That was what Einstein thought. So he set up a thought experiment to prove that quantum was only apparently random, called the Einstein-Rosen-Podolsky paradox. Turned out, after Aspect ran the experiment, that Einstein was wrong. Reality was more random than he thought. It still might be the case that there's an order behind the quantum randomness, but that's currently more an article of faith than scientific insight.
With these numbers, the argument for vaccination is very thin. The same article finds that kids that got vaccinations for one year, are more prone to get flu the next year. So there are side-effects. It's not that clear-cut as you think it is.
Where's the proof that the flu vaccines work? There's a new vaccine every year. Where's the double blind? Was the flu vaccine of 2008 better than that of 2009? Please try to find out. Although we are capable of creating a vaccine for practically any strain, we cannot predict which strain will be dominant. Taking a flu shot every year is not supported by any evidence, as this vaccine is not tied particularly well to what flu will be dominant that year. I live in a country where we do not have a tradition of taking flu shots. So we do not have herd immunity. We don't die of flu more often than the US though. Even funnier, one of our scientists just recently created a strain of influenza that has a projected death toll of 50%. Will you start to get yearly shots of the vaccine for that one? Every year? What strains will you get shots for. Do you actually know?
Herd immunity doesn't hold for influenza. This because, in contrast to Polio, tuberculosis, measles, plague and many more, influenza vaccines do not work in general. They work for particular strains, but there is no guarantee that that particular strain will come in your neighbourhood. It's going to be a different strain. Influenza mutates too fast to be caught by a vaccine that needs to be created, and mass-produced. Getting your flu shot is a ritual, not a vaccine.
I don't think the guy thinks that you don't need to be vaccinated against polio and measles, but that you don't need to be vaccinated against influenza. Whole different story. Do you truly believe that the pharmaceutical industry really has a good handle on influenza? Strains the come out of nowhere and pop up at random? There's very little clinical evidence that vaccines against influenza work. Very little. There's plenty of evidence that it works for polio, tuberculosis and such. Being skeptical about the unproven, doesn't imply being skeptical about the proven.
Proof that influenza vaccines actually work is very scarce, and highly disputed. Proof that influenza kills healthy people is non-existent (the Spanish flu of 1918 wreaked havoc on starving people, not healthy ones). Healthy people get sick, get better, and get back to work. I've had my two days of influenza break already this year.
Think about it for a minute. Every winter, three or more new and unique influenza viruses become effective. These epidemics start somewhere on the globe, and spread from there. In some way, the pharmaceutical industry has a handle on this, and knows exactly which viruses will comprise this season's spread of influenza, and creates vaccines that can deal with these strains.
If you buy this story, I've got a bridge to sell you. I believe in vaccinations for small-pox, tuberculosis, and all these other vaccines that work against a particular virus. I don't buy vaccines against a quickly and erratically mutating virus such as influenza. We might get a handle on one strain, but we'll never be sure that this strain is the one that's going to be the one making people sick.
I doubt it would work. You see, this virus is man-made, intelligent designed if you will, not thoroughly tested such as any virus that went through the rigors of natural selection. For real influenza, billions of mutants have been tried, all but a few have failed to get any traction. How many tries did we give this virus? How many has it killed already? None? Not very convincing for a real influenza.
So chances are, that even though this virus has a theoretical killing rate of 50%, I highly doubt it would be able to spread effectively. The moment it gets out, it probably will get eaten by a passing microbe or so. Burp and gone.
Intelligent design is just not up to the task of dealing with real life.
She's been a major figure in Dutch politics for a few terms. Been minister and all that. She's been the European commisioner for competition for 4 years. Here she sued Microsoft and won. Something no one in the US had been able to do. Despite the fact that another party in Holland really argued they had a better claim for having a representative in the European commision, she was nominated for this role of commisioner for iT. This is one of the final roles in her career, and also here, she is generally acclaimed to be a success, even by her political opponents. Her career is going just fine, to the level of having streets and airports named after her.
You talk about combining ideas. Patents are not about ideas, patents are about implementations. A true test for obviousness would be to give a few engineers the idea (use GPS to find a webservice), and see what they come up with as implementations. If it's the same as the patent, then the implementation can be considered obvious, and the patent invalid.
A "bad patent" is a patent that protects an idea by the one and only obvious implementation of the idea. One-click buying was a bad patent, and if this patent is merely combining two ideas, instead of combining two technologies in a new whole, it is a bad patent.
It seems that the patent system is broken to the point that people expect that ideas can be patented. They should not.
Again, does the Koch commisioned research count as skeptical, while the exact same research done by regular channels count as pro-AGW? If I try to set up a temperature curve, am I doing pro-AGW research? If I try to setup a temperature curve, am I doing skeptical of AGW reasearch? Research is research: all you can claim is that a lot of money goes into climate research.
Your stat is bullshit, it equates research with convictions: it assumes that all climate research is done by pro-AGW people, that the money going into climate research (directly or indirectly!!!) is 50 billion, therefore 50 billion goes into pro-AGW research. Utter nonsense.
Let's look at how much grant money has been spent on elementary particle physics w.r.t. phlogiston research. What can we conclude from that? Probably that elementary particle physics is a rather good approximation of the truth, and phlogisten is not. Grant money goes into an area for research, not into conclusions.
Consider this recent study commissioned by the Koch brothers. This was an study that was done to check the measurements. It was one of many (as there have been lots of these studies, but done with 'tainted' money, namely research money). The conclusion was the global warming is real. Now do you conclude that the Koch brothers are now on record for funding pro global-warming ideas? No, they actually funded research. This is probably the last time they'll do that, as real research concentrates on the truth, not on some agenda.
Sorry, nice try, but flat out wrong. I live in what is called a constitutional monarchy, not a democratic republic (and certainly not a people's republic). Both are representative democracies, as these are the words we use to describe the types of governments that we have in the first world.
Most of the stuff read in this thread is about code generation. That's trivial. But you're right, I didn't read the article, at the time of my original post. All I know is that Microsoft has a great tradition of providing completely new concepts to the industry, rather than rehashing or buying stuff that innovators have been doing..NET in particular is completely novel an new and has no resemblance to predecessors, especially not with Java. So please, this Andres dude, what did he propose to be new to.NET that has never been seen before. (You do have Lisp, and SmallTalk (Squeek) to compare to)
In this case, the dude had a physical based model that is known to describe the phenomenon. Simplifying this model to fit the data is a bit weird in this situation, though it might be a fun suggestion that physicists should provide models that can be calibrated with different number of data points. Oh, if you have only 10 points, use Kepler's law, between 10 and 100 you can use Newton, more than one 100, use Einstein.
Calibration is the word used in physics to describe the process of estimating the value of free parameters in models that are based upon first principles. Such models are things like the Navier-Stokes equations, relativity laws and other physical models that are conjured up by physicists to describe reality. I can understand that you're unfamiliar with it, but that does make you ignorant, not right. The article does not claim that he's changing the data, he's truly changing the parameters.
The only way that you can use out of sample data to use a model is that you use if once. You either accept the model, or you continue to live without using a model. The moment that you use that out of sample data a second time, because the first model failed, that out of sample data is no longer out of sample, and part of your development data.
Now for the argument by authority: I've been building both physical based models as well as statistical models for about twenty years now, and your lack of perspective on the subtleties in creating models on data shows to me that you've still got a long way to go before you can call yourself a pro.
Look at this from the angle of a linear system: if you have more unknowns than datapoints, you cannot recover your unknowns from the data. This is really all the article is asserting: these (economical) models are simply too complex given the amount of data we have. They are too detailed and therefore useless.
It's easier for 147 corporations to collude in bribing the regulators than for 14.700.
Please remember what happened to the SEC. Every official in the SEC has a choice: work with Wall street and make a killing in consultancy afterwards, or don't work with them and stay working for government pay. Wall street makes sure that they pay what they promise.
Yeah, he's probably be bought. And even if he isn't bought, a few hundred of his or the other party's denomination voted into congress and senate are bought to make sure any stray individual can't do too much damage.
tiny C compiler has been doing this for a while. Give it a string containing the function, and it returns a function pointer. And of course, Lisp had a 'compile' function (taking an S-expression of course) for almost forever.
Please danbert8, please explain to me what type of government I live in. In my country, we have a head of state, we address her as Her Majesty. We also have an elected house of representatives and a senate. Her Majesty has no political power whatsoever -- her job is purely ceremonial. She inherited the job from her mom. This obviously is not a democracy as -- per your assertion -- we do not have direct majority rule, but have representatives. It just as obviously is not a republic because we have a queen. It can't be a monarchy because the queen doesn't have any political power. What did your teachers teach you what this for of government is?
I disagree. The facts are that we've seen Unix, an operating system that is extremely well-designed, one that is alive and kicking 40 years after creation. In these forty years we've seen IBM creating alternative systems, we've seen Microsoft creating alternative systems, and we've seen Apple creating alternative systems. Both Apple and IBM have seen the errors in their ways (though Apple is trying it again with iOS, and IBM is still pushing Cobol/mainframe), and now mainly run Unix. Microsoft is hurting badly because their system is not even nearly as sophisticated as Ritchies system developed two decades prior. Given this actual history of computing, I shudder to think what computing would look like these days without Unix being around. Unix is at the heart of computing, purely based on merit, not corporate muscle. Without Ritchie, this would not have happened.
Slashdot Mirror
That was what Einstein thought. So he set up a thought experiment to prove that quantum was only apparently random, called the Einstein-Rosen-Podolsky paradox. Turned out, after Aspect ran the experiment, that Einstein was wrong. Reality was more random than he thought. It still might be the case that there's an order behind the quantum randomness, but that's currently more an article of faith than scientific insight.
You say, for arguments sake, that it stops 90% of the influenza infection. It doesn't. It saves 1.5%. A recent article in the Lancet (link to an article describing the original article: "Efficacy and effectiveness of influenza vaccines: a systematic review and meta-analysis", The Lancet, 27 October 2011) states that it saves 1.5 per 100 cases (from 2.7% to 1.18%). This is spinned by the industry as reducing risk by 60%.
With these numbers, the argument for vaccination is very thin. The same article finds that kids that got vaccinations for one year, are more prone to get flu the next year. So there are side-effects. It's not that clear-cut as you think it is.
Where's the proof that the flu vaccines work? There's a new vaccine every year. Where's the double blind? Was the flu vaccine of 2008 better than that of 2009? Please try to find out. Although we are capable of creating a vaccine for practically any strain, we cannot predict which strain will be dominant. Taking a flu shot every year is not supported by any evidence, as this vaccine is not tied particularly well to what flu will be dominant that year. I live in a country where we do not have a tradition of taking flu shots. So we do not have herd immunity. We don't die of flu more often than the US though. Even funnier, one of our scientists just recently created a strain of influenza that has a projected death toll of 50%. Will you start to get yearly shots of the vaccine for that one? Every year? What strains will you get shots for. Do you actually know?
Herd immunity doesn't hold for influenza. This because, in contrast to Polio, tuberculosis, measles, plague and many more, influenza vaccines do not work in general. They work for particular strains, but there is no guarantee that that particular strain will come in your neighbourhood. It's going to be a different strain. Influenza mutates too fast to be caught by a vaccine that needs to be created, and mass-produced. Getting your flu shot is a ritual, not a vaccine.
I don't think the guy thinks that you don't need to be vaccinated against polio and measles, but that you don't need to be vaccinated against influenza. Whole different story. Do you truly believe that the pharmaceutical industry really has a good handle on influenza? Strains the come out of nowhere and pop up at random? There's very little clinical evidence that vaccines against influenza work. Very little. There's plenty of evidence that it works for polio, tuberculosis and such. Being skeptical about the unproven, doesn't imply being skeptical about the proven.
Proof that influenza vaccines actually work is very scarce, and highly disputed. Proof that influenza kills healthy people is non-existent (the Spanish flu of 1918 wreaked havoc on starving people, not healthy ones). Healthy people get sick, get better, and get back to work. I've had my two days of influenza break already this year.
Think about it for a minute. Every winter, three or more new and unique influenza viruses become effective. These epidemics start somewhere on the globe, and spread from there. In some way, the pharmaceutical industry has a handle on this, and knows exactly which viruses will comprise this season's spread of influenza, and creates vaccines that can deal with these strains.
If you buy this story, I've got a bridge to sell you. I believe in vaccinations for small-pox, tuberculosis, and all these other vaccines that work against a particular virus. I don't buy vaccines against a quickly and erratically mutating virus such as influenza. We might get a handle on one strain, but we'll never be sure that this strain is the one that's going to be the one making people sick.
I doubt it would work. You see, this virus is man-made, intelligent designed if you will, not thoroughly tested such as any virus that went through the rigors of natural selection. For real influenza, billions of mutants have been tried, all but a few have failed to get any traction. How many tries did we give this virus? How many has it killed already? None? Not very convincing for a real influenza. So chances are, that even though this virus has a theoretical killing rate of 50%, I highly doubt it would be able to spread effectively. The moment it gets out, it probably will get eaten by a passing microbe or so. Burp and gone.
Intelligent design is just not up to the task of dealing with real life.
If it's a white guy, you say "he's stupid". If it's a black guy, you say "he might be smart, but he's still a chimp". That's the difference.
She's been a major figure in Dutch politics for a few terms. Been minister and all that. She's been the European commisioner for competition for 4 years. Here she sued Microsoft and won. Something no one in the US had been able to do. Despite the fact that another party in Holland really argued they had a better claim for having a representative in the European commision, she was nominated for this role of commisioner for iT. This is one of the final roles in her career, and also here, she is generally acclaimed to be a success, even by her political opponents. Her career is going just fine, to the level of having streets and airports named after her.
You talk about combining ideas. Patents are not about ideas, patents are about implementations. A true test for obviousness would be to give a few engineers the idea (use GPS to find a webservice), and see what they come up with as implementations. If it's the same as the patent, then the implementation can be considered obvious, and the patent invalid.
A "bad patent" is a patent that protects an idea by the one and only obvious implementation of the idea. One-click buying was a bad patent, and if this patent is merely combining two ideas, instead of combining two technologies in a new whole, it is a bad patent.
It seems that the patent system is broken to the point that people expect that ideas can be patented. They should not.
Again, does the Koch commisioned research count as skeptical, while the exact same research done by regular channels count as pro-AGW? If I try to set up a temperature curve, am I doing pro-AGW research? If I try to setup a temperature curve, am I doing skeptical of AGW reasearch? Research is research: all you can claim is that a lot of money goes into climate research.
Your stat is bullshit, it equates research with convictions: it assumes that all climate research is done by pro-AGW people, that the money going into climate research (directly or indirectly!!!) is 50 billion, therefore 50 billion goes into pro-AGW research. Utter nonsense.
Let's look at how much grant money has been spent on elementary particle physics w.r.t. phlogiston research. What can we conclude from that? Probably that elementary particle physics is a rather good approximation of the truth, and phlogisten is not. Grant money goes into an area for research, not into conclusions.
Consider this recent study commissioned by the Koch brothers. This was an study that was done to check the measurements. It was one of many (as there have been lots of these studies, but done with 'tainted' money, namely research money). The conclusion was the global warming is real. Now do you conclude that the Koch brothers are now on record for funding pro global-warming ideas? No, they actually funded research. This is probably the last time they'll do that, as real research concentrates on the truth, not on some agenda.
Sorry, nice try, but flat out wrong. I live in what is called a constitutional monarchy, not a democratic republic (and certainly not a people's republic). Both are representative democracies, as these are the words we use to describe the types of governments that we have in the first world.
Most of the stuff read in this thread is about code generation. That's trivial. But you're right, I didn't read the article, at the time of my original post. All I know is that Microsoft has a great tradition of providing completely new concepts to the industry, rather than rehashing or buying stuff that innovators have been doing. .NET in particular is completely novel an new and has no resemblance to predecessors, especially not with Java. So please, this Andres dude, what did he propose to be new to .NET that has never been seen before. (You do have Lisp, and SmallTalk (Squeek) to compare to)
In this case, the dude had a physical based model that is known to describe the phenomenon. Simplifying this model to fit the data is a bit weird in this situation, though it might be a fun suggestion that physicists should provide models that can be calibrated with different number of data points. Oh, if you have only 10 points, use Kepler's law, between 10 and 100 you can use Newton, more than one 100, use Einstein.
Calibration is the word used in physics to describe the process of estimating the value of free parameters in models that are based upon first principles. Such models are things like the Navier-Stokes equations, relativity laws and other physical models that are conjured up by physicists to describe reality. I can understand that you're unfamiliar with it, but that does make you ignorant, not right. The article does not claim that he's changing the data, he's truly changing the parameters.
The only way that you can use out of sample data to use a model is that you use if once. You either accept the model, or you continue to live without using a model. The moment that you use that out of sample data a second time, because the first model failed, that out of sample data is no longer out of sample, and part of your development data.
Now for the argument by authority: I've been building both physical based models as well as statistical models for about twenty years now, and your lack of perspective on the subtleties in creating models on data shows to me that you've still got a long way to go before you can call yourself a pro.
Please, learn something about data analysis and statistical modeling before making more stupid comments.
Look at this from the angle of a linear system: if you have more unknowns than datapoints, you cannot recover your unknowns from the data. This is really all the article is asserting: these (economical) models are simply too complex given the amount of data we have. They are too detailed and therefore useless.
Is it? The media, in particular Fox news, seem to do fine steering the behavior of millions of people.
It's easier for 147 corporations to collude in bribing the regulators than for 14.700.
Please remember what happened to the SEC. Every official in the SEC has a choice: work with Wall street and make a killing in consultancy afterwards, or don't work with them and stay working for government pay. Wall street makes sure that they pay what they promise.
Yeah, he's probably be bought. And even if he isn't bought, a few hundred of his or the other party's denomination voted into congress and senate are bought to make sure any stray individual can't do too much damage.
tiny C compiler has been doing this for a while. Give it a string containing the function, and it returns a function pointer. And of course, Lisp had a 'compile' function (taking an S-expression of course) for almost forever.
Please danbert8, please explain to me what type of government I live in. In my country, we have a head of state, we address her as Her Majesty. We also have an elected house of representatives and a senate. Her Majesty has no political power whatsoever -- her job is purely ceremonial. She inherited the job from her mom. This obviously is not a democracy as -- per your assertion -- we do not have direct majority rule, but have representatives. It just as obviously is not a republic because we have a queen. It can't be a monarchy because the queen doesn't have any political power. What did your teachers teach you what this for of government is?
I disagree. The facts are that we've seen Unix, an operating system that is extremely well-designed, one that is alive and kicking 40 years after creation. In these forty years we've seen IBM creating alternative systems, we've seen Microsoft creating alternative systems, and we've seen Apple creating alternative systems. Both Apple and IBM have seen the errors in their ways (though Apple is trying it again with iOS, and IBM is still pushing Cobol/mainframe), and now mainly run Unix. Microsoft is hurting badly because their system is not even nearly as sophisticated as Ritchies system developed two decades prior. Given this actual history of computing, I shudder to think what computing would look like these days without Unix being around. Unix is at the heart of computing, purely based on merit, not corporate muscle. Without Ritchie, this would not have happened.
I wouldn't say that Jobs popularized it, that honor belongs to Gates. Jobs packaged decades of results of computing and put it into an appliance.