I agree it's useful. But there are other ways of thinking about it that achieve the same result. E.g., without changing any of the actual worked out (as opposed to lazily evaluated) math, you could use the exact same steps, but instead of saying "infinitesimal" you could just say "and this continues to be true further than we can measure", A would would need to be invented, or revived, with that meaning. Possibly even a redefinition of infinitesimal would work.
Similarly for infinity. That could just mean "and we can't see an end to it". I'm not objecting to "lazily evaluated definitions", I'm objection to definitions that thingify the endpoint of those definitions.
This would throw out much of math, I admit. IIRC GÃdel's incompleteness depends on actual infinity. Only finite systems would be allowed, but the ability to define the bounds wouldn't be required. This is not at all the same as saying there are no bounds.
The US system is both understaffed and underfunded, but it's also true that the funding that exists if improperly allocated. Class sizes need to be reduced. That means more teachers. Ideally a class below the junior year of high school should have 15 or fewer students, with the lower grades needing smaller class sizes. Performance tends to start degrading above about 12 students, but sometimes you need the larger class sizes for some projects, and up to around 15 students the degradation in quality doesn't start becoming severe in high school students. The Junior and Senior years of high school can handle some classes as small lectures, with around 100 students, but you need several aids and the classes MUST be electives. If the students don't already want to learn the material, then it won't work. And you still need smaller classes for consolidation of the learning.
Schools should have a on-site nurse. And there should be a couple of administrators. And the schools shouldn't be too large. (Again, size depends on age of the students.) Etc.
This cannot be done within the current budget, so schools are underfunded. But you will notice that I didn't mention anything about electronics. It is my contention that most electronics interfere with education except in self-directed study. Computers are only appropriate if you are building them in class. (That's not unreasonable. Simple computers can be built without using electronics.)
That said, the limitations on electronics should start being relaxed around 6th grade, with various shop classes...mandatory for all students. By their junior year in high school everyone should have rebuilt a simple four-cycle engine, but elementary chemistry and thermodynamics related to it should be handled in elective classes. Wood working should be handled in 7th grade, or perhaps earlier. Simple cooking in 6th grade, or perhaps 7th. Etc.
Elementary education should be broad, though shallow. Building and using an abacus should be 2nd grade. Theoretical considerations around the practicals should be electives. Build an autoharp or some such. (I'm not sure how difficult an instrument should be chosen, but kits make this pretty simple. But it does need to be a stringed instrument.) And accompany it with electives on math, music theory, etc. All this at a very simple level.
Yes, but why would anyone believe in infinity? Just imagine the idiocy involved in taking all the integers and painting one copy of the red, and all the even numbers in another copy blue.
YOU CAN'T PAINT A NUMBER!! That's a invalid operation on that type of entity.
Infinity was a good idea when it meant "I can't put any limit on it". Taking that kind of functional definition, and a lazy evaluator, and objectifying them is blindingly idiotic. It's a very convenient way of thinking about things too large or too small to hold in your mind, but that doesn't make it real.
<rant> Only the a finite subset of the integers and rational numbers formed by them have any physical mapping. When you try to invoke the real number line you end up having to hypothecate continuity, which is not demonstrable and which I do not believe to be a fact in any area of the actual universe. There is argument whether the fabric of space-time has any meaning below about 10^-33 cm, and I'm willing to accept that, or even something smaller, as the base level of reality, but to assume continuity is unreasonable. And I assume that there's also a largest number. My estimate of this is the powerset of the number of energy states that can be contained within the universe. It's an extremely large number, but it's not at all the same as saying there's no upper limit. I admit that I could have underestimated the size of the largest number, but I'd need proof before I'd accept that there was no upper bound. And having a lazy evaluator for a successor function doesn't give you permission to assume that it keeps working forever.
Well, the old poll tests were racist because the person giving the test could look at your skin before deciding whether you had passed or not, so they really WERE racist. Just having a correlation with race doesn't make something racist, but being determined by "race" does.
P.S.: Race is a social construct, not a genetic or physiological one. It has as much to do with the shape of the nose and texture of the hair as color of the skin, and there is NO reliable genetic difference. There are population mean differences, but that's not the same thing. E.g., very few "white" or "oriental" people have sickle-cell anemia genes. But some do.
And because I'm generally law abiding, I've never bought an e-book. I have a few thousand, mainly from The Gutenberg Project. (Well perhaps I've bought one or two that clearly had no DRM. Baen Books http://www.baen.com/ used to sell some that way, but I find on-line advertising for books to be nearly intolerable. Somehow the people who do it don't get the idea.)
Of course, a contributing factor is that I don't like the form factor. I bought a Nook through Radio Shack, back before it had turned into a phone store, and it sort of works, but it doesn't feel comfortable to read. People talk about reading on their phones, but those screens are so small that I can't understand it. Also e-books commonly botch up graphics horribly...and that's weird. The same book in html will often be fine. Perhaps commercial works are put together better, but until they stop using DRM I'll never know.
But *do* remember that they aren't just going to collect the information, they're going to sell it to others who will correlate it with information collected on other sites. Anonymized data generally isn't. Usually you can determine a particular person from only a very few pieces of nominally anonymous data.
OTOH, you also have to ask yourself "How much is this invasion of privacy going to add to what has already been collected about me?" You can't really consider even officially secret information, like say health records, to be safe. It's usually allowed that some "sufficiently anonymized" extract of the data can be shared, and that "sufficiently anonymized" extract is usually enough to tie them to you personally. How much groups bother to do this isn't clear. So far all I've heard is of populations being targeted, not individual people, and for that even actual anonymous data would probably suffice. (OTOH, another poster claimed that custom ads were targeted at individuals. If you believe him, then the anonymity is frequently violated...and it seems readily doable.)
Please note, though, that much of this information about you you can't control, because it's in the hands of other people, and the groups that they are part of care even less for your privacy than the laws require them to.
I don't follow digital currencies, so I consider it possible that the government accusations are correct. Unfortunately, they lie so often that my default assumption has become that they are lying.
So I'm assuming that this is another blatant abuse of power. I'm not sure, but what trustworthy source could I check against? If the accusations are correct, neither he nor his supporters are going to admit it. And the government statements are uselessly unreliable.
The claim that that's always been true is false. There have been times when vigilante justice was much worse then official justice. But they weren't the times when the official justice had a lot of power behind it.
But they knew about it ahead of time, so they intentionally broke some users systems. You can claim that the users systems were operating in an unsupported manner, but that doesn't deny the claim that they intentionally broke them.
Each time there was a Tsunami it was unlikely to be as large as the one that damaged Fukushima. So people correctly guessed that they were protecting against a larger Tsunami than was likely to hit. And they were right. But it was the unlikely event that hit them.
Now afterwards it's easy to look and say there are historical records of even larger ones. But it's almost always easy to find ignored earlier signs of danger...afterwards. Next time it won't be a Tsunami, it will be something else, a cliff collapsing on the plant, too many alarms going off at once, *something* else that was low probability. And people are lousy at evaluating long term risks. They're even lousy at making tradeoffs of where they should add additional protection, and where they're already over supplied. I have my doubts that the solid concrete domes around US plants are the proper trade-off, when that same money invested in other protection measures might be much more useful. But I don't know. Clearly you need protection, but how important is it that you be able to survive a direct strike by a large airliner?
I actually *do* see evidence that MS is breaking installed systems on purpose. In fact there's proof. Now as to whether the way that they are doing it is legally actionable, I have no clue. Corporations seem to be able to get away with lots of things that would be criminal if an individual did so.
P.S.: (from the summary) "...Microsoft knew ahead of time the update would cause problems for some users but decided to do nothing about it. The update fixes a problem that stops BitLocker encrypting drives because of service crashes in svhost.exe. The update only causes a problem with ASUS..."
Well, the question is "how accurate is the map?". I agree with you that it won't understand the words, only the relationship between them. Understanding would require both a body and a compatible goal structure. But if you have an accurate map, then you can learn the meaning quickly once you get the appropriate equipment. The problem is, I'm not sure the map is any good. Of course, that depends entirely on how they abstract it.
Well, actually my guess is that after it masters the romance fiction genre it will be generalized to other genres. Possibly even eventually math texts. Imagine how that could improve word problems.
Actually, wood is sustainable. But you can't use it too fast or it stops being sustainable. Traditional farms had a wood lot, and the farmers sized the wood lot to how much wood they expected to use on the average. Done right this was sustainable indefinitely. Some farms kept this up for centuries, and they didn't stop because the system wasn't sustainable, but because their kids didn't want to be farmers, or taxes got too high, or cash crops were too attractive.
This isn't saying that wood is a clean fuel. It can be if burned properly, but that's a major project in and of itself. And if done right it's environmentally neutral, neither positive nor negative.
Now you can find lots of poor implementations of this idea, but that doesn't prove that it's inherently bad, it proves that most people who do it either don't know how to do it, don't care how to do it, or can't afford to do it right.
AGW isn't an existential threat to humanity, but it may well be an existential threat to technological civilization. (Probably not, but that this is true isn't obvious.)
OTOH, nuclear power, if you except weapons, also isn't an existential threat to humanity, but it could sure do a lot of damage.
But I happen to like technological civilization and would like to keep it around. This means that we need to be careful about things like nuclear power and genetic modification. It emphatically doesn't mean to avoid them, but it means to take reasonably sufficient precautions. And nobody seems willing to do that.
I'm sorry, but until they get waste reprocessing plants working and reprocessing the waste, safety concerns are quite reasonable. Actually, even afterwards, but then they can be dealt with.
I will admit that there is lots of fear-mongering WRT nuclear power, but that doesn't mean that the proper response is to close your eyes and go la-la-la.
And NOBODY properly calculated decommissioning costs.
The difference is that the environmental costs of solar are born before you start using it. Many of the environmental costs of nuclear are only apparent when you try to retire the plant, or deal with spent fuel.
FWIW, these aren't problems that can't be addressed, they are rather problems that it's always unprofitable to address, and which aren't being addressed. Costs of decommissioning plants are inevitably so low-balled as to be totally fictitious. And the people who made them are either dead or retired by the time the fictitious nature becomes evident, so there's no down-side to *them*. And accurate estimates would make management unhappy.
The point about electric cars is very valid. There is going to be a tremendous demand for additional electric power. I'm not sure that solar isn't the answer, but roof-top solar is certainly insufficient. That molten salt plant being built in Dubai sounds promising. California, Texas, New Mexico, and Arizona have deserts that could handle several plants of that sort, and they *DO* produce power at night. But transmission lines would need to be improved.
OTOH, if Lockheed's fusion power isn't just PR, AND it's actually low in radioactive waste production, then that might be a better answer. Perhaps. Too many unknowns to estimate. (On both sides.)
But for fission plants to be a better answer, there are lots of long-term problems that need to be successfully addressed rather than ignored.
The problem is that every incident actually *IS* a fluke. And flukes are a lot more common that most people are willing to accept.
So saying "this incident was a fluke" isn't a comment saying it wasn't to be expected, but only "most reasonable people wouldn't expect it". Which is true, but not very useful.
When you NEED something to be secure against failure for a prolonged period of time, you need to guard against very low probability events. Probably in this case not improbable on the scale of giant meteor impact, but maybe half-way to that on a logarithmic scale. And that means you can't depend on what people see at a trivial risk, because people are extremely poor at making that kind of assessment.
IOW, it's not just "short-sighted economics driven management", though of course that makes things worse, but even careful engineers will underestimate long-term risks. The only way we have to handle this kind of thing is mathematical analysis, and that requires that we have a valid model that includes all risks. Whoops!
People are unable to accurately evaluate long term risks. This has been proven repeatedly. They also undervalue long-term gains. And this has also been proven repeatedly. And this makes accurately balancing them impossible....except in the cases where you have an accurate model that includes all costs and profits, with an accurate estimate of probabilistic frequency. Which we have only managed for simplified situations. (Even there we find strict limits. Chaotic systems aren't readily predictable. And that includes planetary orbits. Simpler systems, though, are more readily predicted "accurately within stated limits". Unfortunately, I don't believe that nuclear plants count as simple systems.
A lot of the problem with Fukishima was spend reactor rods being stored on site in storage intended for temporary use. Are you willing to bet that equivalent problems don't exist for US plants? To me it appears that the odds are that safety concerns are frequently avoided in the interest of economic operation. Also because there's nowhere better to put them.
Slashdot Mirror
I agree it's useful. But there are other ways of thinking about it that achieve the same result. E.g., without changing any of the actual worked out (as opposed to lazily evaluated) math, you could use the exact same steps, but instead of saying "infinitesimal" you could just say "and this continues to be true further than we can measure", A would would need to be invented, or revived, with that meaning. Possibly even a redefinition of infinitesimal would work.
Similarly for infinity. That could just mean "and we can't see an end to it". I'm not objecting to "lazily evaluated definitions", I'm objection to definitions that thingify the endpoint of those definitions.
This would throw out much of math, I admit. IIRC GÃdel's incompleteness depends on actual infinity. Only finite systems would be allowed, but the ability to define the bounds wouldn't be required. This is not at all the same as saying there are no bounds.
And you think that because you work on classified material, nobody else should be allowed to work where anyone might be able to see them?
Or what is your point?
The US system is both understaffed and underfunded, but it's also true that the funding that exists if improperly allocated. Class sizes need to be reduced. That means more teachers. Ideally a class below the junior year of high school should have 15 or fewer students, with the lower grades needing smaller class sizes. Performance tends to start degrading above about 12 students, but sometimes you need the larger class sizes for some projects, and up to around 15 students the degradation in quality doesn't start becoming severe in high school students. The Junior and Senior years of high school can handle some classes as small lectures, with around 100 students, but you need several aids and the classes MUST be electives. If the students don't already want to learn the material, then it won't work. And you still need smaller classes for consolidation of the learning.
Schools should have a on-site nurse. And there should be a couple of administrators. And the schools shouldn't be too large. (Again, size depends on age of the students.) Etc.
This cannot be done within the current budget, so schools are underfunded. But you will notice that I didn't mention anything about electronics. It is my contention that most electronics interfere with education except in self-directed study. Computers are only appropriate if you are building them in class. (That's not unreasonable. Simple computers can be built without using electronics.)
That said, the limitations on electronics should start being relaxed around 6th grade, with various shop classes...mandatory for all students. By their junior year in high school everyone should have rebuilt a simple four-cycle engine, but elementary chemistry and thermodynamics related to it should be handled in elective classes. Wood working should be handled in 7th grade, or perhaps earlier. Simple cooking in 6th grade, or perhaps 7th. Etc.
Elementary education should be broad, though shallow. Building and using an abacus should be 2nd grade. Theoretical considerations around the practicals should be electives. Build an autoharp or some such. (I'm not sure how difficult an instrument should be chosen, but kits make this pretty simple. But it does need to be a stringed instrument.) And accompany it with electives on math, music theory, etc. All this at a very simple level.
Yes, but why would anyone believe in infinity? Just imagine the idiocy involved in taking all the integers and painting one copy of the red, and all the even numbers in another copy blue.
YOU CAN'T PAINT A NUMBER!! That's a invalid operation on that type of entity.
Infinity was a good idea when it meant "I can't put any limit on it". Taking that kind of functional definition, and a lazy evaluator, and objectifying them is blindingly idiotic. It's a very convenient way of thinking about things too large or too small to hold in your mind, but that doesn't make it real.
<rant>
Only the a finite subset of the integers and rational numbers formed by them have any physical mapping. When you try to invoke the real number line you end up having to hypothecate continuity, which is not demonstrable and which I do not believe to be a fact in any area of the actual universe. There is argument whether the fabric of space-time has any meaning below about 10^-33 cm, and I'm willing to accept that, or even something smaller, as the base level of reality, but to assume continuity is unreasonable. And I assume that there's also a largest number. My estimate of this is the powerset of the number of energy states that can be contained within the universe. It's an extremely large number, but it's not at all the same as saying there's no upper limit. I admit that I could have underestimated the size of the largest number, but I'd need proof before I'd accept that there was no upper bound. And having a lazy evaluator for a successor function doesn't give you permission to assume that it keeps working forever.
Well, the old poll tests were racist because the person giving the test could look at your skin before deciding whether you had passed or not, so they really WERE racist. Just having a correlation with race doesn't make something racist, but being determined by "race" does.
P.S.: Race is a social construct, not a genetic or physiological one. It has as much to do with the shape of the nose and texture of the hair as color of the skin, and there is NO reliable genetic difference. There are population mean differences, but that's not the same thing. E.g., very few "white" or "oriental" people have sickle-cell anemia genes. But some do.
And because I'm generally law abiding, I've never bought an e-book. I have a few thousand, mainly from The Gutenberg Project. (Well perhaps I've bought one or two that clearly had no DRM. Baen Books http://www.baen.com/ used to sell some that way, but I find on-line advertising for books to be nearly intolerable. Somehow the people who do it don't get the idea.)
Of course, a contributing factor is that I don't like the form factor. I bought a Nook through Radio Shack, back before it had turned into a phone store, and it sort of works, but it doesn't feel comfortable to read. People talk about reading on their phones, but those screens are so small that I can't understand it. Also e-books commonly botch up graphics horribly...and that's weird. The same book in html will often be fine. Perhaps commercial works are put together better, but until they stop using DRM I'll never know.
But *do* remember that they aren't just going to collect the information, they're going to sell it to others who will correlate it with information collected on other sites. Anonymized data generally isn't. Usually you can determine a particular person from only a very few pieces of nominally anonymous data.
OTOH, you also have to ask yourself "How much is this invasion of privacy going to add to what has already been collected about me?" You can't really consider even officially secret information, like say health records, to be safe. It's usually allowed that some "sufficiently anonymized" extract of the data can be shared, and that "sufficiently anonymized" extract is usually enough to tie them to you personally. How much groups bother to do this isn't clear. So far all I've heard is of populations being targeted, not individual people, and for that even actual anonymous data would probably suffice. (OTOH, another poster claimed that custom ads were targeted at individuals. If you believe him, then the anonymity is frequently violated...and it seems readily doable.)
Please note, though, that much of this information about you you can't control, because it's in the hands of other people, and the groups that they are part of care even less for your privacy than the laws require them to.
I don't follow digital currencies, so I consider it possible that the government accusations are correct. Unfortunately, they lie so often that my default assumption has become that they are lying.
So I'm assuming that this is another blatant abuse of power. I'm not sure, but what trustworthy source could I check against? If the accusations are correct, neither he nor his supporters are going to admit it. And the government statements are uselessly unreliable.
Plea bargaining was an evil innovation, but it wasn't a patch on the unconstitutional fostering of corruption that asset forfeiture is.
The claim that that's always been true is false. There have been times when vigilante justice was much worse then official justice. But they weren't the times when the official justice had a lot of power behind it.
That's not "Vigilante justice", because it's done under the cover of authority.
But they knew about it ahead of time, so they intentionally broke some users systems. You can claim that the users systems were operating in an unsupported manner, but that doesn't deny the claim that they intentionally broke them.
Each time there was a Tsunami it was unlikely to be as large as the one that damaged Fukushima. So people correctly guessed that they were protecting against a larger Tsunami than was likely to hit. And they were right. But it was the unlikely event that hit them.
Now afterwards it's easy to look and say there are historical records of even larger ones. But it's almost always easy to find ignored earlier signs of danger...afterwards. Next time it won't be a Tsunami, it will be something else, a cliff collapsing on the plant, too many alarms going off at once, *something* else that was low probability. And people are lousy at evaluating long term risks. They're even lousy at making tradeoffs of where they should add additional protection, and where they're already over supplied. I have my doubts that the solid concrete domes around US plants are the proper trade-off, when that same money invested in other protection measures might be much more useful. But I don't know. Clearly you need protection, but how important is it that you be able to survive a direct strike by a large airliner?
You've got a long memory. I'd nearly forgotten about those episodes.
What's the missing word:
"Windows isn't done until _________ won't run." I can't remember. I want to say Lotus, but I'm not sure.
I actually *do* see evidence that MS is breaking installed systems on purpose. In fact there's proof. Now as to whether the way that they are doing it is legally actionable, I have no clue. Corporations seem to be able to get away with lots of things that would be criminal if an individual did so.
P.S.: (from the summary) ..."
"...Microsoft knew ahead of time the update would cause problems for some users but decided to do nothing about it. The update fixes a problem that stops BitLocker encrypting drives because of service crashes in svhost.exe. The update only causes a problem with ASUS
Well, the question is "how accurate is the map?". I agree with you that it won't understand the words, only the relationship between them. Understanding would require both a body and a compatible goal structure. But if you have an accurate map, then you can learn the meaning quickly once you get the appropriate equipment. The problem is, I'm not sure the map is any good. Of course, that depends entirely on how they abstract it.
Harry Harrison did a "viking time travel romance novel". The Technicolor Time Machine http://www.fictiondb.com/autho...
Well, actually my guess is that after it masters the romance fiction genre it will be generalized to other genres. Possibly even eventually math texts. Imagine how that could improve word problems.
Perhaps it depends on how you evaluate your potential clientÃle.
Actually, wood is sustainable. But you can't use it too fast or it stops being sustainable. Traditional farms had a wood lot, and the farmers sized the wood lot to how much wood they expected to use on the average. Done right this was sustainable indefinitely. Some farms kept this up for centuries, and they didn't stop because the system wasn't sustainable, but because their kids didn't want to be farmers, or taxes got too high, or cash crops were too attractive.
This isn't saying that wood is a clean fuel. It can be if burned properly, but that's a major project in and of itself. And if done right it's environmentally neutral, neither positive nor negative.
Now you can find lots of poor implementations of this idea, but that doesn't prove that it's inherently bad, it proves that most people who do it either don't know how to do it, don't care how to do it, or can't afford to do it right.
AGW isn't an existential threat to humanity, but it may well be an existential threat to technological civilization. (Probably not, but that this is true isn't obvious.)
OTOH, nuclear power, if you except weapons, also isn't an existential threat to humanity, but it could sure do a lot of damage.
But I happen to like technological civilization and would like to keep it around. This means that we need to be careful about things like nuclear power and genetic modification. It emphatically doesn't mean to avoid them, but it means to take reasonably sufficient precautions. And nobody seems willing to do that.
I'm sorry, but until they get waste reprocessing plants working and reprocessing the waste, safety concerns are quite reasonable. Actually, even afterwards, but then they can be dealt with.
I will admit that there is lots of fear-mongering WRT nuclear power, but that doesn't mean that the proper response is to close your eyes and go la-la-la.
And NOBODY properly calculated decommissioning costs.
The difference is that the environmental costs of solar are born before you start using it. Many of the environmental costs of nuclear are only apparent when you try to retire the plant, or deal with spent fuel.
FWIW, these aren't problems that can't be addressed, they are rather problems that it's always unprofitable to address, and which aren't being addressed. Costs of decommissioning plants are inevitably so low-balled as to be totally fictitious. And the people who made them are either dead or retired by the time the fictitious nature becomes evident, so there's no down-side to *them*. And accurate estimates would make management unhappy.
The point about electric cars is very valid. There is going to be a tremendous demand for additional electric power. I'm not sure that solar isn't the answer, but roof-top solar is certainly insufficient. That molten salt plant being built in Dubai sounds promising. California, Texas, New Mexico, and Arizona have deserts that could handle several plants of that sort, and they *DO* produce power at night. But transmission lines would need to be improved.
OTOH, if Lockheed's fusion power isn't just PR, AND it's actually low in radioactive waste production, then that might be a better answer. Perhaps. Too many unknowns to estimate. (On both sides.)
But for fission plants to be a better answer, there are lots of long-term problems that need to be successfully addressed rather than ignored.
The problem is that every incident actually *IS* a fluke. And flukes are a lot more common that most people are willing to accept.
So saying "this incident was a fluke" isn't a comment saying it wasn't to be expected, but only "most reasonable people wouldn't expect it". Which is true, but not very useful.
When you NEED something to be secure against failure for a prolonged period of time, you need to guard against very low probability events. Probably in this case not improbable on the scale of giant meteor impact, but maybe half-way to that on a logarithmic scale. And that means you can't depend on what people see at a trivial risk, because people are extremely poor at making that kind of assessment.
IOW, it's not just "short-sighted economics driven management", though of course that makes things worse, but even careful engineers will underestimate long-term risks. The only way we have to handle this kind of thing is mathematical analysis, and that requires that we have a valid model that includes all risks. Whoops!
People are unable to accurately evaluate long term risks. This has been proven repeatedly. They also undervalue long-term gains. And this has also been proven repeatedly. And this makes accurately balancing them impossible....except in the cases where you have an accurate model that includes all costs and profits, with an accurate estimate of probabilistic frequency. Which we have only managed for simplified situations. (Even there we find strict limits. Chaotic systems aren't readily predictable. And that includes planetary orbits. Simpler systems, though, are more readily predicted "accurately within stated limits". Unfortunately, I don't believe that nuclear plants count as simple systems.
A lot of the problem with Fukishima was spend reactor rods being stored on site in storage intended for temporary use. Are you willing to bet that equivalent problems don't exist for US plants? To me it appears that the odds are that safety concerns are frequently avoided in the interest of economic operation. Also because there's nowhere better to put them.