This is a day old/. thread, so nobody is left reading, but how do you get that by applying a patch you are in violation of copyright? If you were, it would indeed be contributory infringement, but where do you get the former from?
The GPL does not restrict what you, an end-user, do with a piece of GPL code you have on your machine. You are free to do with it as you wish, create derivative works, etc. as long as you don't distribute them.
Legally, it is not necessary for the GPL to state it doesn't take away fair use, but would you not agree that ethically, it should not give the illusion that it does, when it says that nothign else lets you copy the program.
This (now-old) thread was about a lawsuit involving IBM and Caldera, two U.S. companies, over U.S. contracts and U.S. law. That's the relevant law here.
Actual damages are the money lost by the copyright holder. My understanding is that because non-monetary things are so hard to quantify, that's what the statutory damages are there for.
Outside the USA, laws give much stronger accord to the "moral rights" -- right of control over what's done with your art. The US tradition is much more financial.
The question of how much money you lost because somebody else gave away binaries without source... Well, there's no saying a court couldn't try to figure it out but we won't get far predicting how they will do it.
Oh, indeed there are implicit ways of being bound to a contract. But we're talking about parties who don't follow allong with the GPL way. We're talking people who take the code, modify it and just release the binaries of their new work.
They aren't following the GPL in any way. They are plain and simple violating the copyright.
People violate copyrights all the time, of course. I venture that some/. users may have violated the copyrights of RIAA members, and distributed songs in violation of copyright. Possibly. Are you open to the idea that if those songs had a licence on them which said, "The only legal means to distribute this song is by accepting the terms of this licence, which requires you to do XYZ"
If you were then caught distributing such a song on Kazaa, would you want the law to be able to bind you to the terms on it? Is the GPL only OK because software is text and the software distros have a file in them called "LISENSE" or "read.me" that you might be expected to read, and music doesn't have a means to do that?
Truth is you make copies all the time. Every time you visit a web site you read it by copying it with an HTTP fetch to local buffers, and then it is copied again to render it and display it. Most of those pages come with policy statements saying "use of this web site is subject to various conditions. By using the site you agree to them."
Shrinkwrap contracts are worse, but this is still too much power to give to copyright. Now, I can see a claim if you can prove that somebody did this knowing full well what the GPL means, but even lawyers debate exactly what the GPL means.
Consider this. I publish an item and in the licence I say, "You may not copy this unless you give me one million dollars!" And you copy it. Do you now owe me a million dollars? No, you owe me what I can extract for copyright infringment. Even though you "agreed" to the terms I put there, just like the GPL.
Let's make it worse. Say you know about my licence but decide your use is a fair use. Example, publishing the crucial part of Gerald Ford's book where he reveals why he pardoned Nixon. Very famous case.
Turns out the court says, "no, it's not a fair use." So now we go back to the regime you claim, which says I knew about the terms and I published it, so now I am bound by them. Can I have my million?
Nope. Copyright law only has so much power. The only power it has is to make you pay for copyright infringment. Not to bind you to an arbitrary contract.
That is indeed what could happen to a work where the copyright was not properly registered. However, on top of that would be the injunction to stop using it, which would be a serious penalty if the subject program depends highly on the code.
If however it's a pluggable module -- I take out the GPL library and plug in a commercial one -- then indeed the violator may not feel much pain, though any lawsuit is a lot of pain, really.
Among the remedies are actual damages. So yes, if the party sold the program for money, some or all of that money could be awarded to the plaintiff (the owner of the GPLd program.) Combined with money they lost, which might be zero for many programs that are not sold.
If you gave away the program but, in violation of the GPL, didn't give away the source, the figuring of actual damages would be challenging! Courts care much more about money than principles of free software.
As for the settling, it is indeed an option to settle by following the GPL. But the truth is that statement might mislead you. Anything the copyright holder will accept is an option when settling. You can settle by standing on your head and going boop-boop if that is acceptable to the copyright holder.
As such, the terms of the GPL don't really enter into it. The GPL is not, I believe, a "here's what happens to you if you use GPL code" licence. It's a "here's what won't happen to you if you follow the GPL" licence. What won't happen is you won't be sued.
The GPL is called a viral licence because it seems to bind further people down the chain to its terms. It doesn't. It just tells people down the chain they won't be violating copyright if they follow the terms. It misleadingly suggests that's the only way to avoid getting sued.
In fact, many authors of GPL programs offer their programs under other licences too, including commercial ones.
Actually, that's exactly how it works. You do what you want, and claim fair use, and they then must sue you to stop you and, in the suit, prove you wrong about that.
Now I agree though that it's not that simple. In that once in court the burden will be on you to show the fair use. Their standard of proof is low, only as high as you set it with your own case as to why it's fair use.
It is common to talk about fair use rights, even though statutorily fair use is as you say a defense against a claim of infringement.
That's because fair use is there to stop the copyright act from violating the 1st amendment. It is declared, as the first amendment is interpreted, that you have a very strong right to criticise, and that no law may abridge it. The fair use doctrine is, in part, an affirmation of those rights, and so it is not inappropriate to call them fair use rights.
As for the rest, I am not sure why you think it's flamebait, I intend it as reasoned argument. I'm not interested in debate where people sit and call the other's postings utter garbage, however, so no need to reply if that's your style!
GCC may be a bad example. Consider instead a GPLd library (not LGPL, real GPL) statically linked with a program.
Some think that doing so puts the program under the GPL, that you can be compelled to release the source of that program.
I doubt that. I think what you're doing is violating the copyright of the library. When you are sued for doing so, you could avoid the suit by releasing the program under GPL, or settle it by doing so, but you can't be _forced_ to release it under GPL.
At most, you can be forced to stop using the library, pay any money lost to the author of the library, and possibly pay statutory damages.
Again, the statutories give teeth to the GPL if the copyright is properly registered before the suit. But to be strict, you can't force the GPL on the enclosing program.
I believe this would be the same if I took a GPL program, heavily modified it and improved it and sold it in a proprietary manner. They could not force me to publish my changes. They could make me pay damages so high I might decide I would rather publish, but the choice to pay or publish would remain mine.
Fair use rights are much more vague than that. THey are of course not a blanket right to do anything, but they are also not simply to make backups. (They are not, by the way, to give copies to friends sand family. USEing the software does not infringe copyright unless you copy it.)
Fair use rights, for example, would include the right to write a review of a GPLd program, by taking it and adding all sorts of comments to it about how it works or why it sucks. You could then legally sell that review, and your review and new comments would NOT be covered by the GPL. Others could not redistribute them without your permission.
Fair use rights would allow a teacher to modify a GPLd program to illustrate an educational point in computer science, and to distribute that program to all his students, but the modified program would NOT be subject to the GPL.
Thus, the GPL should not say nothing else grants you permission. Ok, to get strict, fair use is copying without permission, so fair use rights don't "grant you permission" -- they make your non-permitted copying not a copyright violation.
But that's a minor quibble. The GPL, properly written, would say that you retain your fair use rights regardless of the GPL, and that the code may have other licences which could grant you permission.
People incorporate public domain code into their programs all the time, and sometimes they GPL the programs. However, the public domain portion is still public domain, and I don't have to go out and find another public domain copy, not strictly, though I might want to check it to be sure I am safe.
And as to the last sentence, you've almost got it. The GPL assures you that if you follow the GPL rules in any distribution of the software, you will not be violating the copyright. That is what it does. It does not bind you to follow those rules, it just says if you do follow them you won't be sued
Correct, except add to that c) Are exercising fair use rights d) Have some other licence to the part of the code you used, or it is in the public domain.
But the main point is I don't think there is much conflict between (a) and (b). B wins. You are violating the copyright.
To suggest you have "accepted the GPL implicitly" would imply that, for example, somebody else could take your code and distribute it as though it were under GPL. Or somebody could sue you not to demand remedy for infringement, but for breach of licence contract. I doubt the latter.
Not as I currently read the GPL, which mostly talks about restrictions on you redistributing your modified GPLed program.
I can take a GPL program and modify it to my heart's content on my machine, and keep the source secret, if I don't give the program to others. So far that's been the clear message from most GPL authorities.
The loophole might be that I can also tell you how to take the GPL program and modify it the same, and neither you nor I distribute the modified program. In fact, I "tell" you how to do it by selling you a program that does it.
That the GPL doesn't forbid, not directly. The one claim they could try would be to claim (as the mysql company does) that the program that does the modifying is a derivative work. That's a murky, less tested area.
The GPL attempts to do this. It is a court test that would find out if it has done this.
Many would dispute GPL's clause that since you can only copy the program under GPL terms, anybody who copies the program is agreeing to GPL terms. That is not correct. If you copy a GPL program in a way not permitted in the GPL, you are violating the copyright. That's it. That's all. You are not agreeing to a contract. The GPL wants you to think that you are, and the statutory penalties for violating copyrights are high enough that you might later wish to agree to the contract to settle, but you can't be made to agree to a contract because you copied something.
At least I certainly hope so. Courts might rule either way, but I think we would not want them to rule in favor of the GPL, because of all the other ways such implicit contracts are used (no reverse engineering, must donate firstborn son, etc.)
"5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it."
Here's where the GPL is stating the law incorrectly. First of all, fair use rights allow you to modify and distribute the program regardless of the copyright licence, so the GPL should not say that "nothing else grants you permission." In addition, the claimed GPLd work may be in the public domain or covered by another licence, and so again such a statement is false.
Next, I believe they are wrong in saying that if you copy the program you are accepting the terms of the GPL. If you copy the program, you are violating copyright, not accepting the gpl.
Smart judges are not supposed to see the difference. The law must be applied consistently, and not nicely for things we like and badly for things we don't. Copyright law says you can't make copies. It doesn't say you can be bound to a contract if you do, only that you are infringing copyright. A judge shouldn't change that just because the contract you are alleged to have agreed with does nice things.
Copyright law does essentially one thing. It gives the copyright holder exclusive right to make copies, and thus to get a court to punish and stop those who make copies of something without permission.
The punishments are specific. Injunctions (stop copying!) Actual damages (pay me for what I lost because you copied it.) Statutory damages.
In extreme cases (wilful infringement that really pissed off the court) statutory damages can be up to $150,000 per copy. That's a lot of leverage which can get you to make people obey the GPL.
But copyright itself does not list among remedies, "Make them release their code under the GPL."
That's the point. There is a lot of contention over clauses that say that "merely by doing something, you indicate your acceptance of an agreement."
Most/.ers have been among those opposing such rules, by the way, in shrinkwrap contracts, and contracts on web sites that say that by using the web site you are bound to its terms.
A typical clause of high contention have been those that say that you can't reverse engineer the program, for example.
The reality is using and copying something can't bind you to arbitrary terms. Nobody seriously thinks they can. Copying any future articles of mine, by the way indicates acceptance of the terms on my web site, which include that you must pay me ONE MILLION DOLLARS for each article.
The debate is over what you can bind people to with such contracts. Some court cases have gone one way, others have gone other ways.
I would be amazed if a court were to say, "You put GCC in with your operating system, thus your entire OS must now be released under the GPL." I would be absolutely astounded.
What a court might say is, "You violated the copyright of GCC when you bundled it in your proprietary OS, and therefor the remedies of copyright law are available to the FSF."
Those remedies are:
a) Actual damages -- how much the FSF lost because you did this. (likely to be hard to make large.)
b) Injunctive relief -- remove GCC from your OS now, do what you can to remove it from copies you already released.
c) Statutory damages -- at the court's discretion if GCC had its coypright registered with the LoC before the suit, these can be quite hefty financial damages.
Now A won't scare anybody, but B and C could indeed force the OS vendor to do quite a bit, not because of the GPL or the court ordered relief, but to settle the copyright suit.
You need to go further here. What I'm talking about can be viewed like a black box. It takes a tarball of GPL source code (which is distributed in full accordance with the GPL.) The black box is, among other things, a compiler (or a tool that invokes an existing compiler) to turn that source code not into the usual.o files, but into modified ones, as per your proprietary modifications.
You do not distribute the modified GPL program, and as such, almost none of the terms of the GPL apply to you.
I've always had doubts about the enforceability of the GPL in court.
It seems to me it would be possible to release a proprietary program which takes a GPL'd source program, patches it and links it with independent binaries to make a new proprietary program. One could sell this program which does the patching, and the libraries, and provide the GPL code under GPL terms while keeping the modifications proprietary, as long as the modified code is never distributed. I would be concerned with any claim that the modifying program is itself a derivative work, though the mySQL folks make similar risky claims.
However, I don't see this coming up in the IBM lawsuit. What might be tested there is just what it means to agree to a licence implicitly. We don't want that to be too strong. We don't want to add a lot of strength to those thousands of programs and web pages that say, "Use of this program indicates acceptance of these terms." Only deliberately agreeing to a contract should bind you to a contract.
If you violate the GPL, you are not guilty of violating a contract, you are at most liable for infringing copyright. Which can result in a suit to stop you from doing the infringing, and for actual damages (hard to enumerate with free code) and statutory damages for the packages that properly registered their copyright (now you're talking.)
In the latter case (the statutory damages) and with the injunction, you can then put pressure on somebody using your GPLd code to get out of the violation judgement by following the GPL. And indeed, the GPL says that if you follow the GPL, you are inherently not violating the copyright.
However, the GPL itself can't make another person's code covered by the GPL. The fact that another person's non-GPL distributing of code is a copyright violation can be a tool to help you win a copyright suit, and that victory, or the threat of it, can make you put the screws to the defendant to do -- well, anything. Including giving you cash, or releasing their code under the GPL. It's actually up to you, the real owner of the GPLd code. If the FSF is the owner of GPLd code, it would probably use its power to force the new code to be released under the GPL, but that is its own philosophical decision. This is not inherent in the GPL.
All the GPL says is, "If you modify and copy this code, and you release your modified program under the terms of the GPL, you're not infringing the copyright on it." It does not say, as some people think it does, that if you modify the code and copy it, your new work is under the GPL.
Holding electronic elections is a sign of a backwards country, not an advanced one. Electronic elections are more prone to fraud and breakdown unless done with great care. See this site for details.
In the early days of personal computing, people would try to make up examples of computer use in the home, such as keeping recipies in the kitchen etc.
I think we're finally coming to it slowly. I do it with an old surplus laptop, which sits in the kitchen. It runs a slide show when doing nothing else, has a sheetfed scanner for all the receipts and house paperwork, and is a web browsing station (including recipies from epicurious!) for when I'm in the kitchen.
Plus it can play music and web radio station streaming feeds.
I have 802.11b on it, and generally have the disk in it spin down, working instead with samba over 802.11.
You can add this to the lindows computer but it is usually cheaper to build in, so I think that's the next step. People in homes don't want wires, so I recommend that this be upgraded a bit to add 802.11 and even expect to use a remote network disk for its disk operations. It should be whisper quiet, with as silent a fan as possible.
Another cheap thing to add is an FM transmitter on the sound outputs. Let it send FM to your stereo so again no wires are needed, or even any hookup.
That's where I would want to see testing. I would presume that there is not as much variation as suggested. Get the reminder inkblots. Have 100 people think up the descriptions of what it is. Run the words through a thesaurus.
Two letters gives you 576 combinations. If you can come up with fewer than that, the password is not as strong as thought. Especially since they believe you have to let the user get one wrong to make it reliable.
Slashdot Mirror
Yes it does. Section 2. Full right to modify the program. Only if you distribute any modified program must you put it under the GPL.
That's the tricky part -- what it really says is if you distribute any program derived from the program.
It is an interesting question, is a patch a derivative work? That needs some more research.
The telling recent case is the cleanflicks case.
If it is, it would imply that you can't publish a bug fix to a program without the consent of the original author.
This is a day old /. thread, so nobody is left reading, but how do you get that by applying a patch you are in violation of copyright? If you were, it would indeed be contributory infringement, but where do you get the former from?
The GPL does not restrict what you, an end-user, do with a piece of GPL code you have on your machine. You are free to do with it as you wish, create derivative works, etc. as long as you don't distribute them.
Legally, it is not necessary for the GPL to state it doesn't take away fair use, but would you not agree that ethically, it should not give the illusion that it does, when it says that nothign else lets you copy the program.
This (now-old) thread was about a lawsuit involving IBM and Caldera, two U.S. companies, over U.S. contracts and U.S. law. That's the relevant law here.
Actually, I'm talking about U.S. law here, the law where the GPL was drafted, not my own country's law.
This would be a whole new way of looking at it.
Actual damages are the money lost by the copyright holder. My understanding is that because non-monetary things are so hard to quantify, that's what the statutory damages are there for.
Outside the USA, laws give much stronger accord to the "moral rights" -- right of control over what's done with your art. The US tradition is much more financial.
The question of how much money you lost because somebody else gave away binaries without source... Well, there's no saying a court couldn't try to figure it out but we won't get far predicting how they will do it.
Oh, indeed there are implicit ways of being bound to a contract. But we're talking about parties who don't follow allong with the GPL way. We're talking people who take the code, modify it and just release the binaries of their new work.
/. users may have violated the copyrights of RIAA members, and distributed songs in violation of copyright. Possibly. Are you open to the idea that if those songs had a licence on them which said, "The only legal means to distribute this song is by accepting the terms of this licence, which requires you to do XYZ"
They aren't following the GPL in any way. They are plain and simple violating the copyright.
People violate copyrights all the time, of course. I venture that some
If you were then caught distributing such a song on Kazaa, would you want the law to be able to bind you to the terms on it? Is the GPL only OK because software is text and the software distros have a file in them called "LISENSE" or "read.me" that you might be expected to read, and music doesn't have a means to do that?
Truth is you make copies all the time. Every time you visit a web site you read it by copying it with an HTTP fetch to local buffers, and then it is copied again to render it and display it. Most of those pages come with policy statements saying "use of this web site is subject to various conditions. By using the site you agree to them."
They give you the site free. Are you bound?
Shrinkwrap contracts are worse, but this is still too much power to give to copyright. Now, I can see a claim if you can prove that somebody did this knowing full well what the GPL means, but even lawyers debate exactly what the GPL means.
Consider this. I publish an item and in the licence I say, "You may not copy this unless you give me one million dollars!" And you copy it. Do you now owe me a million dollars? No, you owe me what I can extract for copyright infringment. Even though you "agreed" to the terms I put there, just like the GPL.
Let's make it worse. Say you know about my licence but decide your use is a fair use. Example, publishing the crucial part of Gerald Ford's book where he reveals why he pardoned Nixon. Very famous case.
Turns out the court says, "no, it's not a fair use." So now we go back to the regime you claim, which says I knew about the terms and I published it, so now I am bound by them. Can I have my million?
Nope. Copyright law only has so much power. The only power it has is to make you pay for copyright infringment. Not to bind you to an arbitrary contract.
That is indeed what could happen to a work where the copyright was not properly registered. However, on top of that would be the injunction to stop using it, which would be a serious penalty if the subject program depends highly on the code.
If however it's a pluggable module -- I take out the GPL library and plug in a commercial one -- then indeed the violator may not feel much pain, though any lawsuit is a lot of pain, really.
Among the remedies are actual damages. So yes, if the party sold the program for money, some or all of that money could be awarded to the plaintiff (the owner of the GPLd program.) Combined with money they lost, which might be zero for many programs that are not sold.
If you gave away the program but, in violation of the GPL, didn't give away the source, the figuring of actual damages would be challenging! Courts care much more about money than principles of free software.
As for the settling, it is indeed an option to settle by following the GPL. But the truth is that statement might mislead you. Anything the copyright holder will accept is an option when settling. You can settle by standing on your head and going boop-boop if that is acceptable to the copyright holder.
As such, the terms of the GPL don't really enter into it. The GPL is not, I believe, a "here's what happens to you if you use GPL code" licence. It's a "here's what won't happen to you if you follow the GPL" licence. What won't happen is you won't be sued.
The GPL is called a viral licence because it seems to bind further people down the chain to its terms. It doesn't. It just tells people down the chain they won't be violating copyright if they follow the terms. It misleadingly suggests that's the only way to avoid getting sued.
In fact, many authors of GPL programs offer their programs under other licences too, including commercial ones.
Actually, that's exactly how it works. You do what you want, and claim fair use, and they then must sue you to stop you and, in the suit, prove you wrong about that.
Now I agree though that it's not that simple. In that once in court the burden will be on you to show the fair use. Their standard of proof is low, only as high as you set it with your own case as to why it's fair use.
It is common to talk about fair use rights, even though statutorily fair use is as you say a defense against a claim of infringement.
That's because fair use is there to stop the copyright act from violating the 1st amendment. It is declared, as the first amendment is interpreted, that you have a very strong right to criticise, and that no law may abridge it. The fair use doctrine is, in part, an affirmation of those rights, and so it is not inappropriate to call them fair use rights.
As for the rest, I am not sure why you think it's flamebait, I intend it as reasoned argument. I'm not interested in debate where people sit and call the other's postings utter garbage, however, so no need to reply if that's your style!
GCC may be a bad example. Consider instead a GPLd library (not LGPL, real GPL) statically linked with a program.
Some think that doing so puts the program under the GPL, that you can be compelled to release the source of that program.
I doubt that. I think what you're doing is violating the copyright of the library. When you are sued for doing so, you could avoid the suit by releasing the program under GPL, or settle it by doing so, but you can't be _forced_ to release it under GPL.
At most, you can be forced to stop using the library, pay any money lost to the author of the library, and possibly pay statutory damages.
Again, the statutories give teeth to the GPL if the copyright is properly registered before the suit. But to be strict, you can't force the GPL on the enclosing program.
I believe this would be the same if I took a GPL program, heavily modified it and improved it and sold it in a proprietary manner. They could not force me to publish my changes. They could make me pay damages so high I might decide I would rather publish, but the choice to pay or publish would remain mine.
Fair use rights are much more vague than that. THey are of course not a blanket right to do anything, but they are also not simply to make backups. (They are not, by the way, to give copies to friends sand family. USEing the software does not infringe copyright unless you copy it.)
Fair use rights, for example, would include the right to write a review of a GPLd program, by taking it and adding all sorts of comments to it about how it works or why it sucks. You could then legally sell that review, and your review and new comments would NOT be covered by the GPL. Others could not redistribute them without your permission.
Fair use rights would allow a teacher to modify a GPLd program to illustrate an educational point in computer science, and to distribute that program to all his students, but the modified program would NOT be subject to the GPL.
Thus, the GPL should not say nothing else grants you permission. Ok, to get strict, fair use is copying without permission, so fair use rights don't "grant you permission" -- they make your non-permitted copying not a copyright violation.
But that's a minor quibble. The GPL, properly written, would say that you retain your fair use rights regardless of the GPL, and that the code may have other licences which could grant you permission.
People incorporate public domain code into their programs all the time, and sometimes they GPL the programs. However, the public domain portion is still public domain, and I don't have to go out and find another public domain copy, not strictly, though I might want to check it to be sure I am safe.
And as to the last sentence, you've almost got it. The GPL assures you that if you follow the GPL rules in any distribution of the software, you will not be violating the copyright. That is what it does. It does not bind you to follow those rules, it just says if you do follow them you won't be sued
Correct, except add to that
c) Are exercising fair use rights
d) Have some other licence to the part of the code you used, or it is in the public domain.
But the main point is I don't think there is much conflict between (a) and (b). B wins. You are violating the copyright.
To suggest you have "accepted the GPL implicitly" would imply that, for example, somebody else could take your code and distribute it as though it were under GPL. Or somebody could sue you not to demand remedy for infringement, but for breach of licence contract. I doubt the latter.
Not as I currently read the GPL, which mostly talks about restrictions on you redistributing your modified GPLed program.
I can take a GPL program and modify it to my heart's content on my machine, and keep the source secret, if I don't give the program to others. So far that's been the clear message from most GPL authorities.
The loophole might be that I can also tell you how to take the GPL program and modify it the same, and neither you nor I distribute the modified program. In fact, I "tell" you how to do it by selling you a program that does it.
That the GPL doesn't forbid, not directly. The one claim they could try would be to claim (as the mysql company does) that the program that does the modifying is a derivative work. That's a murky, less tested area.
The GPL attempts to do this. It is a court test that would find out if it has done this.
Many would dispute GPL's clause that since you can only copy the program under GPL terms, anybody who copies the program is agreeing to GPL terms. That is not correct. If you copy a GPL program in a way not permitted in the GPL, you are violating the copyright. That's it. That's all. You are not agreeing to a contract. The GPL wants you to think that you are, and the statutory penalties for violating copyrights are high enough that you might later wish to agree to the contract to settle, but you can't be made to agree to a contract because you copied something.
At least I certainly hope so. Courts might rule either way, but I think we would not want them to rule in favor of the GPL, because of all the other ways such implicit contracts are used (no reverse engineering, must donate firstborn son, etc.)
"5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it."
Here's where the GPL is stating the law incorrectly. First of all, fair use rights allow you to modify and distribute the program regardless of the copyright licence, so the GPL should not say that "nothing else grants you permission." In addition, the claimed GPLd work may be in the public domain or covered by another licence, and so again such a statement is false.
Next, I believe they are wrong in saying that if you copy the program you are accepting the terms of the GPL. If you copy the program, you are violating copyright, not accepting the gpl.
Smart judges are not supposed to see the difference. The law must be applied consistently, and not nicely for things we like and badly for things we don't. Copyright law says you can't make copies. It doesn't say you can be bound to a contract if you do, only that you are infringing copyright. A judge shouldn't change that just because the contract you are alleged to have agreed with does nice things.
Copyright law does essentially one thing. It gives the copyright holder exclusive right to make copies, and thus to get a court to punish and stop those who make copies of something without permission.
The punishments are specific. Injunctions (stop copying!) Actual damages (pay me for what I lost because you copied it.) Statutory damages.
In extreme cases (wilful infringement that really pissed off the court) statutory damages can be up to $150,000 per copy. That's a lot of leverage which can get you to make people obey the GPL.
But copyright itself does not list among remedies, "Make them release their code under the GPL."
That's the point. There is a lot of contention over clauses that say that "merely by doing something, you indicate your acceptance of an agreement."
/.ers have been among those opposing such rules, by the way, in shrinkwrap contracts, and contracts on web sites that say that by using the web site you are bound to its terms.
Most
A typical clause of high contention have been those that say that you can't reverse engineer the program, for example.
The reality is using and copying something can't bind you to arbitrary terms. Nobody seriously thinks they can. Copying any future articles of mine, by the way indicates acceptance of the terms on my web site, which include that you must pay me ONE MILLION DOLLARS for each article.
The debate is over what you can bind people to with such contracts. Some court cases have gone one way, others have gone other ways.
I would be amazed if a court were to say, "You put GCC in with your operating system, thus your entire OS must now be released under the GPL." I would be absolutely astounded.
What a court might say is, "You violated the copyright of GCC when you bundled it in your proprietary OS, and therefor the remedies of copyright law are available to the FSF."
Those remedies are:
a) Actual damages -- how much the FSF lost because you did this. (likely to be hard to make large.)
b) Injunctive relief -- remove GCC from your OS now, do what you can to remove it from copies you already released.
c) Statutory damages -- at the court's discretion if GCC had its coypright registered with the LoC before the suit, these can be quite hefty financial damages.
Now A won't scare anybody, but B and C could indeed force the OS vendor to do quite a bit, not because of the GPL or the court ordered relief, but to settle the copyright suit.
You need to go further here. What I'm talking about can be viewed like a black box. It takes a tarball of GPL source code (which is distributed in full accordance with the GPL.) The black box is, among other things, a compiler (or a tool that invokes an existing compiler) to turn that source code not into the usual .o files, but into modified ones, as per your proprietary modifications.
You do not distribute the modified GPL program, and as such, almost none of the terms of the GPL apply to you.
I've always had doubts about the enforceability of the GPL in court.
It seems to me it would be possible to release a proprietary program
which takes a GPL'd source program, patches it and links it with
independent binaries to make a new proprietary program. One could
sell this program which does the patching, and the libraries, and
provide the GPL code under GPL terms while keeping the modifications
proprietary, as long as the modified code is never distributed.
I would be concerned with any claim that the modifying program is itself
a derivative work, though the mySQL folks make similar risky claims.
However, I don't see this coming up in the IBM lawsuit. What might
be tested there is just what it means to agree to a licence implicitly.
We don't want that to be too strong. We don't want to add a lot of
strength to those thousands of programs and web pages that say, "Use of
this program indicates acceptance of these terms." Only deliberately
agreeing to a contract should bind you to a contract.
If you violate the GPL, you are not guilty of violating a contract,
you are at most liable for infringing copyright. Which can result
in a suit to stop you from doing the infringing, and for actual damages
(hard to enumerate with free code) and statutory damages for the packages
that properly registered their copyright (now you're talking.)
In the latter case (the statutory damages) and with the injunction, you
can then put pressure on somebody using your GPLd code to get out of
the violation judgement by following the GPL. And indeed, the GPL says
that if you follow the GPL, you are inherently not violating the copyright.
However, the GPL itself can't make another person's code covered by
the GPL. The fact that another person's non-GPL distributing of
code is a copyright violation can be a tool to help you win a copyright
suit, and that victory, or the threat of it, can make you put the screws
to the defendant to do -- well, anything. Including giving you cash, or
releasing their code under the GPL. It's actually up to you, the real
owner of the GPLd code. If the FSF is the owner of GPLd code, it would
probably use its power to force the new code to be released under the
GPL, but that is its own philosophical decision. This is not inherent in
the GPL.
All the GPL says is, "If you modify and copy this code, and you release
your modified program under the terms of the GPL, you're not infringing
the copyright on it." It does not say, as some people think it does,
that if you modify the code and copy it, your new work is under the GPL.
Holding electronic elections is a sign of a backwards country, not an advanced one. Electronic elections are more prone to fraud and breakdown unless done with great care. See this site for details.
In the early days of personal computing, people would try to make up examples of computer use in the home, such as keeping recipies in the kitchen etc.
I think we're finally coming to it slowly. I do it with an old surplus laptop, which sits in the kitchen. It runs a slide show when doing nothing else, has a sheetfed scanner for all the receipts and house paperwork, and is a web browsing station (including recipies from epicurious!) for when I'm in the kitchen.
Plus it can play music and web radio station streaming feeds.
I have 802.11b on it, and generally have the disk in it spin down, working instead with samba over 802.11.
You can add this to the lindows computer but it is usually cheaper to build in, so I think that's the next step. People in homes don't want wires, so I recommend that this be upgraded a bit to add 802.11 and even expect to use a remote network disk for its disk operations. It should be whisper quiet, with as silent a fan as possible.
Another cheap thing to add is an FM transmitter on the sound outputs. Let it send FM to your stereo so again no wires are needed, or even any hookup.
That's where I would want to see testing. I would presume that there is not as much variation as suggested. Get the reminder inkblots. Have 100 people think up the descriptions of what it is. Run the words through a thesaurus.
Two letters gives you 576 combinations. If you can come up with fewer than that, the password is not as strong as thought. Especially since they believe you have to let the user get one wrong to make it reliable.