Re: GPL and linking
[Humberto Massa] It had equated the two of them in the first part of the phrase. [Raul Miller] The GPL did not use the word equals. Neither that is to say nor namely are equal to equals. Are we to understand that your argument hinges on such fine semantic distinctions as claiming that that is to say does not connote equivalency? Have you nothing better with which to prop up your point of view? (I'd come up with an analogy for how absurd this is beginning to sound, but by now I suspect you'd entirely miss the point, purposely or not.) signature.asc Description: Digital signature
Re: GPL and linking
[Raul Miller] However, I can present my point of view without resorting to this argument: ... Does that make sense? Much clearer, thanks. I was annoyed by the increasingly fine hair-splitting - thanks for bringing the level back to the realm of the meaningful. signature.asc Description: Digital signature
Re: GPL and linking
On 5/11/05, Peter Samuelson [EMAIL PROTECTED] wrote: The GPL did not use the word equals. Neither that is to say nor namely are equal to equals. Are we to understand that your argument hinges on such fine semantic distinctions as claiming that that is to say does not connote equivalency? Have you nothing better with which to prop up your point of view? I'm disputing an argument which seems to require a number of such fine points. It is difficult for me to raise such disputes without mentioning the the points themselves. However, I can present my point of view without resorting to this argument: Let's say that we have a court case which involves some contested GPLed work. How should we proceed? First, let's consider a work which doesn't have any binaries. This would be no different from any other copyright case -- you have to show that the work in question is copyrighted under the GPL, and you'd have to show that the terms of the GPL are being violated. This should be relatively simple, and we can neglect sections 2 and 3 (which are clearly being complied with if the rest of the license is being followed). Now let's imagine that we've got a case which involves binaries. What do we have to do? First, we need exhibits: the sources, and the binaries. Out of consideration for the court, we want to pick examples which are as simple as possible while representing all of the important contested issues. So let's imagine we have Exhibit A (the sources) and Exhibit B (the binary). [We need to also show that this binary is representative of something which is being distributed, but that's not really different from what you have to do in other copyright cases, so I'll ignore that part.] Second, we need to show that Exhibit B is derived from Exhibit A. Again, we want to present this in a simple and easily understandable form, and we want to also present complete information. Once we've shown that B is derived from A, we can start examining the terms of the GPL to make sure that they are being followed. For example, let's say now that we're the defending party, and we want to show that the mere aggregation clause applies. To do this, we would show that the disputed work could be replaced by something trivial, and that having done so, the program is still the same program -- we might do this by showing that it still has the same behavior. Switching sides again, if someone asserted that the mere aggregation clause applied, and used program behavior to make that assertion, and I believed that mere aggregation did not apply, I would show how the program failed to operate in some independent context, with the disputed section removed. Is that clear enough? Now, back to the argument: an argument has been raised that the GPL is flawed because a work based on the Program defined in two parts, where the first part asserts that work based on the Program is a derivative work. The assertion has been made that the second part of that definition is meaningless. Let's assume that this assertion is true, that the second part of that definition is meaningless. Let's further assume that I can construct an example case where a work isn't covered by the GPL because the second part of that definition is meaningless. What would that mean? Since Section 0 says that the GPL grants you license to distribute this work, and since there's no part of the GPL that grants you license where Section 0 does not apply, in our hypothetical case we would have shown that the GPL does not grant you license to distribute this work. At this point, either: A) Copyright law doesn't apply, so it doesn't matter that you don't have license, or B) The GPL doesn't apply, so it doesn't matter that the GPL doesn't grant you license, of C) Distributing the work is prohibited by law. My argument is that if you reach C) by ignoring the second half of the definition of work based on the Program, that you're doing something wrong. Does that make sense? -- Raul
Re: GPL and linking
On 5/11/05, Raul Miller [EMAIL PROTECTED] wrote: [an argument, much of which would make sense in a parallel universe where the GPL is on the law books as 17 USC 666] I am not a lawyer (or a fortiori a judge), so all that I can do to explain why this isn't valid legal reasoning is to point you at documents to which you and I both have access. To the extent that the arguments that I have made involve fine points, I have backed them up with more valid binding case law than you can shake a stick at. You have offered me the instruction sheet for a copyright registration form and some definitions from random online dictionaries. So I'm not going to say that your point of view isn't perfectly valid as your own point of view; but I don't have any reason to believe that it's a good predictor of how a court case involving the FSF suing FooSoft for linking against GNU readline would be argued. Cheers, - Michael
Re: GPL and linking
On 5/11/05, Michael K. Edwards [EMAIL PROTECTED] wrote: So I'm not going to say that your point of view isn't perfectly valid as your own point of view; but I don't have any reason to believe that it's a good predictor of how a court case involving the FSF suing FooSoft for linking against GNU readline would be argued. Of course, a court case does not have to be argued that way. However, I believe that a person who holds a GPL copyright who neglects these points in court is likely to lose. A judge can ignore issues which are not raised in court, and will focus on issues which are raised and contested in court. -- Raul
Re: GPL and linking
On 5/11/05, Raul Miller [EMAIL PROTECTED] wrote: Of course, a court case does not have to be argued that way. No, but if it's to have a prayer of winning, it has to be argued in terms of the law that is actually applicable, not as if the court were obliged to construe the GPL so that every word has meaning and then proceed directly to copyright law. However, I believe that a person who holds a GPL copyright who neglects these points in court is likely to lose. Erroneous beliefs are among the liberties granted to humankind by the universe. One or both of us holds some very erroneous beliefs. A judge can ignore issues which are not raised in court, and will focus on issues which are raised and contested in court. A judge cannot ignore law which doesn't happen to be in one of the parties' briefs. Cheers, - Michael
Re: GPL and linking
On 5/11/05, Michael K. Edwards [EMAIL PROTECTED] wrote: On 5/11/05, Raul Miller [EMAIL PROTECTED] wrote: Of course, a court case does not have to be argued that way. No, but if it's to have a prayer of winning, it has to be argued in terms of the law that is actually applicable, not as if the court were obliged to construe the GPL so that every word has meaning and then proceed directly to copyright law. The law as written says that you do not have permission to copy except as granted by a license. Thus the GPL's license grant is not only applicable, it's the issue which is most likely to be contested in such a case. A judge can ignore issues which are not raised in court, and will focus on issues which are raised and contested in court. A judge cannot ignore law which doesn't happen to be in one of the parties' briefs. In principle, at least, the parties will not be contesting the law. In principle, one party will be asserting that unlicensed copies are being distributed, and will be asking for monetary compensation for the resulting damages. The other party will be asserting that the copies were licensed (or, perhaps, simply settling out of court). Of course, I did gloss over a number of other issues which you would have to address in a real court case. For example, I didn't say anything about how to determine damages for the case -- for that you'd probably have to put a value on development time and show how the issue has cost you development time. -- Raul
Re: GPL and linking
Fine. I have been goaded into rebutting this specimen. On 5/11/05, Raul Miller [EMAIL PROTECTED] wrote: I'm disputing an argument which seems to require a number of such fine points. It is difficult for me to raise such disputes without mentioning the the points themselves. However, I can present my point of view without resorting to this argument: Let's say that we have a court case which involves some contested GPLed work. How should we proceed? First, let's consider a work which doesn't have any binaries. This would be no different from any other copyright case -- you have to show that the work in question is copyrighted under the GPL, and you'd have to show that the terms of the GPL are being violated. This should be relatively simple, and we can neglect sections 2 and 3 (which are clearly being complied with if the rest of the license is being followed). Nope. Under US law at least (IANALIAJ), you'd have to show: 1. that you, yourself, hold a valid registered copyright to a specific portion of the copyrightable expression in a particular work A; and 2. that a portion of your contribution to A has been copied to work B, using the Computer Associates v. Altai abstraction-filtration-comparison standard, and that the amount of _copyrightable_ material that has been copied exceeds de minimis; and 3. that the distributor of B does not have license from you to copy that material from A to B, or that the distributor's conduct exceeds the scope of the license (e. g. creation of a derivative work when the license extends only to verbatim copies), or that the license has been terminated for material breach not otherwise reparable under the applicable contract law standard; After which, the distributor of B has an opportunity to demonstrate: 4. that some statutory or judicially created affirmative defense, such as fair use, justifies the distributor's conduct; or 5. that public policy or a principle of equity demands that the distributor's conduct be sanctioned despite the unavailability of any defense under current law. Then, and only then, you may be entitled to some relief under copyright law. That relief may be as little as one dollar of damages. Now let's imagine that we've got a case which involves binaries. What do we have to do? First, we need exhibits: the sources, and the binaries. Out of consideration for the court, we want to pick examples which are as simple as possible while representing all of the important contested issues. So let's imagine we have Exhibit A (the sources) and Exhibit B (the binary). [We need to also show that this binary is representative of something which is being distributed, but that's not really different from what you have to do in other copyright cases, so I'll ignore that part.] Second, we need to show that Exhibit B is derived from Exhibit A. Again, we want to present this in a simple and easily understandable form, and we want to also present complete information. Once we've shown that B is derived from A, we can start examining the terms of the GPL to make sure that they are being followed. For example, let's say now that we're the defending party, and we want to show that the mere aggregation clause applies. To do this, we would show that the disputed work could be replaced by something trivial, and that having done so, the program is still the same program -- we might do this by showing that it still has the same behavior. This has no bearing on the definition of work based on the Program or of mere aggregation or on any other relevant ambiguity in the construction of the contract. The only sense in which I can see it having any relevance is if the only theory under which B is derived from A is characters and mise en scene, as in Micro Star v. FormGen; in which case the existence of a reasonable alternative to A, under which B does something similarly useful, may be a successful defense. Switching sides again, if someone asserted that the mere aggregation clause applied, and used program behavior to make that assertion, and I believed that mere aggregation did not apply, I would show how the program failed to operate in some independent context, with the disputed section removed. Is that clear enough? Clear as mud. What do you mean, used program behavior to make that assertion? Even though this is an offer of contract, its drafter harps on one copyright note. Mere aggregation is a phrase with no legal meaning (there is a single usage of this phrase in all of the appellate law accessible to FindLaw, and it refers to members of a school prayer club). According to FindLaw, Merriam-Webster's Dictionary of Law defines aggregation as: 1: the collecting of individual units (as damages) into a whole 2: a collection of separate parts that is unpatentable because no integrated mechanism or new and useful result is produced I think it is vanishingly improbable, even if this were a
Re: GPL and linking
On 5/11/05, Michael K. Edwards [EMAIL PROTECTED] wrote: Fine. I have been goaded into rebutting this specimen. Most of this is focused on contract law issues. I've written a separate post suggesting the obvious alternative (Tort law) Since Section 0 says that the GPL grants you license to distribute this work, and since there's no part of the GPL that grants you license where Section 0 does not apply, in our hypothetical case we would have shown that the GPL does not grant you license to distribute this work. Wrongo. The GPL grants you license to copy, modify, and distribute A under the applicable terms. Whether by mere aggregation or by reductio ad absurdum, you may distribute some collections containing A; and there is no basis in the text of the GPL for enforcing on the licensee any division into some permitted collections and some forbidden collections. So B may be distributed so long as the applicable covenants of specific performance with respect to A are honored. I'm assuming that we're talking about a case involving binaries for the work A+B, which means we're talking about a case where either 1) The applicable terms are being followed, and B is available under GPL terms 1a) B is merely aggregated with A in the context of these binaries, or 2) The applicable terms are not being followed, and B is not available under GPL terms, and the work A+B is a significant work in the context of copyright law. At this point, either: A) Copyright law doesn't apply, so it doesn't matter that you don't have license, or B) The GPL doesn't apply, so it doesn't matter that the GPL doesn't grant you license, of C) Distributing the work is prohibited by law. My argument is that if you reach C) by ignoring the second half of the definition of work based on the Program, that you're doing something wrong. Does that make sense? No. Ok, I'm looking for how you think this doesn't make sense. Copyright law applies to the copying of A. True. And to the copying of B. And, to the copying of A+B. The distributor of B claims license under the GPL to copy A. This requires that B do so under certain terms, which is I think where our dispute lies, but continuing... The court construes the terms of that license, settles all other relevant questions of fact, and either decides that the plaintiff is entitled to some relief or that he is not. No disagreement here. It is then so ordered, and there's a path for appeals on points of law. Prohibited by law doesn't mean jack. It's true that the court can (and will) interpret the law. However, Prohibited by law does in fact have meaning. -- Raul
Re: GPL and linking
Raul Miller wrote: On 5/9/05, Humberto Massa [EMAIL PROTECTED] wrote: You can't re-state something saying a different thing. GPL#0 says that a work based on the Program is a derivative work under copyright law, and then says that is to say, a work containing..., which is NOT a re-statement of a derivative work under copyright law. That's another re-statement of what a work based on the Program means. The GPL just equated the two, before the colon! It states, clearly, that the a work based on the program is a derivative work under copyright law, and then, using a colon and the introductory phrase that is to say, it states that a work based on the program is a work containing My point is that the second statement is not stating the same thing, so it can NOT be a re-statement. It must be something else. Yes and no. The GPL is the authoritative document on whatever it wants to define and whatever it CAN define (the GPL CANNOT define what is a derivative work under copyright law, for instance)... but IF AND ONLY IF it defines it without ambiguity. The GPL is not defining what a derivative work under copyright law means. It's defining what a work based on the Program means. It had equated the two of them in the first part of the phrase. What the GPL actually does is defining a cat this way: '' a cat is the animal on the page 3 of the Domestic Pets Handbook, that is to say, an animal with four legs and whiskers. ''. Does this defines all animals with four legs and whiskers as being cats? Not actually. Cats are outside the scope of copyright law. But cats are not outside the scope of the Domestic Pets Handboook. If you were not trying to win the argument at all costs, you would see that my paragraph in quotes, above, has EXACTLY the same grammatical structure as GPL#0. And the interpretation you are giving to this disposition of the GPL#0 is exactly the same I am giving for cats. You *are* saying that every work that contains the Program or a portion of it is a work based on the Program, as per the GPL. But it's not! Now, every derivative work under copyright law is a work based on the Program... nothing more, nothing less... -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: GPL and linking
On 5/10/05, Humberto Massa [EMAIL PROTECTED] wrote: Raul Miller wrote: That's another re-statement of what a work based on the Program means. The GPL just equated the two, before the colon! It states, clearly, that the a work based on the program is a derivative work under copyright law, and then, using a colon and the introductory phrase that is to say, it states that a work based on the program is a work containing My point is that the second statement is not stating the same thing, so it can NOT be a re-statement. It must be something else. According to Wordnet, that is to say means namely. It had equated the two of them in the first part of the phrase. The GPL did not use the word equals. Neither that is to say nor namely are equal to equals. -- Raul
Re: GPL and linking
Raul Miller wrote: On 5/6/05, Humberto Massa [EMAIL PROTECTED] wrote: ??? Let's try again: '' The GPL tries to define work based on the Program in terms of derivative work under copyright law, and then, after this definition and a colon, it tries to explain what is a derivative work under copyright law, but gives a wrong explanation, which would remain wrong even if only USC 17 was considered as a global copyright law. '' See? The GPL says, in its section 0, caput, with [] braces mine: Except what you're calling a paraphrase of the derivative work concept is a restatement of the work based on the Program concept. You can't re-state something saying a different thing. GPL#0 says that a work based on the Program is a derivative work under copyright law, and then says that is to say, a work containing..., which is NOT a re-statement of a derivative work under copyright law. It would be a re-statement if it said: '' a work based on the Program is a derivative work under copyright law, that is to say, in most jurisdictions, any intellectually-novel work that results from a transformation made on the Program, like a translation to another language etc. etc. etc. '' THIS is a re-statement. I say one thing, then I say the SAME thing with other words. Then again, other things you say, such as 'The GPL tries to define' shows that you're not really interested in talking about what the GPL is or what it's saying. The GPL does define work based on the Program. There is no element of try here. The GPL -- not your email -- is the authoritative document about what the GPL does and does not define. Yes and no. The GPL is the authoritative document on whatever it wants to define and whatever it CAN define (the GPL CANNOT define what is a derivative work under copyright law, for instance)... but IF AND ONLY IF it defines it without ambiguity. What the GPL actually does is defining a cat this way: '' a cat is the animal on the page 3 of the Domestic Pets Handbook, that is to say, an animal with four legs and whiskers. ''. Does this defines all animals with four legs and whiskers as being cats? This is not a definition, because of the ambiguity of the terms. When you study the GPL deeply, and start digging on hermeneutics books, you'll see that the that is to say part is only an explanation or example, and NOT part of an authoritative definition. Especially *because* any ambiguity is construed against the offerer, the only possible *legal* reading of the GPL is that a work based on the Program is exactly defined as a derivative work under (your local) copyright law. Finally, I want to say that I am NOT against the GPL. Only I disagree with its interpretation given in the FSF GPL FAQ and I think that, in the courtroom, (I am pretty sure as far as Brazilian courts are involved, really) considering any collective works as works based on the Program would NOT stick. HTH, Massa -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: GPL and linking
Batist Paklons wrote: This however doesn't really change a lot about our discussion about the GPL. It is my belief that the GPL is horribly drafted. One should either choose the simplistic beauty of a BSD style license, or choose a carefully drafted legalese text, such as the IBM Public License. I grew up in a french culture, which chooses for the former, on the belief that it is impossible to predict everything, so it is better to leave out the details and set forth only general principles. The GPL just fails short on both sides. Another concern is that subsequent versions of the GPL cannot improve the language that easily, in spite of the any later version clause. I cannot believe that any jurisdiction would reasonably allow a I offer you this on these conditions, but a third party may change these conditions at will clause. There is simply no consensus on those future conditions. It is effectively a license change, thus a change of contract, with every possible consequence of notice and so on. Batist, I think you are mistaken about the meaning of the any later version copyright license... the terms are precisely '' This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. '' and they mean that said program is dually-triply-etc licensed under the GPLv2 or v3 or v4 or any other upcoming FSF-GPL, at the licensee's discretion. I am a defender of the GPLv2. I am not a defender of the GPLv3 because I don't know its terms yet... :-) I don't know why would anyone license their work under yet-undisclosed terms, but... HTH, Massa -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: GPL and linking
On 5/9/05, Humberto Massa [EMAIL PROTECTED] wrote: You can't re-state something saying a different thing. GPL#0 says that a work based on the Program is a derivative work under copyright law, and then says that is to say, a work containing..., which is NOT a re-statement of a derivative work under copyright law. That's another re-statement of what a work based on the Program means. Yes and no. The GPL is the authoritative document on whatever it wants to define and whatever it CAN define (the GPL CANNOT define what is a derivative work under copyright law, for instance)... but IF AND ONLY IF it defines it without ambiguity. The GPL is not defining what a derivative work under copyright law means. It's defining what a work based on the Program means. What the GPL actually does is defining a cat this way: '' a cat is the animal on the page 3 of the Domestic Pets Handbook, that is to say, an animal with four legs and whiskers. ''. Does this defines all animals with four legs and whiskers as being cats? Not actually. Cats are outside the scope of copyright law. -- Raul
Re: GPL and linking
I haven't replied in detail to Batist yet because I am still digesting the hash that Babelfish makes out of his Dutch article. And I don't entirely agree that the GPL is horribly drafted, by comparison with the kind of dog's breakfast that is the typical license contract. In the past, I have tried to draft something with similar legal meaning myself, and on review I did a really lousy job. I have used the GPL, and will probably use it again (emphatically without the upgrade option) the next time it comes up, as the default license under which I provide source code for software I write primarily for a client's internal use, insofar as work made for hire provisions do not apply. As such, I have gone out on quite a limb in this discussion, possibly giving a future legal opponent grounds for estopping me from making certain arguments in a courtroom. So be it. On 5/9/05, Humberto Massa [EMAIL PROTECTED] wrote: [snip] Batist, I think you are mistaken about the meaning of the any later version copyright license... the terms are precisely '' This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. '' and they mean that said program is dually-triply-etc licensed under the GPLv2 or v3 or v4 or any other upcoming FSF-GPL, at the licensee's discretion. I used to think it extroardinarily unlikely that this formula, with regard to as-yet-unwritten offers of contract, would have legal force in any jurisdiction. The prevalence of similar terms in shrink-wrap software licenses nowadays -- which I abhor, and blame directly on RMS, Eben Moglen, and the FSF -- has eroded that confidence to some degree. If it were ever to come up in a court case in which I personally was involved, I envision disputing its validity to the last breath. (I reserve the right to do otherwise, of course.) I am a defender of the GPLv2. I am not a defender of the GPLv3 because I don't know its terms yet... :-) I don't know why would anyone license their work under yet-undisclosed terms, but... I too am a defender of the GPLv2 under an interpretation which I believe to be correct under the law in the jurisdiction in which I reside. As to gambling on future license texts: I find it uncomfortable enough to live in a society in which disputes on all scales are frequently settled by reference to a corpus of law of which no human being can possibly retain more than a small fraction in his or her brain, and which is perpetually being evolved and ramified by legislatures, courts, and unspoken consensus. The existence of persons who would knowingly further complicate their lives by handing over additional liberties to a person who publishes opinions such as http://www.gnu.org/philosophy/enforcing-gpl.html appalls me but has ceased to amaze me. Cheers, - Michael
Re: GPL and linking
On Mon, May 09, 2005 at 06:25:46PM -0700, Michael K. Edwards wrote: On 5/9/05, Humberto Massa [EMAIL PROTECTED] wrote: [snip] Batist, I think you are mistaken about the meaning of the any later version copyright license... the terms are precisely '' This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. '' and they mean that said program is dually-triply-etc licensed under the GPLv2 or v3 or v4 or any other upcoming FSF-GPL, at the licensee's discretion. I used to think it extroardinarily unlikely that this formula, with regard to as-yet-unwritten offers of contract, would have legal force in any jurisdiction. The prevalence of similar terms in shrink-wrap software licenses nowadays -- which I abhor, and blame directly on RMS, Eben Moglen, and the FSF -- has eroded that confidence to some degree. If it were ever to come up in a court case in which I personally was involved, I envision disputing its validity to the last breath. (I reserve the right to do otherwise, of course.) I'm confused. Why would an optional upgrade clause (party X may offer alternate terms for this software, which you can accept at your option) like the GPL's be used in a shrink-wrap license? I also don't understand why you're so opposed to it. Why should I not be able to say you can distribute under these conditions; in addition, John may offer you a new license in the future, terms which you may accept or ignore? -- Glenn Maynard -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: GPL and linking
On 07/05/05, Michael K. Edwards [EMAIL PROTECTED] wrote: Again, that's not how it works. In the presence of a valid license contract, one is entitled to contract-law standards of the reasonableness of one's attempts to cure a breach when notified. The automatic termination clause is probably unenforceable in most jurisdictions; I think (IANAL) few would even read it as authority to terminate on inadvertent (non-material) breach, let alone on the licensor's idea of breach if the licensee's (reasonable) construction makes it not a breach. Automatic termination clauses are quite common, and generally held valid. It is often only what constitutes a breach that can lead to such termination that is disputed in court. In my opinion that is one of the few GPL license terms that is quite sound, only the grounds on which that termination happens seem extremely flakey to me. You're quite right; I didn't really mean unenforceable, I meant ineffective as a means of circumventing a court's authority to interpret the contract and set standards of breach and remedy. As in the MySQL case, where the judge decided that the definitional issue was a matter of fair dispute, and thus MySQL could not meet the standard of likely to prevail on the facts; and even if MySQL's interpretation was upheld the breach might well have been cured (leaving the contract intact) by Progress's conduct subsequent to notice of breach; and even if it weren't cured, MySQL could show neither the prospect of irreparable harm nor that the balance of harms favored it, given the conduct pledged by Progress. Hence the already pledged conduct would constitute sufficient remedy pending a full trial of fact, even though the only remedy specified in the GPL is termination. What I really should have written is that automatic termination clauses only affect the date from which the license is deemed to have been terminated in the event that a court determines material breach, but don't give the offeror or drafter any additional authority to interpret whether a breach has occurred. From this perspective, an automatic termination clause isn't so much a way of strengthening the licensor's authority to terminate as it is a declaration that the licensee waives any waivable statutory provisions about notice of termination in the event of breach. It might also affect whether a court-ordered remedy at the conclusion of a full trial includes license termination (i. e., an injunction against continued exercise of rights granted by the license) or merely damages for any conduct to date that fell outside the license. This is in contrast to in the sole judgment of the licensor language, which as I understand it can only take effect upon notice in most jurisdictions, and amounts to termination at will plus a covenant not to terminate without a reasonable belief that one of the termination conditions has been met. Such language (which is not present in the GPL) places the burden upon the licensee to demonstrate, in the event of notice of termination, that the licensor did not have a reasonable basis for belief that there was reason to terminate. Is that how it works in your jurisdiction, more or less? More or less: giving notice is an imperative that only in very few cases is deemed unnecessary. This is in the rather obvious case when notice has no use any more. Waiving the right of notice has to be drafted in a very precise clause without any ambiguity. This clause is certainly not present in the GPL. Thus should a licensor seek a cure for a breach, he still should give notice, even if the termination is automatic. This to give the licensee reasonable time to cure the breach himself (or to start doing his legal research ;) ). The in the sole judgment of the licensor language, is to my opinion irrelevant. Even in the case of automatic termination there is still some action required on the side of the licensor, action which also needs his judgement. Terminating without reasonable belief that a breach occurred amounts to abuse of law and reckless litigation, irregardless of any contract clause (this on the principle of good faith). The clause isn't precise enough either to move the burden of proof, but perchance it could be used in context to interpret the will of the parties. But that context better be more precise. This however doesn't really change a lot about our discussion about the GPL. It is my belief that the GPL is horribly drafted. One should either choose the simplistic beauty of a BSD style license, or choose a carefully drafted legalese text, such as the IBM Public License. I grew up in a french culture, which chooses for the former, on the belief that it is impossible to predict everything, so it is better to leave out the details and set forth only general principles. The GPL just fails short on both sides. Another concern is that subsequent versions of the GPL cannot
Re: GPL and linking
[Note: IALNAP (I am lawyer, not a programmer), arguing solely in Belgian/European context, and english is not my native language.] On 07/05/05, Michael K. Edwards [EMAIL PROTECTED] wrote: Again, that's not how it works. In the presence of a valid license contract, one is entitled to contract-law standards of the reasonableness of one's attempts to cure a breach when notified. The automatic termination clause is probably unenforceable in most jurisdictions; I think (IANAL) few would even read it as authority to terminate on inadvertent (non-material) breach, let alone on the licensor's idea of breach if the licensee's (reasonable) construction makes it not a breach. Automatic termination clauses are quite common, and generally held valid. It is often only what constitutes a breach that can lead to such termination that is disputed in court. In my opinion that is one of the few GPL license terms that is quite sound, only the grounds on which that termination happens seem extremely flakey to me. As to the whole derivative work discussion, my opinion is that a judge would rather easily decide something isn't a derived work. The linux kernel, e.g., wouldn't need those notes of Linus to allow use of the API and so on, on the simple reason that the kernel is designed to do just that. In Europe at least one has an automatic license to do everything that is necessary to run a program for the purpose it is intended to, unless explicitly otherwise agreed to. I believe for the GPL to rule this out, it has to draft a clause that says: you cannot link to this program in such and such a way, unless it is also GPL'ed. In general exceptions to a rule have to be very precise, lest they become the rule and the rule the exception. I am reasoning from a legal background, and I believe that is also wat a judge would do. It is my general opinion, following Michael, that large portions of the FSF FAQ are simply wrong. I have written some more elaborate papers on that topic, albeit discussing intellectual property in more general terms, focussed on Open Source. See http://m9923416.kuleuven.be for that (unfortunately, the most interesting one is written in dutch, and I do not have time to translate). Kind Regards Batist
Re: GPL and linking
On 5/7/05, Batist Paklons [EMAIL PROTECTED] wrote: [Note: IALNAP (I am lawyer, not a programmer), arguing solely in Belgian/European context, and english is not my native language.] It's really cool to have an actual lawyer weigh in, even if TINLAIAJ. :-) On 07/05/05, Michael K. Edwards [EMAIL PROTECTED] wrote: Again, that's not how it works. In the presence of a valid license contract, one is entitled to contract-law standards of the reasonableness of one's attempts to cure a breach when notified. The automatic termination clause is probably unenforceable in most jurisdictions; I think (IANAL) few would even read it as authority to terminate on inadvertent (non-material) breach, let alone on the licensor's idea of breach if the licensee's (reasonable) construction makes it not a breach. Automatic termination clauses are quite common, and generally held valid. It is often only what constitutes a breach that can lead to such termination that is disputed in court. In my opinion that is one of the few GPL license terms that is quite sound, only the grounds on which that termination happens seem extremely flakey to me. You're quite right; I didn't really mean unenforceable, I meant ineffective as a means of circumventing a court's authority to interpret the contract and set standards of breach and remedy. As in the MySQL case, where the judge decided that the definitional issue was a matter of fair dispute, and thus MySQL could not meet the standard of likely to prevail on the facts; and even if MySQL's interpretation was upheld the breach might well have been cured (leaving the contract intact) by Progress's conduct subsequent to notice of breach; and even if it weren't cured, MySQL could show neither the prospect of irreparable harm nor that the balance of harms favored it, given the conduct pledged by Progress. Hence the already pledged conduct would constitute sufficient remedy pending a full trial of fact, even though the only remedy specified in the GPL is termination. What I really should have written is that automatic termination clauses only affect the date from which the license is deemed to have been terminated in the event that a court determines material breach, but don't give the offeror or drafter any additional authority to interpret whether a breach has occurred. From this perspective, an automatic termination clause isn't so much a way of strengthening the licensor's authority to terminate as it is a declaration that the licensee waives any waivable statutory provisions about notice of termination in the event of breach. It might also affect whether a court-ordered remedy at the conclusion of a full trial includes license termination (i. e., an injunction against continued exercise of rights granted by the license) or merely damages for any conduct to date that fell outside the license. This is in contrast to in the sole judgment of the licensor language, which as I understand it can only take effect upon notice in most jurisdictions, and amounts to termination at will plus a covenant not to terminate without a reasonable belief that one of the termination conditions has been met. Such language (which is not present in the GPL) places the burden upon the licensee to demonstrate, in the event of notice of termination, that the licensor did not have a reasonable basis for belief that there was reason to terminate. Is that how it works in your jurisdiction, more or less? As to the whole derivative work discussion, my opinion is that a judge would rather easily decide something isn't a derived work. The linux kernel, e.g., wouldn't need those notes of Linus to allow use of the API and so on, on the simple reason that the kernel is designed to do just that. In Europe at least one has an automatic license to do everything that is necessary to run a program for the purpose it is intended to, unless explicitly otherwise agreed to. I believe for the GPL to rule this out, it has to draft a clause that says: you cannot link to this program in such and such a way, unless it is also GPL'ed. In general exceptions to a rule have to be very precise, lest they become the rule and the rule the exception. Woohoo. Yes, that's how I understand it under US law as well (IANALIAJ), with a couple of asterisks about estoppel and laches. I am reasoning from a legal background, and I believe that is also wat a judge would do. It is my general opinion, following Michael, that large portions of the FSF FAQ are simply wrong. I have written some more elaborate papers on that topic, albeit discussing intellectual property in more general terms, focussed on Open Source. See http://m9923416.kuleuven.be for that (unfortunately, the most interesting one is written in dutch, and I do not have time to translate). I suppose that if I profess to be able to read legalese, I ought to be able to tackle Dutch, with a little help from Google and/or Babelfish. :-) Kind
Re: GPL and linking (was: Urgently need GPL compatible libsnmp5-dev replacement :-()
On 5/5/05, Michael K. Edwards [EMAIL PROTECTED] wrote: Sorry to spam debian-devel -- and with a long message containing long paragraphs too, horrors! -- in replying to this. Who is sorry? How sorry? Let's assume, for the sake of argument, that this sorry-ness is not something that matters enough to you to avoid posting long and elliptical messages to debian-devel. On Wed, May 04, 2005 at 11:51:51PM -0500, Peter Samuelson wrote: The GPL simply defers to copyright law to define derivative work. Actually, it tries to define work based on the Program in terms of derivative work under copyright law, and then incorrectly paraphrases that definition. It's probably worth noting that derivative work and work based on the Program are spelled differently. What's not clear, to me, is whether the word that refers to the d phrase or the w phrase. Careful study sheds no insight into this burning issue. [If I read the GPL, I can't find where it paraphrases the d phrase. On the other hand I can't figure out how someone could claim that the GPL incorrectly paraphrases the w phrase.] There has been so much silliness written about this topic ... Agreed. -- Raul
Re: GPL and linking
Raul Miller wrote: Actually, it tries to define work based on the Program in terms of derivative work under copyright law, and then incorrectly paraphrases that definition. It's probably worth noting that derivative work and work based on the Program are spelled differently. What's not clear, to me, is whether the word that refers to the d phrase or the w phrase. Careful study sheds no insight into this burning issue. ??? Let's try again: '' The GPL tries to define work based on the Program in terms of derivative work under copyright law, and then, after this definition and a colon, it tries to explain what is a derivative work under copyright law, but gives a wrong explanation, which would remain wrong even if only USC 17 was considered as a global copyright law. '' See? The GPL says, in its section 0, caput, with [] braces mine: ''a work based on the Program means either the Program or any derivative work under copyright law [definition #1]: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language [explanation, #2].'' I don't know if the meaning of paraphrase is the same in English as its Portuguese cognate, so maybe the misuse of this word is the only error in his analysis... [If I read the GPL, I can't find where it paraphrases the d phrase. On the other hand I can't figure out how someone could claim that the GPL incorrectly paraphrases the w phrase.] There has been so much silliness written about this topic ... Agreed. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: GPL and linking (was: Urgently need GPL compatible libsnmp5-dev replacement :-()
On 5/6/05, Raul Miller [EMAIL PROTECTED] wrote: On 5/5/05, Michael K. Edwards [EMAIL PROTECTED] wrote: Sorry to spam debian-devel -- and with a long message containing long paragraphs too, horrors! -- in replying to this. Who is sorry? How sorry? Let's assume, for the sake of argument, that this sorry-ness is not something that matters enough to you to avoid posting long and elliptical messages to debian-devel. As I wrote, debian-devel is where the Urgently need GPL compatible libsnmp5-dev replacement discussion is happening. Andrew's somewhat disingenuous This part of the thread belongs on -legal notwithstanding, it had not previously been moved to -legal, just copied there. I was uncertain whether to remove -devel from my reply, but eventually decided to leave it as it was; was there some onus on me to remove -devel? I am hardly a major source of -devel noise, by message count or by bandwidth. But perhaps -devel is reserved for short, erroneous, discourteous messages? (That's not really aimed at Raul, actually.) On Wed, May 04, 2005 at 11:51:51PM -0500, Peter Samuelson wrote: The GPL simply defers to copyright law to define derivative work. Actually, it tries to define work based on the Program in terms of derivative work under copyright law, and then incorrectly paraphrases that definition. It's probably worth noting that derivative work and work based on the Program are spelled differently. What's not clear, to me, is whether the word that refers to the d phrase or the w phrase. Careful study sheds no insight into this burning issue. [If I read the GPL, I can't find where it paraphrases the d phrase. On the other hand I can't figure out how someone could claim that the GPL incorrectly paraphrases the w phrase.] Second sentence in Section 0: The Program, below, refers to any such program or work, and a work based on the Program means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. As I read it, the phrase after the colon is a paraphrase of the ether/or clause it follows, i. e., an attempt to restate it in layman's terms. And it's incorrect, as I explained, and for which I have previously given references to treaty, several countries' statutes, and lots of case law, in messages on -legal to which you responded (generally constructively and courteously, I might add). Ignoring the actual definintion and taking the paraphrase would mean that the largest possible work containing GPL licensed material would still be subject to GPL constraints (modulo the mere aggregation clause, which, if it has legal meaning, applies only to Section 2). And yes, anything copyrightable under the Berne Convention is a work, including (for instance) a Debian CD set. That's obviously problematic, it's obviously not what any GPL licensee believes (GPL section 3 0wns my distro? yeah, right), and it's obviously not a reading any court would accept, even absent the rule of construction against the offeror. There has been so much silliness written about this topic ... Agreed. Lots of sarcasm and cheap shots, too; of which I have sometomes been guilty as well. But they do not constitute negative silliness, and are not something I have associated with your by-line in the past. Cheers, - Michael
Re: GPL and linking
Humberto Massa [EMAIL PROTECTED] writes: ??? Let's try again: All of this discussion of legal minutia misses (and perhaps supports) what, to my mind, is the most compelling argument for accepting the FSF's position on the subject. The fact is that the question does depend on a lot of legal minutia that almost all of us aren't qualified to have an opinion on. So unless it's a make-or-break issue for Debian (which I just don't see), the obvious thing to do is to take the agreeable, safe position. So the question of whether or not the FSF is actually *right* doesn't matter. We should only disagree with them if we have to for the sake of Debian -- in which case we're probably in trouble and should hire a lawyer ASAP. -- Jeremy Hankins [EMAIL PROTECTED] PGP fingerprint: 748F 4D16 538E 75D6 8333 9E10 D212 B5ED 37D0 0A03 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: GPL and linking
On 5/6/05, Jeremy Hankins [EMAIL PROTECTED] wrote: All of this discussion of legal minutia misses (and perhaps supports) what, to my mind, is the most compelling argument for accepting the FSF's position on the subject. The fact is that the question does depend on a lot of legal minutia that almost all of us aren't qualified to have an opinion on. So unless it's a make-or-break issue for Debian (which I just don't see), the obvious thing to do is to take the agreeable, safe position. You may not be qualified (as I am not) to offer legal advice. But you're certainly qualified to have an opinion. And there isn't necessarily an agreeable, safe position. If your livelihood depends on your continued ability to work in the software field, I think it helps to have the ability to read deeply into a contract. Sometimes that requires a review of the law applicable to you personally. I know people who really, really wish they hadn't accepted EULA X, let alone Shared Source Agreement Y. Subtle issues of what constitutes contract acceptance in a given jurisdiction, whose interpretation of an ambiguity prevails, and what things can only be agreed to in writing (or can't be made binding at all) do matter. My experience with lawyers has been quite positive overall, but I have learned two cautionary lessons. One: a lawyer's research is always focused on either backing or influencing his or her client's position, and his or her thinking about the arguments on the other side is often limited to finding counter-arguments for them. Two: in the absence of a lawyer who's on your payroll -- not your company's, not your friend's, not a trusted third party's -- you are your own best legal researcher. Actually, the lawyer I respect most says that's true even when he is on my payroll. Use the primary literature; it's not really that hard, though you might have to do a lot of background reading. (Same goes for medicine and algorithms, and almost all science that merits the name.) So the question of whether or not the FSF is actually *right* doesn't matter. We should only disagree with them if we have to for the sake of Debian -- in which case we're probably in trouble and should hire a lawyer ASAP. The FSF has its own agenda, and it's not principally about keeping people out of the courtroom. Many Debian contributors have said that one recent FSF action or another has seriously damaged their trust in the FSF as a steward of the portion of the software commons that they have acquired by copyright assignment, let alone of all software offered under the GPL. Note that the FSF is not unique in this (RedHat, XFree86, and the Mozilla Foundation are other recent examples), and I still think they're on the side of the angels most of the time. Lots of people rely on Debian to have made the most informed judgment its members can about legal issues. That doesn't mean just the SPI's legal counsel, the -legal regulars, or the ftpmasters; that means the DDs and, to a lesser extent, fellow travelers like me. Oh, with respect to Debian as such it doesn't necessarily mean me; IANAL, TINLA, IANADD, and all that. But when it comes to other entities that accept my recommendation of Debian for their IT or product platform, it's my judgment (among others') that they rely on. In the primary literature I trust; all others pay cash. Cheers, - Michael
Re: GPL and linking (was: Urgently need GPL compatible libsnmp5-dev replacement :-()
On 5/6/05, Michael K. Edwards [EMAIL PROTECTED] wrote: On 5/6/05, Raul Miller [EMAIL PROTECTED] wrote: On 5/5/05, Michael K. Edwards [EMAIL PROTECTED] wrote: On Wed, May 04, 2005 at 11:51:51PM -0500, Peter Samuelson wrote: The GPL simply defers to copyright law to define derivative work. Actually, it tries to define work based on the Program in terms of derivative work under copyright law, and then incorrectly paraphrases that definition. It's probably worth noting that derivative work and work based on the Program are spelled differently. What's not clear, to me, is whether the word that refers to the d phrase or the w phrase. Careful study sheds no insight into this burning issue. [If I read the GPL, I can't find where it paraphrases the d phrase. On the other hand I can't figure out how someone could claim that the GPL incorrectly paraphrases the w phrase.] Second sentence in Section 0: The Program, below, refers to any such program or work, and a work based on the Program means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. I believe you're objecting to the that is to say phrase, which restates what work based on the Program: means. As I read it, the phrase after the colon is a paraphrase of the ether/or clause it follows, i. e., an attempt to restate it in layman's terms. Yes. And that either/or clause says what work based on the Program means. And it's incorrect, as I explained, and for which I have previously given references to treaty, several countries' statutes, and lots of case law, in messages on -legal to which you responded (generally constructively and courteously, I might add). I disagree: work based on the Program is not the same thing as derivative work. The definition of work based on the Program uses the derivative work concept, but builds on that concept. I think claiming they're equivalent is silly. -- Raul
Re: GPL and linking
On 5/6/05, Humberto Massa [EMAIL PROTECTED] wrote: ??? Let's try again: '' The GPL tries to define work based on the Program in terms of derivative work under copyright law, and then, after this definition and a colon, it tries to explain what is a derivative work under copyright law, but gives a wrong explanation, which would remain wrong even if only USC 17 was considered as a global copyright law. '' See? The GPL says, in its section 0, caput, with [] braces mine: Except what you're calling a paraphrase of the derivative work concept is a restatement of the work based on the Program concept. Then again, other things you say, such as 'The GPL tries to define' shows that you're not really interested in talking about what the GPL is or what it's saying. The GPL does define work based on the Program. There is no element of try here. The GPL -- not your email -- is the authoritative document about what the GPL does and does not define. -- Raul
Re: GPL and linking (was: Urgently need GPL compatible libsnmp5-dev replacement :-()
On 5/6/05, Raul Miller [EMAIL PROTECTED] wrote: On 5/6/05, Michael K. Edwards [EMAIL PROTECTED] wrote: [snip] Second sentence in Section 0: The Program, below, refers to any such program or work, and a work based on the Program means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. I believe you're objecting to the that is to say phrase, which restates what work based on the Program: means. Attempts to, anyway. As I read it, the phrase after the colon is a paraphrase of the ether/or clause it follows, i. e., an attempt to restate it in layman's terms. Yes. And that either/or clause says what work based on the Program means. Yep. That phrase is, in its entirety: either the Program or any derivative work under copyright law. And that's the definition of work based on the Program for the duration of the GPL, as far as I'm concerned. And it's incorrect, as I explained, and for which I have previously given references to treaty, several countries' statutes, and lots of case law, in messages on -legal to which you responded (generally constructively and courteously, I might add). I disagree: work based on the Program is not the same thing as derivative work. The definition of work based on the Program uses the derivative work concept, but builds on that concept. I think claiming they're equivalent is silly. Right. either the Program or any derivative work under copyright law \superset derivative work. But collections containing the Program don't fit. That is to say introduces an (incorrect) paraphrase -- not a further expansion of the category. To read otherwise is to do violence to both the grammar and the legal sense of the definition; and as I wrote, would result in an unacceptable scope for the license (any work containing GPL material, up to and including an entire CD set and the shelf of books bundled with it). People who say publicly and often enough that they accept the FSF FAQ's statement that programs using GPL libraries must be released under the GPL ( http://www.fsf.org/licensing/licenses/gpl-faq.html#IfLibraryIsGPL ) may well be estopped from arguing otherwise in court. I prefer not to be numbered among them. (And no, before you say it, I'm not trolling to build a defense for some court case.) But that's completely different from affecting the legal meaning of the license (see Linus's LKML post again). I'd be sorry to see, say, a GR swearing allegiance to the FSF FAQ; that would probably estop Debian in perpetuity from linking GPL against non-GPL, trigger the automatic termation provision immediately and retrospectively due to any of a zillion inadvertent build bugs in the past decade, and lead to the Death Of Debian (TM). But it wouldn't have any effect on what license terms I or any Debian user or derivative would be obligated to accept. Cheers, - Michael
Re: GPL and linking
Michael K. Edwards [EMAIL PROTECTED] writes: You may not be qualified (as I am not) to offer legal advice. But you're certainly qualified to have an opinion. Sure. But it's not relevant to this discussion -- despite what many of the participants seem to believe. And there isn't necessarily an agreeable, safe position. Are you saying there's not? So who's going to sue me (or Debian) for adopting an overbroad idea of what constitutes a derivative? Hey, you decided to abide by my license terms when you didn't have to. I'm gonna sue! (Standing? What's that?) Conversely, if our idea of what constitutes a derived work is too narrow we could end up violating someone's copyright. -- Jeremy Hankins [EMAIL PROTECTED] PGP fingerprint: 748F 4D16 538E 75D6 8333 9E10 D212 B5ED 37D0 0A03 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: GPL and linking
On 5/6/05, Jeremy Hankins [EMAIL PROTECTED] wrote: Michael K. Edwards [EMAIL PROTECTED] writes: You may not be qualified (as I am not) to offer legal advice. But you're certainly qualified to have an opinion. Sure. But it's not relevant to this discussion -- despite what many of the participants seem to believe. Did you read any of the rest of my message? This particular sentence of mine disagrees with your claim that almost all of us aren't qualified to have an opinion on license issues. Then what are we doing messing around with other people's copyrighted material? And there isn't necessarily an agreeable, safe position. Are you saying there's not? So who's going to sue me (or Debian) for adopting an overbroad idea of what constitutes a derivative? Hey, you decided to abide by my license terms when you didn't have to. I'm gonna sue! (Standing? What's that?) It's not particularly agreeable or safe to say, we, Debian, interpret the GPL to recursively follow the depends/reverse-depends relationships of GPLed packages, crossing most of the individual API and package boundaries within the work called Debian, and therefore the strong set within Debian is being offered to our users under the GPL alone, even if the individual packages contain MIT/BSD/whatever licenses in debian/copyright. That's probably a little stronger than the estoppel one risks in saying the Debian consensus is that dynamically linked Quagga - NetSNMP - OpenSSL is illegal (disallowed under the GPL), but not much. My take on it is that such relationships are perfectly legal, but that as a courtesy to the FSF we undertake to resolve such situations when they are discovered, either by efforts to obtain unambiguous license compatibility or by package removal. And if it were me, I'd keep building Quagga against NetSNMP while proceeding with reasonable dispatch, but not in a panic, to request that the Quagga upstream get it together with respect to an OpenSSL exemption. The risk in publicly acknowledging the FSF FAQ as a standard of legitimacy is not that anyone will sue you but that Debian will unwittingly provide a stalking-horse for some GPL copyright holder (not necessarily the FSF) to attack Debian users and derivatives. Say, for instance, I write a program that uses an LGPL library whose upstream doesn't follow a copyright assignment policy, and then someone claims that their GPL code was pasted in a while ago. I watch helplessly while Debian relabels it GPL and purges all GPL-incompatible engineering relationships to that library -- and knowing that they have done so might put me at risk of being estopped along with Debian even if I don't agree with the FSF FAQ myself. That would not be a good situation. (By the way, my undying thanks to the Debian X Strike Force for handling the XFree86 license situation the way they have. No panic, no sudden abandonment of the XFree86 code base, just a decision to decline contributions not available under the MIT/X11 license even if they're from upstream, and to move to an alternate upstream fork after sarge. And a carefully written FAQ, not over-commital on legal issues.) Conversely, if our idea of what constitutes a derived work is too narrow we could end up violating someone's copyright. Again, that's not how it works. In the presence of a valid license contract, one is entitled to contract-law standards of the reasonableness of one's attempts to cure a breach when notified. The automatic termination clause is probably unenforceable in most jurisdictions; I think (IANAL) few would even read it as authority to terminate on inadvertent (non-material) breach, let alone on the licensor's idea of breach if the licensee's (reasonable) construction makes it not a breach. Consider how it worked in Progress Software v. MySQL. The FSF's affidavit on MySQL's behalf claimed that Progress's license was terminated, but the judge didn't buy it, and upheld Progress's right to go on distributing MySQL's GPL material. The judge called the derivative work issue a matter of fair dispute -- and hence not a deliberate breach -- noted that it was arguably cured anyway, that MySQL had not demonstrated irreparable harm, and that the balance of harms favored Progress, and denied the request for preliminary injunction on GPL/copyright grounds. For legal purposes, it often matters not only what you do and don't do but why you say you're (not) doing it. Saying in public that you're trying to do X less often because you believe it's illegal is injudicious at best. Doubly so if you go on to say that you believe that you permanently lost your rights under a license every time you did X. Cheers, - Michael
Re: GPL and linking (was: Urgently need GPL compatible libsnmp5-dev replacement :-()
On 5/6/05, Michael K. Edwards [EMAIL PROTECTED] wrote: On 5/6/05, Raul Miller [EMAIL PROTECTED] wrote: On 5/6/05, Michael K. Edwards [EMAIL PROTECTED] wrote: [snip] Second sentence in Section 0: The Program, below, refers to any such program or work, and a work based on the Program means either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications and/or translated into another language. I believe you're objecting to the that is to say phrase, which restates what work based on the Program: means. Attempts to, anyway. I think this attempts to quip is meaningless. As I read it, the phrase after the colon is a paraphrase of the ether/or clause it follows, i. e., an attempt to restate it in layman's terms. Yes. And that either/or clause says what work based on the Program means. Yep. That phrase is, in its entirety: either the Program or any derivative work under copyright law. And that's the definition of work based on the Program for the duration of the GPL, as far as I'm concerned. To recap: W: work based on the program D: derivative work E: either/or phrase C: phrase after the colon. W means E C paraphrases E Thus, you have concluded, C attempts to paraphrase D Should we keep going back and forth on this, trying to show why you believe C attempts to paraphrase D? Also, either: (1) Your other paragraphs are logically based on this concept (C attempts to paraphrase D), and therefore are based on a false premise, or (2) Your other paragraphs are not related to this paragraph by theme or logic, and thus there's little point in continuing unless they contain some worthwhile independent theme (personally, I've not spotted one -- they just seem like a bunch of statements with little cohesive logic).] Or something else? I don't know why it's important that all this be sent to debian-devel. After this post, I'm probably going to delete debian-devel from my followups (and a great sigh of relief is heard throughout the land). -- Raul
Re: GPL and linking (was: Urgently need GPL compatible libsnmp5-dev replacement :-()
On 5/6/05, Raul Miller [EMAIL PROTECTED] wrote: On 5/6/05, Michael K. Edwards [EMAIL PROTECTED] wrote: On 5/6/05, Raul Miller [EMAIL PROTECTED] wrote: I believe you're objecting to the that is to say phrase, which restates what work based on the Program: means. Attempts to, anyway. I think this attempts to quip is meaningless. How would you like me to say it? Purports to? Professes to? Makes an honest but flawed effort to? Do you not understand my interpretation that the use of quotes around work based on the Program means that the writer is defining it as shorthand for either the Program or any derivative work under copyright law? And that an attempt is then made to paraphrase (restate, whatever) the latter phrase, and that restatement is just plain wrong? You don't have to agree with it, of course, but surely you get it now. Yes. And that either/or clause says what work based on the Program means. Yep. That phrase is, in its entirety: either the Program or any derivative work under copyright law. And that's the definition of work based on the Program for the duration of the GPL, as far as I'm concerned. To recap: W: work based on the program D: derivative work E: either/or phrase C: phrase after the colon. W means E C paraphrases E Thus, you have concluded, C attempts to paraphrase D No. E defines W, which appears in quotes in the original to indicate that it is being given a formal meaning. C is grammatically a paraphrase of E. However, C and E are not the same thing according to law; and grammatically and legally, E is the definition of W, and C is not. Neither is C \union E, C - D, or some other way to assign W a meaning based on the wording of W, the content of an unrelated document, or the distance to the moon. Should we keep going back and forth on this, trying to show why you believe C attempts to paraphrase D? I don't, except insofar as C - the Program attempts to paraphrase E - the Program (= D). Are we done? And if you're going to move it to private e-mail, do it, don't grandstand about it. That is also more characteristic of others around here than it previously has been of you. Cheers, - Michael
Re: GPL and linking (was: Urgently need GPL compatible libsnmp5-dev replacement :-()
I don't, except insofar as C - the Program attempts to paraphrase E - the Program (= D). Oh for Pete's sake, (E - the Program) (= D). What a great place for a word wrap. - Michael
GPL and linking (was: Urgently need GPL compatible libsnmp5-dev replacement :-()
On 5/4/05, Andrew Suffield [EMAIL PROTECTED] wrote: [This part of the thread belongs on -legal] Sorry to spam debian-devel -- and with a long message containing long paragraphs too, horrors! -- in replying to this. But that's where this discussion is actually happening now, and I'm afraid I can't agree with Andrew's implication that this issue is settled on debian-legal in favor of the FSF FAQ's interpretation. This isn't about license text, this is about GPL FUD and Debian's maintainers and other contributors, and debian-devel as a whole needs to hear it once in a while. I argue largely in the context of US law because it's been convenient for me to research, but I welcome counter-arguments from other legal systems -- with concrete references. On Wed, May 04, 2005 at 11:51:51PM -0500, Peter Samuelson wrote: [Paul TBBle Hampson] This of course assumes the phrase derived work is legalese for code dependancy or something. I'm sure the GPL actually defines what _they_ mean by it... The GPL simply defers to copyright law to define derivative work. Actually, it tries to define work based on the Program in terms of derivative work under copyright law, and then incorrectly paraphrases that definition. Under contract law (in most US jurisdictions at least, IANAL, etc.) the recipient is entitled to have this ambiguity construed against the drafter. More below. I might add that claiming a program that uses a library's published API is a derived work is a bit shaky from the get-go. If you actually cut and paste code from the library into your program, it's a lot more clear-cut. We talk about APIs on forums like -legal to save time, because everybody (supposedly) knows what we're talking about there. They aren't directly relevant, it's just that certain aspects of program design will normally have certain legal implications because that's how those things are normally implemented. I think Peter has it right, and I'd like to know what grounds there may be to demur. See my recent posts to debian-legal archives for US case law on the matter, which I (IANAL) would summarize as published APIs are not copyrightable in their role as 'methods of operation' as distinct from their role as creative expression. It's kind of an odd stance for the law to have arrived at -- a difference of usage changes not just whether an act of copying is an infringement but whether the copied material is copyrightable at all. But it makes sense in the context of the prescribed sequence of legal analysis, in which recognizing a protected use too late in the sequence leaves the copier open to lawsuits de novo for subsequent acts of copying the same material. The last time I know of that the US Supreme Court looked at the issue -- an appeal from Lotus Development Corporation v. Borland International, Inc.,49 F.3d 807 (1995) -- they deadlocked 4-4 in one justice's absence. The court transcript is fascinating. The latest and greatest analysis at circuit court level appears to be Lexmark v. Static Control (2004). Yes, the US is not the world. Other legal systems are perfectly within their rights to arrive at different conclusions, and the Berne Convention is far from explicit on the matter. But what actual grounds are there for a belief that some particular country's legal system would rule that the arm's-length use of a published API creates a derivative work? Chapter and verse, folks; even if precedents are not law in your legal system, they're a lot more reliable than reasoning outside a courtroom with no judge in sight. Changing static linking to dynamic, or replacing a linker call with a dlopen() call, *always* has precisely *zero* effect on whether something is a derivative work or not. A work is created derivative, or not, at the time of inception. For source code, this is the time when the code is written. The way in which it is compiled is irrelevant. For a binary, this is the time when the binary is built and linked. A statically linked binary is a derivative work of everything it links because it contains verbatim copies of those things. Every binary, static, dynamic, or other, is a derivative of everything that any part of its source was derived from. I do not think that the binary part of this analysis is correct in any country that implements the Berne Convention. My rebuttal is long enough without the case law references, but you can find them all in the debian-legal archives. Whether statically linked or provided as multiple dynamically linked files, a program composed of separately identifiable independent works of authorship is a collection (in some countries' implementation, compilation) as defined in Article 2 Section 5. Derivative works are defined in Article 2 Section 3 to be [t]ranslations, adaptations, arrangements of music and other alterations of a literary or artistic work. These exist as categories of copyrightable works for completely separate reasons --
