People who have been Mac people for a long time generally don't have that workflow, as the importing of the BSD backend is a fairly recent addition to the Mac world, whereas many of the GUI conventions have been around much longer.
I use mainly Debian (on my desktops) and OS X (on my laptop). There are pros and cons to both, but by far the biggest con of the Mac and pro of Debian is that packing management actually exists and works.
I nearly fell out of my chair when I first saw that the official way to install apps on OS X is to mount a virtual drive image and then manually copy files out of it to somewhere on your hard drive (like the Applications folder), and then unmount the drive image. That's the best they could come up with?
I don't know about there, but in my state (Texas), the statewide Libertarian candidates actually want to abolish Texas's public school system, not just get the federal government out of it. They fundamentally oppose the levying of taxes to provide free universal public education, and believe schooling should be provided solely by private institutions, whether for-profit or charity. But public schooling funded by taxes is hardly something that started in 1980; the common school has been a feature of the U.S. essentially since its inception.
I mean before deposit insurance, that was more or less the system we had. But people didn't all gravitate towards full-reserve banks, despite actually bearing 100% of the risk if a bank went bust. I blame people being short-sighted: they'll go for the fractional-reserve bank if it doesn't seem that likely to fail in the immediate future, and pays better interest. Then when the banks actually do fail, especially a lot of them at once, people demand something be done about it. Which is why we have deposit insurance now, and even the last European holdouts against deposit insurance are instituting it after this round.
I guess I don't have that much confidence things wouldn't work out the same way in another go-around. The fractional-reserve gold banks would become more popular than the full-reserve ones, creating the same money-multiplier issues you describe. Then when some of them went bust, people would demand something be done to restore their savings.
I don't actually see how you can outlaw fractional-reserve banking in a free society, since it requires central power to prohibit it. Say everyone started using gold coins tomorrow. This is actually kind of inconvenient, especially for large sums, so people would start producing "accounts" in which you could store your coins, and you could transfer gold to/from people with tokens, physical or electronic. Since not everyone needs to redeem their gold at the same time, some of these transaction services might lend some of the money out again, not operating at full reserve. As long as you agree to that in the contract when you sign up for an account with them, that seems perfectly legitimate to me.
The cynical explanation is that most people everywhere are populist at heart, which leads to a mish-mash of policies depending on what's popular and who the bogeyman is today. The right- and left-wingers have both typically been better at painting vivid images of bogeymen (whether fat-cat bankers or hordes of immigrants or whatever), and foment a sense of Emergency that Something Has to Be Done. In those conditions, the people promoting principled ideologies and cautious approach are usually steamrollered over by hare-brained solutions to the Emergency.
There's a certain populist ring to the classical liberal ideology of a level playing field with some moderate safety net, anyone can succeed but nobody starves, etc., but it's all a positive feel-good sort of vibe. I guess you can win on that sort of sentiment (it was a lot of Obama's "hope" campaign), but it's trickier. The good bogeymen I can think of focus only on one aspect so are already owned by the right or left; e.g. the opposite of civil libertarianism is 1984-style authoritarianism, but the left is already playing that sort of thing up in opposition to e.g. Guantanamo Bay.
A class-based analysis also gives a sort of population reason for it, since it tends to associate classical liberalism, and the similar European market liberal parties, as the ideology of the merchant classes, i.e. the educated middle and upper-middle class.
The entrenched upper classes tend to be more interested in maintaining their current position and exploiting networks of existing power, so aren't very in favor of dynamic markets and low barriers to entry, market fairness, social safety nets, etc. The lower classes tend to be more interested in immediate improvements to their position, and uncovering misdeeds of the higher classes that are allegedly responsible for their poorer position. Sometimes they also vote for conservative parties for reasons of religiosity or tradition. Neither of those classes, though, lend themselves to classical- or market-liberal types of positions. And there aren't enough educated middle-class folks who really understand and believe in the tradition of Voltaire, John Stuart Mill, Thomas Jefferson, etc.
The Libertarian party supports some principles that, broadly speaking, Americans believe in. More or less, these are classical liberal principles, in the mold of Thomas Jefferson. However, few people support their particular hardline interpretation, which tends to emphasize the anarcho-capitalist aspects, play down Jeffersonian elements that don't fit into that (e.g. Jefferson's view that governments should restrict the power of large corporations), and make few exceptions for any reason. Abolishing free public education, for example, is not a popular position. Neither is privatizing the road system. Some for of social safety nets are also popular---people don't want them abused (e.g. the stereotypical "welfare queens"), but neither do they want them to be totally absent. People also want regulation of private enterprise when its activities can cause negative externalities, such as systemic risks to economies (like banks, where further deregulation, the Libertarian position, is currently extremely unpopular). I could go on for a while.
Now if someone started a political party with positions more similar to those of the editorial line of The Economist newsmagazine, I could see voting for them. That is, support free-market economies with regulation and/or costing of negative externalities (pollution, systemic risks, etc.), a moderate social safety net, and liberal positions on social and civil liberties issues.
I think people are more willing to accept DRM if it seems like you're actually getting some benefit in return: in return for being able to download your games anywhere, you have to "phone home" before playing them to authenticate your local copy. Most DRM seems like you're getting a hassle for no benefit in return, which is sort of insult-to-injury when you actually paid for the content legitimately.
After World War I, the Bayer trademarks in the U.S. were reincarnated in a new company, confusingly also called Bayer, owned by the American firm Sterling Drug. They attempted to enforce the trademarks, but a 1921 court ruling invalidated the "aspirin" mark on grounds that it had come to be used as a generic term for the class of drugs for too long to be recaptured.
I don't use Steam myself (since the Linux client is still supposedly "on the way"), so was misremembering complaints I had read. It's not that Steam ties your account to a particular machine, but that it requires you to "phone home" every time before playing games even if they've already been downloaded. There's an Offline Mode, but there seem to be a lot of complaints about how well it works, and worries that if Steam were ever shut down, your games would basically be useless since they're not playable without authentication.
The points he makes in the Gamasutra summary sound remarkably clueful for the co-founder of a semi-major media firm. He seems to essentially "get it", that when selling content you're in a market, and if you're failing to sell as much as you'd want, the best solution is to figure out how you're failing to succeed in the market rather than whining about pirates.
Basically:
1. Price points are not given from God. There's a supply/demand curve, and if you price things higher, you'll get more profit per item but sell fewer items. What shape this curve takes, and where you ought to locate yourself on it, can vary on a lot of factors, and it's your job as a company selling things to research that, rather than decide "games cost $50/$60, and that's that". Maybe they should cost $20, maybe they should cost $100, maybe it varies based on the game and your goals.
2. There are a lot of people are willing to spend money. Some people will always get your stuff off Bittorrent purely due to the price (because it's free there, and you want money). But this is, contrary to what many media firms think, not the only or main problem. There are a lot of people who are willing to spend money on a lot of things. You'd do best to ask yourself if your company is doing something wrong that's keeping even people who would be willing to give you money from doing so (e.g. region-locked DVDs making it impossible for them to buy a legit copy).
3. Along the lines of #2, DRM can be counter-productive, by making the legit copy seem like a bigger hassle than the cracked copy off Bittorrent. People who are willing to give you money for something they like may not be willing to give you money if you come off seeming like you hate your customers.
Of course, #3 is slightly strange since Valve does in fact use DRM on Steam to authenticate your account to a particular machine. I suppose in their defense it's not nearly as draconian as much DRM, so they at least seem to be making efforts not to piss off their customers. And the existence of Steam in the first place, several years before any other major companies did anything similar, seems to indicate a certain understanding of, "if you make it easy for people to buy your things, they might do so".
I agree that mistakes in sensitive privacy situations can be damaging. But this particular plaintiff, the court found, failed to show that it was damaging in their situation, which is the requirement to sue for damages. They claimed they suffered $25,000 in emotional anguish, and the court held that they didn't provide any plausible legal arguments to support that damage claim.
If we do think, as a matter of public policy, that even harmless violations should be penalized in order to discourage them, there's a way to do that: pass a law that establishes a fine for such violations. The fine, of course, should go to the government, not the plaintiff, unless the plaintiff actually was harmed. Public policy via, you know, actual laws and law enforcement, not ambulance-chasing lawyers and "mental-anguish"-inventing plaintiffs.
Some kinds of research have the problem that they require rather large, long-term bets. The only private-sector businesses that can sustain that are very large ones that can average across even large projects, and very large businesses tend to be managed by revolving doors of managers optimizing for stock price (often short-term, medium-term at best; not 30-year horizons). Smaller businesses tend to have managers in it for the long haul, who do what they think is good for their business rather than their career, but smaller businesses have a cap on what size projects they can take on--- despite Boeing being almost certainly crappily managed, there's a reason no startup developed a competitor to the 787.
I'm not sure that's even possible, though. A very large business optimizing for long-term performance, over multiple very large projects, managing thousands of people, is sort of becoming a quasi-government. The only times I can recall it actually working out are times when it more or less *was* a quasi-government, like the golden era of Bell Labs during AT&T's monopoly period.
I don't see how you can draw your conclusion that "Those who have no intention of releasing code are not going to regardless of what the license says."
There are dozens of examples, one of which I linked, of them doing precisely that: people who had no intention of releasing code, like Monsoon, being forced to do so by the license.
As an example of that viewpoint, this article (PDF), from a 2007 volume of the Journal of Machine Learning Research characterizes open-source license choices from the perspective of scientists releasing their software as follows:
1. "A developer who wants to give away the source code in exchange for proper credit for derivative works, even closed-source ones, could choose the BSD license."
2. "A developer who wants to give away the source code, is comfortable with his source being incorporated into a closed-source product but still wants to receive bug-fixes and changes that are necessary to his source when integrating the code could choose the GNU Lesser General Public License (LGPL)."
3. "A developer who wants to give away the source code and make sure that his program stays open source, that is, any extension (or integration) will require both the original and the derived code to be released as open source, could choose the GNU General Public License (GPL)."
The clearest examples, of course, are the "hardball" ones: code releases where the person in question did not want to release the code for their derived version, but was forced to do so as part of the settlement of a GPL-violation lawsuit. For example, Monsoon Media released their source code only after being sued by BusyBox developers. It seems pretty clear that had BusyBox not been copyleft-licensed, Monsoon Media wouldn't have released their code.
I frequently hear them characterized that way, and I talk mostly to Windows-using academics who're discussing how to release their code, and could hardly care less about Linux or the Free Software Foundation.
The general viewpoint of options that get bandied about is something like:
1. "Research-only" or "non-commercial-use" license: minimal release that will ensure researchers who want it can get it, while retaining all commercial rights that default copyright gives.
2. Copyleft licensing: release that allows code to be used commercially or non-commercially, but only in free-software projects using the same copyleft license.
3. Permissive licensing: release that allows code to be used for any purpose, provided the copyright notice is maintained.
The first is is probably, unfortunately, the most common release mode for research code (though this is changing), since it feels like "giving away" the least as far as potential commercial exploitation goes. The second is usually an easier sell than the third, because it at least guarantees that Microsoft can't put your algorithm into the next version of Excel without paying you, which is the main thing people who get cold feet about releases are worried about (that they're giving away a potential source of livelihood for free).
I agree you'll need to deal with a bit of dependency-resolution, so it's not really for the average desktop user. It's manageable though without huge amounts of knowledge. I've been running it since 2002 (on the same install, no less) and haven't really run into major breakage. I usually don't blindly apt-get dist-update, though, but instead piecemeal update via aptitude, and avoid upgrading packages whose dependencies look to be in flux (e.g. aptitude wanting to remove 10 other packages to install the update).
You get world-class support for bugfixes, reasonable enhancement requests, package-interaction issues, and so on, often with new versions available within days of filing a good bug report. You get some of that in stable, but with a less satisfying lag until the next point release (or with more minor issues until the next major release).
That was really what blew me away when I switched in 2002 from running Windows 2000 full-time to running Debian sid/unstable full time. Complex issues like some program depending on a system behavior that had since changed weren't passed off as someone else's problem, or left for being fixed in the next version. Someone responded, often within hours, asking for details where necessary, we went back and forth by email a few times, they consulted with other maintainers if it was a multi-package issue, and a few days later the problem was fixed. In my day job, I don't get that level of support even for $600 software packages, even after an hour on the phone with inexperienced flunkies.
Unstable is unstable in the sense of changes happening semi-frequently, which you may not want on your production servers. But if your primary problem with Debian stable is that it doesn't get new software often enough, then presumably changes happening semi-frequently is precisely what you do want. And it gets bugfixes and security fixes first.
Despite the name, it's not where totally crazy experimental stuff that is more-likely-broken-than-not happens. There's a separate area, aptly named "experimental", for those packages. For example, the xf86->xorg change was staged in experimental for several months before being pushed to unstable after getting put into pretty good shape. OpenOffice 3 is undergoing a similar process currently, and will presumably be in good shape by the time it gets into unstable.
There is admittedly sometimes breakage in unstable, usually of specific packages, just because it's the newest widely used distribution: something'll never get to testing if it breaks in unstable. You can avoid even that, unless you really are the first person ever to encounter a particular bug, by using apt-listbugs to warn you of packages with major bugs filed against them, and delay upgrading those.
I don't think anyone objected to Microsoft giving customers source code; it's certainly an improvement on their previous practice of not doing so. What irked people was the term "shared source", which was seen as an attempt to confused the term "open source" by introducing a similar-sounding phrase without the same rights attached to it.
Since the federal gas tax hasn't been indexed for inflation or increased since 1993, revenues have declined approximately 35% in real terms over the past decade-and-a-half, while road-maintenance costs have increased faster than inflation. Some states have increased their state gas taxes to keep pace, but the feds sure haven't.
I do agree that a better approach would be to raise the gas tax instead of implementing tolls. But the current gas tax hasn't been indexed for inflation or increased since 1993, so hardly pays for everything. On the other hand, normal car owners *are* pretty much paying their fair share; it's heavy vehicles like trailer trucks that are being grossly undertaxed. Sure, they pay somewhat more gas taxes due to using more gas per mile, and pay some token registration fees, but they cause exponentially more road damage.
The expensive-journal commercial publishers don't have much of a competitive moat: anyone can publish PDFs on the internet with the word "Journal" attached to groups of them, and you've got a journal. If that anyone is well-respected in the field and the PDFs are hosted by a well-known university that also prints off some paper copies for archival, you've got yourself a new journal.
In my area this revolt against the commercial publishers has been quite rapid and successful. The entire board of editors left the journal Machine Learning in 2000, setting up the non-profit, open-access JMLR instead, which is now at least as prestigious (possibly moreso). In more general AI, the open-access, non-profit JAIR now has a much higher impact factor than the old Elsevier journal in the area, "Artificial Intelligence".
Slashdot Mirror
People who have been Mac people for a long time generally don't have that workflow, as the importing of the BSD backend is a fairly recent addition to the Mac world, whereas many of the GUI conventions have been around much longer.
I use mainly Debian (on my desktops) and OS X (on my laptop). There are pros and cons to both, but by far the biggest con of the Mac and pro of Debian is that packing management actually exists and works.
I nearly fell out of my chair when I first saw that the official way to install apps on OS X is to mount a virtual drive image and then manually copy files out of it to somewhere on your hard drive (like the Applications folder), and then unmount the drive image. That's the best they could come up with?
I don't know about there, but in my state (Texas), the statewide Libertarian candidates actually want to abolish Texas's public school system, not just get the federal government out of it. They fundamentally oppose the levying of taxes to provide free universal public education, and believe schooling should be provided solely by private institutions, whether for-profit or charity. But public schooling funded by taxes is hardly something that started in 1980; the common school has been a feature of the U.S. essentially since its inception.
I mean before deposit insurance, that was more or less the system we had. But people didn't all gravitate towards full-reserve banks, despite actually bearing 100% of the risk if a bank went bust. I blame people being short-sighted: they'll go for the fractional-reserve bank if it doesn't seem that likely to fail in the immediate future, and pays better interest. Then when the banks actually do fail, especially a lot of them at once, people demand something be done about it. Which is why we have deposit insurance now, and even the last European holdouts against deposit insurance are instituting it after this round.
I guess I don't have that much confidence things wouldn't work out the same way in another go-around. The fractional-reserve gold banks would become more popular than the full-reserve ones, creating the same money-multiplier issues you describe. Then when some of them went bust, people would demand something be done to restore their savings.
I don't actually see how you can outlaw fractional-reserve banking in a free society, since it requires central power to prohibit it. Say everyone started using gold coins tomorrow. This is actually kind of inconvenient, especially for large sums, so people would start producing "accounts" in which you could store your coins, and you could transfer gold to/from people with tokens, physical or electronic. Since not everyone needs to redeem their gold at the same time, some of these transaction services might lend some of the money out again, not operating at full reserve. As long as you agree to that in the contract when you sign up for an account with them, that seems perfectly legitimate to me.
The cynical explanation is that most people everywhere are populist at heart, which leads to a mish-mash of policies depending on what's popular and who the bogeyman is today. The right- and left-wingers have both typically been better at painting vivid images of bogeymen (whether fat-cat bankers or hordes of immigrants or whatever), and foment a sense of Emergency that Something Has to Be Done. In those conditions, the people promoting principled ideologies and cautious approach are usually steamrollered over by hare-brained solutions to the Emergency.
There's a certain populist ring to the classical liberal ideology of a level playing field with some moderate safety net, anyone can succeed but nobody starves, etc., but it's all a positive feel-good sort of vibe. I guess you can win on that sort of sentiment (it was a lot of Obama's "hope" campaign), but it's trickier. The good bogeymen I can think of focus only on one aspect so are already owned by the right or left; e.g. the opposite of civil libertarianism is 1984-style authoritarianism, but the left is already playing that sort of thing up in opposition to e.g. Guantanamo Bay.
A class-based analysis also gives a sort of population reason for it, since it tends to associate classical liberalism, and the similar European market liberal parties, as the ideology of the merchant classes, i.e. the educated middle and upper-middle class.
The entrenched upper classes tend to be more interested in maintaining their current position and exploiting networks of existing power, so aren't very in favor of dynamic markets and low barriers to entry, market fairness, social safety nets, etc. The lower classes tend to be more interested in immediate improvements to their position, and uncovering misdeeds of the higher classes that are allegedly responsible for their poorer position. Sometimes they also vote for conservative parties for reasons of religiosity or tradition. Neither of those classes, though, lend themselves to classical- or market-liberal types of positions. And there aren't enough educated middle-class folks who really understand and believe in the tradition of Voltaire, John Stuart Mill, Thomas Jefferson, etc.
The Libertarian party supports some principles that, broadly speaking, Americans believe in. More or less, these are classical liberal principles, in the mold of Thomas Jefferson. However, few people support their particular hardline interpretation, which tends to emphasize the anarcho-capitalist aspects, play down Jeffersonian elements that don't fit into that (e.g. Jefferson's view that governments should restrict the power of large corporations), and make few exceptions for any reason. Abolishing free public education, for example, is not a popular position. Neither is privatizing the road system. Some for of social safety nets are also popular---people don't want them abused (e.g. the stereotypical "welfare queens"), but neither do they want them to be totally absent. People also want regulation of private enterprise when its activities can cause negative externalities, such as systemic risks to economies (like banks, where further deregulation, the Libertarian position, is currently extremely unpopular). I could go on for a while.
Now if someone started a political party with positions more similar to those of the editorial line of The Economist newsmagazine, I could see voting for them. That is, support free-market economies with regulation and/or costing of negative externalities (pollution, systemic risks, etc.), a moderate social safety net, and liberal positions on social and civil liberties issues.
I think people are more willing to accept DRM if it seems like you're actually getting some benefit in return: in return for being able to download your games anywhere, you have to "phone home" before playing them to authenticate your local copy. Most DRM seems like you're getting a hassle for no benefit in return, which is sort of insult-to-injury when you actually paid for the content legitimately.
After World War I, the Bayer trademarks in the U.S. were reincarnated in a new company, confusingly also called Bayer, owned by the American firm Sterling Drug. They attempted to enforce the trademarks, but a 1921 court ruling invalidated the "aspirin" mark on grounds that it had come to be used as a generic term for the class of drugs for too long to be recaptured.
I don't use Steam myself (since the Linux client is still supposedly "on the way"), so was misremembering complaints I had read. It's not that Steam ties your account to a particular machine, but that it requires you to "phone home" every time before playing games even if they've already been downloaded. There's an Offline Mode, but there seem to be a lot of complaints about how well it works, and worries that if Steam were ever shut down, your games would basically be useless since they're not playable without authentication.
The points he makes in the Gamasutra summary sound remarkably clueful for the co-founder of a semi-major media firm. He seems to essentially "get it", that when selling content you're in a market, and if you're failing to sell as much as you'd want, the best solution is to figure out how you're failing to succeed in the market rather than whining about pirates.
Basically:
1. Price points are not given from God. There's a supply/demand curve, and if you price things higher, you'll get more profit per item but sell fewer items. What shape this curve takes, and where you ought to locate yourself on it, can vary on a lot of factors, and it's your job as a company selling things to research that, rather than decide "games cost $50/$60, and that's that". Maybe they should cost $20, maybe they should cost $100, maybe it varies based on the game and your goals.
2. There are a lot of people are willing to spend money. Some people will always get your stuff off Bittorrent purely due to the price (because it's free there, and you want money). But this is, contrary to what many media firms think, not the only or main problem. There are a lot of people who are willing to spend money on a lot of things. You'd do best to ask yourself if your company is doing something wrong that's keeping even people who would be willing to give you money from doing so (e.g. region-locked DVDs making it impossible for them to buy a legit copy).
3. Along the lines of #2, DRM can be counter-productive, by making the legit copy seem like a bigger hassle than the cracked copy off Bittorrent. People who are willing to give you money for something they like may not be willing to give you money if you come off seeming like you hate your customers.
Of course, #3 is slightly strange since Valve does in fact use DRM on Steam to authenticate your account to a particular machine. I suppose in their defense it's not nearly as draconian as much DRM, so they at least seem to be making efforts not to piss off their customers. And the existence of Steam in the first place, several years before any other major companies did anything similar, seems to indicate a certain understanding of, "if you make it easy for people to buy your things, they might do so".
I agree that mistakes in sensitive privacy situations can be damaging. But this particular plaintiff, the court found, failed to show that it was damaging in their situation, which is the requirement to sue for damages. They claimed they suffered $25,000 in emotional anguish, and the court held that they didn't provide any plausible legal arguments to support that damage claim.
If we do think, as a matter of public policy, that even harmless violations should be penalized in order to discourage them, there's a way to do that: pass a law that establishes a fine for such violations. The fine, of course, should go to the government, not the plaintiff, unless the plaintiff actually was harmed. Public policy via, you know, actual laws and law enforcement, not ambulance-chasing lawyers and "mental-anguish"-inventing plaintiffs.
Some kinds of research have the problem that they require rather large, long-term bets. The only private-sector businesses that can sustain that are very large ones that can average across even large projects, and very large businesses tend to be managed by revolving doors of managers optimizing for stock price (often short-term, medium-term at best; not 30-year horizons). Smaller businesses tend to have managers in it for the long haul, who do what they think is good for their business rather than their career, but smaller businesses have a cap on what size projects they can take on--- despite Boeing being almost certainly crappily managed, there's a reason no startup developed a competitor to the 787.
I'm not sure that's even possible, though. A very large business optimizing for long-term performance, over multiple very large projects, managing thousands of people, is sort of becoming a quasi-government. The only times I can recall it actually working out are times when it more or less *was* a quasi-government, like the golden era of Bell Labs during AT&T's monopoly period.
I don't see how you can draw your conclusion that "Those who have no intention of releasing code are not going to regardless of what the license says."
There are dozens of examples, one of which I linked, of them doing precisely that: people who had no intention of releasing code, like Monsoon, being forced to do so by the license.
As an example of that viewpoint, this article (PDF), from a 2007 volume of the Journal of Machine Learning Research characterizes open-source license choices from the perspective of scientists releasing their software as follows:
1. "A developer who wants to give away the source code in exchange for proper credit for derivative works, even closed-source ones, could choose the BSD license."
2. "A developer who wants to give away the source code, is comfortable with his source being incorporated into a closed-source product but still wants to receive bug-fixes and changes that are necessary to his source when integrating the code could choose the GNU Lesser General Public License (LGPL)."
3. "A developer who wants to give away the source code and make sure that his program stays open source, that is, any extension (or integration) will require both the original and the derived code to be released as open source, could choose the GNU General Public License (GPL)."
The clearest examples, of course, are the "hardball" ones: code releases where the person in question did not want to release the code for their derived version, but was forced to do so as part of the settlement of a GPL-violation lawsuit. For example, Monsoon Media released their source code only after being sued by BusyBox developers. It seems pretty clear that had BusyBox not been copyleft-licensed, Monsoon Media wouldn't have released their code.
I frequently hear them characterized that way, and I talk mostly to Windows-using academics who're discussing how to release their code, and could hardly care less about Linux or the Free Software Foundation.
The general viewpoint of options that get bandied about is something like:
1. "Research-only" or "non-commercial-use" license: minimal release that will ensure researchers who want it can get it, while retaining all commercial rights that default copyright gives.
2. Copyleft licensing: release that allows code to be used commercially or non-commercially, but only in free-software projects using the same copyleft license.
3. Permissive licensing: release that allows code to be used for any purpose, provided the copyright notice is maintained.
The first is is probably, unfortunately, the most common release mode for research code (though this is changing), since it feels like "giving away" the least as far as potential commercial exploitation goes. The second is usually an easier sell than the third, because it at least guarantees that Microsoft can't put your algorithm into the next version of Excel without paying you, which is the main thing people who get cold feet about releases are worried about (that they're giving away a potential source of livelihood for free).
I agree you'll need to deal with a bit of dependency-resolution, so it's not really for the average desktop user. It's manageable though without huge amounts of knowledge. I've been running it since 2002 (on the same install, no less) and haven't really run into major breakage. I usually don't blindly apt-get dist-update, though, but instead piecemeal update via aptitude, and avoid upgrading packages whose dependencies look to be in flux (e.g. aptitude wanting to remove 10 other packages to install the update).
You get world-class support for bugfixes, reasonable enhancement requests, package-interaction issues, and so on, often with new versions available within days of filing a good bug report. You get some of that in stable, but with a less satisfying lag until the next point release (or with more minor issues until the next major release).
That was really what blew me away when I switched in 2002 from running Windows 2000 full-time to running Debian sid/unstable full time. Complex issues like some program depending on a system behavior that had since changed weren't passed off as someone else's problem, or left for being fixed in the next version. Someone responded, often within hours, asking for details where necessary, we went back and forth by email a few times, they consulted with other maintainers if it was a multi-package issue, and a few days later the problem was fixed. In my day job, I don't get that level of support even for $600 software packages, even after an hour on the phone with inexperienced flunkies.
Unstable is unstable in the sense of changes happening semi-frequently, which you may not want on your production servers. But if your primary problem with Debian stable is that it doesn't get new software often enough, then presumably changes happening semi-frequently is precisely what you do want. And it gets bugfixes and security fixes first.
Despite the name, it's not where totally crazy experimental stuff that is more-likely-broken-than-not happens. There's a separate area, aptly named "experimental", for those packages. For example, the xf86->xorg change was staged in experimental for several months before being pushed to unstable after getting put into pretty good shape. OpenOffice 3 is undergoing a similar process currently, and will presumably be in good shape by the time it gets into unstable.
There is admittedly sometimes breakage in unstable, usually of specific packages, just because it's the newest widely used distribution: something'll never get to testing if it breaks in unstable. You can avoid even that, unless you really are the first person ever to encounter a particular bug, by using apt-listbugs to warn you of packages with major bugs filed against them, and delay upgrading those.
I don't think anyone objected to Microsoft giving customers source code; it's certainly an improvement on their previous practice of not doing so. What irked people was the term "shared source", which was seen as an attempt to confused the term "open source" by introducing a similar-sounding phrase without the same rights attached to it.
Since the federal gas tax hasn't been indexed for inflation or increased since 1993, revenues have declined approximately 35% in real terms over the past decade-and-a-half, while road-maintenance costs have increased faster than inflation. Some states have increased their state gas taxes to keep pace, but the feds sure haven't.
I do agree that a better approach would be to raise the gas tax instead of implementing tolls. But the current gas tax hasn't been indexed for inflation or increased since 1993, so hardly pays for everything. On the other hand, normal car owners *are* pretty much paying their fair share; it's heavy vehicles like trailer trucks that are being grossly undertaxed. Sure, they pay somewhat more gas taxes due to using more gas per mile, and pay some token registration fees, but they cause exponentially more road damage.
Federal grants help pay for mass-transit systems too, and yet I still have to pay $2 to the MTA if I want to ride the New York subway.
The expensive-journal commercial publishers don't have much of a competitive moat: anyone can publish PDFs on the internet with the word "Journal" attached to groups of them, and you've got a journal. If that anyone is well-respected in the field and the PDFs are hosted by a well-known university that also prints off some paper copies for archival, you've got yourself a new journal.
In my area this revolt against the commercial publishers has been quite rapid and successful. The entire board of editors left the journal Machine Learning in 2000, setting up the non-profit, open-access JMLR instead, which is now at least as prestigious (possibly moreso). In more general AI, the open-access, non-profit JAIR now has a much higher impact factor than the old Elsevier journal in the area, "Artificial Intelligence".