So Mr. Valve just gets his brother to make a 1-line cosmetic change to to Dosbox, gets a copy of that with the GPL attached, and then has a licence for a different program that he can redistribute with all his steamy software.
No, it doesn't work that way. You need a license from every copyright holder to copy the software. If any one of them has terminated your license, you can't copy the software anymore; it doesn't matter what else you do.
and it does not taint FUTURE uses of the license (provided they are done in compliance).
Wrong. The GPLv2 explicitly terminates all your rights under the GPL immediately and permanently. That has a number of problems, which is why GPLv3 modified the language somewhat. If you don't believe me, read the discussions leading to the new GPLv3 language.
You're shooting the messenger; I'm simply pointing out the way the GPL (and most commercial licenses) work.
It's also not wise for a company like id/Valve to simply ignore the issue; unless every contributor to DOSbox explicitly reinstates their license, id/Valve always runs the risk that they'll get sued later. id/Valve is free to interpret silence on the part of the copyright holders as "forgiveness", but that's legally unwise.
4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License.
There is no provision to reinstate the license once it's been terminated.
Wrong; it's entirely up to the copyright holder whether to grant you a license, and you may be liable for damages and penalties even if you do not get a license to the software. GPL'ed software is no different than commercial software in this regard, other than that GPL developers usually are nicer about it and less likely to sue you than a commercial vendor.
if an intern at IBM accidentally omits the COPYING file from a minor patch then IBM better be closing its illegal GPL enterprise, quickly!
First of all, you only lose the license to the specific piece of software in question, not all GPL licensed software. Second, IBM has gotten sued over this sort of thing and it has cost them a lot of money, which is why they are careful that this sort of thing isn't happening again. You should be too.
I think that is a very bad FUD to spread.
The only "FUD" is the FUD you're spreading, by falsely claiming that commercial software is any different in this regard.
So, in a worst case scenario, if non-compliance were due to a technical glitch (say, copy.txt was corrupted), what are the legal ramifications? What if the technical glitch was corrected in a day? In an hour?
It doesn't matter; once the license is terminated, you can't get it back except if the authors explicitly grant it back to you.
Of course, much of this depends on the reaction of the authors/licensors and whatever temporary grace they will extend towards the licensee.
It would be foolish for a business to operate under such uncertainty. If you run a business and you violate the GPL on a piece of software, you have to assume that you have lost your rights under the GPL. If the authors remain silent, that doesn't mean that they "extended grace", it means that you continue to operate without a license and can be sued. The only way to remedy that situation is when you go to the authors, explain the situation, and ask them to give you explicit permission to redistribute the software under the GPL again.
4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License.
Small oversight by (on id's part) a hugely prolific developer of GPL'd software. Easily corrected and pushed out to clients straight away.
id is a nice company, but the license is the license. You can't "easily correct" GPL violations: once you screw up, the GPL doesn't apply to you anymore, no matter what you do. The only way to fix that is to get permission from each and every copyright holder. Pointing that out isn't "divisive trolling", it's simply the facts about the GPL.
The third option, which isn't usually available when you screw up with non-free software, is to apologise really fast and comply with the GPL*. Although there are no guarantees free software developers are usually nice folks who can overlook a mistake.
So, just to be clear: you do need to notify the developer once you have failed to comply with the GPL and then want to fix your mistake; by default, you simply lose all rights to the GPL'ed software if you fail to comply with the license, and only the copyright holder can fix that.
The GPL does not allow you to fix problems retroactively: when you violate the license, you lose all rights to using the software. There is no way you can fix it. That's to prevent companies from "forgetting" to acknowledge the software based on the assumption that the worst that can happen to them is to be found out and forced to comply.
So, if Valve/ID violated the GPL, providing the required files at this point is sufficient only if the authors of DOSbox are satisfied with it and grant Valve/ID permission to distribute the software under the GPL again.
These game companies produce proprietary software and take strong measures against copying and copyright infringement. Why do you think it is unreasonable to expect them to comply with other people's software licenses?
Only modifications they might have made to DOSbox will have to be made public.
Please let's get away from this thinking that you can automatically patch up a GPL violation by releasing your modified source code later.
When you violate the GPL, you immediately lose your license to the GPL'ed code and you are liable for your past and future license violations. You cannot make up for that past violation by coming into compliance, and you cannot restore your license to use the code under the GPL license by coming into compliance.
What that liability entails is something that you can negotiate with the authors about, and if you don't reach an agreement, it's for the courts to decide. Theoretically, if the GPL violation is egregious enough, a court might well hand control over other corporate assets, including unrelated software, to the author of the GPL'ed software.
Many GPL authors will be nice and permit you to remedy past GPL violations by coming into compliance, and they may also grant you permission to use the software under the GPL. But all of that is at their sole discretion.
May well leave the system in an inconsistent and unbootable state, and is carried out without warning. This is entirely unacceptable and will leave a stale lockfile in any case.
If this can leave the system in an "inconsistent and unbootable state", then there is something wrong with dpkg; all package operations should amount to atomic transactions.
Human nature in the western culture, you mean. IIRC American Indians, many African cultures, and even our old agricultural society were much respectful of the environment.
The various waves of migrations into the Americas became dominant through genocide. Pre-European Americans were also responsible for numerous extinctions and ecological disasters. Africa and Asia are little different; there are numerous examples of genocide, slavery, infanticide, and man-made ecological disaster. The only groups of humans that managed to live "in harmony" with nature were those lacking the technology and population growth to have any impact.
Current myopic stance started with the industrial revolution, which i suspect was carried off by few powerful people.
Quite to the contrary: European enlightenment was the first time that a major human culture began to understand the responsibility and utility of taking care of their fellow men and the environment. Of course, that didn't abolish greed and intolerance overnight, but it's meant enormous and unprecedented progress over the last several centuries.
You can't tell a professional graphics artist that GIMP should be "good enough" and they "don't need what Photoshop has.
Why not? When Photoshop started out, it was a poorly designed toy with a lousy user interface, yet "professional" users still used it instead of the professional graphics programs of the time. Why? Mostly out of ignorance and partly because of price.
What does that tell us? It tells us that your argument that "professionals" need all the bells and whistles is bullshit. And, in fact, most imaging professionals don't need Photoshop because most imaging professionals at most need simple color corrections and cropping, something for which Photoshop's gimmicks just get in the way.
People modify all three desktop operating systems (Linux, Windows, OS X) to suit their needs, and it doesn't require programming. Windows and OS X desktops usually have dozens of little third party utilities installed, many of which cost money and many of which haven't been tested together. Of the three, Linux probably requires the least amount of tinkering, and all you need is included out of the box.
Neither you nor Microsoft are qualified to determine that since you have no idea what requirements other people in industry may have.
That's why we have standards bodies, and they need to beat away on a standard before it finally gets accepted.
I just hope it doesn't get bogged down in politics and legal wrangling.
I hope it does: Microsoft's approach of creating a spec and then telling people "take it as a standard" is intrinsically unacceptable. Even if, against all odds, they were to produce a good standard, it should get rejected. Microsoft needs to learn to play by industry rules, not attempt to set them.
Designing a new compression format is trivial at this point: the math is well understood, as are the tradeoffs involved. The issue is always patents and standardization.
JPEG2000 has pretty much all the properties you'd want in a next generation digital imaging format, and it's already widely used in the movie industry. There are also hardware compression cores for JPEG2000.
So, I don't see any reason whatsoever to adopt the Microsoft format over JPEG2000.
Of course, Microsoft/Novell don't have to worry about this because prior agreements are explicitly exempted.
For later agreements, the lawyer is missing the point. If Company X distributes under the GPLv3, they are bound by its terms. If Company X later enters into an agreement with Company Y, they need to make sure that they are complying with their obligations under the GPLv3 in that agreement, and that may include imposing conditions on Company Y. If they fail to do so, then they lose the right to distribute the GPLv3 software.
This isn't some bizarre legal theory, lots of software licenses work that way, including Microsoft's own.
Slashdot Mirror
So Mr. Valve just gets his brother to make a 1-line cosmetic change to to Dosbox, gets a copy of that with the GPL attached, and then has a licence for a different program that he can redistribute with all his steamy software.
No, it doesn't work that way. You need a license from every copyright holder to copy the software. If any one of them has terminated your license, you can't copy the software anymore; it doesn't matter what else you do.
You receive a new license every time you download the software. GPLv2 can say that if it wants, but it's not enforceable.
You don't know what you're talking about.
and it does not taint FUTURE uses of the license (provided they are done in compliance).
Wrong. The GPLv2 explicitly terminates all your rights under the GPL immediately and permanently. That has a number of problems, which is why GPLv3 modified the language somewhat. If you don't believe me, read the discussions leading to the new GPLv3 language.
You're shooting the messenger; I'm simply pointing out the way the GPL (and most commercial licenses) work.
It's also not wise for a company like id/Valve to simply ignore the issue; unless every contributor to DOSbox explicitly reinstates their license, id/Valve always runs the risk that they'll get sued later. id/Valve is free to interpret silence on the part of the copyright holders as "forgiveness", but that's legally unwise.
There is no provision to reinstate the license once it's been terminated.
You always can pay and restore the rights
Wrong; it's entirely up to the copyright holder whether to grant you a license, and you may be liable for damages and penalties even if you do not get a license to the software. GPL'ed software is no different than commercial software in this regard, other than that GPL developers usually are nicer about it and less likely to sue you than a commercial vendor.
if an intern at IBM accidentally omits the COPYING file from a minor patch then IBM better be closing its illegal GPL enterprise, quickly!
First of all, you only lose the license to the specific piece of software in question, not all GPL licensed software. Second, IBM has gotten sued over this sort of thing and it has cost them a lot of money, which is why they are careful that this sort of thing isn't happening again. You should be too.
I think that is a very bad FUD to spread.
The only "FUD" is the FUD you're spreading, by falsely claiming that commercial software is any different in this regard.
So, in a worst case scenario, if non-compliance were due to a technical glitch (say, copy.txt was corrupted), what are the legal ramifications? What if the technical glitch was corrected in a day? In an hour?
It doesn't matter; once the license is terminated, you can't get it back except if the authors explicitly grant it back to you.
Of course, much of this depends on the reaction of the authors/licensors and whatever temporary grace they will extend towards the licensee.
It would be foolish for a business to operate under such uncertainty. If you run a business and you violate the GPL on a piece of software, you have to assume that you have lost your rights under the GPL. If the authors remain silent, that doesn't mean that they "extended grace", it means that you continue to operate without a license and can be sued. The only way to remedy that situation is when you go to the authors, explain the situation, and ask them to give you explicit permission to redistribute the software under the GPL again.
Small oversight by (on id's part) a hugely prolific developer of GPL'd software. Easily corrected and pushed out to clients straight away.
id is a nice company, but the license is the license. You can't "easily correct" GPL violations: once you screw up, the GPL doesn't apply to you anymore, no matter what you do. The only way to fix that is to get permission from each and every copyright holder. Pointing that out isn't "divisive trolling", it's simply the facts about the GPL.
The third option, which isn't usually available when you screw up with non-free software, is to apologise really fast and comply with the GPL*. Although there are no guarantees free software developers are usually nice folks who can overlook a mistake.
So, just to be clear: you do need to notify the developer once you have failed to comply with the GPL and then want to fix your mistake; by default, you simply lose all rights to the GPL'ed software if you fail to comply with the license, and only the copyright holder can fix that.
The GPL does not allow you to fix problems retroactively: when you violate the license, you lose all rights to using the software. There is no way you can fix it. That's to prevent companies from "forgetting" to acknowledge the software based on the assumption that the worst that can happen to them is to be found out and forced to comply.
So, if Valve/ID violated the GPL, providing the required files at this point is sufficient only if the authors of DOSbox are satisfied with it and grant Valve/ID permission to distribute the software under the GPL again.
These game companies produce proprietary software and take strong measures against copying and copyright infringement. Why do you think it is unreasonable to expect them to comply with other people's software licenses?
Only modifications they might have made to DOSbox will have to be made public.
Please let's get away from this thinking that you can automatically patch up a GPL violation by releasing your modified source code later.
When you violate the GPL, you immediately lose your license to the GPL'ed code and you are liable for your past and future license violations. You cannot make up for that past violation by coming into compliance, and you cannot restore your license to use the code under the GPL license by coming into compliance.
What that liability entails is something that you can negotiate with the authors about, and if you don't reach an agreement, it's for the courts to decide. Theoretically, if the GPL violation is egregious enough, a court might well hand control over other corporate assets, including unrelated software, to the author of the GPL'ed software.
Many GPL authors will be nice and permit you to remedy past GPL violations by coming into compliance, and they may also grant you permission to use the software under the GPL. But all of that is at their sole discretion.
killall -9 dpkg
May well leave the system in an inconsistent and unbootable state, and
is carried out without warning. This is entirely unacceptable and will
leave a stale lockfile in any case.
If this can leave the system in an "inconsistent and unbootable state", then there is something wrong with dpkg; all package operations should amount to atomic transactions.
I'd rather die than live like that. Unfortunately, this guy doesn't have a choice anymore.
Sort of. Humans didn't create the Sahara desert, but they are responsible for a lot of desertification around the Mediterranean and in Africa.
Human nature in the western culture, you mean. IIRC American Indians, many African cultures, and even our old agricultural society were much respectful of the environment.
The various waves of migrations into the Americas became dominant through genocide. Pre-European Americans were also responsible for numerous extinctions and ecological disasters. Africa and Asia are little different; there are numerous examples of genocide, slavery, infanticide, and man-made ecological disaster. The only groups of humans that managed to live "in harmony" with nature were those lacking the technology and population growth to have any impact.
Current myopic stance started with the industrial revolution, which i suspect was carried off by few powerful people.
Quite to the contrary: European enlightenment was the first time that a major human culture began to understand the responsibility and utility of taking care of their fellow men and the environment. Of course, that didn't abolish greed and intolerance overnight, but it's meant enormous and unprecedented progress over the last several centuries.
You can't tell a professional graphics artist that GIMP should be "good enough" and they "don't need what Photoshop has.
Why not? When Photoshop started out, it was a poorly designed toy with a lousy user interface, yet "professional" users still used it instead of the professional graphics programs of the time. Why? Mostly out of ignorance and partly because of price.
What does that tell us? It tells us that your argument that "professionals" need all the bells and whistles is bullshit. And, in fact, most imaging professionals don't need Photoshop because most imaging professionals at most need simple color corrections and cropping, something for which Photoshop's gimmicks just get in the way.
People modify all three desktop operating systems (Linux, Windows, OS X) to suit their needs, and it doesn't require programming. Windows and OS X desktops usually have dozens of little third party utilities installed, many of which cost money and many of which haven't been tested together. Of the three, Linux probably requires the least amount of tinkering, and all you need is included out of the box.
It's Yahoo! Directory all over again, but this time misusing university employees to do the work.
Hard to say, but it doesn't really matter; on Jupiter, at least, there is likely lots of chemical energy available. Think "deep sea life".
Carl Sagan wrote a paper about possible life on Jupiter.
because you can do almost everything that TIFF can do, but with less complexity
Multipage documents?
It's genuinely a good format.
Neither you nor Microsoft are qualified to determine that since you have no idea what requirements other people in industry may have.
That's why we have standards bodies, and they need to beat away on a standard before it finally gets accepted.
I just hope it doesn't get bogged down in politics and legal wrangling.
I hope it does: Microsoft's approach of creating a spec and then telling people "take it as a standard" is intrinsically unacceptable. Even if, against all odds, they were to produce a good standard, it should get rejected. Microsoft needs to learn to play by industry rules, not attempt to set them.
Designing a new compression format is trivial at this point: the math is well understood, as are the tradeoffs involved. The issue is always patents and standardization.
JPEG2000 has pretty much all the properties you'd want in a next generation digital imaging format, and it's already widely used in the movie industry. There are also hardware compression cores for JPEG2000.
So, I don't see any reason whatsoever to adopt the Microsoft format over JPEG2000.
Of course, Microsoft/Novell don't have to worry about this because prior agreements are explicitly exempted.
For later agreements, the lawyer is missing the point. If Company X distributes under the GPLv3, they are bound by its terms. If Company X later enters into an agreement with Company Y, they need to make sure that they are complying with their obligations under the GPLv3 in that agreement, and that may include imposing conditions on Company Y. If they fail to do so, then they lose the right to distribute the GPLv3 software.
This isn't some bizarre legal theory, lots of software licenses work that way, including Microsoft's own.