The Schulze method is as complicated as it is precisely because it provides a good way of breaking this kind of ties. Every Condorcet method has the property of finding the Condorcet winner if there is one, and determining that is actually very easy: just do the pairwise comparison.
The reason there are many different Condorcet methods is exactly the question of what you do in the face of ties. The Schulze method is merely one method of breaking such ties.
First off, the biggest obstacle to American success is China and their unfair trade practices. By keeping their currency pegged to the US dollar at artificially low rate they are creating trade barriers to real free trade.
Try thinking about this situation in real terms, i.e. comparing the resources available to a country vs. the resources it is able to consume. Because the Chinese government keeps buying US$, the US is able to import cheap goods from China, which means that the US, as a nation, is able to consume more resources than it normally would be able to. If I were the US government, I would be happy about the Chinese policy!
It is true that this competition causes US manufacturing jobs to disappear. However, that's not a bad thing at all, because it means that those workers are freed up for other, more advanced and useful tasks. They can help raise the living standards for US citizens even beyond what they are used to today.
Of course, actually making use of this opportunity would require a government that isn't stuck in neo-liberal economic thinking. For example, it would require a government that realizes that idle resources can simply be employed via higher deficit spending in programs that advance public purpose. In other words, it would require a government that understands Modern Monetary Theory.
In the meantime, those of us who do have an unobstructed view on the fundamentals on how monetary systems and the economy function are keeping up the facepalming in light of the supreme idiocy that comes out of our politicians on both sides of the Atlantic.
The hyperlink expiration is a valid point. I guess most hyperlinks from 10 years ago are no longer valid today, and many are being hijacked by domain squatters. So in that sense, the domain squatters of the future would be required to break public-key cryptographic systems, which is a more formidable barrier than what we have today.
Still, securing hyperlinks to critical sites (e.g. bank websites) is a serious issue. How many bookmarks from 10 years ago do you still have? As long as you visit them regularly enough, a roll-over system could be implemented where a replacement public key gets phased in over the course of some months to one year.
I don't think people - as in, non-Slashdot visiting users - care one bit for hierarchical namespaces. The only thing that the system has to offer them is
1. The ability to click on links on websites and have them work.
2. The ability to type in an URL that they saw in an advertisment, magazine, or other off-line source.
Note that neither of these tasks requires a hierarchical namespace, and only the second task requires a centrally controlled namespace. On top of that, that second ability may be becoming increasingly irrelevant, giving search engines, 2D barcodes, etc.
There is, in fact, a useful way forward that would be any crypto-nerd's dream by reducing the importance of DNS and ridding the world of the mess that is SSL chains of trust at the same time. Forget about a decentrally controlled name space (that is totally unrealistic), but encourage people to use public-key based links. The idea is that - except for task 2 above - we would use hyperlinks that contain public key hashes (or even entire public keys). A decentralized lookup system would map public key hashes to IPs, and upon connection to the given IP, your browser/mail-client/whatever could directly verify whether the server was trusted, because you could verify whether it owns the correct private key.
Such a system cannot solve task 2 above, but a world with the outlined public key based server lookup, combined with a mostly vestigial centrally controlled namespace system that maps names to public keys could work, and it could even work pretty well.
The only problem is that I can see no natural path that takes us from where we are today towards that ideal world. Oh well.
This is the first time I've seen something like this in reality. I am in awe. (And yes, I realize that the two scenarios are not exactly the same, but still...)
Please explain why you support the government wasting over 3 trillion dollars of borrowed Chinese money, and then please explain why people opposed to that make you so angry.
No need, as I don't believe the money is wholly wasted. What makes me angry is the 4-7 trillion dollars we've spent and have accrued liability for with pointless boondoggles in Iraq and Afghanistan, and the people who blindly support those "wars". But that's beside the point.
I agree with that, though I would also add that the expression "3 trillion dollars of borrowed Chinese money" shows a gross misunderstanding of how the US monetary system works. You may be interested in an economic approach to understanding fiat money called the Modern Monetary Theory.
The fact of the matter is that the Chinese earned US$ by selling goods to the US. This money ended up in an account at the central bank (the Fed). This account can be thought of as a checking account. The act of buying US treasury bonds is equivalent to shifting that money from a checking account into an interest earning savings account.
In other words, it is not the US government that spends money borrowed from the Chinese. Rather, the Chinese are saving money that they earned from the US, money that comes from US federal spending.
It is important to understand the distinction. For a first reading, I recommend The 7 Deadly Innocent Frauds by Warren Mosler. Also, billy blog by Australian economist Bill Mitchell makes for interesting reading.
Poor people currently pay either no taxes at all or very low taxes as proportion of their income.
Nonsense. Poor people pay no income tax. There are lots of other types of taxes out there.
What the distribution of the tax burden ultimately ends up being is another question, and I do not have concrete data about the US. But to claim that poor people pay no taxes is just plain wrong and distorts the discussion.
What really killed C++ for me was when a student created a situation like this:
std::string somefunction(){}
As others have pointed out, most compilers give at least a warning about this or even fail to compile, which is well within what the standard allows. And it is quite reasonable for the C++ standard to make this an undefined behaviour, because it is impossible in general for the compiler decide whether the flow of control can leave a function via the closing brace. If a compiler could decide that, then it could decide problems like the halting problem that are known to be undecidable (in a computability sense).
Now of course, the standard could instead have defined the correct behaviour of the compiler to be that if it cannot decide this, it should introduce code that throws an exception. But that would be introducing code that the programmer has not asked for, e.g. if the programmer has proven that control flow cannot fall off the end of the function, but the compiler is unable to follow that proof. That goes against the philosophy of C++ as a systems programming language.
Whether anybody should write code where the compiler is unable to figure out stuff like this is another question, of course, and obviously compilers should do a best effort to catch the obvious cases - but they already do that. So this is a pretty lame argument against C++ as a language.
Things like give me an example where a struct would be a better choice and where a class would be better.
Few people get that one right. The only difference between struct in class in C++ is that struct defaults to public, whereas class defaults to private. In other words, the choice between struct and class is purely cosmetic and can only depend on your coding guidelines.
The only case where it does matter is if you want to mix C and C++. Then any data structure that should be shared between the two obviously has to be a struct, because a C compiler does not recognize the keyword class.
You're confusing mathematics with computations on a calculator. On a pocket calculator, you really cannot compute with pi, because it only computes with limited precision floating point numbers. On the other hand, using a computer algebra system which does symbolic computation, you can even do computations with pi.
Also, you can quite easily do mathematics with pi. How else do you think people prove things like sin(pi) = 0?
Mathematics is not the same as doing computations. In fact, the two things have been different at least since Euclid published his Elements. The relationship between the two things is that some parts of Mathematics provide the foundation for doing Computations, and many (most?) parts of mathematics make heavy use of symbolic computations.
This obviously depends on what you mean by "exist".
Is the natural number 1 an object that physically exists in our world? Clearly the answer to that is "No". There are no "natural number particles" in physics. Yet there can be collections of objects, and they can be counted, and that is one way to represent natural numbers in the real world.
Perhaps the first breakthrough in mathematics (think 4000 or more years ago) was the discovery that natural numbers can also be represented in other, more abstract ways, such as by using place-value systems like our decimal system.
Still, to this day, nobody has ever observed an actual natural number in the physical world. All that is being observed are representations of that number, or collections of the size of that natural number, etc.
I would also add the following to what you wrote - it's not just innovation, but also demand that is key: The economic "experts" in the media are claiming all the time that government must make it easier for companies to get loans so that those companies can then invest. Well guess what? Companies on average don't want to invest in the current economic climate because there is not enough demand for their products and services.
Which sounds good, until you realize that there are times when deficit spending is legitimate and necessary for the good of all those that are concerned. It's just when you start wasting money on things like pointless wars and tax breaks for the rich that you start to run into trouble.
I would go even further: there is significant evidence that on average, deficit spending is a requirement for a well functioning economy in a fiat money system - where by well functioning I mean that there are no resources left to idle, in particular it means there is no waste in the form of idle production capacities and involuntary unemployment.
The key is to understand the basic mechanisms that underlie a modern monetary system, as that will forever change your understanding of the function of federal government debt. I recommend the book by economist Warren Mosler 7 Deadly Innocent Frauds for a start, it's not too long a read and really challenges your understanding of money in a good way. Once you're through with that, I recommend the blog of Australian economist Bill Mitchell, billy blog.
What is and is not money is all up to what people perceive. Money is money because you can spend it. Whatever construct people are willing to use for trade is money, doesn't matter how theoretical it is or if nothing backs it. So they have a system in their heads that has nothing to do with reality.
Modern Monetary Theory actually offers an alternative theory: Money has value because government forces people to pay taxes, thus creating a demand for the money issued by government. At least to me that makes much more sense than the fuzzy circular of reasoning that is "It has value because people say it has value". I suggest you google for "7 Deadly Innocent Frauds" by Warren Mosler, or for billyblog by Bill Mitchell, two economists who seem to "get it".
I won't waste my time trying to explain this on Slashdot, but if you're interested in some eye-opening economics lesson, I suggest you read the book 7 Deadly Innocent Frauds by US economist Warren Mosler. Once you're through that, I recommend billyblog, the blog of Australian economist Bill Mitchell, one of the most prolific proponents of Modern Monetary Theory.
Just to give you an idea of what you will find: Basically, the current economic mainstream hasn't really understood that we no longer have the gold standard. It is implicitly based on false assumptions that are easily rebuked given a proper understanding of the functional workings of a fiat currency.
Greetings from Europe. Your congress is neither moderate nor left-wing. It is also not socially liberal, which would be the more precise term for what you want to say.
To substantiate, I recommend you take a look at politicalcompass.org to learn about the political coordinate system in a more objective light.
Perhaps the most powerful idea of computer science is that computation is computation is computation. This should be obvious to anybody already working in IT, but to outsiders it is a total mystery.
The concept of an algorithm with input and output, where the input can be anything from a bunch of numbers, to a book, to an image, etc. is extremely powerful. Of course, along with it you need to understand that ultimately, any kind of input ends up being represented as a bunch of numbers or as a bit string anyway.
Whether you are manipulating images, or adding numbers, or serving websites, you always have the same theory of computation underlying it. This idea is so fundamental that we inside the profession usually don't even realize that this is a major revelation to anybody not accustomed to it. As a consequence, it's the kind of thing that really needs to be taught in schools.
To the best of my knowledge, yes. The whole term "elliptic curve cryptography" is quite misleading. All it means is that the groups you are using come from elliptic curves instead of from (e.g.) modular arithmetic. The attacks against discrete logarithms are, as far as I know, generic in the sense that the only requirement is that you implement the group multiplication in your quantum computer. So any kind of discrete logarithm-based cryptography can be broken using quantum computers, independent of the underlying group.
Do you have a source on the claim of mutually-infertile couples? I'm genuinely curious, and I would have assumed that humans are mobile enough and intermix enough to avoid that.
If humanity were to split into multiple species before the founding of the Federation of Planets, we'd be in pretty deep trouble. Luckily, if it ever happens, it will be long after our time. Still, it's a fascinating - if somewhat disturbing - line of thought.
Should the Vatican publish the minutes of their discussion in electing a Pope?
Yes. I'm not Catholic, but the Pope still has an enormous amount of influence on the world, and he is not democratically legitimated. All good reasons to keep a watchful eye.
Should the admissions board at Harvard release the logic behind their rejections?
Yes. Not that I care, personally, but it's important to keep high profile organizations accountable.
How about the hockey team - should their initiation ceremony get plastered on the web?
Personally, I don't care and I wouldn't push for it. However, there are certainly reasons to do so (ignoring the hyperbolical "plastered"). Initiation rituals frequently make the press for including extremely violent and abusive practices, so it's not necessarily a bad idea to take some preventive measures.
Note that all your examples are related to organizations. Organizations don't (shouldn't) have a right to privacy.
Privacy of individuals is very different from privacy of organizations. Individuals are humans whose dignity must be preserved. Organizations, on the other hand, are not (as much as many of them would like to think they are). Accordingly, different standards must be applied to different situations.
How do you think such specific legislation comes into being? Through political lobbying. Who do you think can do such political lobbying? That's right, unions.
I realize that many people in IT-related fields especially in the US have some subconscious aversion against unions, but maybe it's time for you to become realistic...
Excuse me, but driving people to work that many hours per week is always exploitation. I would be more inclined to agree with you if what we were talking about was below-McDonald's wages while working regular 40hrs work weeks.
Continuous crunch periods and excessive overtime as one apparently sees in the games industry is exploitation, end of the story. If you can't manage to get the project done without this exploitation, maybe you should hire more people (probably at lower wages, depending on the job market).
Slashdot Mirror
The Schulze method is as complicated as it is precisely because it provides a good way of breaking this kind of ties. Every Condorcet method has the property of finding the Condorcet winner if there is one, and determining that is actually very easy: just do the pairwise comparison. The reason there are many different Condorcet methods is exactly the question of what you do in the face of ties. The Schulze method is merely one method of breaking such ties.
Here are my thoughts on the matter.
First off, the biggest obstacle to American success is China and their unfair trade practices. By keeping their currency pegged to the US dollar at artificially low rate they are creating trade barriers to real free trade.
Try thinking about this situation in real terms, i.e. comparing the resources available to a country vs. the resources it is able to consume. Because the Chinese government keeps buying US$, the US is able to import cheap goods from China, which means that the US, as a nation, is able to consume more resources than it normally would be able to. If I were the US government, I would be happy about the Chinese policy!
It is true that this competition causes US manufacturing jobs to disappear. However, that's not a bad thing at all, because it means that those workers are freed up for other, more advanced and useful tasks. They can help raise the living standards for US citizens even beyond what they are used to today.
Of course, actually making use of this opportunity would require a government that isn't stuck in neo-liberal economic thinking. For example, it would require a government that realizes that idle resources can simply be employed via higher deficit spending in programs that advance public purpose. In other words, it would require a government that understands Modern Monetary Theory.
In the meantime, those of us who do have an unobstructed view on the fundamentals on how monetary systems and the economy function are keeping up the facepalming in light of the supreme idiocy that comes out of our politicians on both sides of the Atlantic.
The hyperlink expiration is a valid point. I guess most hyperlinks from 10 years ago are no longer valid today, and many are being hijacked by domain squatters. So in that sense, the domain squatters of the future would be required to break public-key cryptographic systems, which is a more formidable barrier than what we have today.
Still, securing hyperlinks to critical sites (e.g. bank websites) is a serious issue. How many bookmarks from 10 years ago do you still have? As long as you visit them regularly enough, a roll-over system could be implemented where a replacement public key gets phased in over the course of some months to one year.
I don't think people - as in, non-Slashdot visiting users - care one bit for hierarchical namespaces. The only thing that the system has to offer them is
1. The ability to click on links on websites and have them work.
2. The ability to type in an URL that they saw in an advertisment, magazine, or other off-line source.
Note that neither of these tasks requires a hierarchical namespace, and only the second task requires a centrally controlled namespace. On top of that, that second ability may be becoming increasingly irrelevant, giving search engines, 2D barcodes, etc.
There is, in fact, a useful way forward that would be any crypto-nerd's dream by reducing the importance of DNS and ridding the world of the mess that is SSL chains of trust at the same time. Forget about a decentrally controlled name space (that is totally unrealistic), but encourage people to use public-key based links. The idea is that - except for task 2 above - we would use hyperlinks that contain public key hashes (or even entire public keys). A decentralized lookup system would map public key hashes to IPs, and upon connection to the given IP, your browser/mail-client/whatever could directly verify whether the server was trusted, because you could verify whether it owns the correct private key.
Such a system cannot solve task 2 above, but a world with the outlined public key based server lookup, combined with a mostly vestigial centrally controlled namespace system that maps names to public keys could work, and it could even work pretty well.
The only problem is that I can see no natural path that takes us from where we are today towards that ideal world. Oh well.
This is the first time I've seen something like this in reality. I am in awe. (And yes, I realize that the two scenarios are not exactly the same, but still...)
No need, as I don't believe the money is wholly wasted. What makes me angry is the 4-7 trillion dollars we've spent and have accrued liability for with pointless boondoggles in Iraq and Afghanistan, and the people who blindly support those "wars". But that's beside the point.
I agree with that, though I would also add that the expression "3 trillion dollars of borrowed Chinese money" shows a gross misunderstanding of how the US monetary system works. You may be interested in an economic approach to understanding fiat money called the Modern Monetary Theory.
The fact of the matter is that the Chinese earned US$ by selling goods to the US. This money ended up in an account at the central bank (the Fed). This account can be thought of as a checking account. The act of buying US treasury bonds is equivalent to shifting that money from a checking account into an interest earning savings account.
In other words, it is not the US government that spends money borrowed from the Chinese. Rather, the Chinese are saving money that they earned from the US, money that comes from US federal spending.
It is important to understand the distinction. For a first reading, I recommend The 7 Deadly Innocent Frauds by Warren Mosler. Also, billy blog by Australian economist Bill Mitchell makes for interesting reading.
Poor people currently pay either no taxes at all or very low taxes as proportion of their income.
Nonsense. Poor people pay no income tax. There are lots of other types of taxes out there.
What the distribution of the tax burden ultimately ends up being is another question, and I do not have concrete data about the US. But to claim that poor people pay no taxes is just plain wrong and distorts the discussion.
What really killed C++ for me was when a student created a situation like this:
As others have pointed out, most compilers give at least a warning about this or even fail to compile, which is well within what the standard allows. And it is quite reasonable for the C++ standard to make this an undefined behaviour, because it is impossible in general for the compiler decide whether the flow of control can leave a function via the closing brace. If a compiler could decide that, then it could decide problems like the halting problem that are known to be undecidable (in a computability sense).
Now of course, the standard could instead have defined the correct behaviour of the compiler to be that if it cannot decide this, it should introduce code that throws an exception. But that would be introducing code that the programmer has not asked for, e.g. if the programmer has proven that control flow cannot fall off the end of the function, but the compiler is unable to follow that proof. That goes against the philosophy of C++ as a systems programming language.
Whether anybody should write code where the compiler is unable to figure out stuff like this is another question, of course, and obviously compilers should do a best effort to catch the obvious cases - but they already do that. So this is a pretty lame argument against C++ as a language.
Things like give me an example where a struct would be a better choice and where a class would be better.
Few people get that one right. The only difference between struct in class in C++ is that struct defaults to public, whereas class defaults to private. In other words, the choice between struct and class is purely cosmetic and can only depend on your coding guidelines.
The only case where it does matter is if you want to mix C and C++. Then any data structure that should be shared between the two obviously has to be a struct, because a C compiler does not recognize the keyword class.
You're confusing mathematics with computations on a calculator. On a pocket calculator, you really cannot compute with pi, because it only computes with limited precision floating point numbers. On the other hand, using a computer algebra system which does symbolic computation, you can even do computations with pi.
Also, you can quite easily do mathematics with pi. How else do you think people prove things like sin(pi) = 0?
Mathematics is not the same as doing computations. In fact, the two things have been different at least since Euclid published his Elements. The relationship between the two things is that some parts of Mathematics provide the foundation for doing Computations, and many (most?) parts of mathematics make heavy use of symbolic computations.
Do natural numbers exist in the real world?
This obviously depends on what you mean by "exist".
Is the natural number 1 an object that physically exists in our world? Clearly the answer to that is "No". There are no "natural number particles" in physics. Yet there can be collections of objects, and they can be counted, and that is one way to represent natural numbers in the real world.
Perhaps the first breakthrough in mathematics (think 4000 or more years ago) was the discovery that natural numbers can also be represented in other, more abstract ways, such as by using place-value systems like our decimal system.
Still, to this day, nobody has ever observed an actual natural number in the physical world. All that is being observed are representations of that number, or collections of the size of that natural number, etc.
I'll try this approach the next time I have to explain that problem to somebody.
I wish I had mod points for you right now.
I would also add the following to what you wrote - it's not just innovation, but also demand that is key: The economic "experts" in the media are claiming all the time that government must make it easier for companies to get loans so that those companies can then invest. Well guess what? Companies on average don't want to invest in the current economic climate because there is not enough demand for their products and services.
Which sounds good, until you realize that there are times when deficit spending is legitimate and necessary for the good of all those that are concerned. It's just when you start wasting money on things like pointless wars and tax breaks for the rich that you start to run into trouble.
I would go even further: there is significant evidence that on average, deficit spending is a requirement for a well functioning economy in a fiat money system - where by well functioning I mean that there are no resources left to idle, in particular it means there is no waste in the form of idle production capacities and involuntary unemployment.
The key is to understand the basic mechanisms that underlie a modern monetary system, as that will forever change your understanding of the function of federal government debt. I recommend the book by economist Warren Mosler 7 Deadly Innocent Frauds for a start, it's not too long a read and really challenges your understanding of money in a good way. Once you're through with that, I recommend the blog of Australian economist Bill Mitchell, billy blog.
Actually, compulsory military service is being suspended in Germany right now.
What is and is not money is all up to what people perceive. Money is money because you can spend it. Whatever construct people are willing to use for trade is money, doesn't matter how theoretical it is or if nothing backs it. So they have a system in their heads that has nothing to do with reality.
Modern Monetary Theory actually offers an alternative theory: Money has value because government forces people to pay taxes, thus creating a demand for the money issued by government. At least to me that makes much more sense than the fuzzy circular of reasoning that is "It has value because people say it has value". I suggest you google for "7 Deadly Innocent Frauds" by Warren Mosler, or for billyblog by Bill Mitchell, two economists who seem to "get it".
That the nation's debt, is only getting bigger
This is not really a problem in itself.
I won't waste my time trying to explain this on Slashdot, but if you're interested in some eye-opening economics lesson, I suggest you read the book 7 Deadly Innocent Frauds by US economist Warren Mosler. Once you're through that, I recommend billyblog, the blog of Australian economist Bill Mitchell, one of the most prolific proponents of Modern Monetary Theory.
Just to give you an idea of what you will find: Basically, the current economic mainstream hasn't really understood that we no longer have the gold standard. It is implicitly based on false assumptions that are easily rebuked given a proper understanding of the functional workings of a fiat currency.
Greetings from Europe. Your congress is neither moderate nor left-wing. It is also not socially liberal, which would be the more precise term for what you want to say.
To substantiate, I recommend you take a look at politicalcompass.org to learn about the political coordinate system in a more objective light.
Perhaps the most powerful idea of computer science is that computation is computation is computation. This should be obvious to anybody already working in IT, but to outsiders it is a total mystery.
The concept of an algorithm with input and output, where the input can be anything from a bunch of numbers, to a book, to an image, etc. is extremely powerful. Of course, along with it you need to understand that ultimately, any kind of input ends up being represented as a bunch of numbers or as a bit string anyway.
Whether you are manipulating images, or adding numbers, or serving websites, you always have the same theory of computation underlying it. This idea is so fundamental that we inside the profession usually don't even realize that this is a major revelation to anybody not accustomed to it. As a consequence, it's the kind of thing that really needs to be taught in schools.
To the best of my knowledge, yes. The whole term "elliptic curve cryptography" is quite misleading. All it means is that the groups you are using come from elliptic curves instead of from (e.g.) modular arithmetic. The attacks against discrete logarithms are, as far as I know, generic in the sense that the only requirement is that you implement the group multiplication in your quantum computer. So any kind of discrete logarithm-based cryptography can be broken using quantum computers, independent of the underlying group.
It does not help that the media conditions us to think that "he doesn't drink" is equivalent to "he is an upstanding citizen."
Just tell them that Hitler was a teetotaler. I'm a teetotaler, too, by the way. Oh my God, I'm like Hitler! :P
Do you have a source on the claim of mutually-infertile couples? I'm genuinely curious, and I would have assumed that humans are mobile enough and intermix enough to avoid that.
If humanity were to split into multiple species before the founding of the Federation of Planets, we'd be in pretty deep trouble. Luckily, if it ever happens, it will be long after our time. Still, it's a fascinating - if somewhat disturbing - line of thought.
Should the Vatican publish the minutes of their discussion in electing a Pope?
Yes. I'm not Catholic, but the Pope still has an enormous amount of influence on the world, and he is not democratically legitimated. All good reasons to keep a watchful eye.
Should the admissions board at Harvard release the logic behind their rejections?
Yes. Not that I care, personally, but it's important to keep high profile organizations accountable.
How about the hockey team - should their initiation ceremony get plastered on the web?
Personally, I don't care and I wouldn't push for it. However, there are certainly reasons to do so (ignoring the hyperbolical "plastered"). Initiation rituals frequently make the press for including extremely violent and abusive practices, so it's not necessarily a bad idea to take some preventive measures.
Note that all your examples are related to organizations. Organizations don't (shouldn't) have a right to privacy.
Privacy of individuals is very different from privacy of organizations. Individuals are humans whose dignity must be preserved. Organizations, on the other hand, are not (as much as many of them would like to think they are). Accordingly, different standards must be applied to different situations.
How do you think such specific legislation comes into being? Through political lobbying. Who do you think can do such political lobbying? That's right, unions.
I realize that many people in IT-related fields especially in the US have some subconscious aversion against unions, but maybe it's time for you to become realistic...
Excuse me, but driving people to work that many hours per week is always exploitation. I would be more inclined to agree with you if what we were talking about was below-McDonald's wages while working regular 40hrs work weeks.
Continuous crunch periods and excessive overtime as one apparently sees in the games industry is exploitation, end of the story. If you can't manage to get the project done without this exploitation, maybe you should hire more people (probably at lower wages, depending on the job market).