Citation needed, if you would. Injecting 10mL of any oil into your bloodstream is probably not a good idea, but the LD50 studies I've read put the level somewhere north of 130 mg/kg of pure THC (citing Rosencrantz 1983). Hash oil is potent, but it's unlikely to be better than 60% THC at best, and the higher the potency, the less suitable it would be for injection, since THC is basically a resin. As far as I have been able to determine, hash oil has a density about the same as water, so all told the average person would probably have to shoot up ~14 mL of very pure oil, and even then most of the effects would be due to physical effects of the oil rather than the neurological effects of the drug.
I can definitely see someone dying from injecting oil into their veins. However, as long as you're willing to argue fine technical points, is it still a death from cannabis if the cannabis was immaterial to the cause of death?
You're either underestimating the future change or overestimating the past change. Generally there isn't a goal temperature/CO2 level, just a pre-industrial baseline. We've already done enough to change the planet drastically. At this point we're just hoping that we don't continue to make things worse. At this point, we're still emitting ever-greater amounts of carbon year after year. At what point do you imagine that we should maybe dial back the things that we know raise the equilibrium temperature of the Earth? How quickly do you think plants and animals can adapt to 3-5 degrees of global temperature change? Because it looks like people would rather find out the answer to these things by massive uncontrolled experiment rather than simulation at this point. Buckle up.
There are very few releases based on Debian that can claim to be "bleeding edge", and Debian stable is very much not one of them. Systemd is the default in Mint's upstream distributions and almost all major distributions. There is no good reason to call it "immature". The Mint developers kowtowed to a certain vocal minority in not adopting the upstream system, which is annoying insofar as it is departing from the norm, but they aren't planning on maintaining this separation indefinitely, which annoys the systemd opponents. I don't know why you are bothering to argue that the debate about systemd is that it is buggy or new. That was an argument at one point, but it has always been secondary to concerns about overreach, NIH, binary logs, and monolithic libraries. Choosing to use deprecated software (as opposed to either OpenRC or systemd) increases the maintenance burden.
and now they're not packaging multimedia libraries any more.
So what? Just a couple of clicks and you can download them
This is the case with most other distributions as well. It nullifies one of the main advantages of Linux Mint. Now I can't give someone a Mint USB stick and expect them to have a fully functional system at the end of the install process, I have to explain my country's shitty legal system. Up until now, there has been a Linux distribution which allowed me to give someone basic OS functionality without having to make a political statement as well. There are more people that need to use computers than are willing and capable of understanding technical ramifications of intellectual property laws. In point of fact, most people don't know anything about them, and can't be bothered to learn. Linux Mint just made it significantly harder for most people to install and use. I'm not wildly enthusiastic about the idea of Linux becoming a popular desktop OS, but I do think this is a step backwards for that goal.
It's not that marijuana has never been linked to any deaths, it's that there has never been a lethal marijuana overdose, and estimates of the LD50 are 20,000 to 40,000 times the normal dose. I have no doubt that marijuana in combination with other drugs or other health conditions could be fatal, but [a] the difference between the effective dose and the lethal dose is one of the greatest of all psychoactive substances, and [b] consuming a lethal dose is more than a little impractical. It is completely impossible to consume a lethal dose of marijuana cigarettes. Even taking low estimates for the LD50, it would require smoking more than one cigarette per second for a sustained period of time.
As you no doubt know, the role of cannabis in producing psychosis is debated, and odds are there are genetic factors there as well. I believe it is more fair to say that drug use can produce psychosis in susceptible individuals, without needing to be more specific. Similarly, there are studies on both sides of the violence issue, and using the word "linked" is somewhat disingenuous. I believe further study is necessary to be able to firmly establish either position.
Cannabis has been established as one of the safest recreational drugs. One can even make favorable comparisons to caffeine. As one of the >40% of the US who has tried marijuana, I would say you're being alarmist, and I don't think that balances out the "pro-drug propaganda". Telling people to "Fuck off!" is also not indicative of a desire for honest discussion.
I have Cinnamon/Debian on one machine and Cinnamon/Mint on another. I have no idea what you mean by "added troubleshooting". You understand that most of Mint's packages are pulled directly from upstream, yes? And that the major distros do not generally have a preferred DE? Other distributions also have their own testing and QA processes; I'm not sure if you meant to denigrate those or whether you simply didn't know they existed.
In any case, avoiding bugs does not usually mean using whatever the developers consider to be the latest "stable" release. There is no substitute for a long real-world testing cycle. Even if there were reason to believe that Cinnamon were buggy on other platforms, that would make it buggy software. Which is hardly an advertisement.
LMDE has had inconsistent releases. I'm not sure if they figured out whether they were pulling from testing or rolling their own packages. The Linux Mint website was hacked to distribute malware. Linux Mint devs managed to create some package name conflicts with upstream. I read they are holding off on systemd for now, but plan to switch at some point, a position calculated to annoy everyone. There are equally simple ways to get a distro with Cinnamon, and now they're not packaging multimedia libraries any more.
I'm out of reasons to consider installing Linux Mint, I think. Are there more positives/negatives I'm missing? Or can we just write them off at this point?
There is no indication besides your willingness to believe evil about others that your supposition is plausible. You have presented no evidence. You've thrown out some nonsense about felons and corrupt mayors, and now some nonsense about false standards, to distract from the point that the preponderance of the evidence is very much in opposition to your claims. I am certainly willing to consider any evidence you care to make an argument for. Asking that you support your opinion with objective facts is not an undue burden.
Burdick v. US is not the only reference to pardons in American jurisprudence. If you'll read further than the summary, you will find that the first point was a direction on courtroom procedure, and therefore necessarily presupposed the existence of a court. It reaffirms earlier language in United States v. Wilson, and it was specifically addressing whether Burdick could continue to claim Fifth Amendment protections after having been offered the pardon. You're also suggesting that enumerated precepts are non-severable, which is very much not the case. Each is intended to address a separate question of law, and as regards the statement on guilt, that was intended to address issues raised by the ruling of Brown v. Walker, which involved statutory immunity from the consequences of testimony. The court held in Walker that the statutory immunity was as strong as the Fifth Amendment protection, and thus the government could compel testimony. In Burdick, in addition to setting out acceptance as a condition for the validity of a pardon, the court held that pardons carry the imputation of guilt specifically because they are able to be refused. It is the act of accepting the pardon, not the act of introducing it to a court, which connotes guilt.
You're arguing like a programmer, not a lawyer. You would be far better off trying to argue from e.g. Ex parte Garland that acceptance of a pardon does not imply guilt. You might start here or here. However, whether or not the legal concept of guilt could be strictly said to apply after having accepted a pardon, social opprobrium very much applies, and the political effects are unlikely to be very different.
Your link seems to contradict what you're stating. The pardoned person can't introduce the pardon into court proceedings without there having been charges.
Ford kept a little excerpt from Burdick v. US in his wallet for years after he pardoned Nixon. However, again, he did that without Nixon having been formally charged with any offenses, and the pardon itself makes no mention of any specific crimes. The legality of the pardon is arguably questionable, but various executive officers have more or less unconstrained abilities to pardon individuals, and no one is going to proceed with a trial if the defendant has already obtained a blanket pardon. The Supreme Court in 1915 did not anticipate this exercise of Presidential power, but I don't think anyone has argued that it should be somehow limited. Either way, it happened, so what are you going to do about it?
Repeating your assertion is not equivalent to substantiating it. Find me an example of any felon being elected to federal office, or state executive office. Find me any poll which suggests that it's a remotely plausible situation. Then you can work on explaining why every other person on the previously-linked list had their career ended by their convictions, regardless of political stripe, whether they were pardoned, or what year this took place. If you cannot show strong evidence to support your prejudices, I'll thank you not to continue repeating them.
Tell you what, if you really believe that, find me any instance where a felon has been elected to any federal office or state governor or lieutenant governor. And then we'll weigh that against the far longer list of people for whom conviction instantly ended their political careers.
It is not to your credit that you are willing to believe that your political opposites are so venal as to vote for an ideological proponent no matter what their criminal background. Most people aren't going to let their political opinions overrule their basic morality sensibility, and there is no more universally despised class in this country than felons. There is no rational basis for your assertion.
Do you think the American people would elect someone to the Presidency who had admitted to a felony?
A pardon is an admission of guilt, per Burdick v. United States. Not only does your scenario require a great deal of corruption and malfeasance from both parties, but also the collaboration of the American people in electing a known felon. Are you going to vote for Clinton if she accepts a pardon?
I am going to be generous in assuming that you either live outside of the US in a jurisdiction in which what you say is true. However, if you had consulted the link I gave, you would see that you are very much wrong as concerns US law.
You don't seem to know how pardons work. She doesn't have to plead guilty; Nixon for example was pardoned without ever having been charged. Accepting a pardon, however, is an admission of guilt. There would be no faster way to lose an election. Republicans would be calling for her head on a plate before the ink on the signature dried.
I gave lots of citations in our last discussion. I don't know how you imagine that you retain any sort of moral advantage by simply not reading them.
However, that's not what I asked about. If you refuse to identify any part of the theory of AGW that you think is false, we can only conclude that you do actually think the theory is valid.
I'll waste a few keystrokes reiterating that Sawyer (1972) predicted.6 degrees of warming by the year 2000, which turned out to be exactly correct. I don't lean on that as incontrovertible proof, however. Calculations of global temperature aren't really evidence for or against the underlying theory. So if you will be so good as to highlight where the theory is wrong, I would be more than happy to provide you with the appropriate citation.
So have you decided what part of AGW you don't like? Besides its reporting in the popular press. Is it that CO2 doesn't absorb heat? That it doesn't build up in the atmosphere? Do you know of more than one way to transfer heat to space? You were very evasive in our last discussion.
Also, the only one currently talking about public policy appears to be you. I understand you have a rabid aversion to being told what to do, but that does not invalidate basic physics.
Nuclear fission is fairly simple and produces enormous energy for a given input...but the variables involved make plants large, safety measures redundant (by intent) and costs high. If we can simplify things and bring costs down by better understanding the atomic process then maybe 'too cheap to meter' could actually be a thing one day.
Not that nuclear power should not be cheaper, but "too cheap to meter" will never happen and should never happen. There are negative externalities that need to be priced into the cost of producing power from any source, and eventually we are going to run up into some hard thermodynamic limits to how much energy we can produce without cooking ourselves. We're going to end up (in a few centuries) in a thermodynamic zero sum game. This may be as good as it gets for power costs, and we probably should think carefully about anything that would provide incentives for greater power use per capita.
There is nothing wrong with making nuclear power cheaper, but any power source being "too cheap to meter" would probably do more harm than good.
Of course, these days it's all a mish-mash and a binary can be somewhere - dynamics in/sbin, statics in/bin, executables in/opt and/var, etc.
Which is one reason why Fedora and company are simplifying things by shoving everything in/usr. Unix was not originally designed to have different executables in different places, Thompson and Ritchie simply ran out of disk space, and in the era of small disks it was a sensible enough partitioning scheme. With the BSDs and commercial Unixes, it also makes more sense to distinguish between binaries supplied by the vendor and user-provided binaries, but Linux is more along the lines of "ship it all, and let root sort it out." So certain things have changed in the last forty years that make the rigid hierarchy less important.
I worry about the transition though. Technically, it's easy to just make/bin a symlink to/usr/bin. However, it was probably technically easy to have NTFS pretend to have 8.3 filenames when it needed to, as well. I feel like this was an object lesson in the dangers of "soft deprecation". Similarly, bitcoin seems to be having a lot of issues because of the developers' unwillingness to make breaking changes. Both of these things make me extremely skeptical of the long-term value of backwards compatibility. Are symlinks setting up the wrong user expectations? Probably not, but it still might be best to fail hard and quickly.
For example I think the Linux (POSIX?) file system was written before they invented autocomplete, it's all TLAs like/var/usr/bin/lib/wtf.
In this case it's the file system hierarchy, not the file system. Personally, I think the argument for longer filenames is bogus. Using longer filenames isn't necessarily going to make their purpose any more clear, and for everything outside of the home folder, the novice user should probably not be touching that stuff, any more than they should be poking around in C:\Windows. Being user friendly is not a feature for things that are not intended for casual use. Autocomplete is an even worse argument: I'm not saving any keystrokes by typing/bi[TAB] versus/bin.
However, your example was somewhat poorly chosen in another sense, because while there is no call to make the names longer, at least one major distribution got rid of some of those top-level folders. Fedora likes to move fast and break things anyway, but in this case the historical justification for splitting up the binaries was, well, kind of ridiculous. Thompson and Ritchie created that particular issue a couple years before CP/M inflicted drive letters on us, but forty years later it's still a bug worth fixing. Most of today's code and systems will be pretty hoary in forty years, and I'm not sure I would consider it a virtue if it ran unmodified on my...hmm, well, whatever system exists at that time. One can always use emulation to provide old features, but most of the time I'd rather that not be happening at the OS level.
Given that Windows inherited both 8.3 filenames and drive letters from CP/M, it makes sense to talk about them in the same context. Drive letters are pretty harmless, but having "secret" 8.3 filenames and unremovable folders is probably something that needs to go. Linux definitely doesn't have those kind of problems.
Slashdot Mirror
Citation needed, if you would. Injecting 10mL of any oil into your bloodstream is probably not a good idea, but the LD50 studies I've read put the level somewhere north of 130 mg/kg of pure THC (citing Rosencrantz 1983). Hash oil is potent, but it's unlikely to be better than 60% THC at best, and the higher the potency, the less suitable it would be for injection, since THC is basically a resin. As far as I have been able to determine, hash oil has a density about the same as water, so all told the average person would probably have to shoot up ~14 mL of very pure oil, and even then most of the effects would be due to physical effects of the oil rather than the neurological effects of the drug.
I can definitely see someone dying from injecting oil into their veins. However, as long as you're willing to argue fine technical points, is it still a death from cannabis if the cannabis was immaterial to the cause of death?
Ironically, the author would like you to stop using git.
Like he said, it's the perfect *NIX source control system...
The O'Reilly book, Version Control with Git, for me provided the "ah-ha!" explanation of git's internals that made things made sense.
You're either underestimating the future change or overestimating the past change. Generally there isn't a goal temperature/CO2 level, just a pre-industrial baseline. We've already done enough to change the planet drastically. At this point we're just hoping that we don't continue to make things worse. At this point, we're still emitting ever-greater amounts of carbon year after year. At what point do you imagine that we should maybe dial back the things that we know raise the equilibrium temperature of the Earth? How quickly do you think plants and animals can adapt to 3-5 degrees of global temperature change? Because it looks like people would rather find out the answer to these things by massive uncontrolled experiment rather than simulation at this point. Buckle up.
There are very few releases based on Debian that can claim to be "bleeding edge", and Debian stable is very much not one of them. Systemd is the default in Mint's upstream distributions and almost all major distributions. There is no good reason to call it "immature". The Mint developers kowtowed to a certain vocal minority in not adopting the upstream system, which is annoying insofar as it is departing from the norm, but they aren't planning on maintaining this separation indefinitely, which annoys the systemd opponents. I don't know why you are bothering to argue that the debate about systemd is that it is buggy or new. That was an argument at one point, but it has always been secondary to concerns about overreach, NIH, binary logs, and monolithic libraries. Choosing to use deprecated software (as opposed to either OpenRC or systemd) increases the maintenance burden.
and now they're not packaging multimedia libraries any more.
So what? Just a couple of clicks and you can download them
This is the case with most other distributions as well. It nullifies one of the main advantages of Linux Mint. Now I can't give someone a Mint USB stick and expect them to have a fully functional system at the end of the install process, I have to explain my country's shitty legal system. Up until now, there has been a Linux distribution which allowed me to give someone basic OS functionality without having to make a political statement as well. There are more people that need to use computers than are willing and capable of understanding technical ramifications of intellectual property laws. In point of fact, most people don't know anything about them, and can't be bothered to learn. Linux Mint just made it significantly harder for most people to install and use. I'm not wildly enthusiastic about the idea of Linux becoming a popular desktop OS, but I do think this is a step backwards for that goal.
It's not that marijuana has never been linked to any deaths, it's that there has never been a lethal marijuana overdose, and estimates of the LD50 are 20,000 to 40,000 times the normal dose. I have no doubt that marijuana in combination with other drugs or other health conditions could be fatal, but [a] the difference between the effective dose and the lethal dose is one of the greatest of all psychoactive substances, and [b] consuming a lethal dose is more than a little impractical. It is completely impossible to consume a lethal dose of marijuana cigarettes. Even taking low estimates for the LD50, it would require smoking more than one cigarette per second for a sustained period of time.
As you no doubt know, the role of cannabis in producing psychosis is debated, and odds are there are genetic factors there as well. I believe it is more fair to say that drug use can produce psychosis in susceptible individuals, without needing to be more specific. Similarly, there are studies on both sides of the violence issue, and using the word "linked" is somewhat disingenuous. I believe further study is necessary to be able to firmly establish either position.
Cannabis has been established as one of the safest recreational drugs. One can even make favorable comparisons to caffeine. As one of the >40% of the US who has tried marijuana, I would say you're being alarmist, and I don't think that balances out the "pro-drug propaganda". Telling people to "Fuck off!" is also not indicative of a desire for honest discussion.
I have Cinnamon/Debian on one machine and Cinnamon/Mint on another. I have no idea what you mean by "added troubleshooting". You understand that most of Mint's packages are pulled directly from upstream, yes? And that the major distros do not generally have a preferred DE? Other distributions also have their own testing and QA processes; I'm not sure if you meant to denigrate those or whether you simply didn't know they existed.
In any case, avoiding bugs does not usually mean using whatever the developers consider to be the latest "stable" release. There is no substitute for a long real-world testing cycle. Even if there were reason to believe that Cinnamon were buggy on other platforms, that would make it buggy software. Which is hardly an advertisement.
LMDE has had inconsistent releases. I'm not sure if they figured out whether they were pulling from testing or rolling their own packages. The Linux Mint website was hacked to distribute malware. Linux Mint devs managed to create some package name conflicts with upstream. I read they are holding off on systemd for now, but plan to switch at some point, a position calculated to annoy everyone. There are equally simple ways to get a distro with Cinnamon, and now they're not packaging multimedia libraries any more.
I'm out of reasons to consider installing Linux Mint, I think. Are there more positives/negatives I'm missing? Or can we just write them off at this point?
There is no indication besides your willingness to believe evil about others that your supposition is plausible. You have presented no evidence. You've thrown out some nonsense about felons and corrupt mayors, and now some nonsense about false standards, to distract from the point that the preponderance of the evidence is very much in opposition to your claims. I am certainly willing to consider any evidence you care to make an argument for. Asking that you support your opinion with objective facts is not an undue burden.
Burdick v. US is not the only reference to pardons in American jurisprudence. If you'll read further than the summary, you will find that the first point was a direction on courtroom procedure, and therefore necessarily presupposed the existence of a court. It reaffirms earlier language in United States v. Wilson, and it was specifically addressing whether Burdick could continue to claim Fifth Amendment protections after having been offered the pardon. You're also suggesting that enumerated precepts are non-severable, which is very much not the case. Each is intended to address a separate question of law, and as regards the statement on guilt, that was intended to address issues raised by the ruling of Brown v. Walker, which involved statutory immunity from the consequences of testimony. The court held in Walker that the statutory immunity was as strong as the Fifth Amendment protection, and thus the government could compel testimony. In Burdick, in addition to setting out acceptance as a condition for the validity of a pardon, the court held that pardons carry the imputation of guilt specifically because they are able to be refused. It is the act of accepting the pardon, not the act of introducing it to a court, which connotes guilt.
You're arguing like a programmer, not a lawyer. You would be far better off trying to argue from e.g. Ex parte Garland that acceptance of a pardon does not imply guilt. You might start here or here. However, whether or not the legal concept of guilt could be strictly said to apply after having accepted a pardon, social opprobrium very much applies, and the political effects are unlikely to be very different.
Your link seems to contradict what you're stating. The pardoned person can't introduce the pardon into court proceedings without there having been charges.
Ford kept a little excerpt from Burdick v. US in his wallet for years after he pardoned Nixon. However, again, he did that without Nixon having been formally charged with any offenses, and the pardon itself makes no mention of any specific crimes. The legality of the pardon is arguably questionable, but various executive officers have more or less unconstrained abilities to pardon individuals, and no one is going to proceed with a trial if the defendant has already obtained a blanket pardon. The Supreme Court in 1915 did not anticipate this exercise of Presidential power, but I don't think anyone has argued that it should be somehow limited. Either way, it happened, so what are you going to do about it?
Repeating your assertion is not equivalent to substantiating it. Find me an example of any felon being elected to federal office, or state executive office. Find me any poll which suggests that it's a remotely plausible situation. Then you can work on explaining why every other person on the previously-linked list had their career ended by their convictions, regardless of political stripe, whether they were pardoned, or what year this took place. If you cannot show strong evidence to support your prejudices, I'll thank you not to continue repeating them.
Tell you what, if you really believe that, find me any instance where a felon has been elected to any federal office or state governor or lieutenant governor. And then we'll weigh that against the far longer list of people for whom conviction instantly ended their political careers.
It is not to your credit that you are willing to believe that your political opposites are so venal as to vote for an ideological proponent no matter what their criminal background. Most people aren't going to let their political opinions overrule their basic morality sensibility, and there is no more universally despised class in this country than felons. There is no rational basis for your assertion.
Do you think the American people would elect someone to the Presidency who had admitted to a felony?
A pardon is an admission of guilt, per Burdick v. United States. Not only does your scenario require a great deal of corruption and malfeasance from both parties, but also the collaboration of the American people in electing a known felon. Are you going to vote for Clinton if she accepts a pardon?
I am going to be generous in assuming that you either live outside of the US in a jurisdiction in which what you say is true. However, if you had consulted the link I gave, you would see that you are very much wrong as concerns US law.
You really think people are going to change their vote over some arcane computer geeky technical crime?
I am quite sure that no one who admits to being guilty of a felony will be elected to the Presidency, whatever the details.
You don't seem to know how pardons work. She doesn't have to plead guilty; Nixon for example was pardoned without ever having been charged. Accepting a pardon, however, is an admission of guilt. There would be no faster way to lose an election. Republicans would be calling for her head on a plate before the ink on the signature dried.
You gave precisely zero citations.
Why lie?
Why continue to avoid talking about the theory?
I gave lots of citations in our last discussion. I don't know how you imagine that you retain any sort of moral advantage by simply not reading them.
However, that's not what I asked about. If you refuse to identify any part of the theory of AGW that you think is false, we can only conclude that you do actually think the theory is valid.
I'll waste a few keystrokes reiterating that Sawyer (1972) predicted .6 degrees of warming by the year 2000, which turned out to be exactly correct. I don't lean on that as incontrovertible proof, however. Calculations of global temperature aren't really evidence for or against the underlying theory. So if you will be so good as to highlight where the theory is wrong, I would be more than happy to provide you with the appropriate citation.
So have you decided what part of AGW you don't like? Besides its reporting in the popular press. Is it that CO2 doesn't absorb heat? That it doesn't build up in the atmosphere? Do you know of more than one way to transfer heat to space? You were very evasive in our last discussion.
Also, the only one currently talking about public policy appears to be you. I understand you have a rabid aversion to being told what to do, but that does not invalidate basic physics.
Nuclear fission is fairly simple and produces enormous energy for a given input...but the variables involved make plants large, safety measures redundant (by intent) and costs high. If we can simplify things and bring costs down by better understanding the atomic process then maybe 'too cheap to meter' could actually be a thing one day.
Not that nuclear power should not be cheaper, but "too cheap to meter" will never happen and should never happen. There are negative externalities that need to be priced into the cost of producing power from any source, and eventually we are going to run up into some hard thermodynamic limits to how much energy we can produce without cooking ourselves. We're going to end up (in a few centuries) in a thermodynamic zero sum game. This may be as good as it gets for power costs, and we probably should think carefully about anything that would provide incentives for greater power use per capita.
There is nothing wrong with making nuclear power cheaper, but any power source being "too cheap to meter" would probably do more harm than good.
Of course, these days it's all a mish-mash and a binary can be somewhere - dynamics in /sbin, statics in /bin, executables in /opt and /var, etc.
Which is one reason why Fedora and company are simplifying things by shoving everything in /usr. Unix was not originally designed to have different executables in different places, Thompson and Ritchie simply ran out of disk space, and in the era of small disks it was a sensible enough partitioning scheme. With the BSDs and commercial Unixes, it also makes more sense to distinguish between binaries supplied by the vendor and user-provided binaries, but Linux is more along the lines of "ship it all, and let root sort it out." So certain things have changed in the last forty years that make the rigid hierarchy less important.
I worry about the transition though. Technically, it's easy to just make /bin a symlink to /usr/bin. However, it was probably technically easy to have NTFS pretend to have 8.3 filenames when it needed to, as well. I feel like this was an object lesson in the dangers of "soft deprecation". Similarly, bitcoin seems to be having a lot of issues because of the developers' unwillingness to make breaking changes. Both of these things make me extremely skeptical of the long-term value of backwards compatibility. Are symlinks setting up the wrong user expectations? Probably not, but it still might be best to fail hard and quickly.
I knew that if I didn't put a disclaimer on the last line, people would take it seriously, and look! Vindication.
Thank you slashdot, for living up to the lowest of my expectations.
For example I think the Linux (POSIX?) file system was written before they invented autocomplete, it's all TLAs like /var/usr/bin/lib/wtf.
In this case it's the file system hierarchy, not the file system. Personally, I think the argument for longer filenames is bogus. Using longer filenames isn't necessarily going to make their purpose any more clear, and for everything outside of the home folder, the novice user should probably not be touching that stuff, any more than they should be poking around in C:\Windows. Being user friendly is not a feature for things that are not intended for casual use. Autocomplete is an even worse argument: I'm not saving any keystrokes by typing /bi[TAB] versus /bin.
However, your example was somewhat poorly chosen in another sense, because while there is no call to make the names longer, at least one major distribution got rid of some of those top-level folders. Fedora likes to move fast and break things anyway, but in this case the historical justification for splitting up the binaries was, well, kind of ridiculous. Thompson and Ritchie created that particular issue a couple years before CP/M inflicted drive letters on us, but forty years later it's still a bug worth fixing. Most of today's code and systems will be pretty hoary in forty years, and I'm not sure I would consider it a virtue if it ran unmodified on my...hmm, well, whatever system exists at that time. One can always use emulation to provide old features, but most of the time I'd rather that not be happening at the OS level.
Given that Windows inherited both 8.3 filenames and drive letters from CP/M, it makes sense to talk about them in the same context. Drive letters are pretty harmless, but having "secret" 8.3 filenames and unremovable folders is probably something that needs to go. Linux definitely doesn't have those kind of problems.