Sorry I should have clarified. All of my friends in college were econ majors and I just base my statements on what they've told me. I personally did not go down that road.
That explains why you have such a distorted view of economics. All government intervention falls into one of three cases.
The first cases is where externalities dominate. I don't want to live in a country where poor people can't get health care, because people without health care spread diseases that will make me sick. If there were no public health services to care for the poor, a plague would come to punish the rich for their arrogance. It has happened before and, if people like you get their way, it will happen again.
The second case is where a competitive market is impossible. Utilities are either government owned or heavily regulated because there is only room for one sewer, one power grid and one road network; these are called natural monopolies. If there was a company which could charge whatever it wanted for these things, it would charge tens or hundreds of times what they cost, because it is impossible for competitors to enter the market. I would rather be forced to pay a government that I can vote against than be forced to pay a private company which I have no control over.
The third case is simple altruism, helping the poor, through welfare and other programs. As much as you hate it, this is absolutely necessary. If you try to let people starve, they will steal instead. It is much cheaper to help someone out of poverty than to put them in prison, and make no mistake, that's exactly the choice we have.
If more politicians thought like you do, our society would collapse. Fortunately, our government isn't run by confused armchair economists.
Nice ad hominem, but you're talking nonsense yourself. You say that "the number of implementations that exist, or may exist, for a given algorithm, is not at all dictated by whether or not the algorithm is patented"; this is obviously false, because if an algorithm is patented, it is illegal for anyone but the patent holder to implement it. Of course patented algorithms have fewer implementations; that's the whole point of patenting it in the first place! Fewer implementations mean less competition which means lower quality. This is my thesis, and you haven't actually argued against it.
The relative difficulty of algorithm design vs. implementation isn't really relevant, but writing software is not a matter of translating pseudocode. A typical program has one or two complex algorithms couched in thousands of trivial ones, and the few complex algorithms usually aren't what determines its quality. (And don't presume to know what my background is; I've designed and implemented plenty of original algorithms, plus both of the algorithms you used as examples.)
This is only true if you honestly believe that the true ingenuity lies not in coming up with a novel algorithm for solving some sophisticated problem, but in performing the undergrad-level task of implementing said algorithm in code.
Sorry to burst your bubble, but there's nothing hard about algorithms. Coding up a large and complicated system, however, is very difficult. I don't care how cool your algorithm is; I care only about the program that contains it. And if the algorithm is patented, then there's only one such program, and it's probably shitty.
Your credit card number is not a password, because you have to give it away every time you buy something. If someone wants to steal a credit card number, they can get it from any unscrupulous employee of any business that sells things, which means they'll always succeed. The solution is to replace credit cards with smart cards that use public-key cryptography. That means that your credit card contains a number which you can use to sign transactions and prove that you are authorized to make payments, but you don't have to give every employee of every merchant you buy from the power to impersonate you.
Social security numbers have the same problem, only worse, because you can't just cancel your SSN like you can with a credit card. Banks pretend that your SSN is a password, but there are thousands of people who have access to your social security number and at least one of them will sell it on the black market.
Fixing this mess will cost the banks a lot of money, but they made this mess and it's their responsibility to clean it up. We need the federal government to mandate real security measures, because fraud is quickly becoming the norm.
Here's a thought: instead of fines, revoke the corporate charter for serious crimes. Only in America can people still get the death penalty while corporations can't.
Sounds great in theory, but would be terribly destructive; the actions of a few bad apples could cost thousands of people their jobs. A better solution is to replace the corporation's government; make all outstanding shares non-voting (thus stripping the board and the investors of their power) and hold an election for company CEO, with the old board ineligible and low-level employees making up most of the vote.
The problem with child pornography is that it's too easy to frame someone for. Up to a quarter of all computers may be part of botnets (source). For every compromised computer, there exists a person who could be paid to frame the computer's owner for pedophilia. If we spend a billion dollars on hunting down people with child porn on their computers, we're going to find a lot of people who didn't put it there but can't prove that they didn't. In other words, we will falsely accuse a lot of people, and ruin their lives and reputations. That will be a travesty.
This is certainly a neat concept. However, no one has made one - including the patenters. It won't be possible to make one until a lot of technologies have improved (especially battery technology).
Since it's impossible to make, there can't be prior art. Since it's being patented before it *could* be made, it never will be made. This is a very common, very ridiculous occurrence.
The article's claim is probably not true. While the amount of electricity used by data centers has been greatly increasing, that's only because so many new ones are being built. Eventually, we get to the point where everyone and their dog has their own data center, and the trend stops. Also, Moore's Law means that data centers in the future will do more work with less hardware and less electricity. Trying to predict how much electricity will be used by computers in 2020 is silly, because even Intel can't know that.
I don't think anyone would go to much trouble to tamper with the NJ Presidential primary. There's not enough at stake to justify the risk, and the discrepancies seen involve small numbers of votes. Thus, there are only two possible conclusions.
Either the voting machines are so unreliable that they introduce random errors, or someone is planning to tamper with the general election, and conducting a test run.
If done well, voting machine tampering would leave no evidence at all. We were lucky that some discrepancies showed up this time, because otherwise we would never have known. But now the bad guys know what they need to cover up. That's why it's vitally important that we get rid of these black-box voting machines and go back to a more primitive, more trustworthy system.
designing a language in 1978 which makes structured programming almost impossible is just insane
TeX is not a programming language, it's for typesetting. HTML doesn't support structured programming either, and nobody cares. If you need structured programming in TeX, you're doing it wrong.
As much as the Pentagon and the analysts are scummy liars, the real blame lies with the media outfits. Surely there are enough retired officers and enough military historians to use as a counterpoint. I mean, the news agencies had guys on the ground that, even with the limited access the Army gives them, knew from the beginning the problems with the Pentagon's story.
But only the Pentagon's hand-picked people got to see anything. Retired military officers don't get to tour military bases or get briefings from top generals and the Secretary of Defense. Media outlets had to choose between sources who were biased and sources who didn't know anything.
I've got a better reason; ISPs factor in the average bandwidth use when deciding prices. If 1/2 the bandwidth used by the average connection was stolen through WiFi the average person's internet bill would double, whether or not he had actually secured his connection.
No, it wouldn't! Prices would only go up by about 5%. Internet service is not like heating oil or tap water; it doesn't cost more to provide just because you use it more. If everyone suddenly started using twice as much bandwidth, they'd have to upgrade some routers, and that would be it. Routers are cheap. On the other hand, stringing a cable to your house, paying a techie to answer the phone and a lawyer to deal with the town is expensive.
Net neutrality means you can't bill your competitor's customers. This is absolutely essential to a free market.
See, there are actually four parties involved. The end user, Bob, buys a connection from an ISP, CableCo. Meanwhile, example.com, buys a connection from a different ISP, ExampleOnline. CableCo and ExampleOnline are competitors, but they have a peering agreement, which means that they agree to share the costs of a connection which lets Bob visit example.com. What's happening here is that CableCo is trying to get money from example.com. But example.com is ExampleOnline's customer! If ExampleOnline's customers are generating traffic which CableCo can't handle, then they need to renegotiate their peering agreement, not go after ExampleOnline's customers. That's unethical and possibly illegal.
Lets think of a situation where some guy has really pissed me off. I go onto a forum, identify him with a characteristic that can't be mistaken, the car he will be driving that I "own" (actually in this example his car, but you don't know that) which luckily for me is a very distinguishable car (or bicycle, whatever) claim he stole it and sit back waiting for him to be harrassed by forum members trying to find my stuff.
Then that would be libel, which is a tort. Fortunately, all that the car's owner would have to do to prove his ownership is show the registration certificate. The investigator would then report on the forum that the complaint was in error, the accuser would be discredited, and that's it.
The question is what is someone considering suicide going to do a search for - suicide, suicide consueling, or suicide methods?
They should have asked that question of a bunch of recent suicide attempters first.
They did. From the article, "the researchers collected 12 broad search terms gathered in part from interviews with those who had attempted suicide."
Broadband is too cheap, it's obvious that when you reduce ADSL to a low price comparable to dialup that the price becomes unsustainable if people are using lots of bandwidth.
This isn't obvious at all. Network equipment keeps getting faster and cheaper, and the only thing an ISP needs to provide more bandwidth is more and faster routers. ISPs like to pretend that your using ten times as much bandwidth costs them ten times as much, but it doesn't; it just means they have to upgrade a little sooner.
It's not just for your own good, it's also for everyone else's good. If you're vaccinated against a disease, you can't carry it and can't spread it. Ideally, everyone would be vaccinated, the disease would go extinct, and vaccinations would no longer be necessary. This has already happened for smallpox, and is close to happening for polio and measles. If you have people refusing to take the vaccine, then you can never get rid of the disease entirely, and so the rest of us would need to keep taking the vaccine forever.
"... The Geneva-based Centre on Housing Rights and Evictions estimates that 1.5 million of Beijing's natives will have been displaced from their homes by government edict when the Olympics finally begins." Someone please try to justify evicting one and a half million people for the Olympics.
Please cite your sources. I tracked down your source, the Center on Housing Rights and Evictions Violator Award - Beijing. From that same article, however,
The main areas in which evictions have been carried out within the Municipality of Beijing during the period between 2000 and 2007 are neighbourhoods in the four central districts of the capital where overcrowding and old or dangerous housing is common; namely Dongcheng, Xicheng, Chongwen and Xuanwu. Large-scale evictions have also been carried out in several Chengzhongcun (literally, villages in the city), poor informal settlements comprising housing that has not been approved for construction, does not comply with building codes and typically is not properly serviced.
They aren't evicting people to clear space for new development, they're evicting people from unsafe, overcrowded shanties. Sounds like they should've demolished those buildings a long time ago, but they're just getting around to it now because they don't want the rest of the world to see how bad they were.
This program almost certainly does not infringe on Blizzard's copyright. However, (1) this program exists for the sole purpose of cheating, and (2) cheating is a violation of Blizzard's terms of service. In other words, they're encouraging people to violate their contract with Blizzard, which could be considered tortious interference.
(Disclaimer: I am not a lawyer, and I don't know whether Blizzard is actually arguing this angle.)
I would sincerely like to be able to have a computer display in my glasses that I could view while walking around or standing in line.. at the very least providing something akin to a wearable Garmin gps device. The problem in my daydream has always been; how do I control the silly thing? How am I going to type? Mini keyboards like that on my phone are fine for short messages, but unsuited to any sort of real industrial typing and completely useless if I have to be walking or driving at the same time.
Typing with an EEG will never work. I think the best solution is a folding keyboard, such as one of these, but with a thumb keyboard on the outside. When sitting, you'd have a laptop-sized keyboard; standing, you'd have to use your thumbs. You'd still need your hands free, but at least you wouldn't have to look at your hands to see the screen, and you could put it in your pocket without losing sight of the screen. In order to type with both hands free, you'd probably need an implant of some kind, and that won't be worth it until surgical techniques get a whole lot better.
The main difference between Apple and other companies is who they design products for. Most companies try to identify a market segment, work out what people in that segment want to buy, and then produce a product for that market. Apple tries to produce products for Steve Jobs. He has fairly good taste and so often those are also products that other people want to buy.
It's simpler than that. Most people don't know what they want until they see it, so asking people what they want is a waste of time. Steve Jobs wants the same things most other people do, but he knows what he wants before he's seen it. Thus, it's easier to design for him than for a focus group.
Yeah? It'd also have cleaned up all the "legacy" software people are using. Like iTunes. Not to mention all the actual legacy software like kids educational software, drivers for old hardware, etc. I also don't know why he thinks this would have cleaned up viruses and spyware. These guys adapt fast and the extra anti-patch systems in 64 Vista aren't all that strong.
It would've broken all the old drivers, yes, but they did that anyways. Regular applications would be unaffected, because x86-64 processors can run 32-bit programs just fine.
Re:Wrong marketing did them in, clock *does* matte
on
Is AMD Dead Yet?
·
· Score: 3, Insightful
The only way you could have gotten a performance difference that large is if the Pentium 4 was using an SIMD extension which the AMD CPU wasn't using. In other words, if the test was specifically optimized in favor of the Pentium 4 and not optimized in the same way for AMD.
Yes, clock does matter, but there are tradeoffs, and Intel chose to maximize clock frequency at the expense of all else. AMD had to either explain that to customers, or switch to using an actual benchmark to measure performance. Argue all you like about which benchmark they chose, but it was the right decision.
Slashdot Mirror
The first cases is where externalities dominate. I don't want to live in a country where poor people can't get health care, because people without health care spread diseases that will make me sick. If there were no public health services to care for the poor, a plague would come to punish the rich for their arrogance. It has happened before and, if people like you get their way, it will happen again.
The second case is where a competitive market is impossible. Utilities are either government owned or heavily regulated because there is only room for one sewer, one power grid and one road network; these are called natural monopolies. If there was a company which could charge whatever it wanted for these things, it would charge tens or hundreds of times what they cost, because it is impossible for competitors to enter the market. I would rather be forced to pay a government that I can vote against than be forced to pay a private company which I have no control over.
The third case is simple altruism, helping the poor, through welfare and other programs. As much as you hate it, this is absolutely necessary. If you try to let people starve, they will steal instead. It is much cheaper to help someone out of poverty than to put them in prison, and make no mistake, that's exactly the choice we have.
If more politicians thought like you do, our society would collapse. Fortunately, our government isn't run by confused armchair economists.
Nice ad hominem, but you're talking nonsense yourself. You say that "the number of implementations that exist, or may exist, for a given algorithm, is not at all dictated by whether or not the algorithm is patented"; this is obviously false, because if an algorithm is patented, it is illegal for anyone but the patent holder to implement it. Of course patented algorithms have fewer implementations; that's the whole point of patenting it in the first place! Fewer implementations mean less competition which means lower quality. This is my thesis, and you haven't actually argued against it.
The relative difficulty of algorithm design vs. implementation isn't really relevant, but writing software is not a matter of translating pseudocode. A typical program has one or two complex algorithms couched in thousands of trivial ones, and the few complex algorithms usually aren't what determines its quality. (And don't presume to know what my background is; I've designed and implemented plenty of original algorithms, plus both of the algorithms you used as examples.)
Your credit card number is not a password, because you have to give it away every time you buy something. If someone wants to steal a credit card number, they can get it from any unscrupulous employee of any business that sells things, which means they'll always succeed. The solution is to replace credit cards with smart cards that use public-key cryptography. That means that your credit card contains a number which you can use to sign transactions and prove that you are authorized to make payments, but you don't have to give every employee of every merchant you buy from the power to impersonate you.
Social security numbers have the same problem, only worse, because you can't just cancel your SSN like you can with a credit card. Banks pretend that your SSN is a password, but there are thousands of people who have access to your social security number and at least one of them will sell it on the black market.
Fixing this mess will cost the banks a lot of money, but they made this mess and it's their responsibility to clean it up. We need the federal government to mandate real security measures, because fraud is quickly becoming the norm.
The problem with child pornography is that it's too easy to frame someone for. Up to a quarter of all computers may be part of botnets (source). For every compromised computer, there exists a person who could be paid to frame the computer's owner for pedophilia. If we spend a billion dollars on hunting down people with child porn on their computers, we're going to find a lot of people who didn't put it there but can't prove that they didn't. In other words, we will falsely accuse a lot of people, and ruin their lives and reputations. That will be a travesty.
This is certainly a neat concept. However, no one has made one - including the patenters. It won't be possible to make one until a lot of technologies have improved (especially battery technology).
Since it's impossible to make, there can't be prior art. Since it's being patented before it *could* be made, it never will be made. This is a very common, very ridiculous occurrence.
The article's claim is probably not true. While the amount of electricity used by data centers has been greatly increasing, that's only because so many new ones are being built. Eventually, we get to the point where everyone and their dog has their own data center, and the trend stops. Also, Moore's Law means that data centers in the future will do more work with less hardware and less electricity. Trying to predict how much electricity will be used by computers in 2020 is silly, because even Intel can't know that.
I don't think anyone would go to much trouble to tamper with the NJ Presidential primary. There's not enough at stake to justify the risk, and the discrepancies seen involve small numbers of votes. Thus, there are only two possible conclusions.
Either the voting machines are so unreliable that they introduce random errors, or someone is planning to tamper with the general election, and conducting a test run.
If done well, voting machine tampering would leave no evidence at all. We were lucky that some discrepancies showed up this time, because otherwise we would never have known. But now the bad guys know what they need to cover up. That's why it's vitally important that we get rid of these black-box voting machines and go back to a more primitive, more trustworthy system.
Net neutrality means you can't bill your competitor's customers. This is absolutely essential to a free market.
See, there are actually four parties involved. The end user, Bob, buys a connection from an ISP, CableCo. Meanwhile, example.com, buys a connection from a different ISP, ExampleOnline. CableCo and ExampleOnline are competitors, but they have a peering agreement, which means that they agree to share the costs of a connection which lets Bob visit example.com. What's happening here is that CableCo is trying to get money from example.com. But example.com is ExampleOnline's customer! If ExampleOnline's customers are generating traffic which CableCo can't handle, then they need to renegotiate their peering agreement, not go after ExampleOnline's customers. That's unethical and possibly illegal.
It's not just for your own good, it's also for everyone else's good. If you're vaccinated against a disease, you can't carry it and can't spread it. Ideally, everyone would be vaccinated, the disease would go extinct, and vaccinations would no longer be necessary. This has already happened for smallpox, and is close to happening for polio and measles. If you have people refusing to take the vaccine, then you can never get rid of the disease entirely, and so the rest of us would need to keep taking the vaccine forever.
This program almost certainly does not infringe on Blizzard's copyright. However, (1) this program exists for the sole purpose of cheating, and (2) cheating is a violation of Blizzard's terms of service. In other words, they're encouraging people to violate their contract with Blizzard, which could be considered tortious interference.
(Disclaimer: I am not a lawyer, and I don't know whether Blizzard is actually arguing this angle.)
It would've broken all the old drivers, yes, but they did that anyways. Regular applications would be unaffected, because x86-64 processors can run 32-bit programs just fine.
The only way you could have gotten a performance difference that large is if the Pentium 4 was using an SIMD extension which the AMD CPU wasn't using. In other words, if the test was specifically optimized in favor of the Pentium 4 and not optimized in the same way for AMD.
Yes, clock does matter, but there are tradeoffs, and Intel chose to maximize clock frequency at the expense of all else. AMD had to either explain that to customers, or switch to using an actual benchmark to measure performance. Argue all you like about which benchmark they chose, but it was the right decision.
Wrong. The original copyright was for 14 years, with an optional 14 year extension. Copyright would expire even if the author was still alive.