But there is. If it's predicted that the sea level will rise this much by the end of the century, and given how log it's going to take to actually shut down and clear a nuclear power plant site to a level it's safe to let it flood, the time to start is now for those plants we believe to be effected by such a oceanic rise.
The obstacles to decommissioning a nuclear power plant are entirely administrative; you treat them like laws of nature. That's what makes your alarmism so ridiculous even if the predictions were sound and proven.
Anyway, there is nothing wrong with having a nuclear power plant even below sea level; the Dutch do.
No action is being taken because the people responsible don't look out that far in the future.
Good! I hope they will continue to act rationally rather than give in to FUD.
but they will be dead before it really comes to pass. So it joins the list for "the next generation" to deal with.
It's unfortunate that there really isn't a threat of impending flooding and disaster; maybe if there were, it would spur the bureaucrats and "environmentalists" into action. Causing a bit of sea level rise would be worth it just for that.
In any case, decommissioned or not, these plants won't be producing power anymore in 30 years, let alone 80 years.
NOAA's worst-case scenario, meanwhile, predicts that the oceans will rise nearly 7 feet in the next 86 years.
NOAA worst case for 2046 is 21.48 in, or less than 2 ft. The graph below is misleading, and the "0.9 ft Storm" bar is a guess by the HuffPost journalists, not a scientific result.
The whole thing is pure FUD anyway. Nothing magical happens when nuclear power plants end up below sea level; you can build a dike around them, a negligible expense compared to the rest of the plant. The Dutch have built nuclear power plants below sea level to no ill effect.
Fukishima's problem wasn't flooding, it was poor design and outdated safety measures.
The idea that these nuclear power plants are still relevant in 86 years should scare people more than any sea level rise. All those nuclear power plants are completely obsolete. If they need to be torn down and rebuilt elsewhere with new, safer, more efficient technology, we're all better off.
No, my position is that the technical terms used in an abstract are defined in the paper. You cannot just take a sentence out of an abstract and interpret it in a way that's convenient for an argument you want to make.
Since the paper redefines the meaning of the terms "expressing a position on AGW", "endorse", and "consensus position", the sentence you quoted means something different from what you implied it meant. The error here is yours, not the authors' (the authors can be faulted, however, for being misleading).
(Of course, abstracts quite frequently do contradict the body of a paper; authors try to get away with sensationalizing in the abstract because abstracts tend not to be very carefully checked by reviewers because most scientists, unlike you, understand that the abstract by itself doesn't show anything.)
Furthermore, nowhere have I made the claim that the semantics of Nile in its current form is a perfect replacement for any language for scientific computation, as opposed to the thought that there could be some lessons to be learned.
You assume incorrectly that these are "lessons" people working in these areas don't already know and still need to learn.
On the other hand, you'll find that actually implementing something like Nile correctly and efficiently is extremely hard; so hard, that by the time you're done, the world has already moved on to other tools.
I realize that depriving people of their beloved imperative loops might seem cruel, but designing the language in a way that would make obvious coding styles easily executed on vector machines seems a bit saner to me than allowing people to write random loops and then either hope that the vectorizer will sort it out
People do write large scale numerical code using functional abstractions, automatic vectorization, pipelines, data flow, and all that. There are excellent libraries for those purposes for both C++ and Fortran.
The reason people still stick to C++ and Fortran rather than going to a fully automatic system is threefold. First, it isn't necessary: modern versions of C++ and Fortran already have all the necessary high level support; except for an unfamiliar syntax, Nile doesn't add any functionality. Second, there is tons of legacy code that people want to interoperate with easily and without much hassle; switching to a new language with new conventions makes that hard. Third, with tools like Nile, sooner or later, users hit cases where the automated tools don't work anymore, and then they have to go back to figuring out how to link in C/C++/Fortran anyway.
Language and systems people often have the notion that numerical and parallel software developers do what they do out of ignorance, when in fact it's a rational decision to stick with the tried-and-true existing languages and tools. When improvements are needed, they need to be made incrementally and in a backwards-compatible fashion, and that's how they are happening.
The reasoning behind this "expert's" opinion seems to be that higher temperatures lead to more drought and thereby to more wildfires. But that's far from certain:
All of that is true for a lot of other domains too. Tests and assertions do exactly what you want: "make sure everything seems to be looking and working as you'd expect". The strategy for testing is the same as for a debugger, but with testing you write down what you're doing and you can reproduce it. I think a debugger is a good learning tool, but for experienced programmers, it's a waste of time.
(BTW, most places I have worked would expect applicants to be able to implement A* from memory and without a debugger in less than 1/2h.)
Yes, but the "hard" bugs are the ones that happen at run-time. Those depends on the context such as Networking, Rendering, and AI bugs that can be a real PITA to track down.
Those bugs all come down to some precondition, postcondition, or invariant that you depend on not being satisfied. Instead of spending time in the debugger, create a test that reproduces the problem, then add code that ensures that everything you assume is true actually is true. To track down networking bugs and race conditions, inject bad data and/or delays.
For an experienced programmer in a statically typed language (like C++), there are three things that catch bugs before you even get to a debugger: the static type checker, runtime assertions, and unit tests. If any one of them triggers, it is usually obvious how to fix stuff even without a debugger.
Debuggers should be simple, minimal, and reliable, so that the 1-2 times a year that you need them, they get the job done correctly and without having to remember a lot of complicated stuff.
Well, having seen a couple of generations grow up, unfortunately it's not a guess. Current generations are much less independent and much more supervised than past generations, and they consider it "normal".
Somewher ein the middle is where we have landed, and although not perfect, it does work relatively well (relative to say Iran or Nigeria for example).
Unfortunately, it consumes an ever growing amount of resources without delivering any better results, year after year, and that can't go on forever. Remember that most of the failing countries used to be wealthy and powerful at some point in the past too.
That's the way it works in many places. It sounds good, but ends up creating a different set of corrupt institutions. I think a general tax refund might be a better choice. Also, relying more on community service and less on fines would not only reduce this problem, it's also tends to be a more uniform punishment (given that fines are meaningless to wealthy people).
Both murder rates and traffic fatalities are influenced a great deal more by factors completely beyond our control than by policing and laws. So, the kinds of comparisons you are making are meaningless.
As a devil's advocate, US police are woefully underfunded
US police receive far more money than police forces elsewhere.
They might get a bunch of SWAT stuff from the government, but actual basic policing, substations, and other items needed to process all but murders and attempted murders are not funded.
The "items" needed to process the other stuff are simply people. And when each US police officer receives several times (literally) the salary of an equivalent police officer abroad, then we end up with much less bang for the buck. But in my experience, these kinds of thefts and crimes are actually more frequent in Europe and no more effectively investigated than in the US. Tourist locations in Europe are a hotbed for theft and other crimes, but even smaller cities are affected.
Besides, this kind of policing is a local matter; in many communities, iPhone thefts simply aren't a problem. In places like San Francisco and New York City, they are a result of the social problems and conflicts those cities are creating for themselves. Here, I can leave my bike outside unlocked or lose my phone and I'll get it back.
When these kids grow up, most of them will be totally used to it and many more will accept government paternalism without complaint "for their own good". Many will likely even feel lost without "father" or "big brother" watching them and telling them what they are doing wrong.
I did not pull out a random sentence from the paper. It is from the fucking abstract - which, if you didn't know, is a summary of the paper.
Abstracts, if you didn't know, are not definitive summaries of the paper; they merely exist to help you decide whether it's worth reading the whole paper. You need to read and understand the whole paper in order to even begin to be able to debate what it means and whether its conclusions are valid.
The only "out of context" the statement had, was that it was not in the context of your opinion.
You quoted the sentence in support of a statement you made. Your statement is wrong, and the sentence you quote doesn't support it. In order to understand why it doesn't support it, you need to read the actual paper.
5. Unproven, in that you can't "prove it" before it happens sure. But if you trust the concept of models in science in general, then there is no reason to not trust the climate models that say the temp is going to rise by X amount in Y years, plus/minus Z error range.
Whether one should trust a model depends on (1) whether it has been experimentally tested, and (2) what assumptions go into it. Most scientific models are tested directly experimentally; you cannot do that for climate models. In addition, I have looked at climate models in some detail, and they rely on assumptions and mechanisms that themselves are untested, such as assumptions about the existence of specific positive feedback mechanisms and the absence of unknown negative feedback mechanisms. In addition, some of them are not physical assumptions, but assumptions about population growth and the economy.
6. If you believe the models provide a useful prediction (plus/minus some error range) of something like temperature over time, then why not trust models that are also predicting things like levels of ocean acidification, ice melt, etc.. that we know, really know, what things will happen to the Earth as a result. Like.. all fisheries dying if the ocean is X acidic. That would be fairly devastating.
Even if you believe the predictions themselves, the predicted costs also depend on numerous more assumptions. For example, why is sea level rise necessarily a big problem? One third of the Netherlands is below sea level, but you don't see the Dutch either starving or invading their neighbors; yet that's the prediction AGW activists make for that situation for other countries. And basic economics tell us that almost no disaster 100 years from now warrants taking action now, but AGW activists are quite selective in their use of science and economics.
Personally, I think that if we don't mess up our global economy, two things will happen. First, people will quickly move off fossil fuels all by themselves, because better technologies will come along all by themselves. Second, a century from now, almost the entire world will be wealthier than the Dutch are today, so even if some of the more dire predictions of AGW were to come true, people across the globe would be sufficiently wealthy to deal with it.
As I was saying: Adobe Creative Suite online may well be garbage. However, for many people other cloud software solutions probably end up being more reliable than if they try to manage their own Windows machine.
I'm not pretending to be an "arbiter". I told you specifically why and how the sentence you quoted is misleading, in terms of the "methods" section of the paper. If you disagree with my analysis, then the right thing to do is to look at the methods section and explain why you think I'm wrong. That's how scientific discourse works.
If you think that any random sentence you pull out of context out of a scientific paper proves your point and you choose to ignore the actual methods and data that went into the conclusion, then you are simply not scientifically literate; science doesn't work that way.
"Among abstracts expressing a position on AGW, 97.1% endorsed the consensus position that humans are causing global warming."
So, sorry, but true.
Incorrect. You said:
So of course the 97% of published papers that agree with their position don't have anything to do with their positions.
No, read the paper. 97% of published papers do not agree with their position; the great majority of published papers take no position at all.
And that sentence doesn't mean what you think it means, because it refers to terms previously defined in the paper. "The consensus" the authors refer to is not the IPCC consensus or a consensus on AGW; "expressing an opinion" means "the paper is somehow related to AGW"; and "endorse" means "seems consistent with, but may not have actually stated anything". It's misleading and deceptive. But at least the authors didn't have the nerve to commit outright fraud, so you can decode what they are actually saying by reading the entire paper.
What they are actually saying is:
Among abstracts about topics that we believe are related to AGW and might be used to determine a position, 97.1% did not appear to contradict the idea that humans have contributed to some degree to observed warming.
Again, actually read the paper, instead of quoting out of context.
As of this writing, a sticky note on your local computer says "Windows is currently undergoing maintenance. The drive was infected by a virus you caught from an Adobe Photoshop plugin and needs to be wiped completely. We're trying to recover as much of your data as possible" Even though I've come to like some locally-hosted software, like TextEdit, I don't think I'd want tools for manipulating local media tied even loosely to the uptime of my local computer (or Windows), or endanger my Windows installation by installing anything other than completely minimal software on it.
(Of course, Adobe Create Cloud may still suck and make you dependent on complex local and remote software, but cloud services in general have been a big win, at least for me.)
What do you think will happen if climate change leads to less fresh water being available, or arable land starts shrinking?
Why would it? In fact, it looks to me like total arable land will likely increase. Freshwater will also likely increase, since there will be increased evaporation from oceans.
While bad in itself, you don't think events such as those will lead to war?
About a third of the Netherlands is underwater; are the Dutch attacking Belgium or Germany? Did the starving people of the Sahel invade Europe in murderous attacks?
Slashdot Mirror
The post you responded to asked questions anybody should be asking about any scientific result.
Your response is the response of a Luddite and someone who is ignorant of science.
The obstacles to decommissioning a nuclear power plant are entirely administrative; you treat them like laws of nature. That's what makes your alarmism so ridiculous even if the predictions were sound and proven.
Anyway, there is nothing wrong with having a nuclear power plant even below sea level; the Dutch do.
Good! I hope they will continue to act rationally rather than give in to FUD.
Good!
It's unfortunate that there really isn't a threat of impending flooding and disaster; maybe if there were, it would spur the bureaucrats and "environmentalists" into action. Causing a bit of sea level rise would be worth it just for that.
In any case, decommissioned or not, these plants won't be producing power anymore in 30 years, let alone 80 years.
I did bother to read the article:
NOAA worst case for 2046 is 21.48 in, or less than 2 ft. The graph below is misleading, and the "0.9 ft Storm" bar is a guess by the HuffPost journalists, not a scientific result.
The whole thing is pure FUD anyway. Nothing magical happens when nuclear power plants end up below sea level; you can build a dike around them, a negligible expense compared to the rest of the plant. The Dutch have built nuclear power plants below sea level to no ill effect.
Fukishima's problem wasn't flooding, it was poor design and outdated safety measures.
The idea that these nuclear power plants are still relevant in 86 years should scare people more than any sea level rise. All those nuclear power plants are completely obsolete. If they need to be torn down and rebuilt elsewhere with new, safer, more efficient technology, we're all better off.
No, my position is that the technical terms used in an abstract are defined in the paper. You cannot just take a sentence out of an abstract and interpret it in a way that's convenient for an argument you want to make.
Since the paper redefines the meaning of the terms "expressing a position on AGW", "endorse", and "consensus position", the sentence you quoted means something different from what you implied it meant. The error here is yours, not the authors' (the authors can be faulted, however, for being misleading).
(Of course, abstracts quite frequently do contradict the body of a paper; authors try to get away with sensationalizing in the abstract because abstracts tend not to be very carefully checked by reviewers because most scientists, unlike you, understand that the abstract by itself doesn't show anything.)
You assume incorrectly that these are "lessons" people working in these areas don't already know and still need to learn.
On the other hand, you'll find that actually implementing something like Nile correctly and efficiently is extremely hard; so hard, that by the time you're done, the world has already moved on to other tools.
People do write large scale numerical code using functional abstractions, automatic vectorization, pipelines, data flow, and all that. There are excellent libraries for those purposes for both C++ and Fortran.
The reason people still stick to C++ and Fortran rather than going to a fully automatic system is threefold. First, it isn't necessary: modern versions of C++ and Fortran already have all the necessary high level support; except for an unfamiliar syntax, Nile doesn't add any functionality. Second, there is tons of legacy code that people want to interoperate with easily and without much hassle; switching to a new language with new conventions makes that hard. Third, with tools like Nile, sooner or later, users hit cases where the automated tools don't work anymore, and then they have to go back to figuring out how to link in C/C++/Fortran anyway.
Language and systems people often have the notion that numerical and parallel software developers do what they do out of ignorance, when in fact it's a rational decision to stick with the tried-and-true existing languages and tools. When improvements are needed, they need to be made incrementally and in a backwards-compatible fashion, and that's how they are happening.
The reasoning behind this "expert's" opinion seems to be that higher temperatures lead to more drought and thereby to more wildfires. But that's far from certain:
http://journals.ametsoc.org/do...
All of that is true for a lot of other domains too. Tests and assertions do exactly what you want: "make sure everything seems to be looking and working as you'd expect". The strategy for testing is the same as for a debugger, but with testing you write down what you're doing and you can reproduce it. I think a debugger is a good learning tool, but for experienced programmers, it's a waste of time.
(BTW, most places I have worked would expect applicants to be able to implement A* from memory and without a debugger in less than 1/2h.)
Those bugs all come down to some precondition, postcondition, or invariant that you depend on not being satisfied. Instead of spending time in the debugger, create a test that reproduces the problem, then add code that ensures that everything you assume is true actually is true. To track down networking bugs and race conditions, inject bad data and/or delays.
For an experienced programmer in a statically typed language (like C++), there are three things that catch bugs before you even get to a debugger: the static type checker, runtime assertions, and unit tests. If any one of them triggers, it is usually obvious how to fix stuff even without a debugger.
Debuggers should be simple, minimal, and reliable, so that the 1-2 times a year that you need them, they get the job done correctly and without having to remember a lot of complicated stuff.
Well, having seen a couple of generations grow up, unfortunately it's not a guess. Current generations are much less independent and much more supervised than past generations, and they consider it "normal".
Unfortunately, it consumes an ever growing amount of resources without delivering any better results, year after year, and that can't go on forever. Remember that most of the failing countries used to be wealthy and powerful at some point in the past too.
That's the way it works in many places. It sounds good, but ends up creating a different set of corrupt institutions. I think a general tax refund might be a better choice. Also, relying more on community service and less on fines would not only reduce this problem, it's also tends to be a more uniform punishment (given that fines are meaningless to wealthy people).
Both murder rates and traffic fatalities are influenced a great deal more by factors completely beyond our control than by policing and laws. So, the kinds of comparisons you are making are meaningless.
US police receive far more money than police forces elsewhere.
The "items" needed to process the other stuff are simply people. And when each US police officer receives several times (literally) the salary of an equivalent police officer abroad, then we end up with much less bang for the buck. But in my experience, these kinds of thefts and crimes are actually more frequent in Europe and no more effectively investigated than in the US. Tourist locations in Europe are a hotbed for theft and other crimes, but even smaller cities are affected.
Besides, this kind of policing is a local matter; in many communities, iPhone thefts simply aren't a problem. In places like San Francisco and New York City, they are a result of the social problems and conflicts those cities are creating for themselves. Here, I can leave my bike outside unlocked or lose my phone and I'll get it back.
When these kids grow up, most of them will be totally used to it and many more will accept government paternalism without complaint "for their own good". Many will likely even feel lost without "father" or "big brother" watching them and telling them what they are doing wrong.
Abstracts, if you didn't know, are not definitive summaries of the paper; they merely exist to help you decide whether it's worth reading the whole paper. You need to read and understand the whole paper in order to even begin to be able to debate what it means and whether its conclusions are valid.
You quoted the sentence in support of a statement you made. Your statement is wrong, and the sentence you quote doesn't support it. In order to understand why it doesn't support it, you need to read the actual paper.
Whether one should trust a model depends on (1) whether it has been experimentally tested, and (2) what assumptions go into it. Most scientific models are tested directly experimentally; you cannot do that for climate models. In addition, I have looked at climate models in some detail, and they rely on assumptions and mechanisms that themselves are untested, such as assumptions about the existence of specific positive feedback mechanisms and the absence of unknown negative feedback mechanisms. In addition, some of them are not physical assumptions, but assumptions about population growth and the economy.
Even if you believe the predictions themselves, the predicted costs also depend on numerous more assumptions. For example, why is sea level rise necessarily a big problem? One third of the Netherlands is below sea level, but you don't see the Dutch either starving or invading their neighbors; yet that's the prediction AGW activists make for that situation for other countries. And basic economics tell us that almost no disaster 100 years from now warrants taking action now, but AGW activists are quite selective in their use of science and economics.
Personally, I think that if we don't mess up our global economy, two things will happen. First, people will quickly move off fossil fuels all by themselves, because better technologies will come along all by themselves. Second, a century from now, almost the entire world will be wealthier than the Dutch are today, so even if some of the more dire predictions of AGW were to come true, people across the globe would be sufficiently wealthy to deal with it.
As I was saying: Adobe Creative Suite online may well be garbage. However, for many people other cloud software solutions probably end up being more reliable than if they try to manage their own Windows machine.
I'm not pretending to be an "arbiter". I told you specifically why and how the sentence you quoted is misleading, in terms of the "methods" section of the paper. If you disagree with my analysis, then the right thing to do is to look at the methods section and explain why you think I'm wrong. That's how scientific discourse works.
If you think that any random sentence you pull out of context out of a scientific paper proves your point and you choose to ignore the actual methods and data that went into the conclusion, then you are simply not scientifically literate; science doesn't work that way.
Incorrect. You said:
No, read the paper. 97% of published papers do not agree with their position; the great majority of published papers take no position at all.
And that sentence doesn't mean what you think it means, because it refers to terms previously defined in the paper. "The consensus" the authors refer to is not the IPCC consensus or a consensus on AGW; "expressing an opinion" means "the paper is somehow related to AGW"; and "endorse" means "seems consistent with, but may not have actually stated anything". It's misleading and deceptive. But at least the authors didn't have the nerve to commit outright fraud, so you can decode what they are actually saying by reading the entire paper.
What they are actually saying is:
Again, actually read the paper, instead of quoting out of context.
(Of course, Adobe Create Cloud may still suck and make you dependent on complex local and remote software, but cloud services in general have been a big win, at least for me.)
Why would it? In fact, it looks to me like total arable land will likely increase. Freshwater will also likely increase, since there will be increased evaporation from oceans.
About a third of the Netherlands is underwater; are the Dutch attacking Belgium or Germany? Did the starving people of the Sahel invade Europe in murderous attacks?