What part of a user-mode exploit don't you understand?
It is possible that they could use a subsequent privilege escalation vulnerability to elevate their privileges to administrator level. Sure, it requires that such a hole exists, but I would be surprised if none exists.
If someone wants to use a new license that he doesn't like then somehow attempts to wrangle control from him so that it could be used
We've been though this before. I don't agree that GPLv3 userspace software running on top of a GPLv2 kernel will "wrangle control from him" or "add restrictions to the kernel".
I don't think any of his complaints were based around the intended goals of the GPLv3 but the way they were implemented.
On the contrary, I think it is pretty clear that he opposed the DRM part from the start, and that this microscopic softening w.r.t. GPLv3 is in part caused by a watering-down of that section. Preventing DRM to restrict the freedom of GPL-covered software was one of the primary goals, so I think it is quite wrong to claim that he only opposed the implementation.
Linus isn't an idiot.
I never said he was. He is a very fine engineer, but that does not make him a deity when it comes to legal matters.
Linus had some good objections and had to spell out why v3 code as the license stood couldn't go into GPLv2 software.
I don't think that GPLv3 code going into GPLv2 software is a high priority with the FSF. The important part is that GPLv2 code can go into GPLv3 software (except for code licensed as "GPLv2 only").
If the license is written correctly,
The word correctly is subjective in this context. What is the correct way to write the license to achieve one goal may be incorrect to achieve another goal.
it could because the GPL says the future versions of the license needed to be in the same spirit of the GPL.
I cannot see how GPLv3 code can go into GPLv2 code with anything else than effectively forcing the GPLv3 to become the GPLv2, and thus making it redundant. GPLv3 code will naturally close certain loopholes in the GPLv2, which will make it impossible for GPLv3 code to go into software where those license loopholes are still present.
Hence, code that's "v2 or later" doesn't automatically become v3 and the authors can't just change it to v3 without every contributors' permission.
No? Really? I think you are wrong there (IANAL though). A package licensed under "GPLv2 or later" can be relicensed by anyone as "GPLv3 or later". If it couldn't be relicensed without the explicit permission of the authors, it would effectively be the same as the "GPLv2 only" variant. The permission to relicense under a later version of the GPL is granted by the "or later" part.
It's true that it doesn't grant them as many rights as (say) the BSD license, and has different strings attached, but it's misleading to talk of it taking away rights.
Some people tend to regard the public domain as some kind of baseline for works that aspire to be free, and then view everything that is claimed to be free but not as "free" as works in the public domain, as restricted and "taking away rights". Anyway, I agree with you and not them. I view the "restrictions" in the GPL just like the "restrictions" against killing, raping, robbing or otherwise harming people in society at large. The BSD or the public domain may be more "free" in that they would allow people to kill each other, but that's really not freedom to me. Those who are killed (or otherwise harmed) have their freedom seriously infringed.
The only thing that was a problem is the obligation to release source code, which is probably why, like most of the people, I would opt for a BSD license over a GPL...
Then it's funny that most of the projects aren't BSD-licensed. Oh wait, many people actually like the GPL.
I never tried the higher parts of the temperature range, such as over 80 degrees Celsius (dry sauna), but below that, it can be pretty nice. The air is hot, but it doesn't burn your skin as you might think it would. It is a relaxing experience, and it can be quite fun with friends and a lot of beer.
It is probably better to try it yourself than to have me describe it for you. Additionally, Wikipedia has a long article about saunas.
Another reason why you won't boil is because of the low energy content of air, compared to e.g. water. This can be experienced by comparing a dry sauna with a wet sauna of the same temperature (I suggest staying below 50 degrees C for the comparison, but YMMV). The wet sauna feels much hotter, because the water vapor in the wet sauna transfer much more heat to your body than the air in the dry sauna. Therefore, wet saunas are usually heated to much lower temperatures than dry saunas.
Remember that the GPLv3 will automatically relicense most of the GPLv2 software out there
No, it won't. For the main distribution of a certain package to become GPLv3, the maintainer has to explicitly change the license. The difference from those projects that are "GPLv2 only", is that "GPLv2 or later" can be relicensed by other people than the copyright holder. If you give me a piece of software licensed under the "GPLv2 or later", I can redistribute under "GPLv3 or later" even if you hold the copyright to the software.
This all sounds moot, though. If the Linux Kernel is licensed as GPLv2 (with no additional wording), and there is provision for changing it, then Linus' opinion wasn't even needed? Or am I missing something?
No, it is moot. Some people just continue to insist that Linus has to give his stamp of approval if this license is to take off. If he won't, they mean that the FSF should just throw it away and forget about it. And this despite the fact that Linus can hardly move the kernel to GPLv3 even if he wanted to.
At the risk of sounding inflammatory, GPL is, in fact, source code DRM.
As is every other software license*. Every software license carries a set of rules than the user of covered software must adhere to or refrain from using the software.
*= And please, don't call it DRM, since it is not even close to being DRM. DRM is a computerized version of the police sitting inside your computer telling you what you can and cannot do. Not only does it tell you, but it actually stops you from breaking the rules. Neither the GPL or any other textual software license does this.
If you want to use GPL licensed source you have to sign up to rules much as you have to sign up to rules when you but a tune from Apple. Again, I acknowledge that these are noble and generally positive rules but any rules implicitly impinge on freedom.
Like I said, your version of freedom is the one where you are free to kill anyone you like, and everyone is free to kill you if they would feel like it. I do not agree that this version of freedom gives the maximum freedom to everyone. Those who are killed certainly have their freedom badly infringed. I see the restriction on harming other people as a very slight restriction, and it does not make me feel less free.
For example, let's say Microsoft covenants not to sue RedHat Linux users. Would that mean RedHat could no longer distribute GPLv3 software? What if they are perfectly aware of Microsoft's covenant and Microsoft's vague assertions of general patent nonsense, but they aren't in any business deals or talks with Microsoft? How could the GPLv3 possibly allow this scenario, but prevent an exactly identical scenario with Novell, but with back-room talks and agreements taking place? It seems like a nearly impossible provision to enforce.
I agree, this is a hard problem, and no sensible solution might exist. I was a little bit too quick in my previous comment. If Microsoft could issue a covenant not to sue Red Hat's customers, even without Red hat's approval, and such a clause would be in effect, they could actually shut down Red Hat just by issuing a covenant not to sue. They cannot be allowed to hold such power. Thus, the FSF need to be very careful not to give the GPLv3 unwanted side-effects, such as granting Microsoft another nuclear option.
Do you have some reason to believe that it would switch to a more restrictive GPL when even the previous versions of the GPL were too restrictive for the project?
Those were different restrictions though. The LGPL is a modification of the GPL that allows non-GPL software to link to the covered work. It will surely not become GPLv3, but I see no reason why a new version of LGPL that incorporates some important pieces from GPLv3 (such as the DRM and patent clauses) would be impossible.
In other words, there are many types of restrictions, and many are orthogonal to each other. Instead of a one-dimensional continuum of "restrictedness", "restrictedness" is more accurately described as a multidimensional space. One dimension is linking restriction, another is DRM, a third is patent restrictions, etc. They do not impact each other.
Freedom under the GPL has always been complicated, because it means free-as-in-FSF. As I've observed before, that isn't the same as "free" by any English language definition, which would be more akin to a BSD-style licence.
Would it? The BSD is akin to "You can do anything you want", while the GPL is akin to "You can do anything you want, except killing, raping, robbing or otherwise harming other people". The GPL is free, it just tries to stop people from restricting other peoples' freedom.
It could not prohibit the deal itself, but it could be possible to significantly impact Novell's ability to distribute code covered by the new license, and this is what it is trying to accomplish. Novell is knowingly distributing code that its partner asserts patents against, and the partner provides a promise not to sue to Novell's customers, but nobody else.
The GPLv3 is useless without adoption. If they don't make any concessions the GPLv3 will be still-born, and none of the author's agendas will be pushed.
It will at least be adopted by the entire GNU project. If that still counts as being stillborn by your definition, I don't know.
The GPLv3 (even more so than the GPLv2) is a cleverly worded attempt to control how users use GPLed software. It's just like DRM, but instead of technical measures they are using purely legal measures.
In the first case, the author has a license which restricts what you may do with a given piece of software (you may not disassemble it). In the second, the author has a license which restricts what you may do with a given piece of software (you may not sell it without releasing the source code).
This is called "having your cake and eating it, too."
The difference is that in the first case, further restrictions than those imposed by copyroight law are added in the license. In the second case, some of those restrictions imposed by copyright law are removed if you agree with the license.
Also, the GPL requirements on the distributor are orthogonal to the question whether the First Sale Doctrine trumps the EULA. The GPL imposes certain obligations on the distributor, while the First Sale Doctrine gives certain rights to the receiver of the good. IANAL, but I'd say that there is no contradiction between demanding that the distributor comply with the GPL at the same time as you ignore a restriction in the EULA that may be unenforceable and trumped by the law.
So they better bloody well have done b), which means the offer can be used by anyone and is transferrable (and for that matter, duplicatable).
Not absolutely anyone, but anyone who have received binaries from you, even indirectly. From the GPL FAQ:
What does this "written offer valid for any third party" mean? Does that mean everyone in the world can get the source to any GPL'ed program no matter what?
"Valid for any third party" means that anyone who has the offer is entitled to take you up on it.
If you commercially distribute binaries not accompanied with source code, the GPL says you must provide a written offer to distribute the source code later. When users non-commercially redistribute the binaries they received from you, they must pass along a copy of this written offer. This means that people who did not get the binaries directly from you can still receive copies of the source code, along with the written offer.
The reason we require the offer to be valid for any third party is so that people who receive the binaries indirectly in that way can order the source code from you.
How do you value damages on something you give away for free ?
The FSF recommends that the primary request is for opening the code, not for monetary damages, just because proving monetary damages can be difficult to prove on an application you distribute for free. Thus, the primary options for the defendant is to open up the code or refrain from distribution.
It's nice when companies respect and honor the GPL, but if they don't, it's not like the average 12 year old Debian hacker is going to have resources and desire to sue.
I don't believe they would have to make it widely available either, so they can restrict it to those customers who actually have that particular set-top box.
Except that such restrictions are quite useless. Since the customers would get the right to redistribute (per the GPL), any customer could put up the source on his webpage for all the non-customers in the world to get.
Using your logic, why don't Linux hippies stop reverse engineering closed software when they see the "Do not disassemble" clause?
I don't think reverse engineering network protocols or file formats (which is mostly what the reverse engineering done by the "Linux hippies" is all about) can be counted as disassembling the software. Disassembling the software is just that, applying the reverse of the process of the assembler, to produce an assembly file from the binary. This is rarely, if ever, done to reverse engineer network protocols or file formats.
Slashdot Mirror
It is possible that they could use a subsequent privilege escalation vulnerability to elevate their privileges to administrator level. Sure, it requires that such a hole exists, but I would be surprised if none exists.
I guess this would be a suitable place to make some joke about protected mode vs real mode in Windows, but I can't think of any. :(
Or you misunderstood me.
Again, the "or later" part is for the user of the code, not the author.I cannot see where that contradicts what I wrote.
I wondered when you would show up.
If someone wants to use a new license that he doesn't like then somehow attempts to wrangle control from him so that it could be usedWe've been though this before. I don't agree that GPLv3 userspace software running on top of a GPLv2 kernel will "wrangle control from him" or "add restrictions to the kernel".
I don't think any of his complaints were based around the intended goals of the GPLv3 but the way they were implemented.On the contrary, I think it is pretty clear that he opposed the DRM part from the start, and that this microscopic softening w.r.t. GPLv3 is in part caused by a watering-down of that section. Preventing DRM to restrict the freedom of GPL-covered software was one of the primary goals, so I think it is quite wrong to claim that he only opposed the implementation.
Linus isn't an idiot.I never said he was. He is a very fine engineer, but that does not make him a deity when it comes to legal matters.
Linus had some good objections and had to spell out why v3 code as the license stood couldn't go into GPLv2 software.I don't think that GPLv3 code going into GPLv2 software is a high priority with the FSF. The important part is that GPLv2 code can go into GPLv3 software (except for code licensed as "GPLv2 only").
If the license is written correctly,The word correctly is subjective in this context. What is the correct way to write the license to achieve one goal may be incorrect to achieve another goal.
it could because the GPL says the future versions of the license needed to be in the same spirit of the GPL.I cannot see how GPLv3 code can go into GPLv2 code with anything else than effectively forcing the GPLv3 to become the GPLv2, and thus making it redundant. GPLv3 code will naturally close certain loopholes in the GPLv2, which will make it impossible for GPLv3 code to go into software where those license loopholes are still present.
No? Really? I think you are wrong there (IANAL though). A package licensed under "GPLv2 or later" can be relicensed by anyone as "GPLv3 or later". If it couldn't be relicensed without the explicit permission of the authors, it would effectively be the same as the "GPLv2 only" variant. The permission to relicense under a later version of the GPL is granted by the "or later" part.
You have never heard of self-punishment?
Some people tend to regard the public domain as some kind of baseline for works that aspire to be free, and then view everything that is claimed to be free but not as "free" as works in the public domain, as restricted and "taking away rights". Anyway, I agree with you and not them. I view the "restrictions" in the GPL just like the "restrictions" against killing, raping, robbing or otherwise harming people in society at large. The BSD or the public domain may be more "free" in that they would allow people to kill each other, but that's really not freedom to me. Those who are killed (or otherwise harmed) have their freedom seriously infringed.
Then it's funny that most of the projects aren't BSD-licensed. Oh wait, many people actually like the GPL.
I never tried the higher parts of the temperature range, such as over 80 degrees Celsius (dry sauna), but below that, it can be pretty nice. The air is hot, but it doesn't burn your skin as you might think it would. It is a relaxing experience, and it can be quite fun with friends and a lot of beer.
It is probably better to try it yourself than to have me describe it for you. Additionally, Wikipedia has a long article about saunas.
Another reason why you won't boil is because of the low energy content of air, compared to e.g. water. This can be experienced by comparing a dry sauna with a wet sauna of the same temperature (I suggest staying below 50 degrees C for the comparison, but YMMV). The wet sauna feels much hotter, because the water vapor in the wet sauna transfer much more heat to your body than the air in the dry sauna. Therefore, wet saunas are usually heated to much lower temperatures than dry saunas.
No, it won't. For the main distribution of a certain package to become GPLv3, the maintainer has to explicitly change the license. The difference from those projects that are "GPLv2 only", is that "GPLv2 or later" can be relicensed by other people than the copyright holder. If you give me a piece of software licensed under the "GPLv2 or later", I can redistribute under "GPLv3 or later" even if you hold the copyright to the software.
No, it is moot. Some people just continue to insist that Linus has to give his stamp of approval if this license is to take off. If he won't, they mean that the FSF should just throw it away and forget about it. And this despite the fact that Linus can hardly move the kernel to GPLv3 even if he wanted to.
As is every other software license*. Every software license carries a set of rules than the user of covered software must adhere to or refrain from using the software.
*= And please, don't call it DRM, since it is not even close to being DRM. DRM is a computerized version of the police sitting inside your computer telling you what you can and cannot do. Not only does it tell you, but it actually stops you from breaking the rules. Neither the GPL or any other textual software license does this.
If you want to use GPL licensed source you have to sign up to rules much as you have to sign up to rules when you but a tune from Apple. Again, I acknowledge that these are noble and generally positive rules but any rules implicitly impinge on freedom.Like I said, your version of freedom is the one where you are free to kill anyone you like, and everyone is free to kill you if they would feel like it. I do not agree that this version of freedom gives the maximum freedom to everyone. Those who are killed certainly have their freedom badly infringed. I see the restriction on harming other people as a very slight restriction, and it does not make me feel less free.
I agree, this is a hard problem, and no sensible solution might exist. I was a little bit too quick in my previous comment. If Microsoft could issue a covenant not to sue Red Hat's customers, even without Red hat's approval, and such a clause would be in effect, they could actually shut down Red Hat just by issuing a covenant not to sue. They cannot be allowed to hold such power. Thus, the FSF need to be very careful not to give the GPLv3 unwanted side-effects, such as granting Microsoft another nuclear option.
Those were different restrictions though. The LGPL is a modification of the GPL that allows non-GPL software to link to the covered work. It will surely not become GPLv3, but I see no reason why a new version of LGPL that incorporates some important pieces from GPLv3 (such as the DRM and patent clauses) would be impossible.
In other words, there are many types of restrictions, and many are orthogonal to each other. Instead of a one-dimensional continuum of "restrictedness", "restrictedness" is more accurately described as a multidimensional space. One dimension is linking restriction, another is DRM, a third is patent restrictions, etc. They do not impact each other.
Would it? The BSD is akin to "You can do anything you want", while the GPL is akin to "You can do anything you want, except killing, raping, robbing or otherwise harming other people". The GPL is free, it just tries to stop people from restricting other peoples' freedom.
It could not prohibit the deal itself, but it could be possible to significantly impact Novell's ability to distribute code covered by the new license, and this is what it is trying to accomplish. Novell is knowingly distributing code that its partner asserts patents against, and the partner provides a promise not to sue to Novell's customers, but nobody else.
It will at least be adopted by the entire GNU project. If that still counts as being stillborn by your definition, I don't know.
Could you back that up please?
In the first case, the author has a license which restricts what you may do with a given piece of software (you may not disassemble it). In the second, the author has a license which restricts what you may do with a given piece of software (you may not sell it without releasing the source code).
This is called "having your cake and eating it, too."
The difference is that in the first case, further restrictions than those imposed by copyroight law are added in the license. In the second case, some of those restrictions imposed by copyright law are removed if you agree with the license.
Also, the GPL requirements on the distributor are orthogonal to the question whether the First Sale Doctrine trumps the EULA. The GPL imposes certain obligations on the distributor, while the First Sale Doctrine gives certain rights to the receiver of the good. IANAL, but I'd say that there is no contradiction between demanding that the distributor comply with the GPL at the same time as you ignore a restriction in the EULA that may be unenforceable and trumped by the law.
Are youy trying to imply that this would make source distribution for the GPL:ed components unnecessary?
Not absolutely anyone, but anyone who have received binaries from you, even indirectly. From the GPL FAQ:
What does this "written offer valid for any third party" mean? Does that mean everyone in the world can get the source to any GPL'ed program no matter what?
"Valid for any third party" means that anyone who has the offer is entitled to take you up on it.
If you commercially distribute binaries not accompanied with source code, the GPL says you must provide a written offer to distribute the source code later. When users non-commercially redistribute the binaries they received from you, they must pass along a copy of this written offer. This means that people who did not get the binaries directly from you can still receive copies of the source code, along with the written offer.
The reason we require the offer to be valid for any third party is so that people who receive the binaries indirectly in that way can order the source code from you.
The FSF recommends that the primary request is for opening the code, not for monetary damages, just because proving monetary damages can be difficult to prove on an application you distribute for free. Thus, the primary options for the defendant is to open up the code or refrain from distribution.
It's nice when companies respect and honor the GPL, but if they don't, it's not like the average 12 year old Debian hacker is going to have resources and desire to sue.That's why organizations like gpl-violations.org exist.
Except that such restrictions are quite useless. Since the customers would get the right to redistribute (per the GPL), any customer could put up the source on his webpage for all the non-customers in the world to get.
I don't think reverse engineering network protocols or file formats (which is mostly what the reverse engineering done by the "Linux hippies" is all about) can be counted as disassembling the software. Disassembling the software is just that, applying the reverse of the process of the assembler, to produce an assembly file from the binary. This is rarely, if ever, done to reverse engineer network protocols or file formats.