It's not a measure of efficiency. Nothing in the original post claimed it was. It is, however, a legitimate measure of the social benefit created by the industry. Much has been written about the jobless recovery that has created enormous profits for a small number of people without helping most of the population. If subsidies for fossil fuels merely inflate the already egregious oil company profits, while subsidies for solar energy create jobs, that is a clear advantage to the latter.
This is independent of questions of efficiency (megawatts/dollar), environmental impact, etc.
I agree - "reducing duplication" is not the way to reform academic research. But there are a lot of other ways it could be made more efficient.
Consider that most professors don't actually do much research; instead, they manage research groups. That's fine if they want to be managers, but more often the system just forces them in that direction. The result is that most real research is done by postdocs and new professors in their first few years on the job. A typical researcher gets maybe 10 years to really focus on doing research, and then they get sucked off into other things. The very people who are best qualified to do research aren't allowed to do it.
Then there's the way so many research projects are tied to one particular person (often a grad student or postdoc), and when they move on, the project dies. Their samples get pushed to the back of the freezer, their code languishes on a hard drive somewhere, their equipment gets shoved under a table to gather dust, and eventually it all gets thrown out. Far too many academic groups are organized as "a set of people doing projects that interest them" rather than "a research organization coordinating a set of research projects that are judged to be important."
None of this is to say we should be cutting research funding. But we really should reconsider the whole academic research system and consider how it could be made more efficient.
What I said was, "The solar spectrum is most intense in the visible part of the spectrum." And that is completely and totally correct. The fact that it's only a small part of the spectrum does not change the fact that the maximum solar intensity is right in the middle of that part.
It hardly makes a difference. Take a look at the solar spectrum. That graph only goes up to about 2.5 um, at which point the energy is already pretty insignificant. At 3 um it's even lower.
I think you're getting confused. What matters most is reflectivity, not emissivity. The solar spectrum is most intense in the visible part of the spectrum (no surprise - that's why our eyes evolved to be sensitive to that part of the spectrum), and white paint is much more reflective in that part of the spectrum than black paint (by definition - that's what it means to be white!).
And why are you quoting the emissivity of a particular paint at 3m wavelength? That's radio frequency. Solar energy in that frequency range is completely negligible compared to visible.
Experience definitely counts for a lot in programming. I certainly think I'm a better programmer now than I was ten years ago. I've learned a lot in that time!
On the other hand, I do think there's some truth in the suggestion that better programmers are more likely to stay in the field. The best programmers are the people who love doing it, who come home from a day of programming at work to spend their evening doing their own programming projects, who are always learning about new subjects and techniques just for the fun of it. And those are precisely the people who are most likely to turn down that promotion to management; who, if they get laid off, don't even consider switching fields because why would they ever want to do something else? Mediocre programmers leave the field as they get older, leaving only the better ones.
They are being put into use. Consider this recent article:
The cost of solar cells, the main component in standard panels, has fallen 21 percent so far this year...
Think about that. A 21 percent drop in price so far this year??? New technologies are bringing down the cost of solar at a really incredible rate. That's not some futuristic, maybe it'll be available some day technology. It's what's on the market right now.,
Actually, solar thermal plants require less land than either coal (if you count the land required for mining) or hydroelectric (if you count the land required for storing water behind the dam). And then there's photovoltaics, which can usually be put on the roofs of existing buildings, requiring effectively no land at all.
If you store cash under your bed and someone breaks into your house, its gone. That's why most people don't keep their money under their beds - they put it in a bank. The bank has much better security than you do. If someone robs a bank, it's the bank that loses money, not the customers. And if they lose so much money they can't pay back the customers, then the government covers it.
I wouldn't store large sums of money in a file on my computer, but once my bank offers to let me deposit bitcoins and let them manage the security, then it becomes a real option.
The paper even admits these comparisons are completely invalid. From the "Java Tunings" section:
Jeremy Manson brought the performance of Java on par with the original C++ version. This version is kept in the java_pro directory. Note that Jeremy deliberately refused to optimize the code further, many of the C++ optimizations would apply to the Java version as well.
So they intentionally optimized the Java version much less than the C++ version, and they know the Java version could be sped up by applying the same optimizations they used for the C++ version. And (who would believe it?) the C++ version came out faster than the Java version! What a surprise!
Or a more directly relevant comparison: you can buy an unlocked G2x for $500, or an unlocked Nexus S for $529. Both of those are pretty comparable phones to the iPhone, and arguably are better. For example, the G2x has a dual core processor and supports 4G. It only has 8 GB built in storage, but you can add a 32 GB SD card for $50. Given that, $749 for a 32 GB iPhone seems pretty egregious.
I wouldn't call that "little to do with MS". The only reason they're having trouble competing with Android is that they don't themselves make Android phones. Instead they have their own smartphone OS that can't compete, and they're switching from that to a different smartphone OS that also shows no sign of being able to compete. If they began making Android phones, they would then be competing strictly on hardware. That would have been a bad thing when their OS was a positive differentiator. But today it's become a major liability. Having the same OS as everyone else would now be a big improvement for them.
The problem, of course, is that they can't do that. They've signed a contract with MS agreeing to use WP exclusively.
Hijacking is hardly the only form of terrorism. A nice explosive would do just as well. And maglevs are really ideal for that. If you blow up an airplane, you just destroy that one plane. If you blow up a train, you prevent any train from getting through until the rubble is cleared away and the track is repaired. For an ordinary train track, that's not too hard, but maglev track is much more complicated and incredibly expensive.
Note that T-mobile has two different types of prepaid plans. With "pay as you go" plans, you put in an amount of money that can be used for voice, SMS, or data. With "pay by the month" plans, you pay a fixed amount per month, but with no multi-month commitment. Either one might end up best for you, depending on how much you plan to use the phone.
Why do you believe they'll let people get into their many-billion-dollar, super high tech maglev without requiring them to go through a security scan first?
I don't pretend to understand the proof, but physicists are adamant that hidden variables have been ruled out.
Not true at all. Only specific classes of hidden variables theories have been ruled out (via tests of Bell's inequality). There are two major limitations on these experiments:
1. Bell's inequality only applies to local realistic theories. Think of the words "local" and "realistic" as just being jargon: they have specific technical meanings in this context that don't necessarily match their colloquial, everyday meanings. Suffice to say that some hidden variables theories do not have both those properties.
2. No actual violation of Bell's inequality has every been observed. To compensate for experimental limitations, all the tests so far have had to make additional assumptions beyond those used in deriving Bell's inequality. Those assumptions are not valid for some hidden variables theories. (People who believe in hidden variables tend to speak of these as "supplementary assumptions," while those who don't believe in hidden variables speak of them as "loopholes". But both names refer to the same thing.)
The problem is that when you plunk out your money to buy the shiny new, it is not a sale. Legally, you're buying a "license" to use the game, which gives game companies the ability to dictate and/or change the terms of those license at their will.
Let me tell you a story. This story didn't actually happen in the real world. It only took place in a fantasy world that software companies wished they lived in.
I went into a store that sells software. (Pick your favorite - GameStop, Target, Office Depot, whatever.) They had lots of boxed software on the shelves. I selected the one I wanted, took it up to the register, and said, "I'd like to buy this." The cashier replied, "I'm sorry but that isn't for sale. If you'd like to license it, though, please read this agreement and sign here."
Like I said, that didn't really happen. Here's what actually happened.
I told the cashier, "I'd like to buy this," and they replied, "That will be $30." So I handed over my credit card, they scanned it, and I signed the receipt. Then they handed me the software back and said, "Here you are."
At this point, the sale was complete. The purchase agreement was concluded. So I took the boxed software (which I now owned, since it had legally been sold to me) home and opened it up. Inside was a piece of paper covered with tiny writing that I did not bother to read, since it obviously contained no information of value. But if I had read it, I would have discovered it to be mostly incomprehensible legalese. If I had then taken it to a lawyer, they could have translated it for me as follows: "You know that software you just bought? You didn't really buy it. You actually just licensed it, and if you want to use it, you have to agree to all these other terms we didn't tell you about before you bought it and before we took your money."
The only reasonable response, of course, is "@#$% you! I do not agree to any of your terms. Nonetheless, I will continue to use this software, because it belongs to me and I have every right to use it." A purchase it a contract. Once that contract is complete - once they have taken my money and given me the software - it's too late for them to rewrite the agreement and add in new terms.
Exactly. Just look at the list of "best practices" in the summary:
We all know by now that Test Driven Development is a best practice.
According to who? I certainly don't know that. There are a limited number of cases where TDD makes sense, and a much larger number of cases where it doesn't make sense and shouldn't be used.
And 70% unit test coverage.
Who came up with that number??? Really, it depends enormously on what you're doing. For low level computational code, I feel uncomfortable with anything less than 100% coverage. For UI code, unit tests usually don't make sense and I'm quite happy with 0% coverage.
After reading the article, I think he has some really good points, and a lot of what he says is true. I've seen many times how ingrown disciplines can get, and how ignorant people can be of relevant work being done in the next building over. Each field tends to follow its own course, with everyone doing things the way "everyone else" (that is, everyone in their field) does it because they assume that's the best way. Then someone will "discover" a method that's been widely known for decades in other fields, and it will revolutionize their work. How many times did the FFT get reinvented over the years? And even today, most physics engines use integration methods that were state of the art in 1900. If people who write physics engines talked to numerical analysts a little more often, they would learn that better techniques have been known for decades.
And it's not just math or science. I remember hearing a talk by a psychology professor who was really interested in literature and tried to do a collaboration with a literature professor. After explaining his own research (cognitive psych studies of what happens when you read a fictional story), the literature professor informed him, "That isn't the sort of psychology we're interested in in literature." A little surprised, he asked, "What sort of psychology are you interested in?" And, as he put it, "It quickly became apparent that the sort of psychology they were interested in was a sort no psychologist had believed in for 50 years."
No matter what happens with the economy, no matter what trends develop with off-shoring, no matter what happens in the financial services industry, there is one thing you can be sure of: people with a good education will be better positioned than those without. And a good education in a STEM field offers more possibilities than almost any other. You can mourn the engineering grads going to work in finance, but keep in mind: they have access to those jobs because of their engineering degrees. All else being equal, a person with a strong technical background will almost always have more opportunities than a person with a non-technical background.
According to IGN, their latest information is that it will have a triple core PPC processor, similar to the one in the Xbox 360 but running at a higher clock rate. They also claim it will have an AMD R700 based GPU (that's the Radeon HD 4XXX generation).
Yes, I agree with everything that people are saying. The main causes of poverty are sociological, not technological. Corruption, archaic legal systems, etc. do far more to keep people in poverty then lack of access to tractors. And yet...
Think about the possibilities of this. Create open source designs for a ton of useful industrial devices. Build up a community of people making steady improvements to them. Sure, the initial users won't be impoverished farmers in central Africa. But think of all the people in the developed world with the interest and resources to make use of this. And it can grow from there. Want to start a company to manufacture industrial equipment? Here are the detailed plans, all ready to go. This is where open source software was in the mid-80s: only useful to a handful of people, nearly all of whom lived in wealthy countries. So no, this project isn't likely to cure poverty on its own, but it can still grow into something incredibly valuable.
Slashdot Mirror
It's not a measure of efficiency. Nothing in the original post claimed it was. It is, however, a legitimate measure of the social benefit created by the industry. Much has been written about the jobless recovery that has created enormous profits for a small number of people without helping most of the population. If subsidies for fossil fuels merely inflate the already egregious oil company profits, while subsidies for solar energy create jobs, that is a clear advantage to the latter.
This is independent of questions of efficiency (megawatts/dollar), environmental impact, etc.
I agree - "reducing duplication" is not the way to reform academic research. But there are a lot of other ways it could be made more efficient.
Consider that most professors don't actually do much research; instead, they manage research groups. That's fine if they want to be managers, but more often the system just forces them in that direction. The result is that most real research is done by postdocs and new professors in their first few years on the job. A typical researcher gets maybe 10 years to really focus on doing research, and then they get sucked off into other things. The very people who are best qualified to do research aren't allowed to do it.
Then there's the way so many research projects are tied to one particular person (often a grad student or postdoc), and when they move on, the project dies. Their samples get pushed to the back of the freezer, their code languishes on a hard drive somewhere, their equipment gets shoved under a table to gather dust, and eventually it all gets thrown out. Far too many academic groups are organized as "a set of people doing projects that interest them" rather than "a research organization coordinating a set of research projects that are judged to be important."
None of this is to say we should be cutting research funding. But we really should reconsider the whole academic research system and consider how it could be made more efficient.
And your point is?
What I said was, "The solar spectrum is most intense in the visible part of the spectrum." And that is completely and totally correct. The fact that it's only a small part of the spectrum does not change the fact that the maximum solar intensity is right in the middle of that part.
http://xkcd.com/774
Wrong. Take a look at the spectrum and see for yourself: http://en.wikipedia.org/wiki/File:Solar_Spectrum.png.
It hardly makes a difference. Take a look at the solar spectrum. That graph only goes up to about 2.5 um, at which point the energy is already pretty insignificant. At 3 um it's even lower.
I think you're getting confused. What matters most is reflectivity, not emissivity. The solar spectrum is most intense in the visible part of the spectrum (no surprise - that's why our eyes evolved to be sensitive to that part of the spectrum), and white paint is much more reflective in that part of the spectrum than black paint (by definition - that's what it means to be white!).
And why are you quoting the emissivity of a particular paint at 3m wavelength? That's radio frequency. Solar energy in that frequency range is completely negligible compared to visible.
Experience definitely counts for a lot in programming. I certainly think I'm a better programmer now than I was ten years ago. I've learned a lot in that time!
On the other hand, I do think there's some truth in the suggestion that better programmers are more likely to stay in the field. The best programmers are the people who love doing it, who come home from a day of programming at work to spend their evening doing their own programming projects, who are always learning about new subjects and techniques just for the fun of it. And those are precisely the people who are most likely to turn down that promotion to management; who, if they get laid off, don't even consider switching fields because why would they ever want to do something else? Mediocre programmers leave the field as they get older, leaving only the better ones.
The cost of solar cells, the main component in standard panels, has fallen 21 percent so far this year...
Think about that. A 21 percent drop in price so far this year??? New technologies are bringing down the cost of solar at a really incredible rate. That's not some futuristic, maybe it'll be available some day technology. It's what's on the market right now.,
They also had something just like this at Maker Faire. This is definitely not the first.
Wind and solar take a LOT of space.
Actually, solar thermal plants require less land than either coal (if you count the land required for mining) or hydroelectric (if you count the land required for storing water behind the dam). And then there's photovoltaics, which can usually be put on the roofs of existing buildings, requiring effectively no land at all.
If you store cash under your bed and someone breaks into your house, its gone. That's why most people don't keep their money under their beds - they put it in a bank. The bank has much better security than you do. If someone robs a bank, it's the bank that loses money, not the customers. And if they lose so much money they can't pay back the customers, then the government covers it.
I wouldn't store large sums of money in a file on my computer, but once my bank offers to let me deposit bitcoins and let them manage the security, then it becomes a real option.
Jeremy Manson brought the performance of Java on par with the original C++ version. This version is kept in the java_pro directory. Note that Jeremy deliberately refused to optimize the code further, many of the C++ optimizations would apply to the Java version as well.
So they intentionally optimized the Java version much less than the C++ version, and they know the Java version could be sped up by applying the same optimizations they used for the C++ version. And (who would believe it?) the C++ version came out faster than the Java version! What a surprise!
Truly nothing to see here. Please move along.
Or a more directly relevant comparison: you can buy an unlocked G2x for $500, or an unlocked Nexus S for $529. Both of those are pretty comparable phones to the iPhone, and arguably are better. For example, the G2x has a dual core processor and supports 4G. It only has 8 GB built in storage, but you can add a 32 GB SD card for $50. Given that, $749 for a 32 GB iPhone seems pretty egregious.
I wouldn't call that "little to do with MS". The only reason they're having trouble competing with Android is that they don't themselves make Android phones. Instead they have their own smartphone OS that can't compete, and they're switching from that to a different smartphone OS that also shows no sign of being able to compete. If they began making Android phones, they would then be competing strictly on hardware. That would have been a bad thing when their OS was a positive differentiator. But today it's become a major liability. Having the same OS as everyone else would now be a big improvement for them.
The problem, of course, is that they can't do that. They've signed a contract with MS agreeing to use WP exclusively.
Hijacking is hardly the only form of terrorism. A nice explosive would do just as well. And maglevs are really ideal for that. If you blow up an airplane, you just destroy that one plane. If you blow up a train, you prevent any train from getting through until the rubble is cleared away and the track is repaired. For an ordinary train track, that's not too hard, but maglev track is much more complicated and incredibly expensive.
Note that T-mobile has two different types of prepaid plans. With "pay as you go" plans, you put in an amount of money that can be used for voice, SMS, or data. With "pay by the month" plans, you pay a fixed amount per month, but with no multi-month commitment. Either one might end up best for you, depending on how much you plan to use the phone.
Why do you believe they'll let people get into their many-billion-dollar, super high tech maglev without requiring them to go through a security scan first?
I don't pretend to understand the proof, but physicists are adamant that hidden variables have been ruled out.
Not true at all. Only specific classes of hidden variables theories have been ruled out (via tests of Bell's inequality). There are two major limitations on these experiments:
1. Bell's inequality only applies to local realistic theories. Think of the words "local" and "realistic" as just being jargon: they have specific technical meanings in this context that don't necessarily match their colloquial, everyday meanings. Suffice to say that some hidden variables theories do not have both those properties.
2. No actual violation of Bell's inequality has every been observed. To compensate for experimental limitations, all the tests so far have had to make additional assumptions beyond those used in deriving Bell's inequality. Those assumptions are not valid for some hidden variables theories. (People who believe in hidden variables tend to speak of these as "supplementary assumptions," while those who don't believe in hidden variables speak of them as "loopholes". But both names refer to the same thing.)
The problem is that when you plunk out your money to buy the shiny new, it is not a sale. Legally, you're buying a "license" to use the game, which gives game companies the ability to dictate and/or change the terms of those license at their will.
Let me tell you a story. This story didn't actually happen in the real world. It only took place in a fantasy world that software companies wished they lived in.
I went into a store that sells software. (Pick your favorite - GameStop, Target, Office Depot, whatever.) They had lots of boxed software on the shelves. I selected the one I wanted, took it up to the register, and said, "I'd like to buy this." The cashier replied, "I'm sorry but that isn't for sale. If you'd like to license it, though, please read this agreement and sign here."
Like I said, that didn't really happen. Here's what actually happened.
I told the cashier, "I'd like to buy this," and they replied, "That will be $30." So I handed over my credit card, they scanned it, and I signed the receipt. Then they handed me the software back and said, "Here you are."
At this point, the sale was complete. The purchase agreement was concluded. So I took the boxed software (which I now owned, since it had legally been sold to me) home and opened it up. Inside was a piece of paper covered with tiny writing that I did not bother to read, since it obviously contained no information of value. But if I had read it, I would have discovered it to be mostly incomprehensible legalese. If I had then taken it to a lawyer, they could have translated it for me as follows: "You know that software you just bought? You didn't really buy it. You actually just licensed it, and if you want to use it, you have to agree to all these other terms we didn't tell you about before you bought it and before we took your money."
The only reasonable response, of course, is "@#$% you! I do not agree to any of your terms. Nonetheless, I will continue to use this software, because it belongs to me and I have every right to use it." A purchase it a contract. Once that contract is complete - once they have taken my money and given me the software - it's too late for them to rewrite the agreement and add in new terms.
We all know by now that Test Driven Development is a best practice.
According to who? I certainly don't know that. There are a limited number of cases where TDD makes sense, and a much larger number of cases where it doesn't make sense and shouldn't be used.
And 70% unit test coverage.
Who came up with that number??? Really, it depends enormously on what you're doing. For low level computational code, I feel uncomfortable with anything less than 100% coverage. For UI code, unit tests usually don't make sense and I'm quite happy with 0% coverage.
And keeping your CCN complexity numbers below 20.
You've got to be kidding.
After reading the article, I think he has some really good points, and a lot of what he says is true. I've seen many times how ingrown disciplines can get, and how ignorant people can be of relevant work being done in the next building over. Each field tends to follow its own course, with everyone doing things the way "everyone else" (that is, everyone in their field) does it because they assume that's the best way. Then someone will "discover" a method that's been widely known for decades in other fields, and it will revolutionize their work. How many times did the FFT get reinvented over the years? And even today, most physics engines use integration methods that were state of the art in 1900. If people who write physics engines talked to numerical analysts a little more often, they would learn that better techniques have been known for decades.
And it's not just math or science. I remember hearing a talk by a psychology professor who was really interested in literature and tried to do a collaboration with a literature professor. After explaining his own research (cognitive psych studies of what happens when you read a fictional story), the literature professor informed him, "That isn't the sort of psychology we're interested in in literature." A little surprised, he asked, "What sort of psychology are you interested in?" And, as he put it, "It quickly became apparent that the sort of psychology they were interested in was a sort no psychologist had believed in for 50 years."
No matter what happens with the economy, no matter what trends develop with off-shoring, no matter what happens in the financial services industry, there is one thing you can be sure of: people with a good education will be better positioned than those without. And a good education in a STEM field offers more possibilities than almost any other. You can mourn the engineering grads going to work in finance, but keep in mind: they have access to those jobs because of their engineering degrees. All else being equal, a person with a strong technical background will almost always have more opportunities than a person with a non-technical background.
According to IGN, their latest information is that it will have a triple core PPC processor, similar to the one in the Xbox 360 but running at a higher clock rate. They also claim it will have an AMD R700 based GPU (that's the Radeon HD 4XXX generation).
I think this is an awesome idea.
Yes, I agree with everything that people are saying. The main causes of poverty are sociological, not technological. Corruption, archaic legal systems, etc. do far more to keep people in poverty then lack of access to tractors. And yet...
Think about the possibilities of this. Create open source designs for a ton of useful industrial devices. Build up a community of people making steady improvements to them. Sure, the initial users won't be impoverished farmers in central Africa. But think of all the people in the developed world with the interest and resources to make use of this. And it can grow from there. Want to start a company to manufacture industrial equipment? Here are the detailed plans, all ready to go. This is where open source software was in the mid-80s: only useful to a handful of people, nearly all of whom lived in wealthy countries. So no, this project isn't likely to cure poverty on its own, but it can still grow into something incredibly valuable.