Your comment is spot on about probably saving the government money, up until the "disability" part, which reveals a serious bias against the deaf. Many deaf go to college (and not just the two where sign language is a first class language) and/or get jobs just like hearing people.
On my HDHP plans, the deductible has always been just shy of the legal contribution limits on HSAs. So while your advice might sound great the only way my family could have taken it would have been to not go to the doctor.
Beyond that.... The legal definition of CP also includes depictions of sexual acts with minors who have reached the age of consent in their jurisdiction, but are not yet 18 years of age. So, while it might be legal for an adult to have sex with a 16 year-old in many parts of the USA, it is illegal for that adult to photograph herself doing so. Similarly, in many more parts of the USA it is legal for two minors to have sex with each other, but not for either of them to photograph the activity-- they don't even have to share the images, mere possession is a crime. Finally, you can't even make erotic images of *yourself* if you are under 18! And if you do, you cannot wait until you are 18 and then distribute those images without violating the law.
Actually, pretty soon the only people willing to fly will be those who are so accustomed to being pushed around by little bullies at the airport that when a gang of truly dangerous folks does try to steal a plane the bullies will succeed because the passengers will be unable to think, or act, assertively enough to make a difference.
Microsoft vs. Apple was a copyright thing. Lindows was trademark-related.
Microsoft is often the target of patent lawsuits. I can't recall them starting one. Please show your work.
Apple does not have a monopoly on anything. They made one good deal with the music industry and apparently came up with a compelling value offer for consumers (or the easily duped, take your pick). But nothing prevents consumers from buying music through many, many other channels and putting them on one of the hundreds of different devices in the world that play MP3. Apple also came out with a decent phone and lots of people bought one. But the only "monopoly" here is over delivery of applications to those who own those phones. Lots of other "smart" phones exist. BlackBerry and Windows Mobile have been around a lot longer than iPhone and are still very popular. iPad? Whatever. A niche device for the moment.
If you invest time and money into putting your entire business model into Apple's hands, the only serious issue is your gullibility. Apple have been raving control freaks since at least 1984.
one of the most selfish things a person can do.
Are you crazy? In a world where 16,000 children die of hunger every year, eating is the most selfish thing one can do.
At least the something he is doing is to benifit people that vote, and not a company that puts money in pockets.
Because if there's one thing I hate it is money in pockets! That's why I'm glad Senator Franken gets paid well into the six figures to help scold private businesses for how they treat their non-paying "customers". I guess the money in *his* pockets doesn't count? Because what with the bailouts and health-care bill he voted for, I'm sure going to have a lot less money in *my* pockets, in spite of the best efforts of the company I work for to pay me a decent salary.
Please note: The whole reason companies exist is to put money in pockets. The pockets of those who risk their money to start the company and those who put in their time and ability to make the company successful. If you don't like how Facebook runs their business feel free to not sign on to their system.
They are not like Microsoft, where for the longest time there was no alternative to Windows.
Um, what? There was always an alternative to Windows. It was called MacOS, it was first released on January 24, 1984. That's over a year and a half before Windows 1.0 was available in November 1985.
But I note that any blindly anti-MS sentiment on Slashdot can get modded "insightful" no matter how contrary to fact it is.
Why, if we both pay for the same service level, should your packets get priority just because your protocol wants less latency? That means that you get the service you paid for and I don't. If you want more of the pipe more of the time, then you should pay for the privilege.
And how would that be any different from just using commit hooks in any old source control system? Any idea how much more/less work it will take to set up these "wizards" and code templates correctly in Wave, as opposed to, say, just building them in your IDE of choice?
No. The secret ingredient is marketing skillz. Their products are neither innovative nor particularly well executed. But they are fashionable -- partly because they are designed to look and act as fashion accessories, but also because of marketing.
I'm not suggesting beating your child with a stick
By process of elimination I guess you are suggesting to beat the child with a carrot then. I mean, how many kids do you know that think a carrot is any kind of reward? Most kids I've seen, on the other hand, love to pick up and play with sticks.
From your original post: That means that your "99% accurate" test is wrong 9,999 times out of 10,000!
Bzzzt! Wrong. It is wrong 9,999 times out of 1,000,000. What you have presented here is a blatant failure to do math correctly. I have shown you the correct math and you accuse me of plagiarism? Neither your post nor the Wikipedia entry included any information about why false positives were bad, which I did. Summary: there is no "paradox". What's important here is to focus on problems that arise from false positives. And you don't need to fudge the numbers or twist the meaning of words like "accuracy" to do that. I never did argue against the sentiment you expressed, just took issue with the math involved. But I guess your reading comprehension is as bad as your math.
In your hypothetical disease example, there is no stated reason why a false positive should be considered a bad thing. Who cares if the test gives us 9,999 false positives? Or 999,999? If the treatment is to drink an extra glass of water a day, and nothing bad happens to anyone because of that, then the cost of a false positive is vanishingly small. However, if the treatment is potentially lethal, as I illustrated (and neither you nor Wikipedia did), you need to tread carefully-- perhaps by putting in extra checks or repeating the test (as another poster suggests).
In the case of a terrorism detector, this is a tough call, now isn't it? Even if we don't fudge the numbers and twist the meaning of common words, as you are doing, we still have to decide what's an acceptable outcome when we have false positives--a problem for which this "paradox of the false positive" is completely useless.
you have a 1/10,000 * 1/9,999 chance of the person who dies
Ooops. I mean 1/10,000 * 1/10,000 chance--the odds of two independent events. What I wrote first is the odds of picking any two specific people from the pool of 10,000.
Um. No. If you administer the test 1,000,000 times, then you have to use all 1,000,000 administrations of the test in your analysis of the results. You can't simply discard the negatives! That's called "sleight of hand" or "fudging the numbers". Unless what you really want to do is discuss the odds of a false positive. That's different than making statements about the overall accuracy of the test.
So, if you can't point me to a "competent statistician" other than Cory Doctorow and a pathetic stub page on Wikipedia, then I have to think that maybe I'm right about this. You don't need to fudge the numbers to make your point, though. In fact, a little competent application of statistics will go a long way here. It's critical to understand the underlying sentiment of this "paraox", which is that you have to really look at the consequences of those false positives.
In the disease example, what if the treatment for the disease is particularly nasty and causes 0.01% of the people who undergo it to die? Then out of your 10,000 people who test positive, one of them will die from the treatment. Assuming no correlation between the fatality of the treatment and the fatality of the disease, you have a 1/10,000 * 1/9,999 chance of the person who dies from the treatment being the same person who has the disease. So basically, all the test does is help you kill one person who doesn't have the disease and save the one who does. That is downright stupid. And if the treatment is any more lethal than that, or if the treatment is not a sure thing for the person with the disease, the tests are doing more harm than good.
In the case of terrorism, the problem of any terrorist detector finding an actual terrorist in the act of committing an attack, is probably even worse than 1/1,000,000. I'm guessing that there are something like 1,000,000 airplane rides a day in the USA. In the last five years there have been exactly zero airplane-based terrorist attacks attempted, that I'm aware of. So the real odds of anyone actually being a terrorist are, well, zero. But assume one attack every five years and the odds are something like 1/18,250,000,000. Now what exactly are we going to do with people when the terrorist detector is positive? And are we content to allow that when the odds of a false positive are overwhelming?
That means that your "99% accurate" test is wrong 9,999 times out of 10,000! [snip] If there's one thing the government and our educational institutions could do to keep us safer, it's this: teach us how statistics works
They can start by teaching Cory Doctorow how to count. The hypothetical test is wrong 9,999 times out of 1,000,000. Assuming, of course, that the test only produces false positives, and not also false negatives. That's what 99% accurate means.
Actually, the point of a bicameral legislature such as Congress is to have a built-in check and balance system at the level of law-making in the first place.
Of course, the GP's points were wrong on both the types of checks and balances that do exist, and that telco immunity is some sort of unconstitutional ex post facto law.
However, I agree that, even in the real world, a senator should vote against an entire bill that contains noxious items. Unless, of course, he is trading a "yes" vote here for a colleague's "yes" vote on another bill. And those are the sorts of behind the scenes dealings that are hard to make obvious to those of us outside the legislature... which makes the practice somewhat suspect in my mind. But I recognize it as the art of compromise, as well. The important thing here is to look at an overall trend and what these people tout as their own priorities. Is this immunity clause really go to have any effect or is it really something that should be a top priority? I don't think so. But that's my opinion. I've got other things I care a lot more about in the long run...
Slashdot Mirror
Your comment is spot on about probably saving the government money, up until the "disability" part, which reveals a serious bias against the deaf. Many deaf go to college (and not just the two where sign language is a first class language) and/or get jobs just like hearing people.
On my HDHP plans, the deductible has always been just shy of the legal contribution limits on HSAs. So while your advice might sound great the only way my family could have taken it would have been to not go to the doctor.
Beyond that.... The legal definition of CP also includes depictions of sexual acts with minors who have reached the age of consent in their jurisdiction, but are not yet 18 years of age. So, while it might be legal for an adult to have sex with a 16 year-old in many parts of the USA, it is illegal for that adult to photograph herself doing so. Similarly, in many more parts of the USA it is legal for two minors to have sex with each other, but not for either of them to photograph the activity-- they don't even have to share the images, mere possession is a crime. Finally, you can't even make erotic images of *yourself* if you are under 18! And if you do, you cannot wait until you are 18 and then distribute those images without violating the law.
Actually, pretty soon the only people willing to fly will be those who are so accustomed to being pushed around by little bullies at the airport that when a gang of truly dangerous folks does try to steal a plane the bullies will succeed because the passengers will be unable to think, or act, assertively enough to make a difference.
Microsoft vs. Apple was a copyright thing. Lindows was trademark-related. Microsoft is often the target of patent lawsuits. I can't recall them starting one. Please show your work.
Apple does not have a monopoly on anything. They made one good deal with the music industry and apparently came up with a compelling value offer for consumers (or the easily duped, take your pick). But nothing prevents consumers from buying music through many, many other channels and putting them on one of the hundreds of different devices in the world that play MP3. Apple also came out with a decent phone and lots of people bought one. But the only "monopoly" here is over delivery of applications to those who own those phones. Lots of other "smart" phones exist. BlackBerry and Windows Mobile have been around a lot longer than iPhone and are still very popular. iPad? Whatever. A niche device for the moment. If you invest time and money into putting your entire business model into Apple's hands, the only serious issue is your gullibility. Apple have been raving control freaks since at least 1984.
one of the most selfish things a person can do. Are you crazy? In a world where 16,000 children die of hunger every year, eating is the most selfish thing one can do.
At least the something he is doing is to benifit people that vote, and not a company that puts money in pockets.
Because if there's one thing I hate it is money in pockets! That's why I'm glad Senator Franken gets paid well into the six figures to help scold private businesses for how they treat their non-paying "customers". I guess the money in *his* pockets doesn't count? Because what with the bailouts and health-care bill he voted for, I'm sure going to have a lot less money in *my* pockets, in spite of the best efforts of the company I work for to pay me a decent salary.
Please note: The whole reason companies exist is to put money in pockets. The pockets of those who risk their money to start the company and those who put in their time and ability to make the company successful. If you don't like how Facebook runs their business feel free to not sign on to their system.
Plus it's written in [expletive deleted] .NET, so it will probably never be available for Ubuntu
That's funny, because Ubuntu is including F-Spot, which is written in C#.
http://en.wikipedia.org/wiki/Who
http://en.wikipedia.org/wiki/One_time_pad
They are not like Microsoft, where for the longest time there was no alternative to Windows.
Um, what? There was always an alternative to Windows. It was called MacOS, it was first released on January 24, 1984. That's over a year and a half before Windows 1.0 was available in November 1985.
But I note that any blindly anti-MS sentiment on Slashdot can get modded "insightful" no matter how contrary to fact it is.
Why, if we both pay for the same service level, should your packets get priority just because your protocol wants less latency? That means that you get the service you paid for and I don't. If you want more of the pipe more of the time, then you should pay for the privilege.
OTOH, Mod parent up would be a good name for a rock band.
And how would that be any different from just using commit hooks in any old source control system? Any idea how much more/less work it will take to set up these "wizards" and code templates correctly in Wave, as opposed to, say, just building them in your IDE of choice?
No. The secret ingredient is marketing skillz. Their products are neither innovative nor particularly well executed. But they are fashionable -- partly because they are designed to look and act as fashion accessories, but also because of marketing.
I'm not suggesting beating your child with a stick
By process of elimination I guess you are suggesting to beat the child with a carrot then. I mean, how many kids do you know that think a carrot is any kind of reward? Most kids I've seen, on the other hand, love to pick up and play with sticks.
That may be, but they also don't usually confuse "peddling" with "pedaling".
So you'd prefer to learn lessons from fairy tales than from actual history? Interesting approach. :)
it is a concept that I have never quite been able to grasp.
Try harder then. Seriously.
GPL isn't "weaponized". What does that even mean? Nothing, that's what. GPL is no more pernicious than basic copyright or any other license.
From your original post: That means that your "99% accurate" test is wrong 9,999 times out of 10,000!
Bzzzt! Wrong. It is wrong 9,999 times out of 1,000,000. What you have presented here is a blatant failure to do math correctly. I have shown you the correct math and you accuse me of plagiarism? Neither your post nor the Wikipedia entry included any information about why false positives were bad, which I did. Summary: there is no "paradox". What's important here is to focus on problems that arise from false positives. And you don't need to fudge the numbers or twist the meaning of words like "accuracy" to do that. I never did argue against the sentiment you expressed, just took issue with the math involved. But I guess your reading comprehension is as bad as your math.
In your hypothetical disease example, there is no stated reason why a false positive should be considered a bad thing. Who cares if the test gives us 9,999 false positives? Or 999,999? If the treatment is to drink an extra glass of water a day, and nothing bad happens to anyone because of that, then the cost of a false positive is vanishingly small. However, if the treatment is potentially lethal, as I illustrated (and neither you nor Wikipedia did), you need to tread carefully-- perhaps by putting in extra checks or repeating the test (as another poster suggests).
In the case of a terrorism detector, this is a tough call, now isn't it? Even if we don't fudge the numbers and twist the meaning of common words, as you are doing, we still have to decide what's an acceptable outcome when we have false positives--a problem for which this "paradox of the false positive" is completely useless.
Ooops. I mean 1/10,000 * 1/10,000 chance--the odds of two independent events. What I wrote first is the odds of picking any two specific people from the pool of 10,000.
Um. No. If you administer the test 1,000,000 times, then you have to use all 1,000,000 administrations of the test in your analysis of the results. You can't simply discard the negatives! That's called "sleight of hand" or "fudging the numbers". Unless what you really want to do is discuss the odds of a false positive. That's different than making statements about the overall accuracy of the test.
So, if you can't point me to a "competent statistician" other than Cory Doctorow and a pathetic stub page on Wikipedia, then I have to think that maybe I'm right about this. You don't need to fudge the numbers to make your point, though. In fact, a little competent application of statistics will go a long way here. It's critical to understand the underlying sentiment of this "paraox", which is that you have to really look at the consequences of those false positives.
In the disease example, what if the treatment for the disease is particularly nasty and causes 0.01% of the people who undergo it to die? Then out of your 10,000 people who test positive, one of them will die from the treatment. Assuming no correlation between the fatality of the treatment and the fatality of the disease, you have a 1/10,000 * 1/9,999 chance of the person who dies from the treatment being the same person who has the disease. So basically, all the test does is help you kill one person who doesn't have the disease and save the one who does. That is downright stupid. And if the treatment is any more lethal than that, or if the treatment is not a sure thing for the person with the disease, the tests are doing more harm than good.
In the case of terrorism, the problem of any terrorist detector finding an actual terrorist in the act of committing an attack, is probably even worse than 1/1,000,000. I'm guessing that there are something like 1,000,000 airplane rides a day in the USA. In the last five years there have been exactly zero airplane-based terrorist attacks attempted, that I'm aware of. So the real odds of anyone actually being a terrorist are, well, zero. But assume one attack every five years and the odds are something like 1/18,250,000,000. Now what exactly are we going to do with people when the terrorist detector is positive? And are we content to allow that when the odds of a false positive are overwhelming?
They can start by teaching Cory Doctorow how to count. The hypothetical test is wrong 9,999 times out of 1,000,000. Assuming, of course, that the test only produces false positives, and not also false negatives. That's what 99% accurate means.
Thank you. That's a balance. See how that works?
Of course, the GP's points were wrong on both the types of checks and balances that do exist, and that telco immunity is some sort of unconstitutional ex post facto law.
However, I agree that, even in the real world, a senator should vote against an entire bill that contains noxious items. Unless, of course, he is trading a "yes" vote here for a colleague's "yes" vote on another bill. And those are the sorts of behind the scenes dealings that are hard to make obvious to those of us outside the legislature... which makes the practice somewhat suspect in my mind. But I recognize it as the art of compromise, as well. The important thing here is to look at an overall trend and what these people tout as their own priorities. Is this immunity clause really go to have any effect or is it really something that should be a top priority? I don't think so. But that's my opinion. I've got other things I care a lot more about in the long run...