Before you start leveling such claims, I'd like to see the percentage of road-maintenance costs which come from vehicle registration fees (in the GP's jurisdiction). It has been my impression that most of the cost is covered by gas taxes, in which case the GP is indeed paying for the roads.
It is highly unlikely the GP's total tax burden is less than that required to cover his/her own share of the maintenance cost. Even if that were the case, however, the government started this fight by instituting taxation; only by ending that practice permanently, and making full reparations, can they regain the moral authority to demand that would-be freeloaders stay off their property. Until then it comes down to an individual and an organization which each consider the other's actions criminal; you can't expect to be taken reasonably while appealing to just one party to change its ways (trespassing) and ignoring equivalent or greater offenses (theft) from the other side.
Sure, fast charging would take an unreasonable amount of power if you're just plugging it into a standard outlet. However, you don't need that kind of power 24/7—just for the 10 minutes during which you're charging the vehicle. That implies some kind of buffering; you have a local power station that draws about 10 amps @ 220 volt to recharge its internal capacitors over a 24-hour period, and then dumps all that energy into the vehicle's batteries on demand. A trivial implementation of this would be to have two sets of batteries, one slow-charging and one in use, and just swap them when the in-use batteries are drained. This could reasonably be made a 10-minute job with an efficient design.
I considered that, but the odds that someone could break the hash algorithm to the point that it becomes useless for qualifying the recordings as genuine does not seem to me to be significantly better than the odds that the records are made useless in some other way, e.g. by electronic or physical destruction, or that the recorded conversation was fabricated in the first place.
Note, also, that the methods devised to generate collisions for existing hashes (MD5, SHA1) require control of "invisible" regions in both files. There are no "invisible" areas in an audio recording, if you do it right—hash the raw samples, and limit the hash to the more significant bits—and anyone attempting to fabricate evidence after the fact would have to generate a collision with the original recording, not one of their choice.
Regarding the motivation, competitors can be a great source of negative information to counter a company's (possibly excessive) self-promotion. Between the two you have all the data you need as a starting point for your own fact-checking.
The "government" gives carriers a lot of leeway by protecting them from liability for the content they carry.
Only because they incorrectly consider them liable in the first place. They neither provide the content, nor have any reason to know the significance of the bits passing through their lines. Merely providing Internet service should not require some special "common carrier" (or whatever) status to shield one from liability. The only way one should be able to incur such liability is by choosing to continue providing service after becoming aware that doing so contributes to a specific crime.
It's not always a literal gun to the head that takes choice away.
Perhaps not, but "leveling the playing field" via laws (incl. regulations, taxes, etc.) does involve a literal gun to the head, if you refuse to comply. Nothing but a similar threat of literal force could possibly justify that.
Level the "playing field" all you want by peaceful means—just don't be the first to resort to coercion. In this case there is a simple, non-aggressive solution: if you can garner sufficient popular support to step in and forcibly tell ISPs what to do with their network, surely you can turn that same popular support toward forming a self-funded co-op offering Internet service on your preferred terms. Competition trumps conflict every time.
Right. I wasn't addressing the problem of actually destroying all the copies, just the conflict between immutability for audit-trail purposes and the need to delete certain conversations.
For dealing with the copies, a good start would be the suggestion made my another commenter to encrypt the archived copies of the conversations (with unique keys) and hold the keys in a separate database. That way you could back up the conversations themselves by whatever means are most efficient, and to effectively destroy the data you only have to erase the keys. To ensure full destruction you still need to track all the unencrypted copies, of course—and ensure there are no unauthorized copies of unencrypted recordings or conversation keys—for which there is no simple technical solution. That is a matter for policy, procedure, and manual auditing. Fortunately, most of the recordings will never need to be decrypted, so there shouldn't be many unencrypted copies floating around.
If they can figure out a way to delete a 'prohibited conversation' they could theoretically modify the data too.
Technically there is no need to make the conversations themselves immutable. You just need to be able to verify that the recording you have is the one which was originally recorded. A one-way hash can serve this purpose. For each recording, store the conversation itself in an erasable/mutable medium, but record a hash of the conversation in append-only storage (with multiple distributed backups). If you need to show that the recording is legit, compare it with the hash. If you need to delete something, record the deletion in the append-only medium and then remove it from the mutable storage. The hash will remain, but you can't use the hash to obtain information about the conversation without the original recording.
Bonus: You can recognize unauthorized deletions by comparing the mutable and immutable records.
"Reasonably"? I don't think so. I'm fairly sure that something can't be a derived work if it wouldn't qualify for copyright on its own. For example, if I say "The length of your comment is 71 characters", that sentence is derived from your (copyrighted) comment, but it's just a fact and thus not a derived work. Similarly, a torrent file may contain information (hashes & lengths) derived from a copyrighted work, but that information would not qualify for copyright on its own merits—being nothing more than abstract facts about the referenced data—and thus should not be considered a derived work.
Claiming that torrent files are derived works is analogous to claiming that bibliography entries are derived works; both are purely a set of facts which exist only for identification purposes.
(Notice that there's an odd form of double-think involved in allowing copyright for digital works and not facts when anything in digital form is ultimately nothing more than a collection of facts. Of course, allowing copyright over pure facts would be ridiculous, so the only way to resolve the issue is to recognize that it is inconsistent to allow copyright over anything—not that I expect the pro-copyright group to give up over a little matter like their position being inherently self-contradictory.)
This debate has seemingly devolved to an argument over word choices...
It doesn't matter whether the "set of uniform pools" model is the ideal model or the practical one. The point is that the two models are just different ways of describing the same thing.
What I described is the point of insurance from my perspective. If you want to take plain barter (or charity, or forced redistribution) and call it insurance, fine—but we'll still need a word to describe a contract which trades risk for equivalent periodic cost. Any suggestions? (I like "insurance"; it's traditional.)
The ideal free market—the standard libertarian one, not one of the others people sometimes advocate—does not depend on having perfect information, or on the absence of non-aggressive barriers to entry. In fact, improving the available information is one of the basic functions of the market. It does depend on a complete lack of aggression, however, so achieving it is equivalent to eliminating all crime, which clearly isn't going to happen in your lifetime or mine. Fortunately, we don't need an ideal free market for competition to work. We just need something "close enough", and for what we're discussing "close enough" isn't really all that close at all. A lack of aggressive (regulatory) interference would be quite sufficient on its own.
You assume like-risk, I don't. Furthermore, I don't assume that pools are formed by individuals. I merely assume that a demand exists, and that there is a product to fill that demand.
There is absolutely no significant difference between what you said and what I said in this area. Ignore the bit about "like-risk"; it's just distracting you from the big picture. It comes from the abstract model of insurance where each insurance contract is placed into a "pool" with many other contracts with similar risks and premiums. In practice, of course, insurees just pay premiums in proportion to their risk, and no one bothers to keep the actual reserves separate for different levels of risk. It works out the same in the end. The "individuals" part is similar; I never assumed or claimed that insurance only applies to individuals acting alone, although that was the original context (re: medical insurance). However, only individuals make decisions to take out insurance, whether they're acting on their own or as representatives of multi-national corporations. There's no point in considering the actions of an organization separate from the actions of the individuals who run it.
You also misunderstand the reason that insurance works. It works BECAUSE cost is externalized onto all the people who pay for more insurance than they need.
Costs, yes, at least after the fact. However, that misses the point. What matters with insurance is risk (ex ante probability-weighted projected future costs), and in terms of risk there is no subsidy. People don't know how much they'll need; that's why insurance exists, and why the savings-account concept doesn't work. However, in a free market no one buys pays higher premiums than they feel is necessary to cover their risk. In general, no one can say beforehand that a given insuree is paying more for their insurance than is required to cover their projected cost. If they could then they would offer that person a plan with the same coverage at a lower premium, and competition would eliminate the subsidy.
Finally, you also seem to misunderstand my last comment.... Sometimes, it is cheaper for society as a whole to subsidize the prevention of catastrophic events... rather than mop up after it.
I understood your comment just fine, thank you. Disagreement is not equivalent to misunderstanding. The reason I disagree is simple: when you say that "society as a whole" should do something, when the individuals that make up that society have not freely chosen to do so, what you really mean is "some members of society should force the rest to do this, for the good of the whole (in my opinion)." Acting on that statement, however, is completely incompatible with a free society. It divides people into two classes: those who make the rules, and those who must follow them. Whether these classes are determined by birth or status or majority makes not difference; no one has the right to govern another by force.
Anyway, "society" doesn't own property, and thus does not have a "balance sheet". By pretending it does you are ignoring the boundaries between one individual's property and other's. It may seem less expensive to me for you and I to combine our purchases for a better price, but even if I have every reason to believe that I'm right I still can't make you participate. Which is good, because I may not be right—and the whole point of private property is that the owner makes those decisions. What you are proposing is the same, just at a larger scale.
I'm not a Social Darwinist. A Social Darwinist would argue that charity is wrong as well because it allows the weak to survive. (The label "Social Darwinist" is often used to demonize one's opponent, but you are unlikely to encounter one in practice. There are a few out there, though.)
Your argument is a good reason to donate to charity, but you appear to be taking it as a justification for theft. That is not something which you have any right to authorize. If you can persuade others to donate, great. If not, that is their choice to make, not yours.
Nonsense. The demand for insurance is that it is easier to mitigate risk if you pay a small regular chunk of money you don't need in return for an occasional large chunk of money when you need it. The supply of insurance exists because entities are willing and able to pay more than the compound cost over a large set of entities.
Businesses like it because it simplifies their accounting and long-term planning. Not to mention that it frees up resources that don't have to be held in reserves. Individuals like it because an unlikely, but possible catastrophe doesn't mean that they're fucked financially until they die.
Isn't that exactly what I said? Yes, I believe it was: "...the purpose of insurance, medical or otherwise, is for like-risk individuals to form a pool and trade assessed risk for equivalent, but predictable, premiums." In other words, to "mitigate risk."
There are universal benefits to people and businesses having insurance. To the point that it is cheaper to hand out insurance than deal with the after-effects of not having people and businesses insured.
Sure it's cheaper—when the costs are externalized onto all the people forced to pay for more insurance than they need or want, either directly or indirectly via taxes. Everything seems cheaper when you don't have to fund it yourself. That doesn't even consider the additional dead-weight loss of violating basic social rules like private property and self-determination.
There is no such thing as 'like-risk' -- the question is to what extent we can factor known risk factors into premiums.
The idealized model of insurance places individuals into different "pools" depending on their individual risk, with each pool corresponding to a specific risk and equivalent premium. It is assumed that there are sufficient individuals within each pool for things to average out. However, that's just an abstraction. In the real world risk is a continuous variable, so every individual would be in a separate pool (with no averaging). Instead, everyone pays into a single pool in proportion to their respective risk. It works out the same in the end.
Effectively, those who incur reimbursable expense are subsidized by those who don't.
When you look at things from an ex post cost basis, yes. However, that's an artificial point of view; when considering whether to invest in insurance you don't know the shape of future events. The whole point of insurance is to take that uncertainty and turn it into a stable situation from an ex ante point of view, where risk, not cost, is the determining factor. And in terms of risk there is no subsidy; given all the information known ex ante there is no reason for any insuree to believe they are subsidizing anyone. All insurees within a given pool are equally likely to be compensated. (Or, avoiding the "pool" abstraction, all insurees have a probability-weighted projected compensation which is proportional to their premium.)
No one willingly joins an insurance program expecting to pay in substantially more than their individual risk—not when competitors are free to offer them lower premiums—and no insurance program could survive for long without balancing risk if low-risk individuals are permitted to opt out. Ergo, no voluntary insurance program can retain ex ante subsidies over the long run.
What do you propose that those people whose premiums woud [sic] be impossibly high (or who are insurance pariahs) should do?
If they truly cannot get insurance, then their only option is accept the risk themselves and hope they don't get sick or injured.
If they do anyway, and cannot afford health care even after exhausting all available options for charitable assistance, then they are in exactly the same position as one who has a disease or injury for which there is no known cure—a position which, needless to say, every person who has ever lived (or is currently alive) has faced (or will face) at some point.
If you could predict with 80% accuracy whether somebody's house would burn down, then almost everybody could get dirt-cheap fire insurance (which they wouldn't buy anyway since they wouldn't need it), and a small number of people wouldn't be able to afford it and would lose everything they have in a fire.
Rather, the houses likely to burn down would be unoccupied, and probably demolished, and the remaining houses would be be insurable at much lower rates. Everyone wins. Well, some people would be out the cost of a house and need to find new shelter, but at least they won't be caught in a fire.
In the medical case, of course, you can't leave your "house"—but you can take steps to mitigate any diseases you may be genetically predisposed to. Analogies aside, however, by the time your genetics have been determined it's too late for (real) insurance; either you're known to be within normal risk levels, or you're looking for charity, not insurance. The time to take out insurance against genetic risk factors is before conception. At that point it's your parents' DNA that determines the risk, and thus the cost of insurance.
Any free market libertarian should not be using any type of insurance. Afterall, insurance is a form of wealth redistribution.
This demonstrates a profound misunderstanding of both the concept of "wealth redistribution" in respect to libertarian philosophy and the concept of insurance itself.
First, insurance is not "wealth redistribution" even in the limited sense described below, at least when it's not actually being turned into some sort of forced-"charity" scheme via regulation. The projected value of an insurance subscription is equal to the projected value of the premiums being paid (less overhead and the insurer's profit margin, of course, just as with any other service). You're neither subsidizing nor being subsidized by your fellow insurees. What you get from insurance—what makes it worthwhile enough to justify the overhead—is that people tend to prefer that their future costs be predictable. Insurance takes a high-cost, low-probability future event and, by pooling it with many similar events, turns it into a low-cost, predictable event in the form of periodic insurance premiums. Critically, risk is conserved with respect to each insuree; no one pays extra to subsidize anyone else's above-average projected cost (risk). In retrospect, of course, some will be compensated more than others depending on the actual circumstances, but ex ante no insurees can be said to benefit at others' expense.
Second, "wealth redistribution" is only a problem when it's involuntary. Donating to charity is perfectly consistent with libertarian philosophy, and something many libertarians do frequently. When libertarians speak negatively of "wealth redistribution" they're referring to redistribution by force, involuntarily, which is an entirely separate matter. The force is what makes it wrong, not the redistribution.
This defeats the general purpose of medical insurance (which IS for the healthy to subsidize the sick).
No, the purpose of insurance, medical or otherwise, is for like-risk individuals to form a pool and trade assessed risk for equivalent, but predictable, premiums. Subsidies have nothing to do with it.
If you want to join a program whose purpose is to subsidize health care for high-risk individuals ("the sick"), then donate to a suitable charity. That's what they're there for.
If you want to force others to contribute to such a program, well—that's known as theft, or extortion, or some other equally unpleasant name depending on how you go about applying said force. Gain enough popularity and you many even manage a "respectable" name like tax or regulation, but it's the same thing in the end.
As a libertarian I have no interest in elimination either insurance or charity. My only concerns are that the labels be properly applied, and that no one be forced to participate in either. (Or forced not to participate, as happens when regulations prohibit offering true insurance in favor of some hare-brained charity-mislabeled-as-insurance scheme.)
The bad is that private insurance companies are likely to eventually *require* you to get a DNA sample, and possibly reject you if they determine your genes predispose you to old-age diseases.
That's only "bad" if you turn out to be predisposed, in which case your higher risk will no longer be subsidized and you'll have to pay fair premiums in proportion to your risk. For the majority who lack such predispositions, however, this is good news, as it means the cost of providing normal insurance will decrease. (And that, via competition, the price should decrease as well.)
What makes you think they can keep up this rate of sales? This was a one-time promotional fluke. There are only so many people interested in their game, and most of them have probably taken advantage of this offer to get it at a bargain price, meaning that they won't be paying retail in the future.
So you wasted a bunch of tax money on vaccinations your citizens don't want, creating a wasteful surplus within Canada and making it more difficult for the non-Canadians who actually want the vaccinations to get it? Great going there. That really shows how government management improves the efficiency of health care.
The right to get a disease is an obvious consequence of self-ownership (a.k.a. liberty, self-determination). Others can use force in self-defense if and when your presence creates a reasonable expectation of irreversible harm, and if you pass the disease on the recipient can sue you for damages after the fact. That's it, at least in civilized societies. Preemptive "justice" driven by fear, in the absence of immediate danger, is anything but.
The resolution of any recording medium is naturally finite, even before compression, bounded by the Shannon limit and the signal-to-noise ratio of the equipment. For that matter, even sound in air has a similarly bounded resolution, so even a perfect capture of a real-world tone would not require an infinite number of bits. For perfect quality in the context of a human listener, the maximum necessary bandwidth—given ideal compression—cannot be greater than the throughput of the neural channels communicating sounds from the ears to the brain, which is much less than 1.4 Mbps.
However, just as real-world signals lack mathematical regularity, real-world compression algorithms need not preserve all the information in the original signal. It is sufficient to preserve sufficient information that no human listener can tell the difference between the compressed signal and the original. Modern compression algorithms can generally achieve this with no more than about 160 kilobits/second.
Finally, lossless encoding schemes like FLAC demonstrate that most CD-quality audio can be compressed by 50% or more without any loss of quality whatsoever.
True, there are shorter encodings. However, normally one includes the complexity of the decoder when considering the efficiency of a compression algorithm, and I think you'd find that a practical decoder than can handle arbitrary English descriptions (and particularly shorthand) takes more bits than one that can only handle formulas.
The normal representation for considerations of programmatic complexity is the binary lambda calculus, but a full explanation would be a bit off-topic for this discussion.
Slashdot Mirror
Before you start leveling such claims, I'd like to see the percentage of road-maintenance costs which come from vehicle registration fees (in the GP's jurisdiction). It has been my impression that most of the cost is covered by gas taxes, in which case the GP is indeed paying for the roads.
It is highly unlikely the GP's total tax burden is less than that required to cover his/her own share of the maintenance cost. Even if that were the case, however, the government started this fight by instituting taxation; only by ending that practice permanently, and making full reparations, can they regain the moral authority to demand that would-be freeloaders stay off their property. Until then it comes down to an individual and an organization which each consider the other's actions criminal; you can't expect to be taken reasonably while appealing to just one party to change its ways (trespassing) and ignoring equivalent or greater offenses (theft) from the other side.
Sure, fast charging would take an unreasonable amount of power if you're just plugging it into a standard outlet. However, you don't need that kind of power 24/7—just for the 10 minutes during which you're charging the vehicle. That implies some kind of buffering; you have a local power station that draws about 10 amps @ 220 volt to recharge its internal capacitors over a 24-hour period, and then dumps all that energy into the vehicle's batteries on demand. A trivial implementation of this would be to have two sets of batteries, one slow-charging and one in use, and just swap them when the in-use batteries are drained. This could reasonably be made a 10-minute job with an efficient design.
I considered that, but the odds that someone could break the hash algorithm to the point that it becomes useless for qualifying the recordings as genuine does not seem to me to be significantly better than the odds that the records are made useless in some other way, e.g. by electronic or physical destruction, or that the recorded conversation was fabricated in the first place.
Note, also, that the methods devised to generate collisions for existing hashes (MD5, SHA1) require control of "invisible" regions in both files. There are no "invisible" areas in an audio recording, if you do it right—hash the raw samples, and limit the hash to the more significant bits—and anyone attempting to fabricate evidence after the fact would have to generate a collision with the original recording, not one of their choice.
Motivation aside, it's not slander if it's true.
Regarding the motivation, competitors can be a great source of negative information to counter a company's (possibly excessive) self-promotion. Between the two you have all the data you need as a starting point for your own fact-checking.
The "government" gives carriers a lot of leeway by protecting them from liability for the content they carry.
Only because they incorrectly consider them liable in the first place. They neither provide the content, nor have any reason to know the significance of the bits passing through their lines. Merely providing Internet service should not require some special "common carrier" (or whatever) status to shield one from liability. The only way one should be able to incur such liability is by choosing to continue providing service after becoming aware that doing so contributes to a specific crime.
It's not always a literal gun to the head that takes choice away.
Perhaps not, but "leveling the playing field" via laws (incl. regulations, taxes, etc.) does involve a literal gun to the head, if you refuse to comply. Nothing but a similar threat of literal force could possibly justify that.
Level the "playing field" all you want by peaceful means—just don't be the first to resort to coercion. In this case there is a simple, non-aggressive solution: if you can garner sufficient popular support to step in and forcibly tell ISPs what to do with their network, surely you can turn that same popular support toward forming a self-funded co-op offering Internet service on your preferred terms. Competition trumps conflict every time.
Right. I wasn't addressing the problem of actually destroying all the copies, just the conflict between immutability for audit-trail purposes and the need to delete certain conversations.
For dealing with the copies, a good start would be the suggestion made my another commenter to encrypt the archived copies of the conversations (with unique keys) and hold the keys in a separate database. That way you could back up the conversations themselves by whatever means are most efficient, and to effectively destroy the data you only have to erase the keys. To ensure full destruction you still need to track all the unencrypted copies, of course—and ensure there are no unauthorized copies of unencrypted recordings or conversation keys—for which there is no simple technical solution. That is a matter for policy, procedure, and manual auditing. Fortunately, most of the recordings will never need to be decrypted, so there shouldn't be many unencrypted copies floating around.
If they can figure out a way to delete a 'prohibited conversation' they could theoretically modify the data too.
Technically there is no need to make the conversations themselves immutable. You just need to be able to verify that the recording you have is the one which was originally recorded. A one-way hash can serve this purpose. For each recording, store the conversation itself in an erasable/mutable medium, but record a hash of the conversation in append-only storage (with multiple distributed backups). If you need to show that the recording is legit, compare it with the hash. If you need to delete something, record the deletion in the append-only medium and then remove it from the mutable storage. The hash will remain, but you can't use the hash to obtain information about the conversation without the original recording.
Bonus: You can recognize unauthorized deletions by comparing the mutable and immutable records.
"Reasonably"? I don't think so. I'm fairly sure that something can't be a derived work if it wouldn't qualify for copyright on its own. For example, if I say "The length of your comment is 71 characters", that sentence is derived from your (copyrighted) comment, but it's just a fact and thus not a derived work. Similarly, a torrent file may contain information (hashes & lengths) derived from a copyrighted work, but that information would not qualify for copyright on its own merits—being nothing more than abstract facts about the referenced data—and thus should not be considered a derived work.
Claiming that torrent files are derived works is analogous to claiming that bibliography entries are derived works; both are purely a set of facts which exist only for identification purposes.
(Notice that there's an odd form of double-think involved in allowing copyright for digital works and not facts when anything in digital form is ultimately nothing more than a collection of facts. Of course, allowing copyright over pure facts would be ridiculous, so the only way to resolve the issue is to recognize that it is inconsistent to allow copyright over anything—not that I expect the pro-copyright group to give up over a little matter like their position being inherently self-contradictory.)
This debate has seemingly devolved to an argument over word choices...
It doesn't matter whether the "set of uniform pools" model is the ideal model or the practical one. The point is that the two models are just different ways of describing the same thing.
What I described is the point of insurance from my perspective. If you want to take plain barter (or charity, or forced redistribution) and call it insurance, fine—but we'll still need a word to describe a contract which trades risk for equivalent periodic cost. Any suggestions? (I like "insurance"; it's traditional.)
The ideal free market—the standard libertarian one, not one of the others people sometimes advocate—does not depend on having perfect information, or on the absence of non-aggressive barriers to entry. In fact, improving the available information is one of the basic functions of the market. It does depend on a complete lack of aggression, however, so achieving it is equivalent to eliminating all crime, which clearly isn't going to happen in your lifetime or mine. Fortunately, we don't need an ideal free market for competition to work. We just need something "close enough", and for what we're discussing "close enough" isn't really all that close at all. A lack of aggressive (regulatory) interference would be quite sufficient on its own.
You assume like-risk, I don't. Furthermore, I don't assume that pools are formed by individuals. I merely assume that a demand exists, and that there is a product to fill that demand.
There is absolutely no significant difference between what you said and what I said in this area. Ignore the bit about "like-risk"; it's just distracting you from the big picture. It comes from the abstract model of insurance where each insurance contract is placed into a "pool" with many other contracts with similar risks and premiums. In practice, of course, insurees just pay premiums in proportion to their risk, and no one bothers to keep the actual reserves separate for different levels of risk. It works out the same in the end. The "individuals" part is similar; I never assumed or claimed that insurance only applies to individuals acting alone, although that was the original context (re: medical insurance). However, only individuals make decisions to take out insurance, whether they're acting on their own or as representatives of multi-national corporations. There's no point in considering the actions of an organization separate from the actions of the individuals who run it.
You also misunderstand the reason that insurance works. It works BECAUSE cost is externalized onto all the people who pay for more insurance than they need.
Costs, yes, at least after the fact. However, that misses the point. What matters with insurance is risk (ex ante probability-weighted projected future costs), and in terms of risk there is no subsidy. People don't know how much they'll need; that's why insurance exists, and why the savings-account concept doesn't work. However, in a free market no one buys pays higher premiums than they feel is necessary to cover their risk. In general, no one can say beforehand that a given insuree is paying more for their insurance than is required to cover their projected cost. If they could then they would offer that person a plan with the same coverage at a lower premium, and competition would eliminate the subsidy.
Finally, you also seem to misunderstand my last comment.... Sometimes, it is cheaper for society as a whole to subsidize the prevention of catastrophic events ... rather than mop up after it.
I understood your comment just fine, thank you. Disagreement is not equivalent to misunderstanding. The reason I disagree is simple: when you say that "society as a whole" should do something, when the individuals that make up that society have not freely chosen to do so, what you really mean is "some members of society should force the rest to do this, for the good of the whole (in my opinion)." Acting on that statement, however, is completely incompatible with a free society. It divides people into two classes: those who make the rules, and those who must follow them. Whether these classes are determined by birth or status or majority makes not difference; no one has the right to govern another by force.
Anyway, "society" doesn't own property, and thus does not have a "balance sheet". By pretending it does you are ignoring the boundaries between one individual's property and other's. It may seem less expensive to me for you and I to combine our purchases for a better price, but even if I have every reason to believe that I'm right I still can't make you participate. Which is good, because I may not be right—and the whole point of private property is that the owner makes those decisions. What you are proposing is the same, just at a larger scale.
I'm not a Social Darwinist. A Social Darwinist would argue that charity is wrong as well because it allows the weak to survive. (The label "Social Darwinist" is often used to demonize one's opponent, but you are unlikely to encounter one in practice. There are a few out there, though.)
Your argument is a good reason to donate to charity, but you appear to be taking it as a justification for theft. That is not something which you have any right to authorize. If you can persuade others to donate, great. If not, that is their choice to make, not yours.
Nonsense. The demand for insurance is that it is easier to mitigate risk if you pay a small regular chunk of money you don't need in return for an occasional large chunk of money when you need it. The supply of insurance exists because entities are willing and able to pay more than the compound cost over a large set of entities.
Businesses like it because it simplifies their accounting and long-term planning. Not to mention that it frees up resources that don't have to be held in reserves. Individuals like it because an unlikely, but possible catastrophe doesn't mean that they're fucked financially until they die.
Isn't that exactly what I said? Yes, I believe it was: "...the purpose of insurance, medical or otherwise, is for like-risk individuals to form a pool and trade assessed risk for equivalent, but predictable, premiums." In other words, to "mitigate risk."
There are universal benefits to people and businesses having insurance. To the point that it is cheaper to hand out insurance than deal with the after-effects of not having people and businesses insured.
Sure it's cheaper—when the costs are externalized onto all the people forced to pay for more insurance than they need or want, either directly or indirectly via taxes. Everything seems cheaper when you don't have to fund it yourself. That doesn't even consider the additional dead-weight loss of violating basic social rules like private property and self-determination.
There is no such thing as 'like-risk' -- the question is to what extent we can factor known risk factors into premiums.
The idealized model of insurance places individuals into different "pools" depending on their individual risk, with each pool corresponding to a specific risk and equivalent premium. It is assumed that there are sufficient individuals within each pool for things to average out. However, that's just an abstraction. In the real world risk is a continuous variable, so every individual would be in a separate pool (with no averaging). Instead, everyone pays into a single pool in proportion to their respective risk. It works out the same in the end.
Effectively, those who incur reimbursable expense are subsidized by those who don't.
When you look at things from an ex post cost basis, yes. However, that's an artificial point of view; when considering whether to invest in insurance you don't know the shape of future events. The whole point of insurance is to take that uncertainty and turn it into a stable situation from an ex ante point of view, where risk, not cost, is the determining factor. And in terms of risk there is no subsidy; given all the information known ex ante there is no reason for any insuree to believe they are subsidizing anyone. All insurees within a given pool are equally likely to be compensated. (Or, avoiding the "pool" abstraction, all insurees have a probability-weighted projected compensation which is proportional to their premium.)
No one willingly joins an insurance program expecting to pay in substantially more than their individual risk—not when competitors are free to offer them lower premiums—and no insurance program could survive for long without balancing risk if low-risk individuals are permitted to opt out. Ergo, no voluntary insurance program can retain ex ante subsidies over the long run.
What do you propose that those people whose premiums woud [sic] be impossibly high (or who are insurance pariahs) should do?
If they truly cannot get insurance, then their only option is accept the risk themselves and hope they don't get sick or injured.
If they do anyway, and cannot afford health care even after exhausting all available options for charitable assistance, then they are in exactly the same position as one who has a disease or injury for which there is no known cure—a position which, needless to say, every person who has ever lived (or is currently alive) has faced (or will face) at some point.
If you could predict with 80% accuracy whether somebody's house would burn down, then almost everybody could get dirt-cheap fire insurance (which they wouldn't buy anyway since they wouldn't need it), and a small number of people wouldn't be able to afford it and would lose everything they have in a fire.
Rather, the houses likely to burn down would be unoccupied, and probably demolished, and the remaining houses would be be insurable at much lower rates. Everyone wins. Well, some people would be out the cost of a house and need to find new shelter, but at least they won't be caught in a fire.
In the medical case, of course, you can't leave your "house"—but you can take steps to mitigate any diseases you may be genetically predisposed to. Analogies aside, however, by the time your genetics have been determined it's too late for (real) insurance; either you're known to be within normal risk levels, or you're looking for charity, not insurance. The time to take out insurance against genetic risk factors is before conception. At that point it's your parents' DNA that determines the risk, and thus the cost of insurance.
Any free market libertarian should not be using any type of insurance. Afterall, insurance is a form of wealth redistribution.
This demonstrates a profound misunderstanding of both the concept of "wealth redistribution" in respect to libertarian philosophy and the concept of insurance itself.
First, insurance is not "wealth redistribution" even in the limited sense described below, at least when it's not actually being turned into some sort of forced-"charity" scheme via regulation. The projected value of an insurance subscription is equal to the projected value of the premiums being paid (less overhead and the insurer's profit margin, of course, just as with any other service). You're neither subsidizing nor being subsidized by your fellow insurees. What you get from insurance—what makes it worthwhile enough to justify the overhead—is that people tend to prefer that their future costs be predictable. Insurance takes a high-cost, low-probability future event and, by pooling it with many similar events, turns it into a low-cost, predictable event in the form of periodic insurance premiums. Critically, risk is conserved with respect to each insuree; no one pays extra to subsidize anyone else's above-average projected cost (risk). In retrospect, of course, some will be compensated more than others depending on the actual circumstances, but ex ante no insurees can be said to benefit at others' expense.
Second, "wealth redistribution" is only a problem when it's involuntary. Donating to charity is perfectly consistent with libertarian philosophy, and something many libertarians do frequently. When libertarians speak negatively of "wealth redistribution" they're referring to redistribution by force, involuntarily, which is an entirely separate matter. The force is what makes it wrong, not the redistribution.
This defeats the general purpose of medical insurance (which IS for the healthy to subsidize the sick).
No, the purpose of insurance, medical or otherwise, is for like-risk individuals to form a pool and trade assessed risk for equivalent, but predictable, premiums. Subsidies have nothing to do with it.
If you want to join a program whose purpose is to subsidize health care for high-risk individuals ("the sick"), then donate to a suitable charity. That's what they're there for.
If you want to force others to contribute to such a program, well—that's known as theft, or extortion, or some other equally unpleasant name depending on how you go about applying said force. Gain enough popularity and you many even manage a "respectable" name like tax or regulation, but it's the same thing in the end.
As a libertarian I have no interest in elimination either insurance or charity. My only concerns are that the labels be properly applied, and that no one be forced to participate in either. (Or forced not to participate, as happens when regulations prohibit offering true insurance in favor of some hare-brained charity-mislabeled-as-insurance scheme.)
The bad is that private insurance companies are likely to eventually *require* you to get a DNA sample, and possibly reject you if they determine your genes predispose you to old-age diseases.
That's only "bad" if you turn out to be predisposed, in which case your higher risk will no longer be subsidized and you'll have to pay fair premiums in proportion to your risk. For the majority who lack such predispositions, however, this is good news, as it means the cost of providing normal insurance will decrease. (And that, via competition, the price should decrease as well.)
$100,000 in a week (implying $5.2 million/year)
What makes you think they can keep up this rate of sales? This was a one-time promotional fluke. There are only so many people interested in their game, and most of them have probably taken advantage of this offer to get it at a bargain price, meaning that they won't be paying retail in the future.
So you wasted a bunch of tax money on vaccinations your citizens don't want, creating a wasteful surplus within Canada and making it more difficult for the non-Canadians who actually want the vaccinations to get it? Great going there. That really shows how government management improves the efficiency of health care.
The right to get a disease is an obvious consequence of self-ownership (a.k.a. liberty, self-determination). Others can use force in self-defense if and when your presence creates a reasonable expectation of irreversible harm, and if you pass the disease on the recipient can sue you for damages after the fact. That's it, at least in civilized societies. Preemptive "justice" driven by fear, in the absence of immediate danger, is anything but.
I did say that mine was a contrived example.
The resolution of any recording medium is naturally finite, even before compression, bounded by the Shannon limit and the signal-to-noise ratio of the equipment. For that matter, even sound in air has a similarly bounded resolution, so even a perfect capture of a real-world tone would not require an infinite number of bits. For perfect quality in the context of a human listener, the maximum necessary bandwidth—given ideal compression—cannot be greater than the throughput of the neural channels communicating sounds from the ears to the brain, which is much less than 1.4 Mbps.
However, just as real-world signals lack mathematical regularity, real-world compression algorithms need not preserve all the information in the original signal. It is sufficient to preserve sufficient information that no human listener can tell the difference between the compressed signal and the original. Modern compression algorithms can generally achieve this with no more than about 160 kilobits/second.
Finally, lossless encoding schemes like FLAC demonstrate that most CD-quality audio can be compressed by 50% or more without any loss of quality whatsoever.
I just noticed that Slashdot ate my piecewise-defined formula. The step function is:
u(t) = {
0 when t < 0,
1 when t >= 0
}
Also, my formula was 35 bytes, not 25, although—as others have pointed out—it can be represented in fewer bytes than I showed. (35 bytes = 280 bits)
True, there are shorter encodings. However, normally one includes the complexity of the decoder when considering the efficiency of a compression algorithm, and I think you'd find that a practical decoder than can handle arbitrary English descriptions (and particularly shorthand) takes more bits than one that can only handle formulas.
The normal representation for considerations of programmatic complexity is the binary lambda calculus, but a full explanation would be a bit off-topic for this discussion.