So extracting the core of your argument from the long pointless rant that you've wrapped it in: It's wrong to trade with poor countries, because they derive more advantage from it, and become rich countries.
Err, how is this a bad thing? Using your example of Japan, what the value of bilateral trade between the US and Japan immediately after the second world war, and what has that value risen to after decades of investment in their economy.
The best kind of world for rich industrialised countries to be in, is one in which their competitors are rich industrialised countries with similar production costs.
It's almost as if scientists are pressured by universities to make long-term claims in understandable language, rather than precise short-term claims in technical language. Kind of like it is part of their mission to promote public understanding of science, and one of the criteria they are considered on when funding is allocated.
But if you have an axe to grind against "green tech" then go ahead and pretend it is unique to science with environmental impact.
Would this be the same as claiming that cheaters don't prosper?
It would certainly be a hard claim to disprove: we know that cheaters that have been discovered have lost out, while we don't know how many undiscovered cheaters there are.
No. Not in the slightest. A desalination plant takes in a quantity of salt-water, and consumes a fixed amount of energy to separate it into fresh water and a salty residue. There is not an available shortcut to do "less of this" as you suggest. The salty residue has a higher salt concentration than the salt-water source and hence there is an energy potential that could be exploited. The energy released from an Osmotic power station would reduce the net energy requirement to desalinate the water.
No. A citation is required to demonstrate the authenticity of an empirical claim. This claim is simple logic. It does not require the support of an established figure to support it's authenticity. It can be verified by the reader using simple logic.
I was referring to two of your examples: in your example from scripture you claimed that people cheating would be turned away from. Clearly this requires the people being cheated to be aware that they are being cheated.
In your real-world example you pointed out the consequences for the frauds after they were discovered. Obviously discovery requires knowledge of deceit. In both cases the conclusion that reputation tracking (or knowledge of cheating) was required to win at the game is a simple logical inference, not a claim that seeks authenticity through its source.
Reputation tracking is practical for the average Joe - people tend to remember when they've been defrauded. They also tend to tell everybody who will listen about the events. Basic reputation tracking through memory and gossip have been with us for all of recorded history.
So not only is the "golden rule" a sub-optimal strategy, but there is a genuine need and use for "idle gossip", yet I'm fairly sure that I've seen that proscribed somewhere as well....
You make an interesting argument, although it is somewhat at a tangent from the original point. Redefining advantage to mean "temporary advantage followed by a dire penalty" would render my point incorrect. But that is not what advantage means, and even if some interactions would lead to advantages of this form, that does not make it true of all interactions between a cheater and an altruist in general.
The golden rule does not exist in a vacuum. Jesus also said, "If no one welcomes you or listens to your words, as you leave that house or town, shake its dust off your feet." You could act like a tool, and you would in short order have a very small amount of company.
Oh dear. You are trying to argue against maths by quoting scripture. You cannot disprove specific arguments using vague generalities. Your argument is already covered by what I said last time, perhaps you do not understand the words: "So everybody derives an advantage by switching away from the golden rule. Hence it is a weak equilibrium strongly dominated by (almost) any other strategy".
Both your scripture and your example require on the person being deceiving knowing that they are deceived. This is not true in general, and thus people can derive an advantage through deception. Throwing in "oh my god, somebody think of the sick children" does not strengthen your argument.
One game theory staple where it is beneficial is in repeated play of the Prisoner's Dilemma, where first assuming altruism pays off, on the condition that there are enough other players in the population also first assuming altruism.
Not really a staple though is it? The work that you refer to is an attempt to find a better equilibrium than Nash for the repeated game. Although assuming altruism can get there, so does any assumption that people will avoid mutually destructive equilibria . In fact reputation tracking covers all of these assumptions better than the assumption of altruism.
If we're going to play the game of tossing in so-called "truisms" rather than actual reasoning then it seems clear than informed cynicism beats naive wishful thinking.
True - but Three check identification when you sign up for pay as you go anyway. Or at least, they did when I bought my sim. So they still know who you are.
The claim in the article is that they can't associate IP addresses with customer details. But I wouldn't actually believe that for a second. Not least because all of these prepay mobile offers have hard usage caps, and they must have some way of tracking usage even if they complain publicly that they can't.
You seem to have failed entirely to answer the question that you quoted. The "known experimental results" was in the context of a debate about physics, and yet you've just inserted a huge bunch of unrelated waffle about how nice Jesus was, and how much he worked hard to make everyone happy. I certainly would have let that pass if you had the decency to be correct, but sadly where you have wandered from theology through the social sciences to the far extremes of testable science you have erred.
When you say that the golden rule is "statistically more likely", based on "common sense" and "anecdotal evidence" you really are reaching. A simpler (and more scientific) way of analysing it would be in terms of game theory. Let's consider a world where everyone follows this advice: if I decide to act like a complete tool then nobody is going to change their behaviour towards me, so I can get all of the advantages without suffering any of the consequences. By symmetry this argument applies to anyone else. So everybody derives an advantage by switching away from the golden rule. Hence it is a weak equilibrium strongly dominated by (almost) any other strategy.
More complete work has been published on the game theoretic analysis of truth-telling/lying and altruism. So your imaginary friend not only lacks accurate predictions of known experimental results, but when you try and twist his alleged words into such proofs you run up into the problem that he was wrong. Perhaps more critical thought is required?
That's really expensive. Three will offer 15GB/pm for £15/pm. So in movie terms that is three (H264 compressed) blue ray rips at decent quality. I have no idea how expensive blue ray movies are in the UK, but I'm guessing that is a win purely on the monetary scale. Of course, it requires a 24-month contract and there is no guarantee they will still be unable to trace IPs in two years time.
Of course, functional code ultimately gets translated to imperative code, because that's what our processors do. And that translation is being done by a program, which means it will never match the best that a talented human can do, in much the same way that hand-coded assembler, produced by a developer who knows the processor inside and out, will always beat the best-optimizing C compiler.
If you are thinking of the functional effects of the code then it makes sense to consider it as instructions in an imperative model, but for maximising performance on a modern processor this is a really bad way to think of it. This is one of the main reasons that superscalar processors get a bad rep in terms of understanding the performance effects of different sequences. Once you take into account the re-ordering that the processor is free to do you no longer have an imperative sequence of instructions, rather you have a data-flow graph containing a set of "pure" instructions. The first point is that functional code maps into this setting much better than imperative source code, and the second point is that a compiler can organise that flow graph much more effectively than a human programmer.
Sure, a good programmer can beat the best C compiler - but the problem is not that a compiler can't write good code, the problem is that C is a horrific handicap that holds back the code generator. If you relax the constraints on the source language then the human won't stand a chance.
Essentially. The scope of the pronoun is not actually ambiguous - it always refers to the most recent object of speech. But many people would claim that the commas in the sentence are parenthesis in disguise and that compiler was the most recent object. These people are of course wrong, but common enough that it is best to humour them.
I did say it was minor - you'd have to be really picky to spot it (like most editors). Where he says "it won't spot it", it's obvious that the first "it" refers to the compiler at the beginning of the sentence, but grammatically the "it" refers to "local", as in the local variables he was just discussing. Many would argue that it is not an error as such, but it is ambiguous at best. Still far and away, the most grammatically correct I've ever seen on slashdot.
it sounds like you... care about any copyright protection they have
Do you see what selective quoting is capable of? Kind of like how you removed the section about him being happy to pay directly for their copyrighted works....
And was POV running a global illumination algorithm rather than just vanilla ray tracing? Because the difference in complexity between the two approaches would mean days or months on that cellphone, but it allows for the dynamic lighting changes shown in the video. Last time I saw somebody doing a similar quality of rendering to the demo images they were using Radiance on a relatively modern workstation. Each frame took several hours, on a phone (if you actually had the memory available) the scaling would be much worse than the 10x difference in clock speed because of the difference in floating point operations per cycle. So those rendering passes would take multiple days on a phone hardware - but I suspect that Nvidia is showing something with a more complex lighting model, not least because they have the hardware to throw at it. So weeks (or even months in the extreme) are quite reasonable rendering estimates.
While I don't have any use for the program
on
Microsoft COFEE Leaked
·
· Score: 4, Insightful
It's a bit short-sighted to say that nobody does. I'm sure there are lots of people out there with material on their machines that they wouldn't want a law enforcement officer to find. This tool would be perfect for their needs.
Your description didn't sound like the link that I'd posted, so I've just had to go and refresh my memory of it. Yes, you are in fact largely correct. It is a design flaw, and you could call it a hardware problem as they have picked the wrong hardware for the job.
When you split the possibilities into two cases, you are making the assumption (as I did when I first read it) that they are describing a floating-point register, and not a fixed-point register. In a fixed-point register there is no need to do the accumulation each step in order to increase the error. When the two multiplicands are scaled prior to the multiplication they are truncated into the same range. This is the opposite to floating-point where they are scaled without truncation, and any truncation happens after the operation.
So a difference in magnitude between the constant value and the linearly increasing (integer) value creates an increase in both absolute and relative error. Beautiful - it's almost as if they were designing it to fail.
Bad hardware decisions indeed. I would have gone for either floating-point, or a large enough fixed-point register to tolerate a larger difference in magnitude.
Wow. You got a PR rep to send you a publicly available document? I mean, you do realise that court filings are public, don't you? So, this PR rep, this women who is employed to facilitate Spring Design's communication with the public, sent you a copy of a publicly available document? Wow! Your social engineering is off the chart...
For those interested, that don't have 30 whole seconds to google for it themselves, here it is.
Slashdot Mirror
So extracting the core of your argument from the long pointless rant that you've wrapped it in:
It's wrong to trade with poor countries, because they derive more advantage from it, and become rich countries.
Err, how is this a bad thing? Using your example of Japan, what the value of bilateral trade between the US and Japan immediately after the second world war, and what has that value risen to after decades of investment in their economy.
The best kind of world for rich industrialised countries to be in, is one in which their competitors are rich industrialised countries with similar production costs.
It's almost as if scientists are pressured by universities to make long-term claims in understandable language, rather than precise short-term claims in technical language. Kind of like it is part of their mission to promote public understanding of science, and one of the criteria they are considered on when funding is allocated.
But if you have an axe to grind against "green tech" then go ahead and pretend it is unique to science with environmental impact.
Would this be the same as claiming that cheaters don't prosper?
It would certainly be a hard claim to disprove: we know that cheaters that have been discovered have lost out, while we don't know how many undiscovered cheaters there are.
Will only one of us be in six weeks? Is this some kind of autism boot-camp?
No. Not in the slightest. A desalination plant takes in a quantity of salt-water, and consumes a fixed amount of energy to separate it into fresh water and a salty residue. There is not an available shortcut to do "less of this" as you suggest. The salty residue has a higher salt concentration than the salt-water source and hence there is an energy potential that could be exploited. The energy released from an Osmotic power station would reduce the net energy requirement to desalinate the water.
No. A citation is required to demonstrate the authenticity of an empirical claim. This claim is simple logic. It does not require the support of an established figure to support it's authenticity. It can be verified by the reader using simple logic.
I was referring to two of your examples: in your example from scripture you claimed that people cheating would be turned away from. Clearly this requires the people being cheated to be aware that they are being cheated.
In your real-world example you pointed out the consequences for the frauds after they were discovered. Obviously discovery requires knowledge of deceit. In both cases the conclusion that reputation tracking (or knowledge of cheating) was required to win at the game is a simple logical inference, not a claim that seeks authenticity through its source.
Reputation tracking is practical for the average Joe - people tend to remember when they've been defrauded. They also tend to tell everybody who will listen about the events. Basic reputation tracking through memory and gossip have been with us for all of recorded history.
So not only is the "golden rule" a sub-optimal strategy, but there is a genuine need and use for "idle gossip", yet I'm fairly sure that I've seen that proscribed somewhere as well....
You make an interesting argument, although it is somewhat at a tangent from the original point. Redefining advantage to mean "temporary advantage followed by a dire penalty" would render my point incorrect. But that is not what advantage means, and even if some interactions would lead to advantages of this form, that does not make it true of all interactions between a cheater and an altruist in general.
Oh dear. You are trying to argue against maths by quoting scripture. You cannot disprove specific arguments using vague generalities. Your argument is already covered by what I said last time, perhaps you do not understand the words: "So everybody derives an advantage by switching away from the golden rule. Hence it is a weak equilibrium strongly dominated by (almost) any other strategy".
Both your scripture and your example require on the person being deceiving knowing that they are deceived. This is not true in general, and thus people can derive an advantage through deception. Throwing in "oh my god, somebody think of the sick children" does not strengthen your argument.
Not really a staple though is it? The work that you refer to is an attempt to find a better equilibrium than Nash for the repeated game. Although assuming altruism can get there, so does any assumption that people will avoid mutually destructive equilibria . In fact reputation tracking covers all of these assumptions better than the assumption of altruism.
If we're going to play the game of tossing in so-called "truisms" rather than actual reasoning then it seems clear than informed cynicism beats naive wishful thinking.
True - but Three check identification when you sign up for pay as you go anyway. Or at least, they did when I bought my sim. So they still know who you are.
The claim in the article is that they can't associate IP addresses with customer details. But I wouldn't actually believe that for a second. Not least because all of these prepay mobile offers have hard usage caps, and they must have some way of tracking usage even if they complain publicly that they can't.
You seem to have failed entirely to answer the question that you quoted. The "known experimental results" was in the context of a debate about physics, and yet you've just inserted a huge bunch of unrelated waffle about how nice Jesus was, and how much he worked hard to make everyone happy. I certainly would have let that pass if you had the decency to be correct, but sadly where you have wandered from theology through the social sciences to the far extremes of testable science you have erred.
When you say that the golden rule is "statistically more likely", based on "common sense" and "anecdotal evidence" you really are reaching. A simpler (and more scientific) way of analysing it would be in terms of game theory. Let's consider a world where everyone follows this advice: if I decide to act like a complete tool then nobody is going to change their behaviour towards me, so I can get all of the advantages without suffering any of the consequences. By symmetry this argument applies to anyone else. So everybody derives an advantage by switching away from the golden rule. Hence it is a weak equilibrium strongly dominated by (almost) any other strategy.
More complete work has been published on the game theoretic analysis of truth-telling/lying and altruism. So your imaginary friend not only lacks accurate predictions of known experimental results, but when you try and twist his alleged words into such proofs you run up into the problem that he was wrong. Perhaps more critical thought is required?
So you're an academic running in stealth? Watch out people, he is practising ninja physics...
That's really expensive. Three will offer 15GB/pm for £15/pm. So in movie terms that is three (H264 compressed) blue ray rips at decent quality. I have no idea how expensive blue ray movies are in the UK, but I'm guessing that is a win purely on the monetary scale. Of course, it requires a 24-month contract and there is no guarantee they will still be unable to trace IPs in two years time.
If you are thinking of the functional effects of the code then it makes sense to consider it as instructions in an imperative model, but for maximising performance on a modern processor this is a really bad way to think of it. This is one of the main reasons that superscalar processors get a bad rep in terms of understanding the performance effects of different sequences. Once you take into account the re-ordering that the processor is free to do you no longer have an imperative sequence of instructions, rather you have a data-flow graph containing a set of "pure" instructions. The first point is that functional code maps into this setting much better than imperative source code, and the second point is that a compiler can organise that flow graph much more effectively than a human programmer.
Sure, a good programmer can beat the best C compiler - but the problem is not that a compiler can't write good code, the problem is that C is a horrific handicap that holds back the code generator. If you relax the constraints on the source language then the human won't stand a chance.
OK, I have sacked my spellchecker and will make sure that it never works again ;)
Essentially. The scope of the pronoun is not actually ambiguous - it always refers to the most recent object of speech. But many people would claim that the commas in the sentence are parenthesis in disguise and that compiler was the most recent object. These people are of course wrong, but common enough that it is best to humour them.
I did say it was minor - you'd have to be really picky to spot it (like most editors). Where he says "it won't spot it", it's obvious that the first "it" refers to the compiler at the beginning of the sentence, but grammatically the "it" refers to "local", as in the local variables he was just discussing. Many would argue that it is not an error as such, but it is ambiguous at best. Still far and away, the most grammatically correct I've ever seen on slashdot.
Wow, zero spelling mistakes, and one minor grammatical error - you seem to have poked fate in the eye and survived.
Do you see what selective quoting is capable of? Kind of like how you removed the section about him being happy to pay directly for their copyrighted works....
And was POV running a global illumination algorithm rather than just vanilla ray tracing? Because the difference in complexity between the two approaches would mean days or months on that cellphone, but it allows for the dynamic lighting changes shown in the video. Last time I saw somebody doing a similar quality of rendering to the demo images they were using Radiance on a relatively modern workstation. Each frame took several hours, on a phone (if you actually had the memory available) the scaling would be much worse than the 10x difference in clock speed because of the difference in floating point operations per cycle. So those rendering passes would take multiple days on a phone hardware - but I suspect that Nvidia is showing something with a more complex lighting model, not least because they have the hardware to throw at it. So weeks (or even months in the extreme) are quite reasonable rendering estimates.
It's a bit short-sighted to say that nobody does. I'm sure there are lots of people out there with material on their machines that they wouldn't want a law enforcement officer to find. This tool would be perfect for their needs.
You idiot, the uncertainty principle only kicks in when objects are small enough to be dominated by quantum effects.
Like a European car...
Your description didn't sound like the link that I'd posted, so I've just had to go and refresh my memory of it. Yes, you are in fact largely correct. It is a design flaw, and you could call it a hardware problem as they have picked the wrong hardware for the job.
When you split the possibilities into two cases, you are making the assumption (as I did when I first read it) that they are describing a floating-point register, and not a fixed-point register. In a fixed-point register there is no need to do the accumulation each step in order to increase the error. When the two multiplicands are scaled prior to the multiplication they are truncated into the same range. This is the opposite to floating-point where they are scaled without truncation, and any truncation happens after the operation.
So a difference in magnitude between the constant value and the linearly increasing (integer) value creates an increase in both absolute and relative error. Beautiful - it's almost as if they were designing it to fail.
Bad hardware decisions indeed. I would have gone for either floating-point, or a large enough fixed-point register to tolerate a larger difference in magnitude.
So I guess that high IQ doesn't correlate with the ability to learn new environments without training
What can I say? :)
Wow, you're relativity retarded and a time insensitive clod
Wow. You got a PR rep to send you a publicly available document? I mean, you do realise that court filings are public, don't you? So, this PR rep, this women who is employed to facilitate Spring Design's communication with the public, sent you a copy of a publicly available document? Wow! Your social engineering is off the chart...
For those interested, that don't have 30 whole seconds to google for it themselves, here it is.