It's not a theory. It was not brought up based on the available evidence. People who believe in it decide they believe it and then twist the facts to fit.
I don't think so. Trolls don't have to be false or anything (cf the java is slow trolls), they just have to attract predictable responses, which his post certainly will.
No it doesn't assume that. If the creator exists, even outside of the universe, how come? Was he created, and if not, why do we have to have been created.The grandparent has a valid point wrt legos, it's just overruled by the inability of legos to reproduce, which shows they must have been created.
Their CVS log is not a derivative work any more than a film review is a derivative work of the film it reviews. A CVS log is mere commentary.
It's far closer to a cast list or continuity log. It's based entirely on the code and used when modifying the code. It's a part of the program.
Please be at least marginally informed about the topic. We are not talking about GPLed source.
They are distributing LGPL source, and all the terms about distributing the source are the same between the LGPL and GPL, it's only that the LGPL does not require all derivatives to be distributed under itself.
What is your basis for stating that the legal definition of derivation is "anything that depends on it"? Omniweb, for example, depends on KHTML code, and is entirely proprietary.
That's a simplification of the legal definition but it's pretty accurate. Omniweb is a legal derivative of khtml and only allowed to be propriety because khtml's license permits it to be.
This is nothing of the sort. There's no overflow or to anything. It's an ordinary executable, called something like "pic.exe", then when run it sends itself to your contact list saying "Hey, I look great in this nude pic" (I forget the exact message). The sole blame for such outbreaks is clueless users, the only way to stop them is to make more intelligent users
If the function itself is on the stack you can overwrite it - the function isn't going to be deleted before it's overwritten. If not, you can still get a conventional buffer overflow any time you have a multithreaded program, by overwriting the return address for the higher thread from the lower one. Or get a program to write before the start of an array through integer overflow. I'm sure there are other ways to exploit it.
I think it's anything they use. For a browser, where regressions are so important, you'd be watching the CVS to see if there was anything near the patch you're trying to add that got modified recently, and also which regression tests to use. Any programmer notes, if they go with the program and are looked at, are part of the form of the whole work you use to make modifications to it, and as such they have to be distributed under the gpl. IMO IANAL of course.
IIRC even the DMCA isn't quite that dumb, it's only if the technology lacks a substantial non-infringing use that you can't use it. If MGM wins against Grokster you may have a point. I agree with you on copyrights though. Life of the author makes a certain amount of sense, though not so much. N years makes sense. Life + N years just seems silly.
It pretty much does. It means any derivative of the original gpl work that is used when modifying the program. You can't use it to get IDEs because the IDE is not derived from the original khtml code apple took, but their cvs log is - it's completely based on that source and would make no sense without it. Things like comments and even specifications are part of the source - if you distribute gpled source with the comments stipped or without some accompanying function spec sheet you're violating your license. If it's something derived (legally, which means anything that depends on it) from the program that their developers use to modify the program, it's something they need to distribute
Good point. I think it's normally interpreted as "the form you use yourself for modifying it". Which means if apple people really do just use these humongous patches or modify the source with no scm at all they're ok, but I doubt that's how it works.
Erm, no, that's very hackable - just shoot the fedex guy and nick the dvd. Photons, when used correctly, are absolutely, 100%, laws-of-physics-and-mathematics secure.
You can DOS a quantum channel quite easily by cutting the fiber, or putting a big screen up for in-air transmissions. That's not something it's ever been designed to stop (to the best of my knowledge). For reliability, any wavelength can be used, so if you have a material that can perfectly transmit even one wavelength, that's enough. Fiber will have to be replaced eventually and won't go over too long stretches, but I expect it's thought to be good enough for what people need. Fibers from a decade or so ago go 10km without losing much signal, just coated in rubber and stuff, I think.
The data you send over the quantum channel is just random data you'll use for a one time pad key. If it has been intercepted, you just don't use that key. And interceptors can only intercept half (on average) of the bits.
Proper explanation: the intended recipient doesn't either, so they will only get half of them right, and will mess up the ones they don't measure right in a way that could be detected by someone who did know. However, the half they get right will be completely different from what any man in the middle gets right. Then the recipient explains over an insecure link which measurement they made on each photon, and the transmitter tells them which were right. Now the mitm knows what they needed to know, but there's no way to duplicate a photon so if they'd stopped the photons to measure now the recipient would have not recieved anything. So the two have securely exchanged half of the bits the sender sent. They have no way to control which half they exchange, but that doesn't matter, they just use the half they do as a OTP key.
We have had one time pad ciphers for what, 70 years? When was the last time one was cracked? When some dolt in the kremlin decided to re-use their one time pads. Other than that, it has never been broken. Quantum encryption can be exactly the same - when done right it's unbreakable. Doing it right it hard, but far from impossible
Maybe they can see the day when they're no longer the monopoly and are preparing for it. Picture MS with 40% marketshare and no interoperability - you can only trade MS documents with other MS users, etc. If you use MS you're locking out many of your potential customers, but their office suite etc. is more featureful, so for some people it's worth it. Now picture MS with 40% marketshare and the ability to read/write the standard file formats everyone uses. You just choose on the features of the program, which they are ahead in, so you decide whether you can afford it. Which position do you think they'd rather be in?
Follow the money. It's always in the money. All the things you mention are to get the money. But if they can make money by doing the right thing, they won't hesitate to do it any more than they do about doing the things you list.
Not at all. MS will help the OSS community if and only if they make money from it. Look at where the money is, there you will find the motivation, then you can judge it.
Slashdot Mirror
Nope. If you live forever, you will go to china. It's inevitable. It can take until that bird wears away the mountain, but it will happen.
It's not a theory. It was not brought up based on the available evidence. People who believe in it decide they believe it and then twist the facts to fit.
I don't think so. Trolls don't have to be false or anything (cf the java is slow trolls), they just have to attract predictable responses, which his post certainly will.
No it doesn't assume that. If the creator exists, even outside of the universe, how come? Was he created, and if not, why do we have to have been created.The grandparent has a valid point wrt legos, it's just overruled by the inability of legos to reproduce, which shows they must have been created.
It's far closer to a cast list or continuity log. It's based entirely on the code and used when modifying the code. It's a part of the program.
Please be at least marginally informed about the topic. We are not talking about GPLed source.
They are distributing LGPL source, and all the terms about distributing the source are the same between the LGPL and GPL, it's only that the LGPL does not require all derivatives to be distributed under itself.
What is your basis for stating that the legal definition of derivation is "anything that depends on it"? Omniweb, for example, depends on KHTML code, and is entirely proprietary.
That's a simplification of the legal definition but it's pretty accurate. Omniweb is a legal derivative of khtml and only allowed to be propriety because khtml's license permits it to be.
IME cygwin is reasonably satisfying. It's comparable in performance to a live cd, and easier to use.
If it's code that's been built dynamically it could be there. I agree that it's not very likely though, but that still leaves the other methods
This is nothing of the sort. There's no overflow or to anything. It's an ordinary executable, called something like "pic.exe", then when run it sends itself to your contact list saying "Hey, I look great in this nude pic" (I forget the exact message). The sole blame for such outbreaks is clueless users, the only way to stop them is to make more intelligent users
If the function itself is on the stack you can overwrite it - the function isn't going to be deleted before it's overwritten. If not, you can still get a conventional buffer overflow any time you have a multithreaded program, by overwriting the return address for the higher thread from the lower one. Or get a program to write before the start of an array through integer overflow. I'm sure there are other ways to exploit it.
No it doesn't. KDE dot is html 4 while slashdot is still 3.2. I seem to remember reading they use the kuro5hin system, but I'm not sure.
I think it's anything they use. For a browser, where regressions are so important, you'd be watching the CVS to see if there was anything near the patch you're trying to add that got modified recently, and also which regression tests to use. Any programmer notes, if they go with the program and are looked at, are part of the form of the whole work you use to make modifications to it, and as such they have to be distributed under the gpl. IMO IANAL of course.
IIRC even the DMCA isn't quite that dumb, it's only if the technology lacks a substantial non-infringing use that you can't use it. If MGM wins against Grokster you may have a point. I agree with you on copyrights though. Life of the author makes a certain amount of sense, though not so much. N years makes sense. Life + N years just seems silly.
It pretty much does. It means any derivative of the original gpl work that is used when modifying the program. You can't use it to get IDEs because the IDE is not derived from the original khtml code apple took, but their cvs log is - it's completely based on that source and would make no sense without it. Things like comments and even specifications are part of the source - if you distribute gpled source with the comments stipped or without some accompanying function spec sheet you're violating your license. If it's something derived (legally, which means anything that depends on it) from the program that their developers use to modify the program, it's something they need to distribute
Good point. I think it's normally interpreted as "the form you use yourself for modifying it". Which means if apple people really do just use these humongous patches or modify the source with no scm at all they're ok, but I doubt that's how it works.
Because you can't copy a photon exactly. It's similar to the heisenberg uncertainty principle, where you can't measure position and velocity exactly.
Erm, no, that's very hackable - just shoot the fedex guy and nick the dvd. Photons, when used correctly, are absolutely, 100%, laws-of-physics-and-mathematics secure.
C in diamonds is very slow, that's why they glitter so much (enormous refraction), though nowhere near that slow.
You can DOS a quantum channel quite easily by cutting the fiber, or putting a big screen up for in-air transmissions. That's not something it's ever been designed to stop (to the best of my knowledge). For reliability, any wavelength can be used, so if you have a material that can perfectly transmit even one wavelength, that's enough. Fiber will have to be replaced eventually and won't go over too long stretches, but I expect it's thought to be good enough for what people need. Fibers from a decade or so ago go 10km without losing much signal, just coated in rubber and stuff, I think.
The data you send over the quantum channel is just random data you'll use for a one time pad key. If it has been intercepted, you just don't use that key. And interceptors can only intercept half (on average) of the bits.
Proper explanation: the intended recipient doesn't either, so they will only get half of them right, and will mess up the ones they don't measure right in a way that could be detected by someone who did know. However, the half they get right will be completely different from what any man in the middle gets right. Then the recipient explains over an insecure link which measurement they made on each photon, and the transmitter tells them which were right. Now the mitm knows what they needed to know, but there's no way to duplicate a photon so if they'd stopped the photons to measure now the recipient would have not recieved anything. So the two have securely exchanged half of the bits the sender sent. They have no way to control which half they exchange, but that doesn't matter, they just use the half they do as a OTP key.
We have had one time pad ciphers for what, 70 years? When was the last time one was cracked? When some dolt in the kremlin decided to re-use their one time pads. Other than that, it has never been broken. Quantum encryption can be exactly the same - when done right it's unbreakable. Doing it right it hard, but far from impossible
Maybe they can see the day when they're no longer the monopoly and are preparing for it. Picture MS with 40% marketshare and no interoperability - you can only trade MS documents with other MS users, etc. If you use MS you're locking out many of your potential customers, but their office suite etc. is more featureful, so for some people it's worth it. Now picture MS with 40% marketshare and the ability to read/write the standard file formats everyone uses. You just choose on the features of the program, which they are ahead in, so you decide whether you can afford it. Which position do you think they'd rather be in?
Follow the money. It's always in the money. All the things you mention are to get the money. But if they can make money by doing the right thing, they won't hesitate to do it any more than they do about doing the things you list.
Not at all. MS will help the OSS community if and only if they make money from it. Look at where the money is, there you will find the motivation, then you can judge it.
No it doesn't. Programs on cygwin run pretty fast, not native but far faster than vmware.