That's not a minor mistake: H1Bs are temporary visas, green cards are immigrant visas. Green cards can't lead to salary depression for American workers because people hired on green cards are American workers, with all the same labor rights and mobility. So, your diatribe makes even less sense for green cards. Either you just don't know what you're talking about, or you're deliberately misrepresenting the facts to push your political agenda.
I have an education thank you and my salary is 3x the nation average. So stop being a little dick.
So, you are making $130k/year and you're still whining that you're not being paid enough. And because you're not satisfied with your already big salary, you're willing to bad-mouth companies, keep highly skilled and productive people out of the US, and make the US less competitive.
With people like you around, it's no wonder if the US loses the software industry to China and India, just like we lost the auto industry, steel, TVs, and VCRs to overseas.
That is one of the stupidest fucking things I have ever heard. You're giving open source software a bad name.
You really don't get it, do you? OS X is a layer of Apple-proprietary libraries on top of a big core of software that Apple and NeXT derived from open source. For Apple to write something like Mach kernel, the BSD userland, and the gcc compiler from scratch alone would have forced them to increase their R&D operation severalfold.
WebKit would have cost Apple the same money either open or closed. Because they did it open, it is 100x more compatible with the Web.
So you are saying that for the same amount of money as using KHTML, Apple could have developed a shitty, incompatible HTML component. Well, you're absolutely right. But they actually needed a standards-compliant HTML component, not out of the goodness of their hearts, but because Safari wouldn't stand a chance otherwise. And for them to develop and maintain a proprietary HTML rendering component would be a lot more expensive than using an open source one that already works.
You're missing the point. The intent of the LGPL is that I, as a user, can recompile the code covered by the LGPL and replace the original binary with the recompiled one. That means that I should be able to get the exact code that runs on the iPhone (not just something vaguely related to it), and be able to compile and install it on the iPhone.
As far as I know, the iPhone doesn't allow any third party installation of binaries, so Apple isn't meeting that requirement. They may be able to weasel out of that requirement with the LGPLv2 (even though they know full well that that was the intent), but the LGPLv3 would require them to make this possible.
Apple is clearly in violating of the intent of the LGPL, and they will be in violation of the letter of the LGPL with v3.
Basically for a company to get H1-b applicants, they have to interview American workers and give the American workers an equal shot at the position.
That's a nice fairy tale, but you're confusing H1B visas and green cards. There is no requirement to interview American workers for H1B visas; for H1B visas, a company can simply write a letter saying that they couldn't fill the position with an American worker.
The requirement to interview American workers exists only for green card applications. Green cards remove any hold the company may have over their workers, so they are the exact opposite of what a company would apply for if it wants to keep salaries low by hiring cheap immigrants. Companies are indeed trying to skirt that requirement, but that's not to keep salaries low, it's to avoid losing an employee that has likely been with the company for many years and is very valuable to them.
I just recently read this in the news.
Perhaps your inability to read and understand written materials has something to do with your inability to command a higher salary.
Microsoft is not having trouble finding "employees", it's having trouble finding "employees at the wage they want to pay".
No, Microsoft is having trouble finding employees that are worth hiring at any wage. Even without salary, an average employee costs Microsoft about $400k/year. If you can't generate that much new revenue, you aren't worth hiring even if you work for free.
there is a shortage of people who are interested in being paid next to nothing.
Even if that were the case, so what? People from China and India are willing to work for less, they have the skills, so why shouldn't they get the jobs? It's been the same in many other industries; why should the US try to protect US programmers from global competition when garment, steel, and auto workers haven't been? And why would the effect of such protectionist policies be anything other than to cause industries to move out of the US even faster?
Of course, it's silly to think anyway that American engineers are getting job offers at Microsoft in large numbers and turning them down because $80k/year is not enough. Rather, Microsoft simply rejects most applicants because they don't have the skills that they need. You're right that there is no shortage of US programmers and software engineers, there simply is a shortage of US programmers and software engineers that are worth hiring.
Wrong. Violation of license does not automatically dissolve the license. The copyright owners of KHTML would have to terminate Apple's license by alleging noncompliance, and if Apple disagreed, it would have to be taken to court.
You're playing word games. In fact, if they violate the terms of the license, they do indeed lose the rights they have under the license, period.
That follows, among other things, from the fact that a court would determine the termination of the license to be the point in time at which a company violates the license, not the point in time when the court makes the determination.
Also, if Apple hypothetically violated the LGPL with KHTML, they wouldn't have to stop using/shipping Webkit. They'd have to stop using/shipping WebCore, which is a component of Webkit. They could swap it out for a proprietary rendering engine, swap it out for Gecko, or anything else in short order.
Again, you're playing word games. Right now, WebKit and Safari contain LGPL'ed code, so if Apple violated the license, they'd have to stop shipping them, no ifs or buts. What other software they ship under those names in the future is not relevant.
(And it's pretty naive to think that something like KHTML can "just be swapped out". And, in any case, it wouldn't remove their liability for past violations.)
You keep claiming that he's only upset about people stealing his "idea".
No. I'm sure he is upset about many things. Those may include people copying his bits, which is clearly a copyright violation, and which he has a right to be upset about. He may also be victim of a copybot. But that is not what I'm addressing here.
What I am objecting to is that among the many things that this guy is upset about, he is explicitly saying that copying his business model, his work (in a generic sense), and his ideas are prohibited by copyright law. That is wrong. None of those things are covered by copyright law.
Given that the purpose of his animations is generally to be life-like, there is little creativity in the end product, and not only can people create animations that serve the same functions as his, people should be able to create animations that are nearly indistinguishable from his without violating copyright law--as long as they don't actually copy. That's my point.
Furthermore, given his high prices ($40 for a set of animations), it is hardly surprising that there will be competition that attempts to provide the same functionality at a much lower price and that his business is collapsing. The open source client will accelerate that, not because people will steal his stuff, but because it will make it much easier to create content.
Actually, while the unmodified client makes it difficult to copy objects without permission, there have been various hacks around for years.
We're not talking about copying objects, we're talking about copying animations. And while you can do it in principle, I'm not aware of any hacks that actually make it easy. More importantly, that does not seem to be what he is complaining about because he isn't talking about people copying his data, he is complaining about people copying his "ideas". Well, there is no copyright on ideas.
This case is really comparable to the "He stole my HTML!" and "He stole my animated GIFs!" complaints of the mid-nineties.
No, it is not. First, those complaints were about literal copying while this case doesn't seem to be. Furthermore, HTML was on the web without payment, while he is actually selling his products on-line.
It's common in real life, obviously, but in the world he is voicing a complaint about, it's some proportion of novel and non-obvious.
The ability to create avatar animations is built into the system. Not only is it obvious, it's documented and exists there specifically for this purpose, with the expectation that users create, distribute, and improve life-like animations.
Copyright infringement happens when you create a derivative work from an original as well, importing many characteristics of the original.
That sentence is wrong. Yes, copyright infringement occurs when you create a derivative work, but "importing characteristics of the original" is not necessarily "creating a derivative work".
Let's instead liken this to fan fiction vs. commercial fiction. The original producer of the SL sex device wrote a series of novels, developed characters, etc. in a coherent world and sells them. Someone else comes along and without permission, begins to write other novels using the same characters, world and stories and sell them in competition with the original author. That is, clearly, not permissible.
But that analogy is wrong: there is nothing particularly creative, novel, or distinctive about his animations; he is simply trying to create life-like motions. The proper analogy would be that someone writes a boy-meets-girl-boy-loses-girl novel in ebook format, and then claims that all future such novels violate his copyright--clearly unreasonable.
Furthermore, prohibitions against fan fiction are usually not based on copyright but trademarks (specifically, trademarks on the names of characters and places).
His central point was: "Don't you feel like you're being ripped off paying for the health care of jobless people when you're busting a gut earning a living?"
That's exactly what people do in the US, we just pay a lot more for it: we don't let the uninsured die in the streets, we treat them in expensive emergency rooms.
(A second rip-off in the US is that doctors and patients conspire to make doctors look almighty. Doctors like it because they can get money out of useless or pointless treatments, and patients like it because they can fool themselves into thinking that the incurable can be cured and the unpreventable can be prevented.)
Well, there aren't actually a lot of small, light, hi-res PDAs, and the iPhone's UI is actually better than all of them. Strange as it sounds, the iPhone is probably more valuable to some people as a PDA without an AT&T service contract than as a cell phone.
Still, nice as it is in some ways, as a non-extensible device, it's probably overpriced for that purpose at $600. And it's really stupid that the device can't be used with other carriers anyway, in particular given the price.
a Sex Gen is kind of a machine that manipulates avatars into various positions. [...] Alderman said Volkov Catteneo is not the only avatar who has done him wrong. "A lot of people copy me, copy my work, copy my ideas," he said. "Because it's an anonymous platform where you're an avatar cartoon character, as opposed to a real-life person, people think they can operate with impunity."
He's basically talking about animation files. Now, if people literally copy the bits in his animation files, that would be a copyright violation; he'd have a case. But SL makes it pretty hard to do that, and that doesn't sound like what he is complaining about.
Sounds like he is complaining about that people create animations that are "like" his, not merely his. But that basically means that he claims a copyright on the missionary position and that's not right. Copyright doesn't protect ideas, it only protects specific expressions of those ideas. And generic, common expressions aren't copyrightable either, and it seems like the missionary position is pretty generic and common.
You are not free to take a closed device designed to run on a closed network and change the software it is running.
Under GPLv3, you are. And if some vendor is trying to prevent you from doing that, they are in breach of the license, which means they have to stop distributing the software and may face additional liabilities. If the vendor doesn't like that, they shouldn't use the GPLv3 software.
GPL should cover software... it is disturbing that it also tries to change the behavior of the specific hardware that the software runs on.
Apple's licenses specify strong and clear restrictions on what you can and cannot do with their software in relationship to hardware. Why do you think it's wrong when the GPL is doing the same thing?
For instance, if I distribute a new operating system that is proprietary but I include with it a GPL application, does that make the whole operating system fall under the GPL? I surely hope not!
You're mixing up two things. Let's talk about a hypothetical GPL'ed Foobar software package.
If you create a derivative of Foobar, then your derivative is automatically covered by the GPL. You can't do anything about that. Commercial licenses work the same (if they don't provide derivative works outright).
But if you don't create a derivative, that doesn't mean that you can do with Foobar whatever you like; you still need to comply with the other license terms. If you don't, you lose the license to use Foobar and you may be liable for penalties or damages.
So, if Apple violated the LGPL with KHTML, then the consequence would simply be that they had to stop using or shipping KHTML (i.e., Safari and WebKit) immediately. At that point, legally, it wouldn't even make any difference whether they put all of OS X under the GPL, the violation has already occurred and they don't have a license anymore.
How is it smart for the main proponents of GPL3 to admit what others (like Linus Torvalds) have already said: that GPL3 is a bad idea and nobody is going to use it?
Quite to the contrary: this issues demonstrates again that we need GPLv3.
If KHTML were under LGPLv3, then either Apple would have had to make the iPhone user programmable, or they would have had to pay a lot of money to get some other high-quality HTML rendering library. And that's a good thing.
This is shameless FUD from FSF. So FSF, I'll say the same thing that has been said so many times to SCO: Show us the infringing code or STFU.
No, it's not "shameless FUD". Even if Apple can get away with doing with KHTML what they did, through some clever reading of the LGPLv2, the fact is that Apple is clearly violating the explicit, stated intent of the LGPLv2. That alone, Apple deserves condemnation for. And it's not the first time they have done it. Apple likes to save money on R&D by using FOSS (that, incidentally, also means fewer jobs for programmers), while trying to avoid complying with license obligations. Without FOSS, OS X and iPhone wouldn't exist--Apple simply hasn't had the resources or expertise to do anything like it.
And to address abuses like this, the LGPLv3 contains language more specifically prohibiting what Apple did.
Now, whether Apple is actually violating the LGPLv2 in a legal sense, courts will have to decide. I don't think it's as clear as you say that what they are doing is OK, since even the LGPLv2 is quite explicit about allowing you to rebuild a working application.
A worthy goal, albeit not a new one; calling it by a new name ("synthetic biology") won't make it any easier than when people used to call it "genetic engineering".
Not if you're writing software for a supercomputer.
Your implication that supercomputer users are willing to do anything for performance is wrong. There is a small market segment that has stable needs, has a huge budget, and is happy with porting and tuning their handful of applications to some specific box. NASA and some of the national labs may fall into that category.
Most of the supercomputer market, however, is shared computing facilities, and they are not going to get away with telling their dozens of users that they just blew the supercomputing budget for the next few years on a computer that requires their users to port millions of LOC to a new compiler.
Do you have a source for this? I look at SMP as being quite a bit harder. Unless you think we should all learn erlang, threads are hard.
Threads for fast numerical computations are easily encapsulated into libraries; you automatically get speedups for vector code, SIMD code, and parallel problem instances. That covers image and video processing, databases, content based retrieval, and most large scale simulations.
You don't have to do anything different for superscalar. (At least people didn't)
In high performance computing, you have to do a lot for getting efficiency out of superscalar chips. Worse, what you need to do is often dependent on which chip it is.
Putting the work into software is a very good thing. We can change software and recompile, it's harder to change microcode and it is impossible to change silicon. The biggest problem with EPIC is the hardware upgrade path.
If that's what they were doing, it might have been OK. But EPIC replaces dynamic instruction scheduling done in hardware with static instruction scheduling done in software.
If it's a signature it's probably just attached or inserted somewhere in the file and doesn't actually alter the executable portion in any way so I don't know if a judge would buy "we can't reproduce the executable".
What matters is not whether there is some obscure way in which you can interpret the license to mean one thing or another, what matters is what the intent of the license is and how both parties understood it. The LGPL is quite clear about its intent: users should be able to modify the covered code and use the modified version as part of the whole. The license keys are necessary for that.
Apple may be able to claim extreme stupidity and get away with it with (L)GPLv2, but that's why (L)GPLv3 clarifies this point further.
If there isn't any evidence of GPL violation, why make the accusation?
There is evidence: Apple says that the iPhone runs Safari, which implies that it uses KHTML. That is more than a reasonable cause for a court of law to demand that Apple disclose to the court the source code used in iPhone and prove that they aren't violating the LGPL. That's the way software copyright enforcement works in these cases.
Furthermore, for people to give Apple a chance to fix this by talking about it publicly, they are doing Apple a favor. If Apple really is in violation of the LGPL, Apple has already lost all rights to distributing KHTML, meaning they'd have to remove Safari from both the iPhone and OS X in the next update.
Apple didn't start with GPL code at all--they started with BSD and proprietary licensed code,
Actually, development of the iPhone almost certainly involved a lot of GPL'ed code (including gcc); they may simply have avoided shipping it.
The LGPL isn't anti-proprietary, anti-lockdown, or anti-anti-freedom.
Neither is the GPL.
Either way, the LGPL code in the iPhone is just as carefully contained as it is on current Macs
The LGPL requires that people can replace the LGPL'ed library with their own version. So, if KHTML is on the iPhone, Apple needs to make it possible to install a new, user-compiled version of KHTML.
So, Apple, where is the source code to the version of KHTML that runs on the iPhone, how do I compile it, and how do I install the modified version on the iPhone?
(Of course, once I can compile and install a modified KHTML, I effectively can program the iPhone in any way I choose.)
Whoa there! FSF makes an accusation and you swallow it without question? There is no proof that there is any GPL software in the iPhone and until such proof becomes available, how are you any better than RIAA or SCO in assuming otherwise?
Slashdot Mirror
I meant green cards, my bad.
That's not a minor mistake: H1Bs are temporary visas, green cards are immigrant visas. Green cards can't lead to salary depression for American workers because people hired on green cards are American workers, with all the same labor rights and mobility. So, your diatribe makes even less sense for green cards. Either you just don't know what you're talking about, or you're deliberately misrepresenting the facts to push your political agenda.
I have an education thank you and my salary is 3x the nation average. So stop being a little dick.
So, you are making $130k/year and you're still whining that you're not being paid enough. And because you're not satisfied with your already big salary, you're willing to bad-mouth companies, keep highly skilled and productive people out of the US, and make the US less competitive.
With people like you around, it's no wonder if the US loses the software industry to China and India, just like we lost the auto industry, steel, TVs, and VCRs to overseas.
That is one of the stupidest fucking things I have ever heard. You're giving open source software a bad name.
You really don't get it, do you? OS X is a layer of Apple-proprietary libraries on top of a big core of software that Apple and NeXT derived from open source. For Apple to write something like Mach kernel, the BSD userland, and the gcc compiler from scratch alone would have forced them to increase their R&D operation severalfold.
WebKit would have cost Apple the same money either open or closed. Because they did it open, it is 100x more compatible with the Web.
So you are saying that for the same amount of money as using KHTML, Apple could have developed a shitty, incompatible HTML component. Well, you're absolutely right. But they actually needed a standards-compliant HTML component, not out of the goodness of their hearts, but because Safari wouldn't stand a chance otherwise. And for them to develop and maintain a proprietary HTML rendering component would be a lot more expensive than using an open source one that already works.
You're missing the point. The intent of the LGPL is that I, as a user, can recompile the code covered by the LGPL and replace the original binary with the recompiled one. That means that I should be able to get the exact code that runs on the iPhone (not just something vaguely related to it), and be able to compile and install it on the iPhone.
As far as I know, the iPhone doesn't allow any third party installation of binaries, so Apple isn't meeting that requirement. They may be able to weasel out of that requirement with the LGPLv2 (even though they know full well that that was the intent), but the LGPLv3 would require them to make this possible.
Apple is clearly in violating of the intent of the LGPL, and they will be in violation of the letter of the LGPL with v3.
Basically for a company to get H1-b applicants, they have to interview American workers and give the American workers an equal shot at the position.
That's a nice fairy tale, but you're confusing H1B visas and green cards. There is no requirement to interview American workers for H1B visas; for H1B visas, a company can simply write a letter saying that they couldn't fill the position with an American worker.
The requirement to interview American workers exists only for green card applications. Green cards remove any hold the company may have over their workers, so they are the exact opposite of what a company would apply for if it wants to keep salaries low by hiring cheap immigrants. Companies are indeed trying to skirt that requirement, but that's not to keep salaries low, it's to avoid losing an employee that has likely been with the company for many years and is very valuable to them.
I just recently read this in the news.
Perhaps your inability to read and understand written materials has something to do with your inability to command a higher salary.
Microsoft is not having trouble finding "employees", it's having trouble finding "employees at the wage they want to pay".
No, Microsoft is having trouble finding employees that are worth hiring at any wage. Even without salary, an average employee costs Microsoft about $400k/year. If you can't generate that much new revenue, you aren't worth hiring even if you work for free.
there is a shortage of people who are interested in being paid next to nothing.
Even if that were the case, so what? People from China and India are willing to work for less, they have the skills, so why shouldn't they get the jobs? It's been the same in many other industries; why should the US try to protect US programmers from global competition when garment, steel, and auto workers haven't been? And why would the effect of such protectionist policies be anything other than to cause industries to move out of the US even faster?
Of course, it's silly to think anyway that American engineers are getting job offers at Microsoft in large numbers and turning them down because $80k/year is not enough. Rather, Microsoft simply rejects most applicants because they don't have the skills that they need. You're right that there is no shortage of US programmers and software engineers, there simply is a shortage of US programmers and software engineers that are worth hiring.
Wrong. Violation of license does not automatically dissolve the license. The copyright owners of KHTML would have to terminate Apple's license by alleging noncompliance, and if Apple disagreed, it would have to be taken to court.
You're playing word games. In fact, if they violate the terms of the license, they do indeed lose the rights they have under the license, period.
That follows, among other things, from the fact that a court would determine the termination of the license to be the point in time at which a company violates the license, not the point in time when the court makes the determination.
Also, if Apple hypothetically violated the LGPL with KHTML, they wouldn't have to stop using/shipping Webkit. They'd have to stop using/shipping WebCore, which is a component of Webkit. They could swap it out for a proprietary rendering engine, swap it out for Gecko, or anything else in short order.
Again, you're playing word games. Right now, WebKit and Safari contain LGPL'ed code, so if Apple violated the license, they'd have to stop shipping them, no ifs or buts. What other software they ship under those names in the future is not relevant.
(And it's pretty naive to think that something like KHTML can "just be swapped out". And, in any case, it wouldn't remove their liability for past violations.)
You keep claiming that he's only upset about people stealing his "idea".
No. I'm sure he is upset about many things. Those may include people copying his bits, which is clearly a copyright violation, and which he has a right to be upset about. He may also be victim of a copybot. But that is not what I'm addressing here.
What I am objecting to is that among the many things that this guy is upset about, he is explicitly saying that copying his business model, his work (in a generic sense), and his ideas are prohibited by copyright law. That is wrong. None of those things are covered by copyright law.
Given that the purpose of his animations is generally to be life-like, there is little creativity in the end product, and not only can people create animations that serve the same functions as his, people should be able to create animations that are nearly indistinguishable from his without violating copyright law--as long as they don't actually copy. That's my point.
Furthermore, given his high prices ($40 for a set of animations), it is hardly surprising that there will be competition that attempts to provide the same functionality at a much lower price and that his business is collapsing. The open source client will accelerate that, not because people will steal his stuff, but because it will make it much easier to create content.
So, if they provide free movie downloads, does that mean I can legally keep it?
Geez, silly me, what I meant was:
The guy is explicitly complaining about people "copying his ideas" and "copying him". Nowhere does he complain about people copying his data or files.
Actually, while the unmodified client makes it difficult to copy objects without permission, there have been various hacks around for years.
We're not talking about copying objects, we're talking about copying animations. And while you can do it in principle, I'm not aware of any hacks that actually make it easy. More importantly, that does not seem to be what he is complaining about because he isn't talking about people copying his data, he is complaining about people copying his "ideas". Well, there is no copyright on ideas.
This case is really comparable to the "He stole my HTML!" and "He stole my animated GIFs!" complaints of the mid-nineties.
No, it is not. First, those complaints were about literal copying while this case doesn't seem to be. Furthermore, HTML was on the web without payment, while he is actually selling his products on-line.
It's common in real life, obviously, but in the world he is voicing a complaint about, it's some proportion of novel and non-obvious.
The ability to create avatar animations is built into the system. Not only is it obvious, it's documented and exists there specifically for this purpose, with the expectation that users create, distribute, and improve life-like animations.
Copyright infringement happens when you create a derivative work from an original as well, importing many characteristics of the original.
That sentence is wrong. Yes, copyright infringement occurs when you create a derivative work, but "importing characteristics of the original" is not necessarily "creating a derivative work".
Let's instead liken this to fan fiction vs. commercial fiction. The original producer of the SL sex device wrote a series of novels, developed characters, etc. in a coherent world and sells them. Someone else comes along and without permission, begins to write other novels using the same characters, world and stories and sell them in competition with the original author. That is, clearly, not permissible.
But that analogy is wrong: there is nothing particularly creative, novel, or distinctive about his animations; he is simply trying to create life-like motions. The proper analogy would be that someone writes a boy-meets-girl-boy-loses-girl novel in ebook format, and then claims that all future such novels violate his copyright--clearly unreasonable.
Furthermore, prohibitions against fan fiction are usually not based on copyright but trademarks (specifically, trademarks on the names of characters and places).
His central point was: "Don't you feel like you're being ripped off paying for the health care of jobless people when you're busting a gut earning a living?"
That's exactly what people do in the US, we just pay a lot more for it: we don't let the uninsured die in the streets, we treat them in expensive emergency rooms.
(A second rip-off in the US is that doctors and patients conspire to make doctors look almighty. Doctors like it because they can get money out of useless or pointless treatments, and patients like it because they can fool themselves into thinking that the incurable can be cured and the unpreventable can be prevented.)
Well, there aren't actually a lot of small, light, hi-res PDAs, and the iPhone's UI is actually better than all of them. Strange as it sounds, the iPhone is probably more valuable to some people as a PDA without an AT&T service contract than as a cell phone.
Still, nice as it is in some ways, as a non-extensible device, it's probably overpriced for that purpose at $600. And it's really stupid that the device can't be used with other carriers anyway, in particular given the price.
a Sex Gen is kind of a machine that manipulates avatars into various positions. [...] Alderman said Volkov Catteneo is not the only avatar who has done him wrong. "A lot of people copy me, copy my work, copy my ideas," he said. "Because it's an anonymous platform where you're an avatar cartoon character, as opposed to a real-life person, people think they can operate with impunity."
He's basically talking about animation files. Now, if people literally copy the bits in his animation files, that would be a copyright violation; he'd have a case. But SL makes it pretty hard to do that, and that doesn't sound like what he is complaining about.
Sounds like he is complaining about that people create animations that are "like" his, not merely his. But that basically means that he claims a copyright on the missionary position and that's not right. Copyright doesn't protect ideas, it only protects specific expressions of those ideas. And generic, common expressions aren't copyrightable either, and it seems like the missionary position is pretty generic and common.
You are not free to take a closed device designed to run on a closed network and change the software it is running.
Under GPLv3, you are. And if some vendor is trying to prevent you from doing that, they are in breach of the license, which means they have to stop distributing the software and may face additional liabilities. If the vendor doesn't like that, they shouldn't use the GPLv3 software.
GPL should cover software... it is disturbing that it also tries to change the behavior of the specific hardware that the software runs on.
Apple's licenses specify strong and clear restrictions on what you can and cannot do with their software in relationship to hardware. Why do you think it's wrong when the GPL is doing the same thing?
For instance, if I distribute a new operating system that is proprietary but I include with it a GPL application, does that make the whole operating system fall under the GPL? I surely hope not!
You're mixing up two things. Let's talk about a hypothetical GPL'ed Foobar software package.
If you create a derivative of Foobar, then your derivative is automatically covered by the GPL. You can't do anything about that. Commercial licenses work the same (if they don't provide derivative works outright).
But if you don't create a derivative, that doesn't mean that you can do with Foobar whatever you like; you still need to comply with the other license terms. If you don't, you lose the license to use Foobar and you may be liable for penalties or damages.
So, if Apple violated the LGPL with KHTML, then the consequence would simply be that they had to stop using or shipping KHTML (i.e., Safari and WebKit) immediately. At that point, legally, it wouldn't even make any difference whether they put all of OS X under the GPL, the violation has already occurred and they don't have a license anymore.
How is it smart for the main proponents of GPL3 to admit what others (like Linus Torvalds) have already said: that GPL3 is a bad idea and nobody is going to use it?
Quite to the contrary: this issues demonstrates again that we need GPLv3.
If KHTML were under LGPLv3, then either Apple would have had to make the iPhone user programmable, or they would have had to pay a lot of money to get some other high-quality HTML rendering library. And that's a good thing.
This is shameless FUD from FSF. So FSF, I'll say the same thing that has been said so many times to SCO: Show us the infringing code or STFU.
No, it's not "shameless FUD". Even if Apple can get away with doing with KHTML what they did, through some clever reading of the LGPLv2, the fact is that Apple is clearly violating the explicit, stated intent of the LGPLv2. That alone, Apple deserves condemnation for. And it's not the first time they have done it. Apple likes to save money on R&D by using FOSS (that, incidentally, also means fewer jobs for programmers), while trying to avoid complying with license obligations. Without FOSS, OS X and iPhone wouldn't exist--Apple simply hasn't had the resources or expertise to do anything like it.
And to address abuses like this, the LGPLv3 contains language more specifically prohibiting what Apple did.
Now, whether Apple is actually violating the LGPLv2 in a legal sense, courts will have to decide. I don't think it's as clear as you say that what they are doing is OK, since even the LGPLv2 is quite explicit about allowing you to rebuild a working application.
A worthy goal, albeit not a new one; calling it by a new name ("synthetic biology") won't make it any easier than when people used to call it "genetic engineering".
Not if you're writing software for a supercomputer.
Your implication that supercomputer users are willing to do anything for performance is wrong. There is a small market segment that has stable needs, has a huge budget, and is happy with porting and tuning their handful of applications to some specific box. NASA and some of the national labs may fall into that category.
Most of the supercomputer market, however, is shared computing facilities, and they are not going to get away with telling their dozens of users that they just blew the supercomputing budget for the next few years on a computer that requires their users to port millions of LOC to a new compiler.
Do you have a source for this? I look at SMP as being quite a bit harder. Unless you think we should all learn erlang, threads are hard.
Threads for fast numerical computations are easily encapsulated into libraries; you automatically get speedups for vector code, SIMD code, and parallel problem instances. That covers image and video processing, databases, content based retrieval, and most large scale simulations.
You don't have to do anything different for superscalar. (At least people didn't)
In high performance computing, you have to do a lot for getting efficiency out of superscalar chips. Worse, what you need to do is often dependent on which chip it is.
Putting the work into software is a very good thing. We can change software and recompile, it's harder to change microcode and it is impossible to change silicon. The biggest problem with EPIC is the hardware upgrade path.
If that's what they were doing, it might have been OK. But EPIC replaces dynamic instruction scheduling done in hardware with static instruction scheduling done in software.
If it's a signature it's probably just attached or inserted somewhere in the file and doesn't actually alter the executable portion in any way so I don't know if a judge would buy "we can't reproduce the executable".
What matters is not whether there is some obscure way in which you can interpret the license to mean one thing or another, what matters is what the intent of the license is and how both parties understood it. The LGPL is quite clear about its intent: users should be able to modify the covered code and use the modified version as part of the whole. The license keys are necessary for that.
Apple may be able to claim extreme stupidity and get away with it with (L)GPLv2, but that's why (L)GPLv3 clarifies this point further.
If there isn't any evidence of GPL violation, why make the accusation?
There is evidence: Apple says that the iPhone runs Safari, which implies that it uses KHTML. That is more than a reasonable cause for a court of law to demand that Apple disclose to the court the source code used in iPhone and prove that they aren't violating the LGPL. That's the way software copyright enforcement works in these cases.
Furthermore, for people to give Apple a chance to fix this by talking about it publicly, they are doing Apple a favor. If Apple really is in violation of the LGPL, Apple has already lost all rights to distributing KHTML, meaning they'd have to remove Safari from both the iPhone and OS X in the next update.
Apple didn't start with GPL code at all--they started with BSD and proprietary licensed code,
Actually, development of the iPhone almost certainly involved a lot of GPL'ed code (including gcc); they may simply have avoided shipping it.
The LGPL isn't anti-proprietary, anti-lockdown, or anti-anti-freedom.
Neither is the GPL.
Either way, the LGPL code in the iPhone is just as carefully contained as it is on current Macs
The LGPL requires that people can replace the LGPL'ed library with their own version. So, if KHTML is on the iPhone, Apple needs to make it possible to install a new, user-compiled version of KHTML.
So, Apple, where is the source code to the version of KHTML that runs on the iPhone, how do I compile it, and how do I install the modified version on the iPhone?
(Of course, once I can compile and install a modified KHTML, I effectively can program the iPhone in any way I choose.)
Whoa there! FSF makes an accusation and you swallow it without question? There is no proof that there is any GPL software in the iPhone and until such proof becomes available, how are you any better than RIAA or SCO in assuming otherwise?
The proof is that Apple says so: iPhone features a rich HTML client and Safari. Safari is a Cocoa wrapper around KHTML.