radio?
If you find a neighbor's wifi network is open, that doesn't give you carte blanche to use it and snoop their devices and data
I do not have a right to use it, nor is anybody arguing that you should. Use requires broadcasting and interfering with the access points. I should have a right to passively listen and record whatever my neighbor broadcasts, just like any other radio transmission.
Your private data has to be afforded some legal protection regardless of the amount or strength of encryption. The dividing line has to be whether the user had an expectation of privacy when transmitting that data
If you broadcast your private information over the radio, if you print it on fliers, if you have loud conversations about it sitting at Starbucks, it ceases to be private. Whether information is private is not defined by the information, it's always defined by what you have done with it in the past. If you publish it, it ceases to be private.
I think wifi is new enough for non-tech people that for a home network, most courts would agree the owner had an expectation of privacy even if he failed to turn encryption on.
Catering to people's stupidity isn't going to protect them from criminals. The only way to keep your private information from criminals is for people to use encryption. So, we could outlaw unencrypted connections, or we could require companies to print big warning labels. But making listening in to broadcast over the public airwaves illegal has no benefits and only threatens basic and important rights that we all have.
My 'twisted' world view is called 'capitalism'. And yes, if you want to stay employed, you do what the person signing your paycheck tells you to do.
No, your world-view is Marxist because you view employees as little more than slaves. In a capitalist system, people change jobs when their employer doesn't treat them right. And Milner would have had no problems finding another job if he didn't want to record this data; skilled networking software developers are in high demand.
The evidence does not agree with your 'world view'. Also, although cliche, I have to say "Citation needed."
You fabricate the idea that Milner was forced to collect that data. Then you fabricate the idea that Milner will be fired over this. And you fabricate the idea that Milner is now unemployable. Where is your evidence?
Future employers won't take the risk. Taking the moral high ground is not without its consequences; That is why so few people these days do it.
If you think that recording this data was wrong, then the "moral high ground" would have been to quit before recording the data, in which case Milner wouldn't have this on his record.
Ooooh, directly influenced. As in "if we stick his name on it, it gives our work more cache than its crappy quality deserves."
And even if that were true, Campbell's own original ideas are clever soundbites without much substance. Most of his actual contribution has probably in getting people to look at the original myths again, where scores of Hollywood screenwriters, devoid of their own ideas, have copied plots and personalities from.
Sci-fi fandom is especially guilty of pushing this sort of treacle.
Really? A lot of SciFi is about fun and adventure. A lot is about technological possibilities. Some of it may be about "identity", but not in the moronic philosophical sense in which most classical literature deals with the topic.
If anybody is "especially guilty", it is pompous high literature and its academic devotees, the kind of people who traditionally are offended by SciFi.
Most of art and literature deals with "exploring human identity". Star Wars has to be one of the crappiest examples of that and gives SciFi a bad name.
What kind of twisted world view do you have in which corporate employees are transformed into mindless minions that have to obey every command? As an employee, you still have moral and legal responsibilities. If he had thought that this was the wrong thing to do, he could have said "no". In fact, the way Google works, he probably could have said "no" without consequences.
I think what rather happened is that he thought this was an OK thing to do. Good for him! I hope he makes that argument stick, because I think he's right and it's the principled position to take. However, it still comes down to the fact that he made that choice to go ahead, and he needs to now deal with the consequences. And the long term consequences may still be good for him and for all of us, in that people may come to realize that we shouldn't have useless and ineffective legal restrictions on recording publicly broadcast data.
Well, I don't think it happened in this case, but pretending to keep someone's name anonymous and then leaking it is a pretty common strategy in politics.
If you broadcast information publicly and without sufficient encryption, the public can listen in and record it.
Apart from the question of who is right in the abstract, punishing Google or other people isn't going to deter anybody who actually wants to do you harm, since passive listening is pretty much impossible to detect. What we might restrict and punish is the use of such information, for example rebroadcasting it, using it in legal proceedings without a prior warrant, or reselling it.
The real question we should be asking is how people are punished that broadcast private information (e.g., hospitals that use unencrypted networks).
Why would you assume that "a project like this would have better algorithms"? Do you think that art historians have some storehouse of algorithms that are just unknown to Google's stable of top engineers and scientists?
Linux is widely used on the desktop, in education, software development, and various other markets. It hasn't displaced Windows, but it's one of the three big desktop operating systems. If you just look at desktops, there is probably more Linux systems than OS X systems.
I think Linux will continue to grow steadily on the desktop. The big problem is really laptops, but even that is getting better.
Yes. Hence, as I was saying: you cannot reduce your income tax by moving abroad; there are no tax benefits to moving abroad. Companies, in contrast, can shop around for lower corporate taxes abroad and benefit from them.
Face recognition algorithms (at least the ones that work well enough for practical use) recognize faces based on exact shape and appearance. Humans often use characteristics and qualitative deviations from "normal" faces. That's why humans have no problems recognizing caricatures like these: http://tinyurl.com/d8pq9f6 Computers can't do that yet. Paintings tend to be more like caricatures, not photos.
And face recognition in Facebook usually only has a few dozen people to choose from, with a high probability that the same faces occur again. In art, you have tens of thousands of faces and a low probability that any one occurs multiple times.
Pointing this out is not about whether Apple did anything wrong, it's about whether we should change laws so that they can't keep behaving this way.
Furthermore, Apple likes to project a particular kind of image, and their customers should know what the company is really doing: they are a profit maximizing entity, nothing more. All the shiny packaging, talk of innovation, and fancy stores is just a facade to extract the maximum amount of money from your pocket. That's alright, but people need to be reminded of it periodically.
As a US citizen, you can not reduce your income tax by moving abroad. Furthermore, individuals generally impose fewer costs on the government as they become richer, but companies impose higher costs as they get bigger.
And just because it is legal doesn't mean it is ethical, and the point of these kinds of examples is usually not to ask an entity to voluntarily pay more, it is to talk about raising taxes on it in the future.
That kind of project is a sham. Face recognition software works by precise geometric measurements and by identifying unique and precise skin patterns. Neither of those are present in paintings. Paintings vary a lot more and still require human abilities to interpret facial characteristics.
Well, that's nice. Nevertheless, the fact remains that there is no empirical support for the implicit social, psychological or economic theories that underlie Marx's work, and every attempt to put them into practice has failed.
The burden of proof that Marxism is anything other than pompous pseudo-scientific claptrap is still on the proponents of his system. Of course, the Neo-Marxists (Habermas and the like) are even further removed from science and reason than Marx was.
That is not the consensus or the belief of experts in the field.
I did not say that. I said that "end of civilization" or "extinction" claims are _not_ supported by the IPCC report itself. I said that in the very paragraph you quote. Can't you at least read and understand what you quote?
the consensus is that there is a high risk of a very damaging outcome, in both lives (up to hundreds of millions) and money,
That is a complete misrepresentation of the IPCC report. First of all, nowhere does the report state that there is a high risk of hundreds of millions of lives lost (and if it did, that would be wrong). The monetary damages are quantified in Section 5.7: "global mean losses could be 1 to 5% of GDP for 4C of warming", which are about the same as the cost of taking action right now (Table 5.2). So, according to the IPCC report itself, there is no quantified economic reason to prefer taking action over not taking action.
Slashdot Mirror
There is nothing new-fangled about dynamic languages. Some of the oldest and most proven programming languages, Smalltalk, Lisp, and APL, are dynamically typed. For practical purposes, Java is dynamically typed as well; Java's static type system is so badly designed that it might as well not be there.
Furthermore, dynamic typing's strength isn't in quick hacks, it's in building large, long-running systems. You cannot build those systems without reflection, introspection, duck typing, and other dynamic features. When people try to build them in strictly statically typed languages, they implement their own dynamic type systems, and they usually do a worse job than if it were built in.
Perl's problem is not that it is dynamically typed, it is that it is weakly typed (in addition to being generally poorly designed).
I'm not just talking about a mismatch; Fortran 77 and C are not that different as languages. Fortran 2008 has a ton of features for HPC, optimization, genericity, and parallelism that Java is completely lacking. Translating Fortran 2008 into Java is like translating it into Fortran 66, and the code still won't run as fast.
Benjymouse explained it very well. I was referring to multidimensional arrays, which are essential to a lot of scientific work. There is a long list of performance-relevant functionality in which the JVM is deficient relative to other languages.
Whether a large codebase is easy to understand or not depends on documentation, architecture, and unit testing, not on static type checking.
That's because the JVM has a lot of restrictions and people only write benchmarks that conform to those restrictions. Try translating an idiomatically written modern Fortran 2008 (or C++) HPC program into Java: it's extremely hard (because Java lacks a lot of functionality), and getting anything like Fortran/C++ performance out of the Java is even harder.
The JVM may be faster than the CLR on microbenchmarks, but that doesn't mean it's faster on real code, since the JVM has less performance-relevant functionality than the CLR (e.g., generics, arrays, pointers, etc.).
Imagine, in the limit, you could come up with a really simple VM with just two or three instructions that performed spectacularly well on a very restricted set of microbenchmarks, but it wouldn't be any good for writing large programs in.
Your view of Milner is that he is exploited by his company, alienated from his work, and unable to change jobs because of a labor surplus. That is, you hold the classical Marxist view of labor; that's just an objective fact. If you didn't know that was Marxist, start doing your homework.
The capitalist view of labor is that it operates in a free market, so workers can negotiate fair wages, prefer their current work to the alternatives, and can change jobs if another job offers them a better overall value.
One can argue about whether the Marxist view of labor applies to unskilled factory workers (and that observation alone wouldn't amount to a full-scale endorsement of Marxist ideology), but it certainly does not apply to Google engineers. Milner had a choice whether to collect this data or not, and he wouldn't have starved if he had chosen not to.
So it looks like you are now adding fabricated outrage to fabricated facts.
But we have a long-standing principle that if you use the public airwaves, people have a right to listen. Why do you want to change that principle and suddenly criminalize behavior that's been legal for as long as we have had radio?
I do not have a right to use it, nor is anybody arguing that you should. Use requires broadcasting and interfering with the access points. I should have a right to passively listen and record whatever my neighbor broadcasts, just like any other radio transmission.
If you broadcast your private information over the radio, if you print it on fliers, if you have loud conversations about it sitting at Starbucks, it ceases to be private. Whether information is private is not defined by the information, it's always defined by what you have done with it in the past. If you publish it, it ceases to be private.
Catering to people's stupidity isn't going to protect them from criminals. The only way to keep your private information from criminals is for people to use encryption. So, we could outlaw unencrypted connections, or we could require companies to print big warning labels. But making listening in to broadcast over the public airwaves illegal has no benefits and only threatens basic and important rights that we all have.
No, your world-view is Marxist because you view employees as little more than slaves. In a capitalist system, people change jobs when their employer doesn't treat them right. And Milner would have had no problems finding another job if he didn't want to record this data; skilled networking software developers are in high demand.
You fabricate the idea that Milner was forced to collect that data. Then you fabricate the idea that Milner will be fired over this. And you fabricate the idea that Milner is now unemployable. Where is your evidence?
If you think that recording this data was wrong, then the "moral high ground" would have been to quit before recording the data, in which case Milner wouldn't have this on his record.
Democrats are concerned with civil liberties and the rights of the individual. We need change! Oh, wait...
Ooooh, directly influenced. As in "if we stick his name on it, it gives our work more cache than its crappy quality deserves."
And even if that were true, Campbell's own original ideas are clever soundbites without much substance. Most of his actual contribution has probably in getting people to look at the original myths again, where scores of Hollywood screenwriters, devoid of their own ideas, have copied plots and personalities from.
Really? A lot of SciFi is about fun and adventure. A lot is about technological possibilities. Some of it may be about "identity", but not in the moronic philosophical sense in which most classical literature deals with the topic.
If anybody is "especially guilty", it is pompous high literature and its academic devotees, the kind of people who traditionally are offended by SciFi.
Most of art and literature deals with "exploring human identity". Star Wars has to be one of the crappiest examples of that and gives SciFi a bad name.
What kind of twisted world view do you have in which corporate employees are transformed into mindless minions that have to obey every command? As an employee, you still have moral and legal responsibilities. If he had thought that this was the wrong thing to do, he could have said "no". In fact, the way Google works, he probably could have said "no" without consequences.
I think what rather happened is that he thought this was an OK thing to do. Good for him! I hope he makes that argument stick, because I think he's right and it's the principled position to take. However, it still comes down to the fact that he made that choice to go ahead, and he needs to now deal with the consequences. And the long term consequences may still be good for him and for all of us, in that people may come to realize that we shouldn't have useless and ineffective legal restrictions on recording publicly broadcast data.
Well, I don't think it happened in this case, but pretending to keep someone's name anonymous and then leaking it is a pretty common strategy in politics.
If you broadcast information publicly and without sufficient encryption, the public can listen in and record it.
Apart from the question of who is right in the abstract, punishing Google or other people isn't going to deter anybody who actually wants to do you harm, since passive listening is pretty much impossible to detect. What we might restrict and punish is the use of such information, for example rebroadcasting it, using it in legal proceedings without a prior warrant, or reselling it.
The real question we should be asking is how people are punished that broadcast private information (e.g., hospitals that use unencrypted networks).
Why would you assume that "a project like this would have better algorithms"? Do you think that art historians have some storehouse of algorithms that are just unknown to Google's stable of top engineers and scientists?
Linux is widely used on the desktop, in education, software development, and various other markets. It hasn't displaced Windows, but it's one of the three big desktop operating systems. If you just look at desktops, there is probably more Linux systems than OS X systems.
I think Linux will continue to grow steadily on the desktop. The big problem is really laptops, but even that is getting better.
You're confusing face detection and face recognition.
Yes. Hence, as I was saying: you cannot reduce your income tax by moving abroad; there are no tax benefits to moving abroad. Companies, in contrast, can shop around for lower corporate taxes abroad and benefit from them.
Face recognition algorithms (at least the ones that work well enough for practical use) recognize faces based on exact shape and appearance. Humans often use characteristics and qualitative deviations from "normal" faces. That's why humans have no problems recognizing caricatures like these: http://tinyurl.com/d8pq9f6 Computers can't do that yet. Paintings tend to be more like caricatures, not photos.
And face recognition in Facebook usually only has a few dozen people to choose from, with a high probability that the same faces occur again. In art, you have tens of thousands of faces and a low probability that any one occurs multiple times.
Pointing this out is not about whether Apple did anything wrong, it's about whether we should change laws so that they can't keep behaving this way.
Furthermore, Apple likes to project a particular kind of image, and their customers should know what the company is really doing: they are a profit maximizing entity, nothing more. All the shiny packaging, talk of innovation, and fancy stores is just a facade to extract the maximum amount of money from your pocket. That's alright, but people need to be reminded of it periodically.
As a US citizen, you can not reduce your income tax by moving abroad. Furthermore, individuals generally impose fewer costs on the government as they become richer, but companies impose higher costs as they get bigger.
And just because it is legal doesn't mean it is ethical, and the point of these kinds of examples is usually not to ask an entity to voluntarily pay more, it is to talk about raising taxes on it in the future.
That kind of project is a sham. Face recognition software works by precise geometric measurements and by identifying unique and precise skin patterns. Neither of those are present in paintings. Paintings vary a lot more and still require human abilities to interpret facial characteristics.
Well, that's nice. Nevertheless, the fact remains that there is no empirical support for the implicit social, psychological or economic theories that underlie Marx's work, and every attempt to put them into practice has failed.
The burden of proof that Marxism is anything other than pompous pseudo-scientific claptrap is still on the proponents of his system. Of course, the Neo-Marxists (Habermas and the like) are even further removed from science and reason than Marx was.
I did not say that. I said that "end of civilization" or "extinction" claims are _not_ supported by the IPCC report itself. I said that in the very paragraph you quote. Can't you at least read and understand what you quote?
That is a complete misrepresentation of the IPCC report. First of all, nowhere does the report state that there is a high risk of hundreds of millions of lives lost (and if it did, that would be wrong). The monetary damages are quantified in Section 5.7: "global mean losses could be 1 to 5% of GDP for 4C of warming", which are about the same as the cost of taking action right now (Table 5.2). So, according to the IPCC report itself, there is no quantified economic reason to prefer taking action over not taking action.