So treat it like a school / uni essay. You know the "marker" will spend ~30 seconds reading it, so make sure it's "well structured", so a busy reader can get the gist of your argument with next to no effort..
But make sure it's well padded, so they know you put some effort in.
See, school has been useful. It's taught you how to "communicate"!
"What is the best treatment for stage 1 prostate cancer".
Obviously, there's a number of factors. "Wait and see" is best for an old man with a heart condition, as their heart will give out before the cancer grows. But usually, the best treatment will be something else.
Arguably, the question should include all the context, but it's not always feasible. Also, when it's a close call, the question is probably a bad question.
China and India. Fraud and plagiarism are pretty prevalent in both.
Of course, they don't get much fraudulent or plagiarized work into big journals, and the big journals prefer researchers with a good reputation. I suspect that analysing the actual retraction data would show who's responsible. Is it hungrier researchers in newly developed countries (who have a greater incentive to lie), or researchers in "good" countries (or "good" universities) who aren't subject to as much scrutiny?
You can probably tell by breaking down the data. Until someone does, it's not worth speculating.
Actually, I'd imagine that professional photographers mostly just bitch about bad clients, discuss places where you can find interesting subjects, talk about how overrated some famous photos are, and so on.
There's violent bits in the Koran. There's violent bits in the Bible. Chinese Confucianism-Taoism-Buddhism has some objectionable bits.
Even if any of it is the word of God, God doen't exactly stop people misinterpreting it, or just making shit up. Even other true believers are usually pretty accepting of really screwy interpretations.
It's a myth that religions have any existential properties, because religions DO NOT EXIST. They are just words, and words can have a whole range of different meanings. They can have works associated with them (holy books), but those books can be so widely interpreted (just look at the shit some people think the Bible says) that the books have only a small influence on how people interpret the religions.
In multiple choice questions, the "most correct" answer is the right one. Otherwise, all answers can be correct, if you argue hard enough (if it's at all subjective).
The problem is, they used a stupid question - you can scientifically test the "softness" or "sweetness" of a flower. There should be one that's obviously "most correct".
For in-class quizzes, it's not so important (as the student can challenge it), but for a state-wide test there shouldn't be any wriggle room.
Most of the advantages of VPNs (fast disk IO, admin rights) are kind of irrelevant for this. Of course, you'll hammer their bandwidth, so they might block you for that.
ML is barely an OS upgrade. It's just a chance for the OSX app team to write some stuff which works with Lion's new features (some of which are interesting), port some iOS stuff, and charge $30 for a bunch of todo / notification apps.
I'd expect the next upgrade to add some new stuff (like Lion and Leopard did), while ML is just consolidation / apps. It's kind of like Intel's "Tick Tock", without any real benefit from the "Tock".
The difference is, Chinese don't often have handlers. They aren't generally spies, but know that if they can knock fo some IP, they can be very successful in a Chinese university.
This is not too different from what other academics do - it's quite common for academics to leave with a USB stick full of the stuff they were working on, which they use in their next gig.
I guess there might be some more active encouragement in strategic stuff.
At $500 / hour, 40 hours a week, 52 weeks a year, that's over 500 lawyer years. No, I don't think they spent anywhere near that much. Probably an order of magnitude less - 50 lawyer years for $70 million.
That doesn't include legal costs and wasted time for Apple, Intel, etc. That's the real cost of patents - not the cost of enforcing good ones, but the costs of transaction which both sides have to carry.
If no-one (except the police, and professional criminals who'll mostly use them in gang wars), everyone will be safer.
But if you have a gun, you'll feel safer, but make everyone else less safe (as a criminal may sell it). This is especially true if street criminals (who probably wouldn't have guns if they weren't so cheap on the blackmarket) don't have them.
It's a prisoner's dilemma. Guns are good for you, because they make you feel a little safer. They are terrible for society, because they are stolen by thieves, who sell them to gangster wannabes who wouldn't normally be able to get them.
In Australia, bills are required to have a title which properly scopes them. If you give a bill a goofy title, there's a risk that a high court challenge will have it invalidated. So they are all given boring names, like "Federal Financial Relations Amendment (National Health Reform) Bill 2012".
Also, were it not for western industrial espionage against China, we wouldn't have paper or porcelain or tea.
The US doesn't steal commercial know-how because they already have plenty. China is decades behind (in some areas), and can benefit a lot from acquiring foreign IP.
In fact, China's subsidies of industrial inputs (land, energy, water, steel, etc) are there to drag in foreign manufacturing. Want to guess why they want everything made in China? It's so they can figure out how to make it themselves.
It's a hell of a lot better than invading resource-rich countries to try to build up your industrial base. And if no-one ever stole secrets, we'd still all be in the dark ages.
It is "good communication". It does add a lot of value. Malcolm Gladwell famous because he's a great communicator. But it would be silly to call him a scientist.
On the other hand, some doctors will work harder to keep potential donors ticking (even if they think they won't recover), because they now they are still worth keeping warm.
If you aren't a donor, and the doctor thinks you won't recover, they'll prefer to pull the plug and move on to someone worth saving.
Mistakes are made at all points in the chain, but I think this one is more likely than when they decide to turn off life support. There's a lot more process for switching a patient off than there is for saying "let's call it, he's not coming back".
> Government figures from the health ministry show that about 1.5 million people in China need transplants, but only 10,000 transplants are performed annually, according to Xinhua.
In other words, prisoner "donations" just aren't enough. FTA, they don't like using condemned prisoner organs, because they aren't usually in good shape anyway. Nobody in China wants to donate, because of cultural reasons (they are selfish? they think it's "icky"? they don't trust doctors? it's not really Buddhist? no idea).
I'm waiting to hear what the "new national donation system" is. Perhaps registered donors will jump to the top of the list, in order of registration date. Everyone will sign up, knowing that if they don't they'll die waiting for a transplant if they need one.
It was also a blatant rip-off of "Darkness at Noon", a thinly-veiled fictionalization of the Moscow Trials. But Koestler was a rapist as well as a genius, so his books haven't really done as well as they should (except the one about rape... go figure).
Slashdot Mirror
Have a look at "Collective Intelligence". It's a great crash course in ML, comparable to Ng's course (though it skimps a little on the theory).
Would an agreement to distribute all derivative work under the GPL be a form of consideration?
So treat it like a school / uni essay. You know the "marker" will spend ~30 seconds reading it, so make sure it's "well structured", so a busy reader can get the gist of your argument with next to no effort..
But make sure it's well padded, so they know you put some effort in.
See, school has been useful. It's taught you how to "communicate"!
So, it could be that the laws of physics are slightly off, and Dark Matter is just an illusion we "see" because of the errors in physics.
Or Dark Matter could be actual matter, and is just ... dark.
"What is the best treatment for stage 1 prostate cancer".
Obviously, there's a number of factors. "Wait and see" is best for an old man with a heart condition, as their heart will give out before the cancer grows. But usually, the best treatment will be something else.
Arguably, the question should include all the context, but it's not always feasible. Also, when it's a close call, the question is probably a bad question.
China and India. Fraud and plagiarism are pretty prevalent in both.
Of course, they don't get much fraudulent or plagiarized work into big journals, and the big journals prefer researchers with a good reputation. I suspect that analysing the actual retraction data would show who's responsible. Is it hungrier researchers in newly developed countries (who have a greater incentive to lie), or researchers in "good" countries (or "good" universities) who aren't subject to as much scrutiny?
You can probably tell by breaking down the data. Until someone does, it's not worth speculating.
Actually, I'd imagine that professional photographers mostly just bitch about bad clients, discuss places where you can find interesting subjects, talk about how overrated some famous photos are, and so on.
There's violent bits in the Koran. There's violent bits in the Bible. Chinese Confucianism-Taoism-Buddhism has some objectionable bits.
Even if any of it is the word of God, God doen't exactly stop people misinterpreting it, or just making shit up. Even other true believers are usually pretty accepting of really screwy interpretations.
It's a myth that religions have any existential properties, because religions DO NOT EXIST. They are just words, and words can have a whole range of different meanings. They can have works associated with them (holy books), but those books can be so widely interpreted (just look at the shit some people think the Bible says) that the books have only a small influence on how people interpret the religions.
In multiple choice questions, the "most correct" answer is the right one. Otherwise, all answers can be correct, if you argue hard enough (if it's at all subjective).
The problem is, they used a stupid question - you can scientifically test the "softness" or "sweetness" of a flower. There should be one that's obviously "most correct".
For in-class quizzes, it's not so important (as the student can challenge it), but for a state-wide test there shouldn't be any wriggle room.
Yes, as opposed to the land of Black Mambas, Lions, Rhinos, and Hippos.
Most of the advantages of VPNs (fast disk IO, admin rights) are kind of irrelevant for this. Of course, you'll hammer their bandwidth, so they might block you for that.
Lion was lackluster, but did add some nice stuff.
ML is barely an OS upgrade. It's just a chance for the OSX app team to write some stuff which works with Lion's new features (some of which are interesting), port some iOS stuff, and charge $30 for a bunch of todo / notification apps.
I'd expect the next upgrade to add some new stuff (like Lion and Leopard did), while ML is just consolidation / apps. It's kind of like Intel's "Tick Tock", without any real benefit from the "Tock".
The difference is, Chinese don't often have handlers. They aren't generally spies, but know that if they can knock fo some IP, they can be very successful in a Chinese university.
This is not too different from what other academics do - it's quite common for academics to leave with a USB stick full of the stuff they were working on, which they use in their next gig.
I guess there might be some more active encouragement in strategic stuff.
Given Samsung practically makes the iPhone (at least, all the profitable components, not the low-profit assembly), it's no surprise.
It sounds like Clare Torry *claimed* half the rights, but settled for "an undisclosed sum".
700 million * $1 = 700 million.
At $500 / hour, 40 hours a week, 52 weeks a year, that's over 500 lawyer years. No, I don't think they spent anywhere near that much. Probably an order of magnitude less - 50 lawyer years for $70 million.
That doesn't include legal costs and wasted time for Apple, Intel, etc. That's the real cost of patents - not the cost of enforcing good ones, but the costs of transaction which both sides have to carry.
Why not?
If no-one (except the police, and professional criminals who'll mostly use them in gang wars), everyone will be safer.
But if you have a gun, you'll feel safer, but make everyone else less safe (as a criminal may sell it). This is especially true if street criminals (who probably wouldn't have guns if they weren't so cheap on the blackmarket) don't have them.
It's a prisoner's dilemma. Guns are good for you, because they make you feel a little safer. They are terrible for society, because they are stolen by thieves, who sell them to gangster wannabes who wouldn't normally be able to get them.
In Australia, bills are required to have a title which properly scopes them. If you give a bill a goofy title, there's a risk that a high court challenge will have it invalidated. So they are all given boring names, like "Federal Financial Relations Amendment (National Health Reform) Bill 2012".
The US doesn't have that kind of rule.
Also, were it not for western industrial espionage against China, we wouldn't have paper or porcelain or tea.
The US doesn't steal commercial know-how because they already have plenty. China is decades behind (in some areas), and can benefit a lot from acquiring foreign IP.
In fact, China's subsidies of industrial inputs (land, energy, water, steel, etc) are there to drag in foreign manufacturing. Want to guess why they want everything made in China? It's so they can figure out how to make it themselves.
It's a hell of a lot better than invading resource-rich countries to try to build up your industrial base. And if no-one ever stole secrets, we'd still all be in the dark ages.
It is "good communication". It does add a lot of value. Malcolm Gladwell famous because he's a great communicator. But it would be silly to call him a scientist.
Moore's Law does not apply to:
1. Bandwidth
2. Battery life
3. Mobile data pricing.
On the other hand, some doctors will work harder to keep potential donors ticking (even if they think they won't recover), because they now they are still worth keeping warm.
If you aren't a donor, and the doctor thinks you won't recover, they'll prefer to pull the plug and move on to someone worth saving.
Mistakes are made at all points in the chain, but I think this one is more likely than when they decide to turn off life support. There's a lot more process for switching a patient off than there is for saying "let's call it, he's not coming back".
> Government figures from the health ministry show that about 1.5 million people in China need transplants, but only 10,000 transplants are performed annually, according to Xinhua.
In other words, prisoner "donations" just aren't enough. FTA, they don't like using condemned prisoner organs, because they aren't usually in good shape anyway. Nobody in China wants to donate, because of cultural reasons (they are selfish? they think it's "icky"? they don't trust doctors? it's not really Buddhist? no idea).
I'm waiting to hear what the "new national donation system" is. Perhaps registered donors will jump to the top of the list, in order of registration date. Everyone will sign up, knowing that if they don't they'll die waiting for a transplant if they need one.
It was also a blatant rip-off of "Darkness at Noon", a thinly-veiled fictionalization of the Moscow Trials. But Koestler was a rapist as well as a genius, so his books haven't really done as well as they should (except the one about rape ... go figure).