Or for the projects where having to manage every allocation and deletion is tedious and gets in the way of actually implementing a program.
next version, we get rid of pointers:P
You say this like it is a joke, but it is actually not so terrible an idea. Dumb pointers introduce a whole host of ways for things to go wrong, and in a lot of cases the compiler is in a position to create a pointer for you (see: Common Lisp) when it is truly what you need. If you are writing code at the lowest level, then sure, you need pointers and it would be stupid not to use them; but a lot of projects out there do not need the sort of low level control over pointers that you wind up dealing with when you write C++ code.
I can't understand why anyone would start a new project in C++. The only reason anyone could seriously consider it is if they have to maintain or enhance an existing C++ code base.
While I agree with the sentiment, the reason people keep using C++ is pretty clear: habit. Everyone knows C++ or some very similar language, there is a lot of legacy C++ code, and people are still being trained to write code in C++ (or some similar language). If you were starting a new project and you said, "Let's use Scheme," half the people in the meeting would say they had never written a single Scheme program, and that they did not have time to learn it, and that it would be easier to just use C++ (or something similar). You could list every technical advantage that Scheme has over C++ and it would not make any difference.
Do not underestimate the friction that you face when it comes to C++ and similar languages.
The problem is that C++ dominates the industry; people who want to use a different language, even when they can list every good reason for making that decision, often find that the pressure to stick with C++ is too much to overcome. A lot of people simply refuse to hear the arguments about better languages for a particular project, or are afraid of being unable to find programmers who know something that is not C++ (or similar enough, like Java), or do not want to have to write interfaces for an existing C++ codebase.
I would say that Common Lisp, with a decent compiler and some attention paid to code efficiency, is most certainly a contender on all of the above. The main selling point of C++ now is habit -- large amounts of existing C++ code and lots of programmers who were trained to use C++.
The Church-Turing thesis unifies different models of computation; while Wikipedia tells you that this means that any computational process can be implemented with a Turing Machine, it is equally true that any computational process can be expressed using lambda calculus, combinator logic, mu-recursion, and a variety of other theoretical models of computation.
As for algorithmic analysis...it is not terribly hard to consider in terms of the lambda calculus, although that is less common in algorithms textbooks or CS courses. It is not generally the case that algorithms which are easily understood in terms of Turing machines are harder to analyze in terms of lambda caluclus, nor is it generally the case that algorithms expressed in imperative languages are more efficient or easier to analyze than those expressed in functional languages.
To describe it as "getting in the way" is a gross misrepresentation of the theory of computation and the theory of algorithms. Functional programming is not inefficient, certainly not in terms of the asymptotic performance of functional code. Functional programming languages have seen quite a bit of use "in the real world," including environments where efficiency (not necessarily raw performance, but efficiency in terms of computational complexity) matters.
Also, as a side note, the Wikipedia article's assertion about false lower bounds is misleading. There are a variety of different delays associated with memory access, which may be address dependent (e.g. seek times on hard drives if you are using swap, cache misses, etc.). If these costs are polynomial in the amount of memory used by the program (assuming some sort of locality effects), then the algorithm will still be a polynomial time algorithm.
Frankly, most programmers deal with some level of abstraction between them and the actual machine. In fact, this usually results in faster code than if the programmer were to spend all of his time hand tuning assembly language code; compilers are generally better at producing efficient code than any individual programmer, with a couple exceptions (instructions that are difficult for compilers to take advantage of, for example, but these do not represent the majority of use cases). Some abstraction must be fixed before an algorithm can be analyzed; a Turing machine is a convenient abstraction, but it is not the only such abstraction.
Computer science is about the theory behind computers
Not to nitpick, but this should really be stated as "the theory behind computation." The theory behind (electronic) computers is what computer and electrical engineers study; everything a computer scientist does applies equally to someone working with a pen and paper (in theory, anyway). It is rather unfortunate that so many CS curricula treat theory as some sort of an afterthought, instead of the primary focus.
People are probably taking Time Warner's side because streaming cable channels to your computer is so obvious that everyone is shocked that the channels are objecting to it. The fact that Time Warner is doing it, and channels are trying to stop them, is pretty much where everyone stops reading.
Really, there should be no litigation here. Customers pay for cable service, which means channels are streamed to them; it should make no difference what type of computer they use to watch those channels.
iPhone and friends do provide a way to import a signing certificate. It just costs money to obtain the certs because the infrastructure needed to support those certs costs money.
...and there is some reason that only Apple's PKI is allowed on these devices? Is there a good reason why knowledgeable users should not be allowed to use a different PKI? Void the warranty if you choose to live outside of Apple's system; nothing wrong with that.
Therefore, the GPL effectively prevents any useful implementation of code signing.
Interesting conclusion you draw there. The GPLv3 only requires you to leave users with the option of living outside of a tightly controlled environment -- essentially, there needs to be some way for users to be able to run their own code. You only have to provide that option upon request; you do not have to make it as easy as "push this button to disable the signature check!" You can warn users that by importing a third party certificate, they are taking a risk with their security. Security and code signing is not equivalent to, "only Apple gets to sign code."
somebody is going to find a way to throw that switch without the user's knowledge.
Not if the switch requires the user to go to an Apple store in person. The switch does not need to be available in software, and even if it is, if prior to throwing the switch the user cannot run unsigned code, it is substantially hard for the attacker to activate it.
I've thought about this a lot, and I really don't see a better way that retains even a modicum of security from code signing.
Apparently your mode of thought treats all of Apple's customers as if they were all part of one big organization, with Apple acting as the IT department for that organization. This is a skewed view of the world; Apple's customers are not one big family, and Apple is not acting out of a desire to best serve their customers' collective interests.
Either way, the point is that it isn't a case of wanting control for control's sake, but rather wanting control to prevent harm to the users.
No, it is a case of control for profit's sake. Apple wants to get a cut of any money that anyone makes selling software for the iPad; they are not content to just make money selling iPads. This is not the first time we have seen this view of computing; it has been the standard for video game consoles for many years. As far back as the 1960s, companies were talking about computation as a utility, planning systems where users would be billing by the CPU-minute and the megabyte of RAM; that was not control for control's sake either, it was control for the sake of profit.
I realize that the notion of absolute freedom meaning absolute anarchy seems like a good idea to some folks, but most folks would prefer at least some limits to protect them from harm.
Which is exactly why importing a third part certificate, or running unsigned code, should be something that is optional and only made available to those users who ask for it. Security does not have to be draconian; Linux distros have been signing their packages for a long time, and there are plenty of users who just stick to their distro's official repositories and never import other certificates or run unsigned code. There is no reason why Apple could not follow a similar approach: sign all code, but allow users to opt out of the security that signing provides. Like I said, it would be perfectly reasonable to require people to show up at the Apple store, have someone warn them about security, agree to void their warranty, and have their tablet unlocked or enable some third party certificate.
Interesting, because when last I checked, the relevant sections of law -- copyright and patent -- were neither property laws nor laws protecting natural rights. Then again, IANAL.
Version 3 which is anti-commercial
Completely false; go check with Red Hat if you think GPLv3 is anti-commercial.
Is RMS completely ignorant of the fact that many of these projects received a lot of commercial support in the past and that Version 3 is basically a big middle finger directed at them? Is he out to destroy the FOSS movement by alienating some of its largest contributors?
No, RMS is neither ignorant nor is the GPLv3 harmful to business. The Free Software Foundation has always had the protection of user freedoms as its primary goal, and while the GPLv2 served that goal for a long time, there were certain loopholes that some companies were exploiting -- loopholes that the GPLv3 corrected. Having the right to study and modify code is meaningless if your computer prevents you from running your modified code, and software patents could be used to deny you the right to redistribute your code.
Corporations are as able as they have ever been to sell, sponsor, and benefit from software licensed under GPLv3; little has changed since the GPLv2. Corporations that seek to maliciously exploit loopholes (TiVo) may have a problem, but most corporations do not do that and will be largely unaffected by the move. If you are granting your users the freedoms the GPL is designed to protect, then GPLv3 is not a problem.
It's not just control freaks that have a problem with it. It's also security-conscious engineering teams. Those bits of GPLv3 betray a fundamental lack of understanding of the need for proper code signing.
This is a red herring; the GPLv3 does not prevent you from using code signing, it just requires you to allow those users who want to run their own code to do so. Imagine if your phone had an option that allowed you to import a different a signing certificate, or to disable signature checks, so that those users who wanted to run their own code were able to do so. You could warn people that it might be harmful to security, you could void their warranties, but why is it necessary to tell someone that they are not allowed to modify their own device?
It is not a question of security, it is a question of control.
Are we going to imprison the people who decided to use Windows as the operating system for a critical, safety-sensitive computer? Why are we acting like the problems here end with this guy? Computers are not some magical object that dark wizards vie for control over; the fact that this guy could have endangered hospital patients because he was interacting with the HVAC computer (and ultimately, that is what he was doing: interacting with the computer) says more about the problems with the HVAC controller than about the hacker.
On the other hand, why was the HVAC system left open to these sorts of attacks? If it is as safety sensitive and critical as the FBI is claiming, one would think that Windows should be low on the list of operating systems to choose.
Today he'd haul you in until your identity could be confirmed. His probable cause is based on who he thinks you are.
Am I the only person who thinks that the solution to that problem is to raise the standard for probable cause, rather than making it easier for the police to check fingerprints?
This could be wonderful for helping local police capture criminals more quickly who are on the run from another jurisdiction. The "Big Brother" aspect of this having the potential to lead to a database of biometric information on EVERYONE is frightening.
These two aspects are more closely related than you make them seem. There would be no problem with surveillance if we could trust the government not to pass Orwellian laws. You say that making the jobs of local police forces easier is a good thing? What happens when it comes time to enforce a law that prohibits you from voicing a particular political stance (such as communism)? You won't want their job to be easier then.
A common argument made by law enforcement is the "limited resources" argument: even if they could technically arrest anyone, they do not have those sorts of resources, and therefore they will only go after people worth arresting. Such an argument becomes pretty difficult to make when you start talking about technology that enables the police to do more with less. If the job of two officers can now be done by one, then police resources have become less limited, and we should expect to see even more people arrested. Suddenly, those laws we passed years ago and said, "well, they will only go after the people who really matter!" have the potential to come back to haunt us.
We already imprison more people than any other country; why are we talking about making it easier for the police to arrest people? I would count "making it easier for the police to arrest people" as a negative, not a positive, until we undertake a monumental effort of legal reform to reduce the number of things people can be arrested for.
This is not a lease. The phone was purchased, plain and simple, and if the carrier wants to give you a discount, to the point of the phone being sold at a loss, then you still purchased it.
No, I do not stop complaining, despite my own views on homosexuality and the gay movement. The way Apple censors programs on the iPad is a bad thing, regardless of whether or not I personally agree with the premise of the programs that are being censored.
Funny, I thought authors did it for the love of the art form, and musicians, and every other type of artist. When last I checked, the majority of authors made very little from the books they write.
Really though, where are the people complaining about how the mass availability of cameras has reduced the monetary incentive to paint images of the world? Where is the call to ban cameras, or to restrict how cameras may be used?
As a final note, copyright is a fairly recent development in humanity's legal history. For millennia, we got on just fine without it, and in fact some of history's most important works were created before any copyright systems existed. It was copying that led to the preservation of some of those works.
Now, am I a fan of throwing out copyrights all together? Not really, to be honest; copyright has its place, and can be used to protect the free flow of information (copyleft). What we need to do is scale back copyright terms, all the way back to the original length of 14 years.
Slashdot Mirror
next version, we get rid of pointers :P
You say this like it is a joke, but it is actually not so terrible an idea. Dumb pointers introduce a whole host of ways for things to go wrong, and in a lot of cases the compiler is in a position to create a pointer for you (see: Common Lisp) when it is truly what you need. If you are writing code at the lowest level, then sure, you need pointers and it would be stupid not to use them; but a lot of projects out there do not need the sort of low level control over pointers that you wind up dealing with when you write C++ code.
I can't understand why anyone would start a new project in C++. The only reason anyone could seriously consider it is if they have to maintain or enhance an existing C++ code base.
While I agree with the sentiment, the reason people keep using C++ is pretty clear: habit. Everyone knows C++ or some very similar language, there is a lot of legacy C++ code, and people are still being trained to write code in C++ (or some similar language). If you were starting a new project and you said, "Let's use Scheme," half the people in the meeting would say they had never written a single Scheme program, and that they did not have time to learn it, and that it would be easier to just use C++ (or something similar). You could list every technical advantage that Scheme has over C++ and it would not make any difference.
Do not underestimate the friction that you face when it comes to C++ and similar languages.
The problem is that C++ dominates the industry; people who want to use a different language, even when they can list every good reason for making that decision, often find that the pressure to stick with C++ is too much to overcome. A lot of people simply refuse to hear the arguments about better languages for a particular project, or are afraid of being unable to find programmers who know something that is not C++ (or similar enough, like Java), or do not want to have to write interfaces for an existing C++ codebase.
I would say that Common Lisp, with a decent compiler and some attention paid to code efficiency, is most certainly a contender on all of the above. The main selling point of C++ now is habit -- large amounts of existing C++ code and lots of programmers who were trained to use C++.
Do not rely on undefined behavior, and these things will not be a problem.
The Church-Turing thesis unifies different models of computation; while Wikipedia tells you that this means that any computational process can be implemented with a Turing Machine, it is equally true that any computational process can be expressed using lambda calculus, combinator logic, mu-recursion, and a variety of other theoretical models of computation.
As for algorithmic analysis...it is not terribly hard to consider in terms of the lambda calculus, although that is less common in algorithms textbooks or CS courses. It is not generally the case that algorithms which are easily understood in terms of Turing machines are harder to analyze in terms of lambda caluclus, nor is it generally the case that algorithms expressed in imperative languages are more efficient or easier to analyze than those expressed in functional languages.
To describe it as "getting in the way" is a gross misrepresentation of the theory of computation and the theory of algorithms. Functional programming is not inefficient, certainly not in terms of the asymptotic performance of functional code. Functional programming languages have seen quite a bit of use "in the real world," including environments where efficiency (not necessarily raw performance, but efficiency in terms of computational complexity) matters.
Also, as a side note, the Wikipedia article's assertion about false lower bounds is misleading. There are a variety of different delays associated with memory access, which may be address dependent (e.g. seek times on hard drives if you are using swap, cache misses, etc.). If these costs are polynomial in the amount of memory used by the program (assuming some sort of locality effects), then the algorithm will still be a polynomial time algorithm.
Frankly, most programmers deal with some level of abstraction between them and the actual machine. In fact, this usually results in faster code than if the programmer were to spend all of his time hand tuning assembly language code; compilers are generally better at producing efficient code than any individual programmer, with a couple exceptions (instructions that are difficult for compilers to take advantage of, for example, but these do not represent the majority of use cases). Some abstraction must be fixed before an algorithm can be analyzed; a Turing machine is a convenient abstraction, but it is not the only such abstraction.
Computer science is about the theory behind computers
Not to nitpick, but this should really be stated as "the theory behind computation." The theory behind (electronic) computers is what computer and electrical engineers study; everything a computer scientist does applies equally to someone working with a pen and paper (in theory, anyway). It is rather unfortunate that so many CS curricula treat theory as some sort of an afterthought, instead of the primary focus.
See: Church-Turing thesis.
the real world
People go to vocational schools to be trained for real world jobs. CMU is not a vocational school.
People are probably taking Time Warner's side because streaming cable channels to your computer is so obvious that everyone is shocked that the channels are objecting to it. The fact that Time Warner is doing it, and channels are trying to stop them, is pretty much where everyone stops reading.
Really, there should be no litigation here. Customers pay for cable service, which means channels are streamed to them; it should make no difference what type of computer they use to watch those channels.
They want you to pay separately for streaming their channel to your iPad. They count anything that is not "expanding revenue streams" as a loss.
iPhone and friends do provide a way to import a signing certificate. It just costs money to obtain the certs because the infrastructure needed to support those certs costs money.
Therefore, the GPL effectively prevents any useful implementation of code signing.
Interesting conclusion you draw there. The GPLv3 only requires you to leave users with the option of living outside of a tightly controlled environment -- essentially, there needs to be some way for users to be able to run their own code. You only have to provide that option upon request; you do not have to make it as easy as "push this button to disable the signature check!" You can warn users that by importing a third party certificate, they are taking a risk with their security. Security and code signing is not equivalent to, "only Apple gets to sign code."
somebody is going to find a way to throw that switch without the user's knowledge.
Not if the switch requires the user to go to an Apple store in person. The switch does not need to be available in software, and even if it is, if prior to throwing the switch the user cannot run unsigned code, it is substantially hard for the attacker to activate it.
I've thought about this a lot, and I really don't see a better way that retains even a modicum of security from code signing.
Apparently your mode of thought treats all of Apple's customers as if they were all part of one big organization, with Apple acting as the IT department for that organization. This is a skewed view of the world; Apple's customers are not one big family, and Apple is not acting out of a desire to best serve their customers' collective interests.
Either way, the point is that it isn't a case of wanting control for control's sake, but rather wanting control to prevent harm to the users.
No, it is a case of control for profit's sake. Apple wants to get a cut of any money that anyone makes selling software for the iPad; they are not content to just make money selling iPads. This is not the first time we have seen this view of computing; it has been the standard for video game consoles for many years. As far back as the 1960s, companies were talking about computation as a utility, planning systems where users would be billing by the CPU-minute and the megabyte of RAM; that was not control for control's sake either, it was control for the sake of profit.
I realize that the notion of absolute freedom meaning absolute anarchy seems like a good idea to some folks, but most folks would prefer at least some limits to protect them from harm.
Which is exactly why importing a third part certificate, or running unsigned code, should be something that is optional and only made available to those users who ask for it. Security does not have to be draconian; Linux distros have been signing their packages for a long time, and there are plenty of users who just stick to their distro's official repositories and never import other certificates or run unsigned code. There is no reason why Apple could not follow a similar approach: sign all code, but allow users to opt out of the security that signing provides. Like I said, it would be perfectly reasonable to require people to show up at the Apple store, have someone warn them about security, agree to void their warranty, and have their tablet unlocked or enable some third party certificate.
Code is property
Interesting, because when last I checked, the relevant sections of law -- copyright and patent -- were neither property laws nor laws protecting natural rights. Then again, IANAL.
Version 3 which is anti-commercial
Completely false; go check with Red Hat if you think GPLv3 is anti-commercial.
Is RMS completely ignorant of the fact that many of these projects received a lot of commercial support in the past and that Version 3 is basically a big middle finger directed at them? Is he out to destroy the FOSS movement by alienating some of its largest contributors?
No, RMS is neither ignorant nor is the GPLv3 harmful to business. The Free Software Foundation has always had the protection of user freedoms as its primary goal, and while the GPLv2 served that goal for a long time, there were certain loopholes that some companies were exploiting -- loopholes that the GPLv3 corrected. Having the right to study and modify code is meaningless if your computer prevents you from running your modified code, and software patents could be used to deny you the right to redistribute your code.
Corporations are as able as they have ever been to sell, sponsor, and benefit from software licensed under GPLv3; little has changed since the GPLv2. Corporations that seek to maliciously exploit loopholes (TiVo) may have a problem, but most corporations do not do that and will be largely unaffected by the move. If you are granting your users the freedoms the GPL is designed to protect, then GPLv3 is not a problem.
It's not just control freaks that have a problem with it. It's also security-conscious engineering teams. Those bits of GPLv3 betray a fundamental lack of understanding of the need for proper code signing.
This is a red herring; the GPLv3 does not prevent you from using code signing, it just requires you to allow those users who want to run their own code to do so. Imagine if your phone had an option that allowed you to import a different a signing certificate, or to disable signature checks, so that those users who wanted to run their own code were able to do so. You could warn people that it might be harmful to security, you could void their warranties, but why is it necessary to tell someone that they are not allowed to modify their own device?
It is not a question of security, it is a question of control.
Is anyone actually surprised by this?
Are we going to imprison the people who decided to use Windows as the operating system for a critical, safety-sensitive computer? Why are we acting like the problems here end with this guy? Computers are not some magical object that dark wizards vie for control over; the fact that this guy could have endangered hospital patients because he was interacting with the HVAC computer (and ultimately, that is what he was doing: interacting with the computer) says more about the problems with the HVAC controller than about the hacker.
On the other hand, why was the HVAC system left open to these sorts of attacks? If it is as safety sensitive and critical as the FBI is claiming, one would think that Windows should be low on the list of operating systems to choose.
Today he'd haul you in until your identity could be confirmed. His probable cause is based on who he thinks you are.
Am I the only person who thinks that the solution to that problem is to raise the standard for probable cause, rather than making it easier for the police to check fingerprints?
Note that the system is also going to check handwriting. Perhaps you could carry around a printed card that says you intend to exercise your rights?
I have the right to remain silent, how are they going to identify me by my voice?
Probably around the same time you inform them that you intend to exercise that right:
http://www.msnbc.msn.com/id/37448356/ns/us_news-crime_and_courts/
This could be wonderful for helping local police capture criminals more quickly who are on the run from another jurisdiction. The "Big Brother" aspect of this having the potential to lead to a database of biometric information on EVERYONE is frightening.
These two aspects are more closely related than you make them seem. There would be no problem with surveillance if we could trust the government not to pass Orwellian laws. You say that making the jobs of local police forces easier is a good thing? What happens when it comes time to enforce a law that prohibits you from voicing a particular political stance (such as communism)? You won't want their job to be easier then.
A common argument made by law enforcement is the "limited resources" argument: even if they could technically arrest anyone, they do not have those sorts of resources, and therefore they will only go after people worth arresting. Such an argument becomes pretty difficult to make when you start talking about technology that enables the police to do more with less. If the job of two officers can now be done by one, then police resources have become less limited, and we should expect to see even more people arrested. Suddenly, those laws we passed years ago and said, "well, they will only go after the people who really matter!" have the potential to come back to haunt us.
We already imprison more people than any other country; why are we talking about making it easier for the police to arrest people? I would count "making it easier for the police to arrest people" as a negative, not a positive, until we undertake a monumental effort of legal reform to reduce the number of things people can be arrested for.
So what? What if I wanted to publish a program that encouraged people to believe that the Earth is flat; should that be banned as well?
What exactly makes it acceptable for Apple to engage in this sort of censorship?
This is not a lease. The phone was purchased, plain and simple, and if the carrier wants to give you a discount, to the point of the phone being sold at a loss, then you still purchased it.
No, I do not stop complaining, despite my own views on homosexuality and the gay movement. The way Apple censors programs on the iPad is a bad thing, regardless of whether or not I personally agree with the premise of the programs that are being censored.
Funny, I thought authors did it for the love of the art form, and musicians, and every other type of artist. When last I checked, the majority of authors made very little from the books they write.
Really though, where are the people complaining about how the mass availability of cameras has reduced the monetary incentive to paint images of the world? Where is the call to ban cameras, or to restrict how cameras may be used?
As a final note, copyright is a fairly recent development in humanity's legal history. For millennia, we got on just fine without it, and in fact some of history's most important works were created before any copyright systems existed. It was copying that led to the preservation of some of those works.
Now, am I a fan of throwing out copyrights all together? Not really, to be honest; copyright has its place, and can be used to protect the free flow of information (copyleft). What we need to do is scale back copyright terms, all the way back to the original length of 14 years.