Re: LGPL module linked with a GPL lib
On 8/4/05, Diego Biurrun [EMAIL PROTECTED] wrote: On Wed, Aug 03, 2005 at 07:24:33PM -0700, Michael K. Edwards wrote: If a public figure as remarkable as RMS does not choose to gather sizable donations to his preferred charity in return for his speaking engagements, then perhaps conference organizers should be prepared for the eccentric behavior that is occasionally reported. You wouldn't - AGAIN - be saying bad things about somebody in public with nothing more than rumors to back it up, would you? It's controversial to say that RMS is occasionally reported to behave eccentrically? And that being a conference speaker doesn't necessarily stop him? Try the CODE conference in 2001, reported at http://www.theregister.co.uk/2001/04/06/free_software_would_have_prevented/ . Bruce Perens was also a speaker; I wonder how he remembers the occasion? Not being a native speaker I'll have to admit that there are subtleties in English that escape me. You are directly questioning another person's character with your - unsubstantiated - theories and spreading those claims in public. This is extremely rude and at some point crosses over into the realm of what dict.leo.org translates into English as slander or libel. Dude, the guy's a public figure, he advocates public policy positions which are intimately tied to his personal fortunes, he participates in and condones the repetition of what my research (IANAL) says is a false set of claims about how copyright licenses work -- claims whose truth or falsity has real consequences -- and if you actually bother to _read_ what I wrote you'll see that I never stated more than a suspicion based on what public evidence I had dug up so far. Slander and libel are serious accusations and if you don't know what they mean it's incautious at best to sling them around. Just as a little reminder, here are the things I wrote that you called slanderous: # Although I have no personal knowledge on the financial side, it # certainly looks to me like it has made them both rich men. Little # snippets in the public record -- Jim Blandy's comment at # http://www.jwz.org/doc/lemacs.html about RMS's luxurious pad on the # fourth floor of posh NE43, the indications from Moglen's letter to # Vidomi and Fluendo's defense of GStreamer that he has ways of # extracting revenues from his role in the FSF, back-of-the-envelope # calculations involving the typical conference speaker fee -- suggest # to me that their tax records would make interesting reading. I could # be completely wrong; they could be scraping by on nominal salaries # from the FSF and Columbia University, plus a MacArthur fellowship here # and there; but it's enough for me to take their bizarre, # uncorroborated assertions about copyright law with a grain of salt. # There's a lot of money to be made in this # area (although it's a pretty hard life if you have close friends and # like your home); and if RMS had a way of laundering the money (don't # give it to me; but donate to the FSF if you like) so as to appear # saintly, he wouldn't be the first. Although I regretted the use of the word laundering immediately on re-reading the latter, and went to some effort to find what facts I could in the public record so as to lay that particular suspicion to rest, you will find neither untruth nor malice in the above. Skepticism, yes, shading over into cynicism in the latter paragraph; but the typical (profit-seeking) conference speaker's fee, and the typical retainer for a legal opinion effectively estopping your prospective opponent's lawyer from arguing otherwise in court, are real money. I have no compunction about saying that I want to know where that money goes -- and if RMS doesn't collect speaker fees, what favors (if any) he asks instead. This should be a dead giveaway that you should not have speculated in the first place, much less on a public mailing list that will be archived from here to eternity. I'd sure rather look back on what I've written, errors and all, than on ignorance and complacency. [snip fairly empty flames from both sides] Diego, I don't care whether you get anything out of what I write or not. Nor do I think that anyone who gives two shakes about what the truth is cares whether _you_ judge [my] evidence poorly researched; they're presumably going to judge for themselves. But I do feel a little down about this particular episode (my cynicism got the better of me, in a mild sort of way, thinking about the enrichment potential of the rubber-chicken circuit), and I don't particularly like being kicked while I'm down. I doubt it impresses anyone else, either. - Michael
Re: LGPL module linked with a GPL lib
On Sat, Aug 06, 2005 at 01:15:22AM -0700, Michael K. Edwards wrote: On 8/4/05, Diego Biurrun [EMAIL PROTECTED] wrote: On Wed, Aug 03, 2005 at 07:24:33PM -0700, Michael K. Edwards wrote: If a public figure as remarkable as RMS does not choose to gather sizable donations to his preferred charity in return for his speaking engagements, then perhaps conference organizers should be prepared for the eccentric behavior that is occasionally reported. You wouldn't - AGAIN - be saying bad things about somebody in public with nothing more than rumors to back it up, would you? It's controversial to say that RMS is occasionally reported to behave eccentrically? And that being a conference speaker doesn't necessarily stop him? No. But this is not what you are implying in the paragraph above. You imply that he takes the liberty to misbehave due to the fact that he does not get money for his speaking engagements. This is controversial and what is worse, you have (again) no facts to back it up. Try the CODE conference in 2001, reported at http://www.theregister.co.uk/2001/04/06/free_software_would_have_prevented/ You're not seriously presenting this polemic diatribe as fact here, are you? Not being a native speaker I'll have to admit that there are subtleties in English that escape me. You are directly questioning another person's character with your - unsubstantiated - theories and spreading those claims in public. This is extremely rude and at some point crosses over into the realm of what dict.leo.org translates into English as slander or libel. Dude, the guy's a public figure, he advocates public policy positions which are intimately tied to his personal fortunes, he participates in and condones the repetition of what my research (IANAL) says is a false set of claims about how copyright licenses work -- claims whose truth or falsity has real consequences -- and if you actually bother to _read_ what I wrote you'll see that I never stated more than a suspicion based on what public evidence I had dug up so far. Slander and libel are serious accusations and if you don't know what they mean it's incautious at best to sling them around. I _have_ read what you wrote and in contrast to you I have never misstated your words. What you are/were doing are serious accusations. For a public figure like RMS his reputation and integrity are everything and attacking them is serious indeed. Thus it's not only uncautious but rude to go around uttering suspicions based on what somebody who once had a quarrel with RMS implied on a random webpage in a public forum like this mailing list. In my legislation you cross the border to slander/libel with these things at some point. But that's not the point. It's serious enough even without the legal implications so let's drop this particular issue. Diego, I don't care whether you get anything out of what I write or not. Nor do I think that anyone who gives two shakes about what the truth is cares whether _you_ judge [my] evidence poorly researched; they're presumably going to judge for themselves. Of course they are, but you are quoting me out of context. You were talking to me directly and I gave you my opinion. But I do feel a little down about this particular episode (my cynicism got the better of me, in a mild sort of way, thinking about the enrichment potential of the rubber-chicken circuit), and I don't particularly like being kicked while I'm down. I doubt it impresses anyone else, either. If you can't stand the heat... Seriously Michael, you have no problem with telling other people off, even in terms that these people find insulting. You happen to be on the receiving end right now. Deal with it. If you wish to avoid this in the future, treat people in a different way. I don't mean to refuse an outstretched hand, though. I'll go on vacation tomorrow and would very much like to put this flamefest to an end. Diego -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Code of conduct and MUAs [was: Re: LGPL module linked with a GPL lib]
Francesco Poli [EMAIL PROTECTED] wrote: IIRC, the code of conduct says that the canonical way to ask to be Cc:ed on replies is setting an appropriate Mail-Followup-To: field. Asking the same in the message body (in natural language) is a useful reminder for users of MUAs that do not automatically honour the Mail-Followup-To: field. Am I correct? Not as far as I can tell. The code of conduct doesn't mention MFT. I'm not surprised, because MFT is a controversial non-working invented header which failed to get standardised and is only supported by a few mailers. Mention your CC wish in your sig if you feel strongly. -- MJR/slef My Opinion Only: see http://people.debian.org/~mjr/ Please follow http://www.uk.debian.org/MailingLists/#codeofconduct -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On 8/6/05, Diego Biurrun [EMAIL PROTECTED] wrote: On Sat, Aug 06, 2005 at 01:15:22AM -0700, Michael K. Edwards wrote: It's controversial to say that RMS is occasionally reported to behave eccentrically? And that being a conference speaker doesn't necessarily stop him? No. But this is not what you are implying in the paragraph above. You imply that he takes the liberty to misbehave due to the fact that he does not get money for his speaking engagements. This is controversial and what is worse, you have (again) no facts to back it up. So you're saying that knowing that his charity's donation is riding on his conduct would not make him feel any more inclined to comport himself with dignity? Now who's being insulting? Try the CODE conference in 2001, reported at http://www.theregister.co.uk/2001/04/06/free_software_would_have_prevented/ You're not seriously presenting this polemic diatribe as fact here, are you? Polemic diatribe? Bill Thompson is a regular BBC correspondent, generally sympathetic to Free Software, opponent of software patents ( http://news.bbc.co.uk/1/hi/technology/3782771.stm ), hardly my idea of an anti-RMS ideologue. Hundreds of people were at that conference, including (as I wrote in the part you snipped) Bruce Perens, who could be asked about his recollection of the event -- if you really think there is anything to dispute; even the most sympathetic of RMS biographers usually mention that he sometimes flies off the handle when opposed on an ethical point. I'm something of a hot-head myself, but then I don't go around inviting critiques of my moral perspective from conference podiums, do I? [snip more flaming, except:] In my legislation you cross the border to slander/libel with these things at some point. I'd like to see you back that up. I'm fortunate to live in a country where it's pretty hard to prove libel against someone who's looking for the truth about a public figure, especially if it's clear that he's not motivated by personal malice. If you are incapable of judging those distinctions when you have an allegiance to that public figure, that's not my problem. - Michael
Re: LGPL module linked with a GPL lib
I wrote: They're a sidetrack to be sure; but kind of an interesting sidetrack. His personal history and philosophy strike me as more reminiscent of Dominic de Guzman or Benedict of Nursia than any modern figure. In any case, I certainly intended no slur on RMS by that, nor on any participant in this discussion. You know, it's funny; I make a comment like that, and a few hours later I run across Nikolai Bezroukov's comment that some of [RMS's] recent letters look like they have been written by a medieval theologian, and I feel dirty. Not because I meant a slur, because I didn't; Dominic and Benedict seem to have been decent, even saintly, men, and leaders of men and women, and comparisons to them are complimentary in a way that, say, a parallel to Francis of Assisi (nice to animals, may have composed a good prayer or two, but a certified kook) wouldn't be. But as thick as my writing style is, I'm sure it's hard to tell the difference between my cumulative critiques and a real hatchet job like Bezroukov's. And once in a while I go off half cocked (not with economic superiority of the free software system, which was a deliberate re-framing of RMS's published philosophy, but with an apparently wrong guess about which non-programming source of income keeps his boat afloat), and I find myself wishing I'd left the whole topic alone. But dammit, this is not a game, this is people's lives and livelihoods. Using deception about the law to claim rights over other people's work is wrong, no matter who is doing it. RMS may sincerely believe that the GPL is a successful hack around contract law and the limits courts have imposed on other software copyright holders; but I don't see how a court could possibly agree with him. Where the money comes from, and where it goes, do have some bearing on whether it's proper to accept the FSF's unsubstantiated assertions on legal matters; and I want to know the truth, and to see it known and acted on by people whose influence over the free software ecosystem is greater than mine. - Michael
Re: LGPL module linked with a GPL lib
I wrote: RMS may sincerely believe that the GPL is a successful hack around contract law and the limits courts have imposed on other software copyright holders; but I don't see how a court could possibly agree with him. Not to be paranoid or anything, but a reminder-disclaimer: The GPL clearly is, and in my non-lawyer view should be, enforceable on people who make and distribute copies of GPLed software, modified or not. It just doesn't, as written, and in my view shouldn't no matter how it is written, compel the use of the GPL on software whose relationship to a GPL work is a matter of economics and engineering (through a published API) rather than modified or adapted expressive content. Tiredly, - Michael
Re: LGPL module linked with a GPL lib
* Michael K. Edwards [EMAIL PROTECTED] [050804 04:24]: And I judge your evidence poorly researched. This does not enhance your credibility when you expound at length (and length and length) on legal affairs. That was a good one. It's really interesting that people who show no evidence of having invested any effort whatsoever themselves in research of any aspect of this topic are so quick to reject, not only the slightest speculation beyond the proven facts, but any evidence I may have brought to bear on any conclusion distasteful to them. I'm sorry. But I (and I think most here) can not rethink everything every time someone writes something. So if you say something I think is wrong, and you want me to believe it, it might be helpful to give proofs, hints or proper arguments. I do not know about the other people here, but I guess most of them will simply ignore long texts. If you have a point, make it. Long self-contradictory texts, inconsistent reasonings, deducing things I do not believe from things I hold even less true or not at all, not to forget your accusing speech will not make me believe you, but only make me feeling guilty of feeding the trolls when answering you. Ignorance is bliss, I guess. Oh, all hail Micheal, the great martyr trying to open our eyes. Better this way? Sorry for writing, but I can only ignore a certain amount of such posts, Bernhard R. Link -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On 8/3/05, Michael K. Edwards [EMAIL PROTECTED] wrote: If we can't even manage this issue in the context of a single paragraph, what hope do we have of codifying protection for newly thought up instances of this issue, in law? That would be the reason that the integrity and competence of judges matters. Your Pyrrhonist (I just learned that nifty word) jump from imperfect certainty to quietism does not persuade me. Nevertheless, intellectual property is fundamentally different from real property, and the differences, in the general case, make it impossible to determine the boundaries of intellectual property. Establishing property rights on intellectual property is like establishing property rights on shadows. Shadows very definitely exist, and you can write laws about them, but they shift and move depending on circumstances. And note that I've never claimed that intellectual property cannot be the subject of law. I've instead been claiming that such law can never equal the laws for real property. There's simply too much uncertainty about the domain for that to work. Answer: codification is easy -- it's easy to put words down on paper and call them law -- but it's unlikely that this codification will ever be meaningful in a general context. Thus, no one really wants to take copyright issues to court, because fundamentally the laws don't make sense. When taken at face value, the concepts are simple enough, but the protected works are not real property. As you point out, they're not even chattels. If you think no one takes copyright issues to court, then you have a very different perspective on the literature than I do. Copyright law mostly makes sense to me, and I rarely feel that I would have decided a recent appellate case differently (though I favor Corey Rusk over Rano and am not that fond of Eldred v. Ashcroft). I do not think this can be attributed to a tendency on my part to accept authority. :-) I said no one wants to take these issues to court, not that they are not taken to court. But I wasn't conveying my thoughts properly -- what I meant to point out that letting the court resolve copyright issues is extremely unpopular -- it's just not wise in most circumstances. (Though I challenge you to show me any cases of real property which does not stake out a physical chunk of the planet. (I'm aware that you can, at least in some cases, move dirt from one location to another, without changing the legal definition of the property boundaries. But my point is: you can determine those property boundaries because of physical properties of matter, such as the fact that mass is conserved, which do not apply in the realm of intellectual property.)) You certainly have a point that the boundaries of a copyright (or patent or trademark) holder's rights are imperfectly defined; but if you have ever owned rural property you may be aware that the same is often true of land, although survey-grade (centimeter-accurate) GPS helps. :-) Modern copyright is an imperfect system, but it sure beats hell out of what preceded it. In those cases, the problem is not one of physics, but one of choice. The properties could be surveyed more precisely, if someone cared enough to do so. With intellectual property, there is nothing to be surveyed. As for xemacs and emacs: RMS has not accepted xemacs code into emacs because the xemacs developers would not, or could not, transfer copyright ownership on that code to the FSF. Which has little to do with (US) copyright law, given precedents such as Aalmuhammed v. Lee;... Compare: http://www.law.cornell.edu/copyright/cases/202_F3d_1227.htm Aalmuhammed never had a written contract... with http://www.gnu.org/prep/maintain/html_node/Copyright-Papers.html#Copyright-Papers Before incorporating significant changes, make sure that the person who wrote the changes has signed copyright papers and that the Free Software Foundation has received and signed them. -- Raul
Re: LGPL module linked with a GPL lib
On 8/4/05, Raul Miller [EMAIL PROTECTED] wrote: Nevertheless, intellectual property is fundamentally different from real property, and the differences, in the general case, make it impossible to determine the boundaries of intellectual property. It's a _little_ more abstract than real property ownership, which is a lot more abstract than possession of a chattel; but it's rather less abstract than, say, ownership of a 401(k) account -- a device where you have limited control of some numbers in a brokerage firm's computer, and the changes you request may or may not result in the actual trading of some mutual fund shares, which in turn once in a while results in the trading of common stock of some companies, which means God knows what. Yet the law has no great difficulty with that kind of property either. If you're looking to set this kind of a limit on the property abstraction, you're about 400 years too late (joint stock companies; 300 for copyright, of course). Establishing property rights on intellectual property is like establishing property rights on shadows. Shadows very definitely exist, and you can write laws about them, but they shift and move depending on circumstances. And note that I've never claimed that intellectual property cannot be the subject of law. I've instead been claiming that such law can never equal the laws for real property. There's simply too much uncertainty about the domain for that to work. Intellectual property law is so much simpler than real estate law (let alone securities law), and so unlike your property rights on shadows straw man, as to make this whole line of inquiry quite fruitless. Copyright, patent, and trademark are imperfect but workable, and they reflect a social consensus that may be fraying in spots (media consumers do like their free beer, and the MPAA/RIAA counterattack has also done quite a bit of damage) but isn't terribly amenable to preaching from either direction. I said no one wants to take these issues to court, not that they are not taken to court. But I wasn't conveying my thoughts properly -- what I meant to point out that letting the court resolve copyright issues is extremely unpopular -- it's just not wise in most circumstances. Nobody wants to go to court if they don't think they have something substantial to gain by it; going to court is expensive and risky. But I certainly have no interest in letting anyone _other_ than legislatures and courts resolve these issues if the parties directly involved cannot -- not DRM schemers, not P2P sharers, not captured regulators, and not RMS either. In those cases, the problem is not one of physics, but one of choice. The properties could be surveyed more precisely, if someone cared enough to do so. With intellectual property, there is nothing to be surveyed. Dude, it's just _not_that_hard_. The copyright, patent, and trademark systems work just fine. How do I know? Because in almost every court case in this area I read about, one side or both is a bunch of assholes. Very, very rarely do you see the bellwether of bad law -- a case where two parties who are minding their own business, coloring within the lines, living and letting live, and peaceably making a buck wind up duking it out in court. You get a few cases where both sides seem sincere about thinking they were in the right, but wound up in a competitive conflict that hinged on a subtle point -- cases like Fogerty v. Fantasy and Lotus v. Borland. But you get a lot more cases where it's hard to have much sympathy for either side, like Sun v. Microsoft and Napster v. RIAA -- cases where greed meets greed and they're either fighting over the scraps of a deal gone wrong (usually an ill-conceived deal in the first place) or tussling over who controls _access_ to a market where square deals are few and far between. Compare: http://www.law.cornell.edu/copyright/cases/202_F3d_1227.htm Aalmuhammed never had a written contract... with http://www.gnu.org/prep/maintain/html_node/Copyright-Papers.html Before incorporating significant changes, make sure that the person who wrote the changes has signed copyright papers and that the Free Software Foundation has received and signed them. Sorry, I didn't make my point clear here, though I've made it on debian-legal before. Aalmuhammed could not exert any of the rights of a copyright holder because his contribution did not rise to the level of authorship on any work that could stand in isolation. (Stand as a coherent chunk of creative expression; please don't go haring off into copyright-irrelevant engineering criteria like whether it can be executed without linking to some other stuff first.) I do not find it remotely plausible (IANAL) that patches, even quite extensive patches, knowingly contributed on GPL terms and incorporated by a maintainer who retains creative control, represent any obstacle to GPL enforcement. RMS had a bad experience with
Re: LGPL module linked with a GPL lib
On 8/4/05, Michael K. Edwards [EMAIL PROTECTED] wrote: It's a _little_ more abstract than real property ownership, which is a lot more abstract than possession of a chattel; but it's rather less abstract than, say, ownership of a 401(k) account -- a device where you have limited control of some numbers in a brokerage firm's computer, and the changes you request may or may not result in the actual trading of some mutual fund shares, which in turn once in a while results in the trading of common stock of some companies, which means God knows what. Yet the law has no great difficulty with that kind of property either. If you're looking to set this kind of a limit on the property abstraction, you're about 400 years too late (joint stock companies; 300 for copyright, of course). I'm not sure how you're quantifying abstractness, but there's a significant difference between 401K accounts and intellectual property. 401K accounts are more more limited in scope, and are much more regular in concept. With intellectual property the rules are, by definition, new for each property. (Well, for copyrights and patents -- trademarks all follow pretty much the same pattern and your analogy is good for that case.) From experience, when you break into new intellectual ground, you're going to make mistakes. Worse, the intellectual breadth of our civilization is so broad that what's new for some community is hundreds of years old for another. And, yet, the thing that legitimizes intellectual property is that it's something new and previously undiscovered. (At least for copyrights and patents.) And note that I've never claimed that intellectual property cannot be the subject of law. I've instead been claiming that such law can never equal the laws for real property. There's simply too much uncertainty about the domain for that to work. Intellectual property law is so much simpler than real estate law (let alone securities law), and so unlike your property rights on shadows straw man, as to make this whole line of inquiry quite fruitless. What's not simple about property rights on shadows? For that matter, why is this a straw man? Remember, this is America. You can sue anyone for anything. http://recenter.tamu.edu/pubs/1092.html (That said, in existing law you still wind up dealing with the boundaries associated with real property.) In those cases, the problem is not one of physics, but one of choice. The properties could be surveyed more precisely, if someone cared enough to do so. With intellectual property, there is nothing to be surveyed. Dude, it's just _not_that_hard_. The copyright, patent, and trademark systems work just fine. How do I know? Because in almost every court case in this area I read about, one side or both is a bunch of assholes. Very, very rarely do you see the bellwether of bad law -- a case where two parties who are minding their own business, coloring within the lines, living and letting live, and peaceably making a buck wind up duking it out in court. You get a few cases where both sides seem sincere about thinking they were in the right, but wound up in a competitive conflict that hinged on a subtle point -- cases like Fogerty v. Fantasy and Lotus v. Borland. But you get a lot more cases where it's hard to have much sympathy for either side, like Sun v. Microsoft and Napster v. RIAA -- cases where greed meets greed and they're either fighting over the scraps of a deal gone wrong (usually an ill-conceived deal in the first place) or tussling over who controls _access_ to a market where square deals are few and far between. Tell that to Mike Jittlov. Or, Courtney Love. Or, ... In other words, the existence of people happy with a situation doesn't mean that all people are happy with all situations. And, ultimately, copyright exists because it makes some people happy. Compare: http://www.law.cornell.edu/copyright/cases/202_F3d_1227.htm Aalmuhammed never had a written contract... with http://www.gnu.org/prep/maintain/html_node/Copyright-Papers.html Before incorporating significant changes, make sure that the person who wrote the changes has signed copyright papers and that the Free Software Foundation has received and signed them. Sorry, I didn't make my point clear here, though I've made it on debian-legal before. Aalmuhammed could not exert any of the rights of a copyright holder because his contribution did not rise to the level of authorship on any work that could stand in isolation. If the facts of all copyright cases were identical, I'd agree with you. However, (for example) given the way interfaces are treated in copyright cases, and given that a computer program is rife with interfaces, this needn't be the triviality you make it out to be. (Stand as a coherent chunk of creative expression; please don't go haring off into copyright-irrelevant engineering criteria like
Re: LGPL module linked with a GPL lib
Apologies to all innocent bystanders for what has degenerated into an offtopic flamefest. Unfortunately MKE has made some statements directed at myself that I feel I cannot leave unanswered. On Wed, Aug 03, 2005 at 07:24:33PM -0700, Michael K. Edwards wrote: On 8/3/05, Diego Biurrun [EMAIL PROTECTED] wrote: If a public figure as remarkable as RMS does not choose to gather sizable donations to his preferred charity in return for his speaking engagements, then perhaps conference organizers should be prepared for the eccentric behavior that is occasionally reported. You wouldn't - AGAIN - be saying bad things about somebody in public with nothing more than rumors to back it up, would you? There's a lot of money to be made in this area (although it's a pretty hard life if you have close friends and like your home); and if RMS had a way of laundering the money (don't give it to me; but donate to the FSF if you like) so as to appear saintly, he wouldn't be the first. You're again bordering on slander, I'd tread more carefully if I were you. Still far from slander; Not being a native speaker I'll have to admit that there are subtleties in English that escape me. You are directly questioning another person's character with your - unsubstantiated - theories and spreading those claims in public. This is extremely rude and at some point crosses over into the realm of what dict.leo.org translates into English as slander or libel. but I confess that I regretted this immediately, and was relieved to be able to find enough public evidence in a matter of a couple of hours to refute my own speculation, at least as regards the FSF. This should be a dead giveaway that you should not have speculated in the first place, much less on a public mailing list that will be archived from here to eternity. Treading carefully does not, if I may say so, seem to be _your_ specialty. I beg your pardon? I have never insulted anybody around here, nor do I question anybody's character. I especially do not question the integrity of public figures with what amounts to nothing to back up my claims, only to be rebutted by everybody including myself immediately. And I judge your evidence poorly researched. This does not enhance your credibility when you expound at length (and length and length) on legal affairs. It's really interesting that people who show no evidence of having invested any effort whatsoever themselves in research of any aspect of this topic are so quick to reject, not only the slightest speculation beyond the proven facts, but any evidence I may have brought to bear on any conclusion distasteful to them. Ignorance is bliss, I guess. What you said about RMS and speakers at free software conferences etc was poorly researched as you even found out yourself after a bit of digging. Shooting from the hip like this has done serious damage to your credibility when talking about other topics. It's not something I'm making up to spite you, it was stated explicitly in some of the replies to your messages in this thread. Ignorance of these facts is indeed bliss. Diego -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On 8/3/05, Michael K. Edwards [EMAIL PROTECTED] wrote: I wrote: So yes, inquiring minds want to know. And this inquiring mind is now satisfied as to what probably pays RMS's rent lately -- the ~$268K Takeda Award he received in 2001. (You couldn't keep a family in Cambridge for four years on that, but RMS doesn't have that problem.) Me, I'd be kind of ashamed to preach the economic superiority of the free software system while living on grant money and conference banquets; but YMMV. RMS doesn't preach the economic superiority of free software. If you fail to understand even such a well-explained position I wonder what your references to all kinds of precedents and such are worth.
RE: LGPL module linked with a GPL lib
** Raul :: On 8/2/05, Michael K. Edwards [EMAIL PROTECTED] wrote: I'm just telling you how it looks to me, and pointing you to where I got what evidence I have so that you can judge for yourself. The FSF is notoriously unforthcoming about their financial dealings, and the cash flows involved are not chump change (see the numbers disclosed by Jamie Zawinski in the Lucid Emacs saga). Whether or not you think RMS and Eben Moglen are cashing in personally (about which I have no evidence), if you are willing to take their uncorroborated claims about the legal strategy at the heart of their enterprise at face value, you are a more trusting man than I. This sounds like something appropriate for the scandal column of a tabloid. But what's the relevance of this issue to debian-legal? IMHO its relevance to d-l is that, if such suspicions are indeed founded, the FSF GPL FAQ should not be taken by face value and that Debian should re-evaluate its position about GPL and linking. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On 8/2/05, Michael K. Edwards [EMAIL PROTECTED] wrote: Mostly I care about the freedom to pursue what is for me both an intellectual interest and a trade, on terms which more or less reflect an accurate perspective on the surrounding law and economics. Misrepresentations and charlatanry draw my scrutiny, whether they come from saints or sinners. I have no problem with you pursuing any such intellectual interests. However, this probably isn't the right list for posting random facts. I'd suggest debian-curiosa. Or, if anyone wants to create it: debian-saints. -- Raul
Re: LGPL module linked with a GPL lib
On 8/3/05, Humberto Massa Guimarães [EMAIL PROTECTED] wrote: IMHO its relevance to d-l is that, if such suspicions are indeed founded, the FSF GPL FAQ should not be taken by face value and that Debian should re-evaluate its position about GPL and linking. Why? Personally, I've quoted this faq as evidence of things that the FSF has done. I don't see any need to re-evaluate those kinds of issues. Beyond that, I believe it's mostly used to provide succinct heuristics for dealing with issues where there is no case law. Because these are heuristics, they can only be approximations and thus we already have to be prepared to evaluate those issues for specific cases. But I see no need to re-evaluate those issues for the general case. -- Raul
Re: LGPL module linked with a GPL lib
On 8/2/05, Patrick Herzig [EMAIL PROTECTED] wrote: RMS doesn't preach the economic superiority of free software. If you fail to understand even such a well-explained position I wonder what your references to all kinds of precedents and such are worth. You've got a fair point, in that RMS doesn't see his arguments as preaching economic superiority; and certainly many commentators have contrasted RMS's ethical perspective with, say, ESR's economic perspective. I don't entirely agree with the way this contrast is portrayed, and in particular I think the ethical/economic dichotomy is a false one. Ethical/financial, perhaps; but that's a calculus of personal motivations that isn't really all that fruitful to discuss. Implicit in my perspective is the view that ethics is the study of human motivation, and economics is the application of the fruits of this study to the public sphere; finance is just probabilities and algebra. So, as I say, I don't fail to understand RMS's attitude that his arguments are ethical arguments and trump economic considerations; I just don't agree. He is welcome to that position with regard to his own choice to publish his source code, and welcome to exhort others to go and do likewise (as I have done, and am likely to do again, from time to time.) But when he asks for the legal power to compel others to do so, in exchange for something he has done or offers to do, he is well into the economics zone. RMS rejects the phrase intellectual property, mostly for reasons of legislative history and philosophy which I consider insightful and with which I agree, but also partly out of a belief that whatever exclusive privileges a creator of knowledge should have over his work should not be codified as property rights. (He also seems to think that they aren't currently codified as property rights, which perplexes me; but that's another line of argument.) But he doesn't believe in laissez faire, either. Now, in his view, there is an a priori ethical imperative to share knowledge, and in the case of software products (which are a sort of distillate of knowledge yet capable of being sold in a form where that knowledge is inaccessible), there is an ethical obligation to disclose the secrets of their making to all who use them. He would like to persuade the world that this ethical imperative should be made law; but failing that, he wants to retain a sort of non-property-based control over the terms on which others use his work (and works whose authors signify their alliance with him by attaching the GPL to them). He demands (or perhaps just assumes) the power to apply his ethical calculus when a work that he controls has played a significant role in the creation of another's work, not just through literal borrowing but by building on the utility of the existing work -- i. e., over any work bearing an economic relationship to his. This brings us to the crux of the matter. RMS seems to think that he has, or should have, this power as a natural right; and I (obviously) don't. I see an author's (or inventor's) rights over his creation as entirely socially created; I think that the law as it stands does a pretty accurate job of capturing that social consensus; and I don't favor attempts at extra-legal end runs around the legislative and judicial process, irrespective of the end in view. There's a philosophical difference here that goes back at least to Hobbes and Hume, if not to Pythagoras and Protagoras. This is perhaps not the forum in which to debate this well-trodden topic further. :-) So does the law in this area follow Hobbes or Hume? Judge for yourself. In any case, I wholly encourage you to deny my selection and exposition of precedents any force of authority. If they are useful to you in making up your own mind, so much to the good; if not, that's fine too. Cheers, - Michael (IANAPhilosopher, either)
Re: LGPL module linked with a GPL lib
On Wed, 2005-07-27 at 14:44 -0700, Michael K. Edwards wrote: How many participants in the KDE/Qt brouhaha actually cited relevant case law? I recall that quite a bit of case law was discussed. Perhaps the debian-legal archives could tell you more. In any case, there's a perfectly good argument that for Debian to piss off the FSF is not a good idea whether or not they have a legal leg to stand on. I personally would be ashamed to lend my good name to their conduct in recent years, but YMMV. In this case, why do you continue to argue with what debian-legal thinks is the prudent course of action for Debian to take, especially when you admit that you may not agree with Debian's goals? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On Sat, 2005-07-30 at 03:55 -0700, Michael K. Edwards wrote: Let me try again. Eben Moglen has a J. D. from Yale. He has been admitted to the bar in New York and before the Supreme Court. He has clerked in district court and for Justice Thurgood Marshall. He has held a professorship of law and legal history at Columbia for over a decade. He is not ignorant of the law. It is my opinion that he knows damn well that there is no such thing as copyright-based license and never has been. It's very useful as a propaganda device to make it appear that there is some rich vein of unmined law in this area, and therefore some difficulty in applying the mountain of case law relevant to any given fact pattern involving the GPL. But the truth as I see it (and I am not alone) is that the GPL is a somewhat unconventionally drafted but otherwise completely routine contract of adhesion. If this is in fact the truth, then many of the things that he, and other attorneys closely associated with the FSF, say in public about the GPL are untrue, perhaps even deliberately misleading. That doesn't inspire my respect. Is that a bald enough statement for you? It is. And, from my perspective, it completely destroys your credibility. It could be the case that everyone who disagrees with you whom you think should know better has ulterior motives. However, I think you need to consider the possibility that you simply do not understand the subject matter as well as you think you do. That you stoop to character assassination as a defense suggests that you are incapable of holding such a low opinion of yourself. I suggest that you may need to find yourself a more credible champion for your position if you want us to accept it. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On 8/3/05, Jeff Licquia [EMAIL PROTECTED] wrote: On Sat, 2005-07-30 at 03:55 -0700, Michael K. Edwards wrote: Let me try again. Eben Moglen has a J. D. from Yale. It is. And, from my perspective, it completely destroys your credibility. What makes your opinion more credible than that of Eben Moglen? Or am I missing something here? -- Chris With the way things are starting to go in this country, if forced to choose between being caught with a van full of pirated DVDs or heroin you'd actually have to pause and think about it. -- Michael Bell, drunkenblog.com
Re: LGPL module linked with a GPL lib
On 8/3/05, Raul Miller [EMAIL PROTECTED] wrote: On 8/2/05, Michael K. Edwards [EMAIL PROTECTED] wrote: Mostly I care about the freedom to pursue what is for me both an intellectual interest and a trade, on terms which more or less reflect an accurate perspective on the surrounding law and economics. Misrepresentations and charlatanry draw my scrutiny, whether they come from saints or sinners. I have no problem with you pursuing any such intellectual interests. However, this probably isn't the right list for posting random facts. Sigh. Did anyone else have trouble understanding that both an intellectual interest and a trade referred to computer programming? Do you not think these random facts are relevant to assessing whether two prominent individuals' preferences and assertions should have the effect that they currently do on your and my pursuit of that trade, and to whether they can be relied on for a disinterested analysis of the applicable law? Have these questions no relevance to debian-legal? I'd suggest debian-curiosa. Or, if anyone wants to create it: debian-saints. Having intimated that I thought it likely that RMS's role in the FSF had made him a rich man, and having been called on that by Diego Bierrun, I felt obliged to report on what facts I could easily find in the public record -- which, as it turns out, don't lend much support to the idea that he is piling up personal assets. (But I doubt he misses a meal very often; I expect his other material needs are more than met, including an ample supply of computer-related toys; as near as I can tell he has done nothing for any reason other than because he feels like it since 1985 or so, if ever; and I count myself as rich partly for similar reasons, whatever one's net worth may be.) I would not have taken the trouble of that particular inquiry except in response to Diego's accusation of slander, which would not have been accurate anyway but did prompt me to go the extra mile. - Michael
Re: LGPL module linked with a GPL lib
On Wed, 2005-08-03 at 13:11 -0700, Michael K. Edwards wrote: On 8/2/05, Patrick Herzig [EMAIL PROTECTED] wrote: RMS doesn't preach the economic superiority of free software. If you fail to understand even such a well-explained position I wonder what your references to all kinds of precedents and such are worth. You've got a fair point, in that RMS doesn't see his arguments as preaching economic superiority; and certainly many commentators have contrasted RMS's ethical perspective with, say, ESR's economic perspective. I don't entirely agree with the way this contrast is portrayed, and in particular I think the ethical/economic dichotomy is a false one. Ethical/financial, perhaps; but that's a calculus of personal motivations that isn't really all that fruitful to discuss. Implicit in my perspective is the view that ethics is the study of human motivation, and economics is the application of the fruits of this study to the public sphere; finance is just probabilities and algebra. I may not be much in the legal department, but you are now commenting on a field I am trained in. Suffice it to say that you have not thought seriously about the implications of your conflation of ethics and economics--or that if you have, then I want nothing to do with you, and you have no business lecturing this group on any subject. If ethics is allowed to be more than sociology, then RMS's position is quite clear: ethical concerns must have priority over economic ones. Stipulating any particular set of ethical standards, I'd say that's not only a clear position, but a rather uncontroversial one. We take a dim view of killing for profit, for example, even if such a decision adversely impacts the hired gun's ability to make a living. Obviously, the question of software freedom is not on the same level as killing for hire, and there are many disagreements regarding the specifics of the ethical questions and their importance. But you seem intent on shutting down the debate (or, less charitably, trying to regain ground from making a grievous error and being called on it) by simply defining it out of existence. If this is your way of handling inconvenient evidence, then I have even less confidence in your legal analysis. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On 8/3/05, Jeff Licquia [EMAIL PROTECTED] wrote: It could be the case that everyone who disagrees with you whom you think should know better has ulterior motives. However, I think you need to consider the possibility that you simply do not understand the subject matter as well as you think you do. That you stoop to character assassination as a defense suggests that you are incapable of holding such a low opinion of yourself. Can you not tell the difference between character assassination and an inquiry into the impartiality of a person held up as an authority? I may well be wrong about the law, and you are welcome to inquire into my motivations as well -- though I claim no authority, only whatever merit the substance of my arguments may carry. I have corresponded (very briefly, on this list) with Eben Moglen and (at greater length) with the FSF, and asked them what basis they have for their position; their argument is, as near as I can tell, _purely_ derived from his personal authority and public stature. To the extent that anyone's motives can ever be deduced from their conduct, don't you think inquiring into his public conduct is fair game under the circumstances? I suggest that you may need to find yourself a more credible champion for your position if you want us to accept it. If not taking Eben Moglen's word over my best effort at understanding the law destroys my credibility in your eyes -- or if I could only retain credibility with you by dissembling the implications for his integrity if he himself knows better -- then I can hardly expect another champion to succeed where I fail. Accept nothing I say without evaluating its evidentiary basis; then my credibility doesn't enter into it. - Michael
Re: LGPL module linked with a GPL lib
On 8/3/05, Jeff Licquia [EMAIL PROTECTED] wrote: I may not be much in the legal department, but you are now commenting on a field I am trained in. Suffice it to say that you have not thought seriously about the implications of your conflation of ethics and economics--or that if you have, then I want nothing to do with you, and you have no business lecturing this group on any subject. You seem to have completely missed my point. Ethics is not dictated by economics; it's the other way around. What possible use is an economics -- a science of human motivation -- that is not informed by ethics? My personal ethics extend into areas that economics cannot reach; but when making an ethical argument for a public policy, I recognize that it must be supported both by an argument from the public good (utility) and an argument that it is fair to individuals. If ethics is allowed to be more than sociology, then RMS's position is quite clear: ethical concerns must have priority over economic ones. Stipulating any particular set of ethical standards, I'd say that's not only a clear position, but a rather uncontroversial one. We take a dim view of killing for profit, for example, even if such a decision adversely impacts the hired gun's ability to make a living. Wow, that's some straw man. If you think you can make an argument from the public good for legalizing assassination for pay, I'd like to hear it. Obviously, the question of software freedom is not on the same level as killing for hire, and there are many disagreements regarding the specifics of the ethical questions and their importance. But you seem intent on shutting down the debate (or, less charitably, trying to regain ground from making a grievous error and being called on it) by simply defining it out of existence. No, I just explained where I was coming from in characterizing RMS's public posture as preach[ing] the economic superiority of the free software system. How you can call this an attempt to shut down the debate is beyond me. If you think it's a grievous error to use the word economics for an inquiry into the conformability of RMS's expectations about GPL enforcement with the balance of public and private interests embodied in the law, then it is an error from which I have no wish to recover. If this is your way of handling inconvenient evidence, then I have even less confidence in your legal analysis. What inconvenient evidence did you have in mind? - Michael
Re: LGPL module linked with a GPL lib
On 8/3/05, Michael K. Edwards [EMAIL PROTECTED] wrote: You've got a fair point, in that RMS doesn't see his arguments as preaching economic superiority; and certainly many commentators have contrasted RMS's ethical perspective with, say, ESR's economic perspective. I don't entirely agree with the way this contrast is portrayed, and in particular I think the ethical/economic dichotomy is a false one. Ethical/financial, perhaps; but that's a calculus of personal motivations that isn't really all that fruitful to discuss. Implicit in my perspective is the view that ethics is the study of human motivation, and economics is the application of the fruits of this study to the public sphere; finance is just probabilities and algebra. I agree with what you've said here (though I use different definitions for the terms ethics and economics than you). And I pretty much agree with the following paragraphs, until we get to a tangent: RMS rejects the phrase intellectual property, mostly for reasons of legislative history and philosophy which I consider insightful and with which I agree, but also partly out of a belief that whatever exclusive privileges a creator of knowledge should have over his work should not be codified as property rights. (He also seems to think that they aren't currently codified as property rights, which perplexes me; but that's another line of argument.) I think his point is that because of the nature of ideas -- that they don't exist in and of themselves, but are abstracts used to describe communication between people -- that it's impossible to codify property rights protecting them. There will always be cracks in the structure. On the one hand, you have property. Property exists as matter, and we have reason to believe (the laws of thermodynamics) that mass is conserved. Ideas do not exist as matter, and while they can be associated with matter there aren't any conservation laws associated with them. There's physically no way to enforce boundaries around them, because in that sense they do not exist. Copyright laws are written to protect the tangible expression of creative ideas, but if you look closely enough there will always be problems determining what is and is not being protected. These boundaries are fluid, because it's simply not possible to survey them or map them -- they have no locations. Now, granted, these laws are sometimes enforced (when there's enough money involved). But, for example, RMS has made it a practice to deliberately avoid dealing with anything which has even a hint of this enforcement associated with it (for example, consider the emacs / xemacs fork). Some people consider him rather poorly for making these kinds of choices, but his loss of credibility in that sense doesn't seem to have much to do with the stuff you're talking about. Well, except that you're indicating that people seriously think of him as a saint. But... there's also plenty of people who think of him as something other than a saint. Personally, I just don't think that issues bearing on sainthood are all that interesting. -- Raul
Re: LGPL module linked with a GPL lib
On 8/3/05, Jeff Licquia [EMAIL PROTECTED] wrote: In any case, there's a perfectly good argument that for Debian to piss off the FSF is not a good idea whether or not they have a legal leg to stand on. I personally would be ashamed to lend my good name to their conduct in recent years, but YMMV. In this case, why do you continue to argue with what debian-legal thinks is the prudent course of action for Debian to take, especially when you admit that you may not agree with Debian's goals? Although I think that argument's perfectly good, I think the arguments on the other side are sometimes better. And as I've written elsewhere, the reasons why one says one is making a particular decision can sometimes have bigger legal consequences than the decision itself. Although I have no direct stake in the outcome of these debates (IANADD), I care enough about Debian's well-being to have put rather a lot of time and thought into the matter. Now, where did I say that I don't agree with Debian's goals? I respect Debian's priorities -- Debian's users and Free Software. My own goals are rarely, if ever, in conflict with them. I am largely satisfied with the definition of Free Software given in the DFSG. But I don't believe that _either_ of Debian's priorities is well served by misunderstanding or misrepresenting the applicable law, by citing fear of legal action rather than courtesy to the FSF as a reason for seeking GPL exemptions from upstream, or by hostility to ISVs who are making an effort to play fair. Cheers, - Michael
RE: LGPL module linked with a GPL lib
On Wed, 2005-08-03 at 10:52 -0300, Humberto Massa Guimarães wrote: IMHO its relevance to d-l is that, if such suspicions are indeed founded, the FSF GPL FAQ should not be taken by face value and that Debian should re-evaluate its position about GPL and linking. If you can prove that the FSF is wrong about copyright law at such fundamental levels, I suspect we will need to re-evaluate far more than our policy regarding the GPL and linking.
Re: LGPL module linked with a GPL lib
On Wed, 2005-08-03 at 15:21 -0700, Michael K. Edwards wrote: No, I just explained where I was coming from in characterizing RMS's public posture as preach[ing] the economic superiority of the free software system. How you can call this an attempt to shut down the debate is beyond me. If you think it's a grievous error to use the word economics for an inquiry into the conformability of RMS's expectations about GPL enforcement with the balance of public and private interests embodied in the law, then it is an error from which I have no wish to recover. I consider it a grievous error to claim that RMS preach[es] the economic superiority of the free software system. You were not calling for an inquiry of any kind in that statement; you were simply snarking. And you were called out for making an incorrect statement. What I'm curious about now is why you felt the need to blather about the nature of ethics and economics, instead of just letting the stupid comment go, and then get even more defensive when someone points out the absurdity of your blathering. You are, of course, free to refuse to admit error, just as we are free to draw whatever conclusions we might from your refusal. But I'm curious to see how far this rabbit hole goes. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On 8/3/05, Raul Miller [EMAIL PROTECTED] wrote: I think his point is that because of the nature of ideas -- that they don't exist in and of themselves, but are abstracts used to describe communication between people -- that it's impossible to codify property rights protecting them. There will always be cracks in the structure. As I understand it, no one in recent centuries has tried to codify property rights protecting ideas as such; neither copyright nor patent nor trademark law presents any barrier to the free dissemination and use of ideas, and trade secret law creates no property right and governs nothing but the breach of duties voluntarily accepted with regard to unpublished information. Imperfect but workable property rights have been created for original expression (at a rather literal level), industrial applications of an invention, and symbols of authenticity; how are these any less amenable to codification than the ownership of a parking lot, a skyscraper, or anything else for which a valuation must be based on the income that may be derived from controlling its use rather than the incremental cost of using it? On the one hand, you have property. Property exists as matter, and we have reason to believe (the laws of thermodynamics) that mass is conserved. A common misapprehension. Chattels are matter. Property is a legal right, i. e., a social convention. I'm not trying to go toe-to-toe with you in the pedantry stakes; but your analogy is faulty right from the outset. The whole _point_ of making copyright, patent, and trademark forms of property is that they become subject to the great body of law that governs legal property rights. They can be sold or transferred without running afoul of continuing performance. They can be used as security for a loan. A non-exclusive right to use them can be offered for a fee or bartered in exchange for a similar right. The owner's exclusive rights may be limited for the sake of the public good, by analogy with public right-of-way and environmental protection laws. [snip] Now, granted, these laws are sometimes enforced (when there's enough money involved). But, for example, RMS has made it a practice to deliberately avoid dealing with anything which has even a hint of this enforcement associated with it (for example, consider the emacs / xemacs fork). I have no idea what you are trying to prove by this example, but you certainly pique my curiosity. What law could RMS have sought to enforce on anyone involved in Lucid or XEmacs? Some people consider him rather poorly for making these kinds of choices, but his loss of credibility in that sense doesn't seem to have much to do with the stuff you're talking about. Actually, the whole Lucid thing excites my sympathy for RMS rather more than the converse. Whatever the facts of the matter may have been, some of the messages in Jamie Zawinski's archive (notably Richard Gabriel's initial public sally) strike me as unnecessarily unkind. In his position I would probably have handled it less gracefully, leaving no one better off. How can I not admire a man whose response to a hostile takeover of his pride and joy begins: quote The long delay in releasing Emacs 19 is the FSF's fault. (In some sense, therefore, mine.) While it's regrettable that there are multiple versions, I can't blame people for filling the gap that the FSF left. One of the goals of the copyleft is to allow people to do this--so that one central maintainer's lapse does not hold back the rest of the community. /quote Well, except that you're indicating that people seriously think of him as a saint. But... there's also plenty of people who think of him as something other than a saint. Personally, I just don't think that issues bearing on sainthood are all that interesting. They're a sidetrack to be sure; but kind of an interesting sidetrack. His personal history and philosophy strike me as more reminiscent of Dominic de Guzman or Benedict of Nursia than any modern figure. In any case, I certainly intended no slur on RMS by that, nor on any participant in this discussion. Cheers, - Michael
Re: LGPL module linked with a GPL lib
On Tue, Aug 02, 2005 at 01:40:42PM -0700, Michael K. Edwards wrote: On 8/2/05, Diego Biurrun [EMAIL PROTECTED] wrote: That RMS gets paid for all the speeches he gives would indeed be news. I have first-hand knowledge that he follows invitations to speak about free software when provided free travel and lodging. Do you know the numbers? As I wrote, I don't. For all I know, RMS never solicits or accepts a speaker fee -- although if so he would be quite extraordinary among conference speakers, even among speakers on free software topics. I believe the typical conference speaker's fee in this area is in the $5K-$20K range (compare http://www.speaking.com/speakerindexes/internet.html ), often more for futurists (several $50K and up speakers at http://www.speaking.com/speakerindexes/future.html ) and actual celebrities with drawing power (usually reported simply as rumored six-figure speaker fee). That would again be news to me. I've just given two talks at LinuxTag (the biggest Linux-related event in Europe) and all I got was two nights in a hotel room. That's what all the speakers get, some do get part of or all of their travel expenses covered, but no more than that. There's a lot of money to be made in this area (although it's a pretty hard life if you have close friends and like your home); and if RMS had a way of laundering the money (don't give it to me; but donate to the FSF if you like) so as to appear saintly, he wouldn't be the first. You're again bordering on slander, I'd tread more carefully if I were you. Speaking of (real) saints: Mother Teresa accepted donations directly and passed them on. There is nothing unethical in that. Your claims are slanderous. I would suggest you to research better before making claims with such serious implications. I'm just telling you how it looks to me, and pointing you to where I got what evidence I have so that you can judge for yourself. And I judge your evidence poorly researched. This does not enhance your credibility when you expound at length (and length and length) on legal affairs. Diego -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On 8/3/05, Jeff Licquia [EMAIL PROTECTED] wrote: I consider it a grievous error to claim that RMS preach[es] the economic superiority of the free software system. You were not calling for an inquiry of any kind in that statement; you were simply snarking. And you were called out for making an incorrect statement. I think my statement is correct as it stands; have you any substantive argument against it? RMS seems to think that it would be a superior economic model -- more conducive to the public good and more fair to individual creators -- if copyright were abolished and copyleft enacted in its place. He also seems to think that the creation of a copyleft microcosm, and the prohibition of its use in conjunction with works outside it, is consistent with the economic bargain embodied in current law. I differ on both points -- and if it's snarking to point out that, by relying on his unique public notoriety to put food on the table, he's not exactly practicing what he preaches, then yes, I'm snarking. What I'm curious about now is why you felt the need to blather about the nature of ethics and economics, instead of just letting the stupid comment go, and then get even more defensive when someone points out the absurdity of your blathering. This point is rather central to my rejection of his claim to the ethical high ground. An ethical stance that fails both economic-model (what if this were the social bargain imposed on everyone?) and economic-tactic (is it at least marginally productive, at an acceptable cost to society, for some people to act thus within the existing social bargain?) tests is no basis for an equitable claim as far as I am concerned. There is of course no reason to demand that any particular activity of his, driven by his personal ethics, be marginally productive; but if he wants to ask that society at large recognize and honor his ethical system with a privilege of exclusivity granted to no other, he's going to have to defend it on utility grounds. You are, of course, free to refuse to admit error, just as we are free to draw whatever conclusions we might from your refusal. But I'm curious to see how far this rabbit hole goes. Now who's snarking? Which is fine by me; but I do not yet find you particularly persuasive on the substantive issues. Cheers, - Michael
Re: LGPL module linked with a GPL lib
On 8/3/05, Michael K. Edwards [EMAIL PROTECTED] wrote: On 8/3/05, Raul Miller [EMAIL PROTECTED] wrote: I think his point is that because of the nature of ideas -- that they don't exist in and of themselves, but are abstracts used to describe communication between people -- that it's impossible to codify property rights protecting them. There will always be cracks in the structure. As I understand it, no one in recent centuries has tried to codify property rights protecting ideas as such; neither copyright nor patent nor trademark law presents any barrier to the free dissemination and use of ideas, and trade secret law creates no property right and governs nothing but the breach of duties voluntarily accepted with regard to unpublished information. Imperfect but workable property rights have been created for original expression (at a rather literal level), industrial applications of an invention, and symbols of authenticity; how are these any less amenable to codification than the ownership of a parking lot, a skyscraper, or anything else for which a valuation must be based on the income that may be derived from controlling its use rather than the incremental cost of using it? There's probably a lesson in here somewhere. information is also a term used to describe how people communicate. You try to draw a distinction between ideas and information. And, indeed, they are spelled differently -- they have different physical representations. And yet, both of these terms refer to the same underlying concept, in this context. If we can't even manage this issue in the context of a single paragraph, what hope do we have of codifying protection for newly thought up instances of this issue, in law? Answer: codification is easy -- it's easy to put words down on paper and call them law -- but it's unlikely that this codification will ever be meaningful in a general context. Thus, no one really wants to take copyright issues to court, because fundamentally the laws don't make sense. When taken at face value, the concepts are simple enough, but the protected works are not real property. As you point out, they're not even chattels. (Though I challenge you to show me any cases of real property which does not stake out a physical chunk of the planet. (I'm aware that you can, at least in some cases, move dirt from one location to another, without changing the legal definition of the property boundaries. But my point is: you can determine those property boundaries because of physical properties of matter, such as the fact that mass is conserved, which do not apply in the realm of intellectual property.)) As for xemacs and emacs: RMS has not accepted xemacs code into emacs because the xemacs developers would not, or could not, transfer copyright ownership on that code to the FSF. -- Raul
Re: LGPL module linked with a GPL lib
On 8/3/05, Diego Biurrun [EMAIL PROTECTED] wrote: That would again be news to me. I've just given two talks at LinuxTag (the biggest Linux-related event in Europe) and all I got was two nights in a hotel room. That's what all the speakers get, some do get part of or all of their travel expenses covered, but no more than that. Then my guess would be that LinuxTag can fill the rooms with people interested in Linux-related topics without having to shell out speaker fees. Which doesn't say that some of the speakers, perhaps even yourself, couldn't earn speaker fees elsewhere if they tried. If a public figure as remarkable as RMS does not choose to gather sizable donations to his preferred charity in return for his speaking engagements, then perhaps conference organizers should be prepared for the eccentric behavior that is occasionally reported. There's a lot of money to be made in this area (although it's a pretty hard life if you have close friends and like your home); and if RMS had a way of laundering the money (don't give it to me; but donate to the FSF if you like) so as to appear saintly, he wouldn't be the first. You're again bordering on slander, I'd tread more carefully if I were you. Still far from slander; but I confess that I regretted this immediately, and was relieved to be able to find enough public evidence in a matter of a couple of hours to refute my own speculation, at least as regards the FSF. Treading carefully does not, if I may say so, seem to be _your_ specialty. Speaking of (real) saints: Mother Teresa accepted donations directly and passed them on. There is nothing unethical in that. Not in the least. It would only be (somewhat) unethical if a large fraction of the donations wound up back in one's own pocket; and that doesn't seem to be the case with RMS. Your claims are slanderous. I would suggest you to research better before making claims with such serious implications. I'm just telling you how it looks to me, and pointing you to where I got what evidence I have so that you can judge for yourself. And I judge your evidence poorly researched. This does not enhance your credibility when you expound at length (and length and length) on legal affairs. It's really interesting that people who show no evidence of having invested any effort whatsoever themselves in research of any aspect of this topic are so quick to reject, not only the slightest speculation beyond the proven facts, but any evidence I may have brought to bear on any conclusion distasteful to them. Ignorance is bliss, I guess. - Michael
Re: LGPL module linked with a GPL lib
On 8/3/05, Raul Miller [EMAIL PROTECTED] wrote: There's probably a lesson in here somewhere. information is also a term used to describe how people communicate. Indeed, among other things; and it is a term sufficiently broad and vague as to have very little utility in law. You try to draw a distinction between ideas and information. And, indeed, they are spelled differently -- they have different physical representations. And yet, both of these terms refer to the same underlying concept, in this context. And yet, the law does distinguish quite successfully between ideas and expression; and disputes about that borderline rarely get as far as an appeals court, usually when someone is deliberately pushing that border for economic gain at another's expense. If we can't even manage this issue in the context of a single paragraph, what hope do we have of codifying protection for newly thought up instances of this issue, in law? That would be the reason that the integrity and competence of judges matters. Your Pyrrhonist (I just learned that nifty word) jump from imperfect certainty to quietism does not persuade me. Answer: codification is easy -- it's easy to put words down on paper and call them law -- but it's unlikely that this codification will ever be meaningful in a general context. Thus, no one really wants to take copyright issues to court, because fundamentally the laws don't make sense. When taken at face value, the concepts are simple enough, but the protected works are not real property. As you point out, they're not even chattels. If you think no one takes copyright issues to court, then you have a very different perspective on the literature than I do. Copyright law mostly makes sense to me, and I rarely feel that I would have decided a recent appellate case differently (though I favor Corey Rusk over Rano and am not that fond of Eldred v. Ashcroft). I do not think this can be attributed to a tendency on my part to accept authority. :-) (Though I challenge you to show me any cases of real property which does not stake out a physical chunk of the planet. (I'm aware that you can, at least in some cases, move dirt from one location to another, without changing the legal definition of the property boundaries. But my point is: you can determine those property boundaries because of physical properties of matter, such as the fact that mass is conserved, which do not apply in the realm of intellectual property.)) You certainly have a point that the boundaries of a copyright (or patent or trademark) holder's rights are imperfectly defined; but if you have ever owned rural property you may be aware that the same is often true of land, although survey-grade (centimeter-accurate) GPS helps. :-) Modern copyright is an imperfect system, but it sure beats hell out of what preceded it. As for xemacs and emacs: RMS has not accepted xemacs code into emacs because the xemacs developers would not, or could not, transfer copyright ownership on that code to the FSF. Which has little to do with (US) copyright law, given precedents such as Aalmuhammed v. Lee; but I probably would have done likewise in his position. I have used both within the past couple of months, and each has its strengths; but if I had the skill and the free time to do so, I think I would rather contribute to GNU Emacs, and would cheerfully assign any copyright I might possess in those contributions to the FSF. For the contributors to Lucid Emacs and XEmacs to have refused to do so strikes me as somewhat churlish, and indeed to reflect a delusional attitude about the value of intellectual property as opposed to customers' and collaborators' trust. Cheers, - Michael
Re: LGPL module linked with a GPL lib
On Wed, Jul 27, 2005 at 04:04:34PM -0700, Michael K. Edwards wrote: Although I have no personal knowledge on the financial side, it certainly looks to me like it has made them both rich men. Little snippets in the public record -- Jim Blandy's comment at http://www.jwz.org/doc/lemacs.html about RMS's luxurious pad on the fourth floor of posh NE43, the indications from Moglen's letter to Vidomi and Fluendo's defense of GStreamer that he has ways of extracting revenues from his role in the FSF, back-of-the-envelope calculations involving the typical conference speaker fee -- suggest to me that their tax records would make interesting reading. I could be completely wrong; they could be scraping by on nominal salaries from the FSF and Columbia University, plus a MacArthur fellowship here and there; but it's enough for me to take their bizarre, uncorroborated assertions about copyright law with a grain of salt. That RMS gets paid for all the speeches he gives would indeed be news. I have first-hand knowledge that he follows invitations to speak about free software when provided free travel and lodging. Your claims are slanderous. I would suggest you to research better before making claims with such serious implications. iego -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On 8/2/05, Diego Biurrun [EMAIL PROTECTED] wrote: That RMS gets paid for all the speeches he gives would indeed be news. I have first-hand knowledge that he follows invitations to speak about free software when provided free travel and lodging. Do you know the numbers? As I wrote, I don't. For all I know, RMS never solicits or accepts a speaker fee -- although if so he would be quite extraordinary among conference speakers, even among speakers on free software topics. I believe the typical conference speaker's fee in this area is in the $5K-$20K range (compare http://www.speaking.com/speakerindexes/internet.html ), often more for futurists (several $50K and up speakers at http://www.speaking.com/speakerindexes/future.html ) and actual celebrities with drawing power (usually reported simply as rumored six-figure speaker fee). There's a lot of money to be made in this area (although it's a pretty hard life if you have close friends and like your home); and if RMS had a way of laundering the money (don't give it to me; but donate to the FSF if you like) so as to appear saintly, he wouldn't be the first. Your claims are slanderous. I would suggest you to research better before making claims with such serious implications. I'm just telling you how it looks to me, and pointing you to where I got what evidence I have so that you can judge for yourself. The FSF is notoriously unforthcoming about their financial dealings, and the cash flows involved are not chump change (see the numbers disclosed by Jamie Zawinski in the Lucid Emacs saga). Whether or not you think RMS and Eben Moglen are cashing in personally (about which I have no evidence), if you are willing to take their uncorroborated claims about the legal strategy at the heart of their enterprise at face value, you are a more trusting man than I. Cheers, - Michael
Re: LGPL module linked with a GPL lib
On 8/2/05, Michael K. Edwards [EMAIL PROTECTED] wrote: I'm just telling you how it looks to me, and pointing you to where I got what evidence I have so that you can judge for yourself. The FSF is notoriously unforthcoming about their financial dealings, and the cash flows involved are not chump change (see the numbers disclosed by Jamie Zawinski in the Lucid Emacs saga). Whether or not you think RMS and Eben Moglen are cashing in personally (about which I have no evidence), if you are willing to take their uncorroborated claims about the legal strategy at the heart of their enterprise at face value, you are a more trusting man than I. This sounds like something appropriate for the scandal column of a tabloid. But what's the relevance of this issue to debian-legal? -- Raul
Re: LGPL module linked with a GPL lib
I wrote: There's a lot of money to be made in this area (although it's a pretty hard life if you have close friends and like your home); and if RMS had a way of laundering the money (don't give it to me; but donate to the FSF if you like) so as to appear saintly, he wouldn't be the first. If there's any laundering going on (hmm, I picked an explosive word, didn't I?), that's not how it's done. I took a quick look at the most recent IRS Form 990 I could find for them (fiscal year ending Sep 2003). Apparently no director (those two included) receives any pay, only one person (the Director of Communications) is paid more than $50K, and there isn't room to hide much of a back-door in the numbers the CPA signed off on ($631K expenses excluding COGS, $346K of which is salaries and the rest of which is accounted for to my unqualified-because-I'm-no-accountant-either satisfaction). Without going to the trouble of more than spot-checking prior years' reports, I'm inclined to say that the FSF per se is fiscally clean (insofar as donation is really the right word for some of those non-sale revenues). Sidetrack: It's very interesting to see that donations doubled to almost $400K in the 1998-1999 fiscal year, without cutting into product sales; but then the revenue mix shifted abruptly towards donations in fiscal 2000 -- a one-year spike, perhaps (I'm guessing) composed of corporate donations related to the KDE/GNOME brouhaha. Fiscal 2001 (Oct 2001 - Sep 2002) was a hard year for the FSF, as it was for almost all US non-profits; but unlike many, the FSF seems to have rebounded nicely. Good for them. On the other hand, no speaker fees appear in the FSF's fiscal 2003 revenues (unless they're inside the $584K in donations or $243K in product sales) -- maybe they go through some other non-profit instead -- and any money received directly by Mr. Moglen from the likes of Vidomi and Fluendo in his capacity as an attorney would be off the FSF's books. Something pays Mr. Stallman's rent, and I'd be pretty surprised if a 1990 MacArthur (nominally given in his capacity as President of the League for Programming Freedom) stretches that far. My skepticism remains less than completely cured. I'm not an investigative journalist, I'm not out to do a hatchet job on the FSF or the individuals involved, and I don't really care how they make their money unless there's something genuinely extortionate about it (for which I have no evidence). There are a lot of things about RMS and the FSF that I respect and, with reservations, even admire. But I repeat that, if you want to know what actual law applies to the GPL, uncorroborated assertions by heavily interested parties are probably not a reliable indicator. Cheers, - Michael
Re: LGPL module linked with a GPL lib
On 8/2/05, Raul Miller [EMAIL PROTECTED] wrote: On 8/2/05, Michael K. Edwards [EMAIL PROTECTED] wrote: I'm just telling you how it looks to me, and pointing you to where I got what evidence I have so that you can judge for yourself. The FSF is notoriously unforthcoming about their financial dealings, and the cash flows involved are not chump change (see the numbers disclosed by Jamie Zawinski in the Lucid Emacs saga). Whether or not you think RMS and Eben Moglen are cashing in personally (about which I have no evidence), if you are willing to take their uncorroborated claims about the legal strategy at the heart of their enterprise at face value, you are a more trusting man than I. This sounds like something appropriate for the scandal column of a tabloid. But what's the relevance of this issue to debian-legal? Scandal columns of tabloids don't generally comb the public record for evidence on both sides of an issue, which I am making a sincere effort to do (see the rebuttal I just sent of my own speculations on FSF finances). This private organization and its affiliates are marketing an approach to my career and livelihood, and presumably that of many debian-legal participants, which rests on assertions about the law that my (unqualified but dogged) research says are false. Millions of dollars have passed through the accounts of entities that RMS and Eben Moglen control over the past 15 years -- entities whose entire justification for existence is the promotion of their views on these topics. Debian-legal has generally taken these two people's views at face value, without asking for their basis -- with some pretty serious consequences for projects like KDE and OpenSSL, and for people who might want to bring ISVs onto Debian and Debian-derived platforms. That is of course entirely within their rights; Debian may be content to toe the FSF line with respect to linking relationships whether or not it has any basis in law. But I don't seem to be the only one around here interested in the question of what might underly that agenda and what risks it might expose Debian to down the line. So yes, inquiring minds want to know. - Michael
Re: LGPL module linked with a GPL lib
I wrote: So yes, inquiring minds want to know. And this inquiring mind is now satisfied as to what probably pays RMS's rent lately -- the ~$268K Takeda Award he received in 2001. (You couldn't keep a family in Cambridge for four years on that, but RMS doesn't have that problem.) Me, I'd be kind of ashamed to preach the economic superiority of the free software system while living on grant money and conference banquets; but YMMV. Again, I don't really care how RMS makes his money; I do care, a little, how Eben Moglen makes his, but only because he uses his status as a law professor at a respected university to bolster claims I find incredible. Mostly I care about the freedom to pursue what is for me both an intellectual interest and a trade, on terms which more or less reflect an accurate perspective on the surrounding law and economics. Misrepresentations and charlatanry draw my scrutiny, whether they come from saints or sinners. ObOnTopicForDebianLegal: This thread was originally about GStreamer. The FAQ at http://gstreamer.freedesktop.org/data/doc/gstreamer/head/faq/html/chapter-legal.html reflects the public posture of that project's principal contributors (and principal corporate funder, Fluendo) on the matter, and claims to be certified by the FSF lawyer team and verified by FSF lawyer and law professor Eben Moglen. Whether or not it accurately reflects the relevant law, I would expect it to form a pretty strong basis for estopping the people who have published it from making conflicting arguments in court. IANAL, TINLA. Cheers, - Michael
Re: LGPL module linked with a GPL lib
I wrote: The contributory / direct infringement difference is kind of interesting from a tactical point of view ... followed by some discussion about the Micro Star opinion that was more inarticulate than usual. The point I was trying to make, in sentences of 25 words or less: The important part of the Micro Star opinion is the exposition of why MAP files are morally equivalent to unauthorized sequels to Duke Nukem. The text of the opinion is complicated by the need to distinguish it on the facts from Galoob. The author in Galoob, in turn, felt the need to distinguish it on the facts from the Seventh Circuit's decision in Midway v. Artic. The result is a thicket of ill-conceived case law harping on technical criteria for whether or not a derivative work has been created. IMHO, was a derivative work created? is the wrong question. The right question is, has the alleged infringer used the original's creative expression in a way that Congress intended to reserve to the copyright holder? Piecewise recycling of existing works, whether by the original author or by enterprising aftermarket tweakers, is becoming more and more common. So perhaps courts should ask, does this boost usability of the existing work without really interfering with the potential for sequels? Or does it just rip the original off as a substitute for going to the trouble of writing new stuff? From this perspective, a speeded-up Galaxian (Midway) is kind of borderline, like a double-time punk cover of a 60's chestnut. Helping Nintendo players cheat (Galoob) has no real effect on the creative expression, but publishing new game levels (Micro Star) does. Artificial criteria about whether an altered display is sufficiently fixed, or whether the alleged infringement is direct or contributory, just get in the way. Maybe that's an outline of another law journal article. :-) Cheers, - Michael (IANAL, TINLA)
Re: LGPL module linked with a GPL lib
On Sat, Jul 30, 2005 at 11:30:15AM -0700, Michael K. Edwards wrote: On 7/30/05, Francesco Poli [EMAIL PROTECTED] wrote: Well, let's say Almost All Rights Reserved. Anyway, it's still really far away from a DFSG-free document: that's basically what I meant... Oddly enough, there is some useful knowledge out there that is not currently available in a DFSG-free document. :-) If I get to the point of being confident that it is a wise thing to do, I may publish some of this analysis in a more convenient form than you already have it (almost all of it is in one message or another in the debian-legal archives), and that publication may come with somewhat more liberal explicit terms about piecewise re-use (few, if any, of my d-l posts come with any more permissions than may be intrinsic to their having been posted to a publicly archived list). I doubt I shall ever attach a DFSG-free license to this work in monograph form; it's not a computer program or the documentation for a computer program, and it's not really something people should be making bug fixes to and re-publishing without a quite strong presumption that they take all of the risk for having changed its meaning or made it look as if the author offered it as authoritative. http://creativecommons.org/license/ You should find a license that fits your needs there. Diego -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On 7/30/05, Michael K. Edwards [EMAIL PROTECTED] wrote: On 7/30/05, Raul Miller [EMAIL PROTECTED] wrote: I count four issues the judge considered, with a bit of detail on each of those issues. I didn't say six issues. I said six reasons why it would be inappropriate to grant a preliminary injunction for breach of the GPL terms, any one of which would have been sufficient. Ok, my mistake. That said:given the nature of common law, precedent, and how narrowly judges like to form their opinions, I believe the intent of the judge was that the case be valid as common law where all six of these reasons are valid, and that any of these reasons taken individually would not be sufficient in and of itself. Eh... it was explicit in the decision that the copyright breach had largely been repaired. So, yeah, in contexts where the breach is repaired, the automatic termination clause can get ignored. There's no such thing as a copyright breach and you can't repair copyright infringement. Sure you can: through contract with the infringed party. If the infringed party does not agree to such a contract, that's when it can't be repaired. Oh, they may well know the truth (as I understand it); but if so, they do not choose to advertise their knowledge. Alternately, if they have a foundation for their claims that my amateur archaeology has been unable to unearth, they seem to have no desire to expose it to public scrutiny. Why should they, when widespread belief in the accuracy of their interpretation gets the job done anyway? As usual, you're implying a lot with what you say, but actually saying very little. Let me try again. Eben Moglen has a J. D. from Yale. He has been admitted to the bar in New York and before the Supreme Court. He has clerked in district court and for Justice Thurgood Marshall. He has held a professorship of law and legal history at Columbia for over a decade. He is not ignorant of the law. This mini-biography seems easy enough to understand. It is my opinion that he knows damn well that there is no such thing as copyright-based license and never has been. This, on the other hand seems rather silly. It's very useful as a propaganda device to make it appear that there is some rich vein of unmined law in this area, and therefore some difficulty in applying the mountain of case law relevant to any given fact pattern involving the GPL. Mountain of case law? There's very little case law with respect to copyrights and computers. Computers operate by making numerous copies of whatever it is that they're processing. The U.S. has some statutory law which addresses what copyright means in a few common cases, and a few copyright cases have appeared in court where computers have been central, but for the most part case law has not come to grips with this issue. I think you're projecting a shadow of this conflict onto the GPL. But this conflict between the way copyright is designed (to protect the tangible expression of ideas from unauthorized copying, with all works receiving some copyright protection by default) and the way computers operate (where copying is required as a fundamental part of any operation) is inherent in copyright law itself. In the U.S., copyright law is explicitly constrained To promote the Progress of Science and useful Arts, and on its face copyright could be construed to prevent any innovation involving computers. That's not the GPL makes some rich unmined vein of law appear for mining except in the sense that the GPL promotes the creation of useful software. But the truth as I see it (and I am not alone) is that the GPL is a somewhat unconventionally drafted but otherwise completely routine contract of adhesion. This is simply false. There are numerous cases where modified forms of the GPL are in use and/or other alternate terms are in use. If you don't mind, I'll cite a few examples: * gcc * linux kernel * perl * ghostscript or, quoting the fsf faq on the gpl: http://www.fsf.org/licensing/licenses/gpl-faq.html#WillYouMakeAnException We do occasionally make license exceptions to assist a project which is producing free software under a license other than the GPL. However, we have to see a good reason why this will advance the cause of free software. More generally, the fact that a license is widely used does NOT make it a contract of adhesion. If this is in fact the truth, then many of the things that he, and other attorneys closely associated with the FSF, say in public about the GPL are untrue, perhaps even deliberately misleading. That doesn't inspire my respect. Is that a bald enough statement for you? Yes, thank you. I believe your opinion to be factually incorrect, but that statement was indeed bald enough for me. Thanks again, -- Raul
Re: LGPL module linked with a GPL lib
On Sun, 31 Jul 2005 14:11:15 +0200 Diego Biurrun wrote: On Sat, Jul 30, 2005 at 11:30:15AM -0700, Michael K. Edwards wrote: On 7/30/05, Francesco Poli [EMAIL PROTECTED] wrote: Well, let's say Almost All Rights Reserved. Anyway, it's still really far away from a DFSG-free document: that's basically what I meant... [...] I doubt I shall ever attach a DFSG-free license to this work in monograph form; it's not a computer program or the documentation for a computer program, and it's not really something people should be making bug fixes to and re-publishing without a quite strong presumption that they take all of the risk for having changed its meaning or made it look as if the author offered it as authoritative. http://creativecommons.org/license/ You should find a license that fits your needs there. Maybe a license that fits Michael's needs, but definitely *not* a DFSG-free one: unfortunately, at the moment, there are no CC licenses that comply with the DFSG... -- :-( This Universe is buggy! Where's the Creator's BTS? ;-) .. Francesco Poli GnuPG Key ID = DD6DFCF4 Key fingerprint = C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4 pgp9vUY5G2wz4.pgp Description: PGP signature
Re: LGPL module linked with a GPL lib
On 7/31/05, Francesco Poli [EMAIL PROTECTED] wrote: Maybe a license that fits Michael's needs, but definitely *not* a DFSG-free one: unfortunately, at the moment, there are no CC licenses that comply with the DFSG... I do not, at present, need a license at all; I am perfectly content with the legal consequences of saying it's unpublished, but feel free to pass a few copies around privately. If there's one thing I'm very _un_likely to do, it's use a license drafted by some random people whose agenda I don't know to say something that can be said in a dozen words. Cheers, - Michael
Re: LGPL module linked with a GPL lib
On 7/28/05, Humberto Massa Guimarães [EMAIL PROTECTED] wrote: Strawman? Fact: the creation of a derivative work is the application of some transformation on the original work. The above snippet (which isn't even copyrightable, for its sheer size and the necessity of expressing the same thing in the same language) is NOT a derivative work of ANY libc implementation, ... It doesn't make sense to extrapolate this is what copyright doesn't protect when you compile uncopyrightable code to cases where you compile copyrightable code. More generally, however, the court will take a number of factors into consideration in a real case, and the hello world program doesn't really represent any of the relevant issues. -- Raul
Re: LGPL module linked with a GPL lib
GPL violators appear to face several potential penalties: On 7/28/05, Michael Poole [EMAIL PROTECTED] wrote: Potential penalties are irrelevant to my question. You assume a priori that such linking is a violation of the GPL. My question was why that assumption is valid. As I explained above, his citation of case law does not fit the facts. Ok, granted -- I didn't really answer your question. You seem to be asking what is it about copyright law that means that the creative efforts represented in a program are protected as opposed to the mechanical trivia. But, ok, you didn't really ask that question either. What you literally asked was What statute or case law supports this position? There isn't much case law covering copyright in the context of computers. And there's not muce statutory law, either. For an interesting exercise, try to determine why it could be legal to view a plain text document (let's assume it's not a plain text document which represents a program) on a computer (given that viewing that document requires the reader make copies of it. That said, copyright law has a very strong element of protecting creative expression and practically no element of protecting technical processes. -- Raul
Re: LGPL module linked with a GPL lib
On 7/28/05, Humberto Massa Guimarães [EMAIL PROTECTED] wrote: The only good answer people in d-l gave me to the question: why is the assumption that such linking is a violation of the GPL valid? is because Eben Moglen said so in the GPL FAQ, and he is a law teacher, so it must be true. If you think that, you haven't been paying attention. The issue is: is a dynamically linked program a creative work? And, factually, you can't make a dynamically linked program without first designing it so that it works (and, in the typical case, spending a fair amount of time polishing it -- debugging, cleaning up the interfaces, etc.) Since this is a factual issue, its resolution is going to require consideration of the facts of the case, and could also require expert testimony about how the particular program in question is normally handled. In other words, you can't generalize from how game cartridges and laser printer cartridges are handled to some blanket rule which covers all cases of dynamic linking. (For instance.) -- Raul
Re: LGPL module linked with a GPL lib
On 7/28/05, Michael K. Edwards [EMAIL PROTECTED] wrote: On 7/28/05, Raul Miller [EMAIL PROTECTED] wrote: I don't think that the point is that people would be going to jail for violating the GPL. Violating the GPL doesn't mean anything. Yes it does -- it means actions in the context of a GPLed program which copyright law doesn't allow and which the GPL license does not allow. GPL violators appear to face several potential penalties: (*) Works they hold copyright on might be released under the terms of the GPL when they thought they could get away with not doing that. This is extraordinarily unlikely except as part of a consent decree, agreed on by the parties and ratified by the court as an alternative to proceeding to judgment. Granted -- copyright cases very rarely are resolved in court, so any court action other than preliminary injunction is extraordinarily unlikely. Nevertheless, this was exactly the outcome of the one case where the GPL was tested in court (with Progress consenting to release their code under an appropriate license), and this has also been the outcome of a number of potential cases which never made it to court. So it may be extraordinarily unlikely except ... but it's still a fairly likely outcome. (*) They might be forced to pay someone else to release their work under GPL terms. This is even more unlikely; Granted. (*) They might be forced to stop distribution of some work where they don't have the proper rights available to them. In other words, injunctive relief -- often granted temporarily pending trial at the preliminary hearing stage, and if granted often followed by out-of-court settlement. As I have explained ad nauseam, I think it very improbable (IANAL, TINLA) that preliminary injunctive relief could be obtained in a linking scenario involving the GPL even under the most egregious of fact patterns; compare Progress Software v. MySQL. I think I made it clear that I think this would not be the first choice of the court -- only in the case where someone was clearly being malicious and refusing to take the easier options would this be granted. Whether that would happen in a preliminary injunction, or not, depends on the particulars of the hypothetical case. Obviously there would be cases where one or more of these would not apply, but if none of these apply that's probably because they're not violating the terms of the GPL. AIUI most copyright infringement suits end with monetary damages, sometimes with attorneys' fees and costs attached, sometimes with impoundment and an added injunction against continued infringement (which makes it quite easy to come back for more penalties if the infringer doesn't cease). But IMHO a competently defended GPL violation suit, if its facts reflect reasonable disagreement about the interpretation of a contract of adhesion, is unlikely to result in any penalty other than the stipulation of future conduct consistent with the contract terms as construed by the court. I think we're loosely agreeing on this point. -- Raul
Re: LGPL module linked with a GPL lib
On 7/28/05, Michael K. Edwards [EMAIL PROTECTED] wrote: On 7/28/05, Raul Miller [EMAIL PROTECTED] wrote: For example, take Progress v. MySql -- here, the stop distribution penalty was not used in part because Progress didn't have anything else -- it would have been destroyed by this penalty. And, Progress had agreed in court to release their software under other terms. This is purely a practical issue (as you'd expect from a preliminary injunction). You still don't seem to understand what happened in that case. The judge cited six reasons why it would be inappropriate to grant a preliminary injunction for breach of the GPL terms, any one of which would have been sufficient. I count four issues the judge considered, with a bit of detail on each of those issues. 1) Likelihood of success: Here, the judge discussed the dynamic linking issue, said that MySQL would probably win, but that the issue was in dispute. 2) Potential for irreparable harm: [This is what i was talking about.] The judge listed a number of particulars about why irreparable harm was not likely. Of particular importance were sworn statements made by Progress that the source code had been released and that the conflicting license requirements were being withdrawn. (This is what I was talking about, above -- the issue was basically resolved at the time the judge wrote that order.) Perhaps, when you say there were six reasons, you've broken this issue down into several pieces? 3) balancing of relevant equities: This favored Progress as suspending distribution of Gemini would have destroyed them. 4) the effect on the public interest: since the copyright issue was basically resolved, this didn't get much discussion. And it was implicit in the structure of the decision that she rejected both the notion that the GPL is a creature of copyright law and the plea in Mr. Moglen's affidavit that the automatic termination clause was central to its enforcement. Eh... it was explicit in the decision that the copyright breach had largely been repaired. So, yeah, in contexts where the breach is repaired, the automatic termination clause can get ignored. Had she taken either of these propositions even a little bit seriously, she would (as she well knows) have been obliged to analyze the request for preliminary injunction according to a completely different standard. This would be rather pointless given that Progress seemed to be committed to repairing the breech. What's amazing are your repeated claims that the FSF doesn't know what it's talking about, legally speaking. Oh, they may well know the truth (as I understand it); but if so, they do not choose to advertise their knowledge. Alternately, if they have a foundation for their claims that my amateur archaeology has been unable to unearth, they seem to have no desire to expose it to public scrutiny. Why should they, when widespread belief in the accuracy of their interpretation gets the job done anyway? As usual, you're implying a lot with what you say, but actually saying very little. -- Raul
Re: LGPL module linked with a GPL lib
On 7/30/05, Raul Miller [EMAIL PROTECTED] wrote: I count four issues the judge considered, with a bit of detail on each of those issues. I didn't say six issues. I said six reasons why it would be inappropriate to grant a preliminary injunction for breach of the GPL terms, any one of which would have been sufficient. Let's walk through them together. 1) Likelihood of success: Here, the judge discussed the dynamic linking issue, said that MySQL would probably win, but that the issue was in dispute. That's not what the opinion said. The two sentences are: Affidavits submitted by the parties' experts raise a factual dispute concerning whether the Gemini program is a derivative or an independent and separate work under GPL para. 2. After hearing, MySQL seems to have the better argument here, but the matter is one of fair dispute. And as such not suitable for judgment as a matter of law. Note that the facts of the case involved: - an essentially _undocumented_ _internal_ interface to which no other outside party had AFAIK ever implemented; - deceit on Progress's part about whether they included the full source of mysqld as distributed on their release CDs (they didn't), and deliberate use of _static_ linking and failure to provide even a binary version of the Gemini table type as .o or .a, leaving customers unable to build a bug-fixed mysqld without losing the table type; - large cash payments from Progress to TCX DataKonsult (predecessor to MySQL AB) which were reported at the time as financing TCX's move to the GPL -- and hence extensive negotiations between the parties about what conduct they both understood the GPL to demand of Progress. Despite all of which, the matter remained one of fair dispute that could not have been resolved at the summary judgment stage even if MySQL's case had been otherwise flawless. So that's blocker #1. #2 was that, even if the evidence had originally been adequate to support the likelihood of success on the merits prong, the judge was not persuaded based on this record that the release of the Gemini source code in July 2001 didn't cure the breach. So much for the automatic termination clause. 2) Potential for irreparable harm: [This is what i was talking about.] The judge listed a number of particulars about why irreparable harm was not likely. Blocker #3: In any event, even if MySQL has shown a likelihood of success on these points, it has not demonstrated that it will suffer any irreparable harm during the pendency of the suit ... The judge is not applying a copyright law standard here, which would have given MySQL an automatic presumption of irreparable harm that it would be up to Progress to rebut. As well she shouldn't have; as there were no sufficient grounds for rescinding the contract, no claim of copyright infringement could succeed. And MySQL was unable to make a showing on the facts that they would suffer _any_ harm during the pendency of the suit, let alone harm that could not be repaired through monetary damages after the fact. Of particular importance were sworn statements made by Progress that the source code had been released and that the conflicting license requirements were being withdrawn. (This is what I was talking about, above -- the issue was basically resolved at the time the judge wrote that order.) ... particularly in light of the sworn statement that all source code for Gemini has been disclosed and the stipulation, given by Progress during the hearing, that the end use license for commercial users will be withdrawn. That would be blockers #4 and #5; either of these voluntary stipulations (nothing to do with a consent decree) would be enough to void the potential for irreparable harm prong of the preliminary injunction test, since proving harm would require both a means by which Progress could extract customer revenues that should have gone to MySQL (difficult if MySQL, along with the rest of the world, also has Progress's source code) and an intention of continuing to do so (difficult if Progress abandons the practice of charging for end-user licenses). Perhaps, when you say there were six reasons, you've broken this issue down into several pieces? 3) balancing of relevant equities: This favored Progress as suspending distribution of Gemini would have destroyed them. Blocker #6: Finally, because the product line using MySQL is a significant portion of NuSphere's business, Progress has demonstrated that the balance of harms tips in its favor regarding the use of the MySQL program under the GPL. This is the third prong of the preliminary injunction test, and MySQL failed this one too. Even if they had demonstrated a likelihood of success on the merits and a probability of irreparable harm if Progress were not immediately enjoined, Progress's position was such that an injunction would do them greater irreparable harm -- and in a contract (not tort) action, that means no injunction. This, too,
Re: LGPL module linked with a GPL lib
On Fri, 29 Jul 2005, Michael K. Edwards wrote: If the GPL lets the user do it, it isn't infringement at all. You can't have contributory infringement if there's no infringement. The GPL is not a new copyright statute with the power to override the meaning of infringement, nor do its drafting oddities render it null. ...you may perhaps say that the GPL explicitly allows the end user to link the software in private. But that is merely a basis for arguing that the copyright holder is estopped from suing end users who are simply using what they have received in good faith. That does not mean that a distributor could not be successfully sued for copyright infringement _if_ it were correct that the act of linking breached rights reserved to the copyright holder. By this reasoning, if linking is normally a breach of rights, I could give you some BSD licensed software and do exactly the same thing. I am estopped from suing you for linking with my BSD software, but I can still prevent other people from helping you link with it, since the linking is still infringement despte the license telling you you can do it. If I give you some software and say that you can link it in private, that's *permission*. The GPL doesn't need to override the meaning of infringement, because it already has a meaning, and that meaning already says that if you have permission to do it, it isn't infringement. If it isn't infringement, helping you do it can't be contributory infingement. Compare, for instance, Micro Star v. FormGen. You could argue that the unauthorized sequel didn't really exist until an end user loaded the MAP file into Duke Nukem, and Micro Star neither authored the MAP files nor distributed them together with Duke Nukem. Oh, come on. FormGen claimed that the MAP files themselves were infringements, because they used the game setting. It didn't matter that the user linked them, because they were unauthorized sequels all by themselves. This was one of the distinctions the court made between it and the Game Genie case, because in their view the Game Genie only lets consumers create derivatives, but the MAP files are derivatives. # More significantly, Nintendo alleged only contributory infringement-- # that Galoob was helping consumers create derivative works; FormGen here # alleges direct infringement by Micro Star, because the MAP files encompass # new Duke stories, which are themselves derivative works. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Code of conduct and MUAs [was: Re: LGPL module linked with a GPL lib]
On Fri, 29 Jul 2005 13:07:00 -0700 Michael K. Edwards wrote: On 7/29/05, Francesco Poli [EMAIL PROTECTED] wrote: [...] P.S.: please do not reply to me directly, while Cc:ing the list, as I didn't ask you to do so... since I'm a debian-legal subscriber, I'd rather not receive messages twice! thanks I went back and checked the code of conduct and you are of course right that I am in error. It's too bad that the code of conduct plus the remailer configuration are inverted relative to the way that MUAs work. Huh? What do you mean by remailer? I assume you are not talking about anonymous remailers... :-? The MUA I use (Sylpheed) seems to automatically reply to the list, unless (as in your case!) a Reply-To: field is set differently. Unfortunately, Sylpheed does not honour the Mail-Followup-To: field, so I have to manage it manually (grin!). Adding a Followup-To: d-l is supposed to be a no-op by the code of conduct, and would make almost all MUAs do the right thing; Are you sure there are *so many* MUAs that honour the Mail-Followup-To: header? Is Sylpheed one of the *few* that don't? I'm told Mutt does the Right Thing(TM) with M-F-T: headers... Which other MUAs behave similarly? but the remailer doesn't do it automatically. And even if I check the headers manually I'm not supposed to assume that a Reply-To: (real mail address) was intentional. I believe my messages come with no Reply-To: field... So I'm supposed to strip the sender from the reply list no matter what Followup-To or Reply-To says, unless I'm not subscribed, please copy me is in the body of the message. I guess I'll do that henceforth, but it really feels broken. IIRC, the code of conduct says that the canonical way to ask to be Cc:ed on replies is setting an appropriate Mail-Followup-To: field. Asking the same in the message body (in natural language) is a useful reminder for users of MUAs that do not automatically honour the Mail-Followup-To: field. Am I correct? -- :-( This Universe is buggy! Where's the Creator's BTS? ;-) .. Francesco Poli GnuPG Key ID = DD6DFCF4 Key fingerprint = C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4 pgplta2556AbI.pgp Description: PGP signature
Re: LGPL module linked with a GPL lib
On Fri, 29 Jul 2005 13:07:00 -0700 Michael K. Edwards wrote: On 7/29/05, Francesco Poli [EMAIL PROTECTED] wrote: In other words, All Rights Reserved... :-( I did say that I would be happy to give you advance permission to circulate a reasonable number of copies privately, which would leave me with no recourse against you unless you set out to misappropriate or grossly misrepresent my work or to defeat its first publication. That's rather different from all rights reserved, and it's the most open I know how to be without seriously compromising its status as an unpublished work still in draft. Well, let's say Almost All Rights Reserved. Anyway, it's still really far away from a DFSG-free document: that's basically what I meant... -- :-( This Universe is buggy! Where's the Creator's BTS? ;-) .. Francesco Poli GnuPG Key ID = DD6DFCF4 Key fingerprint = C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4 pgpTWnDi5HYL3.pgp Description: PGP signature
Re: LGPL module linked with a GPL lib
On 7/30/05, Francesco Poli [EMAIL PROTECTED] wrote: Well, let's say Almost All Rights Reserved. Anyway, it's still really far away from a DFSG-free document: that's basically what I meant... Oddly enough, there is some useful knowledge out there that is not currently available in a DFSG-free document. :-) If I get to the point of being confident that it is a wise thing to do, I may publish some of this analysis in a more convenient form than you already have it (almost all of it is in one message or another in the debian-legal archives), and that publication may come with somewhat more liberal explicit terms about piecewise re-use (few, if any, of my d-l posts come with any more permissions than may be intrinsic to their having been posted to a publicly archived list). I doubt I shall ever attach a DFSG-free license to this work in monograph form; it's not a computer program or the documentation for a computer program, and it's not really something people should be making bug fixes to and re-publishing without a quite strong presumption that they take all of the risk for having changed its meaning or made it look as if the author offered it as authoritative. Cheers, - Michael
Re: LGPL module linked with a GPL lib
On 7/30/05, Ken Arromdee [EMAIL PROTECTED] wrote: By this reasoning, if linking is normally a breach of rights, I could give you some BSD licensed software and do exactly the same thing. I am estopped from suing you for linking with my BSD software, but I can still prevent other people from helping you link with it, since the linking is still infringement despte the license telling you you can do it. But the GPL doesn't tell you in so many words that end users have a license to link. As I wrote, to get to end-user safety without really looking at the interactions between linking and the GPL language, you need to combine a statutory provision that narrows the rights reserved to copyright holders (17 USC 117) with a negative in the GPL (suggesting that, if things that happen during end use don't qualify as modification in a copyright sense, then they're entirely outside the GPL terms). In this view, the deliberate excision of right to use in the GPL v2 (apparently for the sake of bolstering the copyright-based license fiction) leaves end users not with a _license_ -- that's a term in a contract -- but with an _estoppel_ argument against attempts to sue them for doing things that the copyright holder implied they were permitted to do incident to normal use. (End users could also argue, entirely on statutory grounds and without reference to the GPL text at all, that they became the owner of a copy under the Section 109 standard when they obtained it in good faith from a presumptively licensed distributor, and proceed directly to their Section 117 rights without need for estoppel. But this cannot of course be used circularly to protect the distributor.) The BSD and MIT X11 licenses, as I read them, are pure creatures of estoppel and not copyright licenses (in the sense of binding contract terms) at all. Their unilateral grant is so broad as to protect both end use and distribution; but as no one has accepted any significant obligation of return performance, only a reliance to one's own detriment argument supports a claim that the copyright holder can't abrogate that grant at any moment. If I give you some software and say that you can link it in private, that's *permission*. The GPL doesn't need to override the meaning of infringement, because it already has a meaning, and that meaning already says that if you have permission to do it, it isn't infringement. If it isn't infringement, helping you do it can't be contributory infingement. As I wrote, a copyright infringement claim could only be brought home against the distributor _if_ it were correct that the act of linking breached rights reserved to the copyright holder. Suppose the end user's actions are in principle infringing but the copyright holder's statements have led to a sort of half-assed estoppel-based substitute for a license. Then there is no guarantee that someone who _is_ unambiguously engaging in copying and modification (and therefore needs the license offered in the GPL) can rely on the end users' de facto immunity from suit as a defense against claims that end users are being encouraged to bend the rules. That wouldn't really be a copyright infringement claim, though; it would be a claim of breach of the duty to deal in good faith implicit in any contractual relationship, argued as grounds either for recovery of damages under the contract or for rescission of the contract so that a copyright infringement claim can proceed. Compare, for instance, Micro Star v. FormGen. You could argue that the unauthorized sequel didn't really exist until an end user loaded the MAP file into Duke Nukem, and Micro Star neither authored the MAP files nor distributed them together with Duke Nukem. Oh, come on. FormGen claimed that the MAP files themselves were infringements, because they used the game setting. It didn't matter that the user linked them, because they were unauthorized sequels all by themselves. This was one of the distinctions the court made between it and the Game Genie case, because in their view the Game Genie only lets consumers create derivatives, but the MAP files are derivatives. The Game Genie case did _not_ conclude that consumers were creating derivatives; as the Micro Star opinion comments, the discussion of the fair use defense in Galoob was dicta based on a hypothetical. As for the contention that MAP files were unauthorized sequels all by themselves, I do think it's a bit subtler than that, though I can understand your reading the opinion that way. Observe Footnote 5: We note that the N/I MAP files can only be used with D/N-3D. If another game could use the MAP files to tell the story of a mousy fellow who travels through a beige maze, killing vicious saltshakers with paperclips, then the MAP files would not incorporate the protected expression of D/N-3D because they would not be telling a D/N-3D story. This amounts to a declaration that the MAP files are not really unauthorized
Re: LGPL module linked with a GPL lib
On Thu, Jul 28, 2005 at 09:19:15AM -0700, Ken Arromdee wrote: On Thu, 28 Jul 2005, Andrew Suffield wrote: Anyway, the person who recombines the film and track, in the case of dynamic linking, is the *USER*, in the process of using the program, and copyrights protection do not apply at that moment, as per 17USC. You Are Wrong. Under US law, this is Contributory Infringement, which carries a full array of jail terms. SCOTUS just upheld it against Grokster a few weeks ago. Providing an automated system for users to perform infringing acts, with the sole intent of aiding them in performing those acts, is the same as doing them yourself. But that doesn't apply in the case of automatic systems for users to do the link. The GPL allows users to do what they want privately, so the users aren't performing infringing acts themselves. So your theory is that this is okay because users cannot ever possibly do anything aside from private use and other actions that the GPL doesn't restrict? I'm afraid there's a gaping hole in that one... -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | signature.asc Description: Digital signature
Re: LGPL module linked with a GPL lib
On Thu, Jul 28, 2005 at 08:04:40AM -0400, Michael Poole wrote:
Andrew Suffield writes:
On Wed, Jul 27, 2005 at 02:42:35PM -0300, Humberto Massa Guimar?es wrote:
Static linking can *not* create a derived work, because it is an
automatic process. Poster case: is hello, generated from hello.c:
#include stdio.h
int main(int argc, char** argv) {
printf(Hello\n);
return 0;
}
a derivative work of something it's (statically) linked to?
The answer is no, because derivative works, as intelligent
transformations, can only appear when you *create* a work.
This is a FAQ, or more precisely something that people frequently get
wrong. The derivative work was created when you wrote the source code
that needs to be statically linked. Myopically staring at the build
process does nothing to change this.
Comparison to
Grokster et al doesn't hold
Obviously you didn't read the mail you were replying to. Can't see any
point in writing more.
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ |
`. `' |
`- -- |
signature.asc
Description: Digital signature
Re: LGPL module linked with a GPL lib
On Thu, 28 Jul 2005, Michael K. Edwards wrote: But that doesn't apply in the case of automatic systems for users to do the link. The GPL allows users to do what they want privately, so the users aren't performing infringing acts themselves. While Andrew's parallel to Grokster is IMHO inapposite, he is correct that a theory of contributory infringement (also available in other countries under the name vicarious liability) allows recovery from a party whose role is to facilitate and encourage infringement by others. The availability of some sort of personal-use safe harbor (as in European patent law, for instance; see thread on XMMS and MP3) does not necessarily protect a commercial entity whose product or service does not have (or is not actually marketed for the sake of) substantial non-infringing uses. While that's true, the right of users to link the software in private isn't a personal-use safe harbor--it's explicitly allowed by the GPL. If the GPL lets the user do it, it isn't infringement at all. You can't have contributory infringement if there's no infringement. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On Thu, 28 Jul 2005 17:06:58 -0700 Michael K. Edwards wrote: On 7/28/05, Francesco Poli [EMAIL PROTECTED] wrote: What do you mean freely available? Should I request a copy, which license would you send it under? None whatsoever. :-) Just like sending you a paper copy in the mail, with no obligation of confidentiality as such; the copy is yours, feel free to read it, use ideas and facts from it, quote from it under fair use standards, or do anything else you're permitted to by statute and convention. Feel free also to use your statutory right to pass it to another person (not retaining a copy), or request permission (which I will actually be happy to give you in advance) to make a few copies for friends, business associates, whatever. But don't represent it as your own work, don't publish it in a way that would obstruct its publication in a journal (more or less, a way that would have affected its copyright status under pre-1976 law), and don't claim that I authorized you to represent me for any commercial purpose or to quote me as an authority in any legal sense. In other words, All Rights Reserved... :-( P.S.: please do not reply to me directly, while Cc:ing the list, as I didn't ask you to do so... since I'm a debian-legal subscriber, I'd rather not receive messages twice! thanks -- :-( This Universe is buggy! Where's the Creator's BTS? ;-) .. Francesco Poli GnuPG Key ID = DD6DFCF4 Key fingerprint = C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4 pgpkALfXXQWWz.pgp Description: PGP signature
Re: LGPL module linked with a GPL lib
On 7/29/05, Ken Arromdee [EMAIL PROTECTED] wrote: While that's true, the right of users to link the software in private isn't a personal-use safe harbor--it's explicitly allowed by the GPL. If the GPL lets the user do it, it isn't infringement at all. You can't have contributory infringement if there's no infringement. The GPL is not a new copyright statute with the power to override the meaning of infringement, nor do its drafting oddities render it null. It does indeed declare (in v2) that activities other than copying, distribution and modification are not covered by this License; and, taking that statement together with the extent of modification and copying permitted by 17 USC 117 (and its parallels elsewhere), you may perhaps say that the GPL explicitly allows the end user to link the software in private. But that is merely a basis for arguing that the copyright holder is estopped from suing end users who are simply using what they have received in good faith. That does not mean that a distributor could not be successfully sued for copyright infringement _if_ it were correct that the act of linking breached rights reserved to the copyright holder. It makes little difference whether you call this direct or contributory infringement; there is no reason to think that the legal system in any Berne Convention country would find it difficult to bring the consequences home to the party that profits (or would be profiting if it were charging for distribution). Where there is a valid theory under which the copyright holder is suffering a loss not of her own making, any legal system that recognizes tort or a similar concept may be expected to provide a forum for her to demonstrate who is responsible for wronging her. Compare, for instance, Micro Star v. FormGen. You could argue that the unauthorized sequel didn't really exist until an end user loaded the MAP file into Duke Nukem, and Micro Star neither authored the MAP files nor distributed them together with Duke Nukem. FormGen had effectively estopped itself, in its license, from suing either the MAP file authors (who were giving away their work in compliance with the terms of the license) or the end users. Yet FormGen successfully sued Micro Star for copyright infringement, essentially because 1) they were distributing files which amounted to infringing works of fiction and had no substantial non-infringing uses, and 2) they could not claim the non-commercial exemption offered in the license to the MAP file authors themselves. Where the GPL situation differs, of course, is that linking one library to another does not create an unauthorized sequel of either; nor does any other way of using one program's functionality from another without a deliberate intermingling of their creative expression by a human agency. There may well be ways of writing an offer of contract such that the licensee gives up rights with respect to a published work that he would otherwise have under copyright law, in exchange for other rights that are within the copyright holder's power to grant or reserve. But the GPL drafters are in a very poor position to make such an argument. The GPL does not succeed in blocking the creation, distribution, and use of interoperating programs under other licenses. Not, perhaps, because the drafters didn't intend to do so, and certainly not because the GPL is some sort of magical creature of copyright law (it's not). Instead, its interpretation is constrained by its drafters' insistence on claiming that it is designed to give freedoms and take none away, and by their dogged determination to use copyright-law language (however inapposite) to persuade the naive reader that it is free of the contract law's jurisdictional variations and rules of construction. Given the GPL's preamble and its drafters' conduct, I think you would have to go pretty far afield to find a legal system in which a GPL licensor would be permitted to deny a licensee liberties that would be permitted to another with no contractual relationship to the author. And in US law as I understand it, those liberties include the creation, distribution, and use of software that interoperates via published external interfaces, whether or not it is in the economic (or ideological) interests of the original author to permit this. The only exceptions of which I am aware involve quite different causes of action: - misappropriation of characters and mise en scene (Micro Star v. FormGen) - fraud on the copyright office and literal infringement in excess of strict interoperability requirements (Atari v. Nintendo) - misappropriation of trade secrets by former employees (Cadence v. Avant!) - breach of a duly executed contract, and failure of the 17 USC 117 exemption because the licensee did not qualify as the owner of a copy (DSC v. Pulse) - breach of conditions of license in a negotiated contract, and thereafter trademark dilution (Sun v. Microsoft) - trademark
Re: LGPL module linked with a GPL lib
On 7/29/05, Francesco Poli [EMAIL PROTECTED] wrote: In other words, All Rights Reserved... :-( I did say that I would be happy to give you advance permission to circulate a reasonable number of copies privately, which would leave me with no recourse against you unless you set out to misappropriate or grossly misrepresent my work or to defeat its first publication. That's rather different from all rights reserved, and it's the most open I know how to be without seriously compromising its status as an unpublished work still in draft. P.S.: please do not reply to me directly, while Cc:ing the list, as I didn't ask you to do so... since I'm a debian-legal subscriber, I'd rather not receive messages twice! thanks I went back and checked the code of conduct and you are of course right that I am in error. It's too bad that the code of conduct plus the remailer configuration are inverted relative to the way that MUAs work. Adding a Followup-To: d-l is supposed to be a no-op by the code of conduct, and would make almost all MUAs do the right thing; but the remailer doesn't do it automatically. And even if I check the headers manually I'm not supposed to assume that a Reply-To: (real mail address) was intentional. So I'm supposed to strip the sender from the reply list no matter what Followup-To or Reply-To says, unless I'm not subscribed, please copy me is in the body of the message. I guess I'll do that henceforth, but it really feels broken. Cheers, - Michael
Re: LGPL module linked with a GPL lib
On Wed, Jul 27, 2005 at 02:42:35PM -0300, Humberto Massa Guimar?es wrote:
Static linking can *not* create a derived work, because it is an
automatic process. Poster case: is hello, generated from hello.c:
#include stdio.h
int main(int argc, char** argv) {
printf(Hello\n);
return 0;
}
a derivative work of something it's (statically) linked to?
The answer is no, because derivative works, as intelligent
transformations, can only appear when you *create* a work.
This is a FAQ, or more precisely something that people frequently get
wrong. The derivative work was created when you wrote the source code
that needs to be statically linked. Myopically staring at the build
process does nothing to change this.
Additionally, when linking statically, you are performing literal
copying of the original work into the target binary, so the GPL kicks
in there too.
Anyway, the person who recombines the film and track, in the
case of dynamic linking, is the *USER*, in the process of using the
program, and copyrights protection do not apply at that moment, as
per 17USC.
You Are Wrong. Under US law, this is Contributory Infringement, which
carries a full array of jail terms. SCOTUS just upheld it against
Grokster a few weeks ago. Providing an automated system for users to
perform infringing acts, with the sole intent of aiding them in
performing those acts, is the same as doing them yourself.
The rest of the world isn't quite so crazy, but then it doesn't get
into this ridiculous word game in the first place.
--
.''`. ** Debian GNU/Linux ** | Andrew Suffield
: :' : http://www.debian.org/ |
`. `' |
`- -- |
signature.asc
Description: Digital signature
Re: LGPL module linked with a GPL lib
Andrew Suffield writes:
On Wed, Jul 27, 2005 at 02:42:35PM -0300, Humberto Massa Guimar?es wrote:
Static linking can *not* create a derived work, because it is an
automatic process. Poster case: is hello, generated from hello.c:
#include stdio.h
int main(int argc, char** argv) {
printf(Hello\n);
return 0;
}
a derivative work of something it's (statically) linked to?
The answer is no, because derivative works, as intelligent
transformations, can only appear when you *create* a work.
This is a FAQ, or more precisely something that people frequently get
wrong. The derivative work was created when you wrote the source code
that needs to be statically linked. Myopically staring at the build
process does nothing to change this.
What statute or case law supports this position? Comparison to
Grokster et al doesn't hold, for reasons that should have been
obvious: The GPL explicitly allows a user to use and modify code in
any way the user sees appropriate; section 2 only kicks in when you
copy and distribute such modifications. The standard way that one
uses GPLed program is by linking its compiled form with other works at
runtime. It is simply bizarre to suggest that a software author is
liable for contributory copyright infringement by providing a means
for a user to do something that the user is explicitly permitted to
do.
Michael Poole
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On Thu, 28 Jul 2005, Andrew Suffield wrote: Anyway, the person who recombines the film and track, in the case of dynamic linking, is the *USER*, in the process of using the program, and copyrights protection do not apply at that moment, as per 17USC. You Are Wrong. Under US law, this is Contributory Infringement, which carries a full array of jail terms. SCOTUS just upheld it against Grokster a few weeks ago. Providing an automated system for users to perform infringing acts, with the sole intent of aiding them in performing those acts, is the same as doing them yourself. But that doesn't apply in the case of automatic systems for users to do the link. The GPL allows users to do what they want privately, so the users aren't performing infringing acts themselves. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On 7/28/05, Michael Poole [EMAIL PROTECTED] wrote: What statute or case law supports this position? Comparison to Grokster et al doesn't hold, for reasons that should have been obvious: The GPL explicitly allows a user to use and modify code in any way the user sees appropriate; section 2 only kicks in when you copy and distribute such modifications. The standard way that one uses GPLed program is by linking its compiled form with other works at runtime. It is simply bizarre to suggest that a software author is liable for contributory copyright infringement by providing a means for a user to do something that the user is explicitly permitted to do. I don't think that the point is that people would be going to jail for violating the GPL. GPL violators appear to face several potential penalties: (*) Works they hold copyright on might be released under the terms of the GPL when they thought they could get away with not doing that. (*) They might be forced to pay someone else to release their work under GPL terms. (*) They might be forced to stop distribution of some work where they don't have the proper rights available to them. Obviously there would be cases where one or more of these would not apply, but if none of these apply that's probably because they're not violating the terms of the GPL. -- Raul
Re: LGPL module linked with a GPL lib
On 7/27/05, Humberto Massa Guimarães [EMAIL PROTECTED] wrote:
Static linking can *not* create a derived work, because it is an
automatic process. Poster case: is hello, generated from hello.c:
#include stdio.h
int main(int argc, char** argv) {
printf(Hello\n);
return 0;
}
a derivative work of something it's (statically) linked to?
The answer is no, because derivative works, as intelligent
transformations, can only appear when you *create* a work.
As Andrew Suffield has pointed out, this is a strawman
argument. But I'd like to reiterate that point.
(1) the creation happens in the human realm, when we talk
about static linking we're talking about the intent and activities
which lead up to the mechanical process.
(2) the code above is not original. It has been published
before, under an all rights reserved license. In the U.S.
this isn't a problem because of fair use laws. I don't know
what the legal situation is with respect to this particular
code in countries which don't explicitly have that kind of
law on the books. Presumably they have some other
rights for non-copyright-holders which are the real
reason that this code doesn't violate copyright in those
countries.
Note that I'm taking it as a given that -- despite what
arguments people might present -- distribution of the
above hello world code is not going to cause any
problems in any country -- despite copyright treaties,
and despite the license on KR's book.
--
Raul
Re: LGPL module linked with a GPL lib
On 7/27/05, Michael K. Edwards [EMAIL PROTECTED] wrote: Whether or not that agreement purports to bind a developer in ways that copyright law does not, there are limits to what terms a court will permit in a contract of adhesion. Agreed. Then again, the penalties I'd expect the court to apply against someone violating the GPL are relatively mild. Your thesis seems to be that the court will not apply harsh penalties -- and I have no problem with that. But you might consider that the people you claim are making mistakes probably were not thinking that those harsh penalties would apply in the first place. For example, take Progress v. MySql -- here, the stop distribution penalty was not used in part because Progress didn't have anything else -- it would have been destroyed by this penalty. And, Progress had agreed in court to release their software under other terms. This is purely a practical issue (as you'd expect from a preliminary injunction). Anyways, having the restricted software released under GPL compatible terms is the desired outcome -- this is expressed in the GPL, and this has been stated repeatedly by the FSF. What's amazing are your repeated claims that the FSF doesn't know what it's talking about, legally speaking. -- Raul
Re: LGPL module linked with a GPL lib
** Raul Miller ::
On 7/27/05, Humberto Massa Guimarães
[EMAIL PROTECTED] wrote:
Static linking can *not* create a derived work, because it is an
automatic process. Poster case: is hello, generated from hello.c:
#include stdio.h
int main(int argc, char** argv) {
printf(Hello\n);
return 0;
}
a derivative work of something it's (statically) linked to?
The answer is no, because derivative works, as intelligent
transformations, can only appear when you *create* a work.
As Andrew Suffield has pointed out, this is a strawman
argument. But I'd like to reiterate that point.
Strawman?
Fact: the creation of a derivative work is the application of some
transformation on the original work.
The above snippet (which isn't even copyrightable, for its sheer size and the
necessity of expressing the same thing in the same language) is NOT a
derivative work of ANY libc implementation, even if (for some implementation)
it NEEDED to be statically linked. Because when I created the snippet (and I
created it) I did NOT apply any transformation to anything else that is
copyrighted or event copyrightable.
The binary, OTOH, when statically linked, contains (parts of) the linked libc
implementation, copied, so the linked libc license terms must be followed to
distribute such binary, if possible at all.
Now, a dynamic linked binary does NOT contain ANY copyrightable parts of the
libc implementation. So, the distribution of the dynamic linked binary does NOT
depend on authorization from the copyrights holders of the libc implementation.
MORE SO because you can use such binary with another, compatible, libc
implementation.
Even if (as is our real case) you want to distribute together the binary and
the libc implementation, one being GPLd and the other,
GPL-incompatible-licensed, this is a case of mere agreegation because (a) the
GPL says so when you read it correctly, in opposition to a derivative work
under copyright law and (b) because you *are* just aggregating things. If you
develop another libc and install it in your system, and ldconfig away everyone
to it, one of the parts of this equation would have been completely erased.
*This* is the kind of interoperability that has protected in the Lexmark case,
for example.
Re: LGPL module linked with a GPL lib
Raul Miller writes: On 7/28/05, Michael Poole [EMAIL PROTECTED] wrote: What statute or case law supports this position? Comparison to Grokster et al doesn't hold, for reasons that should have been obvious: The GPL explicitly allows a user to use and modify code in any way the user sees appropriate; section 2 only kicks in when you copy and distribute such modifications. The standard way that one uses GPLed program is by linking its compiled form with other works at runtime. It is simply bizarre to suggest that a software author is liable for contributory copyright infringement by providing a means for a user to do something that the user is explicitly permitted to do. I don't think that the point is that people would be going to jail for violating the GPL. GPL violators appear to face several potential penalties: Potential penalties are irrelevant to my question. You assume a priori that such linking is a violation of the GPL. My question was why that assumption is valid. As I explained above, his citation of case law does not fit the facts. Michael Poole -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
** Michael Poole :: Potential penalties are irrelevant to my question. You assume a priori that such linking is a violation of the GPL. My question was why that assumption is valid. As I explained above, his citation of case law does not fit the facts. The only good answer people in d-l gave me to the question: why is the assumption that such linking is a violation of the GPL valid? is because Eben Moglen said so in the GPL FAQ, and he is a law teacher, so it must be true. -- HTH, Massa -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On 7/28/05, Raul Miller [EMAIL PROTECTED] wrote: I don't think that the point is that people would be going to jail for violating the GPL. Violating the GPL doesn't mean anything. The GPL is not a statute. It's just an offer of contract. The only way to enforce it is for a party with standing for a copyright infringement action (i. e., an _author_ -- not just any old contributor of some fragment) to bring suit for copyright infringement with additional claims of breach of contract, to be pressed if the alleged infringer claims license under the GPL for his conduct. The penalties for copyright infringement are laid out in 17 USC Chapter 5; the penalties for breach of contract may include some sort of equitable relief and/or permission to abrogate the contract and proceed to claims of copyright infringement. GPL violators appear to face several potential penalties: (*) Works they hold copyright on might be released under the terms of the GPL when they thought they could get away with not doing that. This is extraordinarily unlikely except as part of a consent decree, agreed on by the parties and ratified by the court as an alternative to proceeding to judgment. The only cases I have seen (but IANAL and this is not an area I have researched closely) that came anywhere near such an order of specific performance prescribed such performance as an alternative to withdrawing a work from circulation altogether. See Sun v. Microsoft, on remand from the Ninth Circuit decision. (*) They might be forced to pay someone else to release their work under GPL terms. This is even more unlikely; the court has no power to compel that someone else to make such an offer. Of course, if the accused thinks that the simplest way to end the dispute is to negotiate such a deal with the third party and offer it as part of a consent decree, that's another story. (*) They might be forced to stop distribution of some work where they don't have the proper rights available to them. In other words, injunctive relief -- often granted temporarily pending trial at the preliminary hearing stage, and if granted often followed by out-of-court settlement. As I have explained ad nauseam, I think it very improbable (IANAL, TINLA) that preliminary injunctive relief could be obtained in a linking scenario involving the GPL even under the most egregious of fact patterns; compare Progress Software v. MySQL. Obviously there would be cases where one or more of these would not apply, but if none of these apply that's probably because they're not violating the terms of the GPL. AIUI most copyright infringement suits end with monetary damages, sometimes with attorneys' fees and costs attached, sometimes with impoundment and an added injunction against continued infringement (which makes it quite easy to come back for more penalties if the infringer doesn't cease). But IMHO a competently defended GPL violation suit, if its facts reflect reasonable disagreement about the interpretation of a contract of adhesion, is unlikely to result in any penalty other than the stipulation of future conduct consistent with the contract terms as construed by the court. Cheers, - Michael
Re: LGPL module linked with a GPL lib
On 7/28/05, Ken Arromdee [EMAIL PROTECTED] wrote: On Thu, 28 Jul 2005, Andrew Suffield wrote: You Are Wrong. Under US law, this is Contributory Infringement, which carries a full array of jail terms. SCOTUS just upheld it against Grokster a few weeks ago. Providing an automated system for users to perform infringing acts, with the sole intent of aiding them in performing those acts, is the same as doing them yourself. But that doesn't apply in the case of automatic systems for users to do the link. The GPL allows users to do what they want privately, so the users aren't performing infringing acts themselves. While Andrew's parallel to Grokster is IMHO inapposite, he is correct that a theory of contributory infringement (also available in other countries under the name vicarious liability) allows recovery from a party whose role is to facilitate and encourage infringement by others. The availability of some sort of personal-use safe harbor (as in European patent law, for instance; see thread on XMMS and MP3) does not necessarily protect a commercial entity whose product or service does not have (or is not actually marketed for the sake of) substantial non-infringing uses. Cheers, - Michael (IANAL, TINLA)
Re: LGPL module linked with a GPL lib
On 7/28/05, Raul Miller [EMAIL PROTECTED] wrote: For example, take Progress v. MySql -- here, the stop distribution penalty was not used in part because Progress didn't have anything else -- it would have been destroyed by this penalty. And, Progress had agreed in court to release their software under other terms. This is purely a practical issue (as you'd expect from a preliminary injunction). You still don't seem to understand what happened in that case. The judge cited six reasons why it would be inappropriate to grant a preliminary injunction for breach of the GPL terms, any one of which would have been sufficient. And it was implicit in the structure of the decision that she rejected both the notion that the GPL is a creature of copyright law and the plea in Mr. Moglen's affidavit that the automatic termination clause was central to its enforcement. Had she taken either of these propositions even a little bit seriously, she would (as she well knows) have been obliged to analyze the request for preliminary injunction according to a completely different standard. What's amazing are your repeated claims that the FSF doesn't know what it's talking about, legally speaking. Oh, they may well know the truth (as I understand it); but if so, they do not choose to advertise their knowledge. Alternately, if they have a foundation for their claims that my amateur archaeology has been unable to unearth, they seem to have no desire to expose it to public scrutiny. Why should they, when widespread belief in the accuracy of their interpretation gets the job done anyway? Cheers, - Michael (IANAL, TINLA)
Re: LGPL module linked with a GPL lib
On Wed, 27 Jul 2005 16:04:34 -0700 Michael K. Edwards wrote: On 7/27/05, Jeff Licquia [EMAIL PROTECTED] wrote: Excuse me for asking, but why is this monograph not freely available? Surely, as a non-lawyer, you have no hope of profiting from it, and having a succint, linkable statement of your arguments would do wonders for preventing such go-arounds as this one, and might even persuade laymen like me. It's freely available by request, but not published, for several reasons. What do you mean freely available? Should I request a copy, which license would you send it under? -- :-( This Universe is buggy! Where's the Creator's BTS? ;-) .. Francesco Poli GnuPG Key ID = DD6DFCF4 Key fingerprint = C979 F34B 27CE 5CD8 DC12 31B5 78F4 279B DD6D FCF4 pgpWjiHaU1o57.pgp Description: PGP signature
Re: LGPL module linked with a GPL lib
On 7/28/05, Francesco Poli [EMAIL PROTECTED] wrote: What do you mean freely available? Should I request a copy, which license would you send it under? None whatsoever. :-) Just like sending you a paper copy in the mail, with no obligation of confidentiality as such; the copy is yours, feel free to read it, use ideas and facts from it, quote from it under fair use standards, or do anything else you're permitted to by statute and convention. Feel free also to use your statutory right to pass it to another person (not retaining a copy), or request permission (which I will actually be happy to give you in advance) to make a few copies for friends, business associates, whatever. But don't represent it as your own work, don't publish it in a way that would obstruct its publication in a journal (more or less, a way that would have affected its copyright status under pre-1976 law), and don't claim that I authorized you to represent me for any commercial purpose or to quote me as an authority in any legal sense. Cheers, - Michael
Re: LGPL module linked with a GPL lib
** Jeff Licquia :: On Tue, 2005-07-26 at 11:14 -0300, Humberto Massa Guimarães wrote: I find this discussion ultimately absurd. Debian is *not* distributing a derivative work. Debian does *not* distribute a work that includes both plugins/libraries. The fact that the things are (dynamically) linked at run time, especially combined with the fact that the plugins are opened with dlopen() and use stable API, is *more* than enough to lift any (inexistent IMHO) no-link requirement of the GPL. I find most of this response confusing. Yes, it is. I was ranting, because this discussion makes me see red. I apologize. First of all, it's clear that Debian *is* distributing a derived work based on GPLed libraries, called Debian GNU/Linux. The First of all, Debian GNU/Linux is *NOT* a derivative work of OpenSSL, GStreamer, nor any of its plugins. A derivative work has a definition in the statute (in the US case, 17USC). specific case in question may fall under the mere aggregation clause of the GPL, but then this is the point you should argue. I The last paragraph holds, independently of the mere aggregation value. abhor imprecision in these discussions, as they are the breeding ground for all kinds of myths and speculation. (Not that I am immune to imprecision, or that I am not occasionally a myth-monger in my own right. But I welcome the correction.) Second, you seem to be asserting that an app and its dynamically linked libraries do not constitute a derived work based on both for the purposes of the GPL. Rather than debate this point, I Yes. There is no derivative work status on the program that uses a library. I and M.K.Edwards, in the last 3 months or so, have brought a lot of arguments and case law to this extent to d-l, and my own and humble conclusion is that: especially in the case of dynamic linking (and more so in the case of dlopen()ing), the distribution by debian of both a program A and a linking-to-A B.so is subject only to the *separate* compliance to the terms of both A and B.so, independently of any terms applied only to derivative works of A or of B.so. think it best to point out that this runs counter to accepted precedent within Debian that dates back a long time; see the KDE/Qt controversy for a famous example. Basing conclusions on this past precedent is not absurd; indeed, it would seem that the onus is on you to prove your assertion. I do not have enough time right now to answer properly (ie, with the links to the discussions, examples, and caselaw that I, amongst others, presented here on d-l), but I trust that you can find them if you are interested. As I said two paragraphs above, I consider that I presented all my arguments in this direction, and (to me, at least) I consider my point proven. That's probably enough for starters. If I am indeed confused and you are correct, then there doesn't seem much point to proceed to the dlopen() question. Ok. -- HTH, Massa
Re: LGPL module linked with a GPL lib
On Wed, 2005-07-27 at 10:05 -0300, Humberto Massa Guimarães wrote: First of all, Debian GNU/Linux is *NOT* a derivative work of OpenSSL, GStreamer, nor any of its plugins. A derivative work has a definition in the statute (in the US case, 17USC). Hmm. I suppose this is part and parcel of the move in the USA to copyright compilations or databases? I suppose I had never thought of it that way. Yes. There is no derivative work status on the program that uses a library. I and M.K.Edwards, in the last 3 months or so, have brought a lot of arguments and case law to this extent to d-l, and my own and humble conclusion is that: especially in the case of dynamic linking (and more so in the case of dlopen()ing), the distribution by debian of both a program A and a linking-to-A B.so is subject only to the *separate* compliance to the terms of both A and B.so, independently of any terms applied only to derivative works of A or of B.so. Mr. Edwards, elsewhere, refers to the GFingerPoken thread, which I had followed. There may be other threads I did not follow, and I will look for them. I confess to not seeing how the manner of linking makes a difference from a copyright point of view. Static linking creates a derived work, in that the resulting binary contains the library, much as how a motion picture film contains its soundtrack. To me, splitting the soundtrack off a movie, and creating a machine to recombine them afterwards, does not cease to make the movie an infringement on the soundtrack's copyright, which is equivalent to the dynamic linking process. Is such a scheme really effective from a legal standpoint in avoiding copyright liability? I do not have enough time right now to answer properly (ie, with the links to the discussions, examples, and caselaw that I, amongst others, presented here on d-l), but I trust that you can find them if you are interested. As I said two paragraphs above, I consider that I presented all my arguments in this direction, and (to me, at least) I consider my point proven. That's great. Other people with legal expertese (the FSF legal team, for example) have done the same, and have come to entirely different conclusions. Others with legal expertese commented, as I recall, on the KDE/Qt controversy back in the day, too, and I don't recall seeing any argument against it that wasn't based on emotion or wishful thinking (the KDE and Qt people are good people, they wouldn't sue anyone). I am not a lawyer, and thus am forced to accept arguments from authority (and regurgitate them when necessary, as was the case in this thread). It seems clear in my interaction with you that my understanding of the copyright process is hopelessly inadequate for evaluating these arguments; there always seems to be some exception to the general rule that people can throw at any position people can take. And, it seems to me, that in the authority face-off, you lose. I've never heard of you outside this forum. Mr. Edwards has already admitted to a lack of formal legal training. The GPL, on the other hand, has a law professor and a team of lawyers behind it, as do other groups promoting free software and open source, and their efforts at enforcing their view of the world have been quite successful to date. Are you seriously telling me that these people don't know what they're talking about regarding the law, and that you do? On what basis can you make such an extraordinary claim? Now, the standard answer when confronted with such a quandary is go hire your own lawyer. Which is a great idea, if you have hundreds of dollars to throw away. If I wanted to throw money at legal crap, I would have stayed in the Windows world. So, you will forgive me, I hope, for continued skepticism. You may be right, and the entire free software community may be wrong, about the way things work. But I'm betting not.
Re: LGPL module linked with a GPL lib
** Jeff Licquia ::
On Wed, 2005-07-27 at 10:05 -0300, Humberto Massa Guimarães wrote:
First of all, Debian GNU/Linux is *NOT* a derivative work of
OpenSSL, GStreamer, nor any of its plugins. A derivative work
has a definition in the statute (in the US case, 17USC).
Hmm. I suppose this is part and parcel of the move in the USA to
copyright compilations or databases? I suppose I had never
thought of it that way.
Derivative works are those that result of a (non-automatable,
intelligent, intellectually novel) _transformation_ of a work.
Compilation works are those that are composed of other works, in
which the intellectual novelty resides in the selection and
arrangement of contents.
Debian, as a whole, is a compilation works on its packages.
Yes. There is no derivative work status on the program that uses
a library. I and M.K.Edwards, in the last 3 months or so, have
brought a lot of arguments and case law to this extent to d-l,
and my own and humble conclusion is that: especially in the case
of dynamic linking (and more so in the case of dlopen()ing), the
distribution by debian of both a program A and a linking-to-A
B.so is subject only to the *separate* compliance to the terms
of both A and B.so, independently of any terms applied only to
derivative works of A or of B.so.
Mr. Edwards, elsewhere, refers to the GFingerPoken thread, which I
had followed. There may be other threads I did not follow, and I
will look for them.
I confess to not seeing how the manner of linking makes a
difference from a copyright point of view. Static linking creates
a derived work, in that the resulting binary contains the library,
Static linking can *not* create a derived work, because it is an
automatic process. Poster case: is hello, generated from hello.c:
#include stdio.h
int main(int argc, char** argv) {
printf(Hello\n);
return 0;
}
a derivative work of something it's (statically) linked to?
The answer is no, because derivative works, as intelligent
transformations, can only appear when you *create* a work.
If I translate Helter Skelter to Portuguese -- not by babelfish --
then I am doing a derivative work. When I write the above hello.c,
not.
much as how a motion picture film contains its soundtrack. To me,
splitting the soundtrack off a movie, and creating a machine to
recombine them afterwards, does not cease to make the movie an
infringement on the soundtrack's copyright, which is equivalent to
the dynamic linking process. Is such a scheme really effective
from a legal standpoint in avoiding copyright liability?
No, because you can't change the soundtrack of a film without
changing the final result. OTOH, you can substitute (almost always)
glibc for dietlibc in a program and have the same result. Such
program does not *depend* on glibc for being what it is, and most
certainly is *not* the result of a transformation applied over
glibc.
Anyway, the person who recombines the film and track, in the
case of dynamic linking, is the *USER*, in the process of using the
program, and copyrights protection do not apply at that moment, as
per 17USC.
I do not have enough time right now to answer properly (ie, with
the links to the discussions, examples, and caselaw that I,
amongst others, presented here on d-l), but I trust that you can
find them if you are interested.
As I said two paragraphs above, I consider that I presented all
my arguments in this direction, and (to me, at least) I consider
my point proven.
That's great. Other people with legal expertese (the FSF legal
team, for example) have done the same, and have come to entirely
different conclusions. Others with legal expertese commented, as
I recall, on the KDE/Qt controversy back in the day, too, and I
don't recall seeing any argument against it that wasn't based on
emotion or wishful thinking (the KDE and Qt people are good
people, they wouldn't sue anyone).
I am not a lawyer, and thus am forced to accept arguments from
authority (and regurgitate them when necessary, as was the case in
this thread). It seems clear in my interaction with you that my
understanding of the copyright process is hopelessly inadequate
for evaluating these arguments; there always seems to be some
exception to the general rule that people can throw at any
position people can take.
And, it seems to me, that in the authority face-off, you lose.
I've never heard of you outside this forum. Mr. Edwards has
already admitted to a lack of formal legal training. The GPL, on
the other hand, has a law professor and a team of lawyers behind
it, as do other groups promoting free software and open source,
and their efforts at enforcing their view of the world have been
quite successful to date. Are you seriously telling me that these
people don't know what they're talking about regarding the law,
and that you do? On what basis can you make such an extraordinary
claim?
Now,
Re: LGPL module linked with a GPL lib
On 7/27/05, Jeff Licquia [EMAIL PROTECTED] wrote: On Wed, 2005-07-27 at 10:05 -0300, Humberto Massa Guimarães wrote: First of all, Debian GNU/Linux is *NOT* a derivative work of OpenSSL, GStreamer, nor any of its plugins. A derivative work has a definition in the statute (in the US case, 17USC). Hmm. I suppose this is part and parcel of the move in the USA to copyright compilations or databases? I suppose I had never thought of it that way. No, the derivative work category (which had a parallel, but no special name or significance, in prior copyright statutes) was created in the 1976 Copyright Act, essentially for the sake of the exceptions to license termination in Sections 203 and 304. It has no special significance under law other than in cases involving these sections, and using it in the text of a license is eccentric at best. Prior US statute and jurisprudence treated translations, adaptations, and so forth as just a subset of works that were both copyrightable as originals and infringing on the older work in the absence of license. Compilations and databases have no relation to derivative works in statute or in history. For a history of the rules about copyright in compilations whose component parts are uncopyrightable facts were established, see the discussion in Feist v. Rural Telephone ( http://www.law.cornell.edu/copyright/cases/499_US_340.htm ); note that compilations were copyrightable under both the 1909 and 1976 Acts, and that decisions like Harper Row (1985) and Feist (1991) _narrowed_ the judicial interpretation about what made them copyrightable, discarding sweat of the brow and focusing on what originality they may contain. I'm not aware of any contemporary move to copyright such things, i. e., to restore the sweat of the brow doctrine; there is some effort to harmonize the treatment of compilations among Berne Convention nations and to codify the Feist standard of thin protection for factual compilations. In short, the GPL drafters could also have attempted to encumber _collective_ works containing a GPL work, which would affect Debian CDs (works of authorship by virtue of the creative expression in the selection and arrangement of their components), but the actual language of the GPL doesn't, when appropriate standards of contract construction are applied. The GPL probably couldn't legally block the _use_ of a GPL component from components under different licenses no matter how it was written, given precedents such as Lotus v. Borland and Lexmark v. Static Control. The copyright monopoly is granted on creative expression, not on function, and is not meant to be leveraged to block competitive interoperability. The League for Programming Freedom's amicus brief in Lotus v. Borland at http://web.archive.org/web/lpf.ai.mit.edu/Copyright/lpf-sc-amicus.html is moderately eloquent on the topic. Maybe you can find a way for Eben Moglen's own words not to damn his position on GPL violation through linking, but I can't. Yes. There is no derivative work status on the program that uses a library. I and M.K.Edwards, in the last 3 months or so, have brought a lot of arguments and case law to this extent to d-l, and my own and humble conclusion is that: especially in the case of dynamic linking (and more so in the case of dlopen()ing), the distribution by debian of both a program A and a linking-to-A B.so is subject only to the *separate* compliance to the terms of both A and B.so, independently of any terms applied only to derivative works of A or of B.so. Mr. Edwards, elsewhere, refers to the GFingerPoken thread, which I had followed. There may be other threads I did not follow, and I will look for them. The message to which I pointed you has a link back into the main fray (threads with titles like Urgently need GPL compatible libsnmp5-dev replacement, GPL and linking, and What makes software copyrightable anyway?). I've put together a 50-some page monograph that contains 'most everything I know about the subject (and about the process of construing the GPL, apart from the warranty disclaimer, about which I've learned more since); copies available for private circulation upon request. I confess to not seeing how the manner of linking makes a difference from a copyright point of view. Static linking creates a derived work, in that the resulting binary contains the library, much as how a motion picture film contains its soundtrack. To me, splitting the soundtrack off a movie, and creating a machine to recombine them afterwards, does not cease to make the movie an infringement on the soundtrack's copyright, which is equivalent to the dynamic linking process. Is such a scheme really effective from a legal standpoint in avoiding copyright liability? The manner of linking doesn't make a difference in principle, and the statically linked case is no more a derived work than the dynamically linked case. Dynamic linking makes it abundantly
Re: LGPL module linked with a GPL lib
On Wed, 2005-07-27 at 14:42 -0300, Humberto Massa Guimarães wrote:
** Jeff Licquia ::
On Wed, 2005-07-27 at 10:05 -0300, Humberto Massa Guimarães wrote:
First of all, Debian GNU/Linux is *NOT* a derivative work of
OpenSSL, GStreamer, nor any of its plugins. A derivative work
has a definition in the statute (in the US case, 17USC).
Hmm. I suppose this is part and parcel of the move in the USA to
copyright compilations or databases? I suppose I had never
thought of it that way.
Derivative works are those that result of a (non-automatable,
intelligent, intellectually novel) _transformation_ of a work.
Compilation works are those that are composed of other works, in
which the intellectual novelty resides in the selection and
arrangement of contents.
Debian, as a whole, is a compilation works on its packages.
Does such compilation in itself give Debian any rights on its own, or is
the compilation seen as non-copyrightable?
Static linking can *not* create a derived work, because it is an
automatic process. Poster case: is hello, generated from hello.c:
#include stdio.h
int main(int argc, char** argv) {
printf(Hello\n);
return 0;
}
a derivative work of something it's (statically) linked to?
The answer is no, because derivative works, as intelligent
transformations, can only appear when you *create* a work.
Hmm. I think this may be the source of a misunderstanding, for which I
will take the credit.
I would agree that hello.c is unencumbered with regards to its
copyright. But the hello executable created through compilation
(statically, to be clear) certainly has some kind of copyright
obligation with regard to the libraries contained within it, doesn't it?
Perhaps derived work is not the right term, but I find it thoroughly
incredible that the copyright of the library is irrelevant with regards
to the executable.
And, if so, then by the terms of the GPL, the whole of the executable
must be itself covered by the terms of the GPL if any component part is.
much as how a motion picture film contains its soundtrack. To me,
splitting the soundtrack off a movie, and creating a machine to
recombine them afterwards, does not cease to make the movie an
infringement on the soundtrack's copyright, which is equivalent to
the dynamic linking process. Is such a scheme really effective
from a legal standpoint in avoiding copyright liability?
No, because you can't change the soundtrack of a film without
changing the final result. OTOH, you can substitute (almost always)
glibc for dietlibc in a program and have the same result. Such
program does not *depend* on glibc for being what it is, and most
certainly is *not* the result of a transformation applied over
glibc.
Well, yes, in the case of glibc. In fact, one of the possible solutions
to the KDE/Qt problem involved reimplementing Qt under the GPL, at which
point the proprietary Qt would not be a prerequisite for KDE.
Anyway, the person who recombines the film and track, in the
case of dynamic linking, is the *USER*, in the process of using the
program, and copyrights protection do not apply at that moment, as
per 17USC.
So does this mean that libraries cannot impose restrictions on the
programs that use them? I think far more groups than the free software
community rely on this being the case; see the license agreement on any
Microsoft SDK for an example.
I think, in hindsight, that Debian (in the Qt/KDE case) fell for
arguments that are not sound, nor based on the law, even if they
were presented as such.
It would have been good for someone to have made those arguments at the
time. Why they did not do so is a mystery. (Of course, those arguments
could have been made, and missed by me, in which case cites for those
arguments would be worthwhile. Obviously, the cites were missed by far
more people than me.)
Re: LGPL module linked with a GPL lib
On Wed, 2005-07-27 at 12:00 -0700, Michael K. Edwards wrote: The message to which I pointed you has a link back into the main fray (threads with titles like Urgently need GPL compatible libsnmp5-dev replacement, GPL and linking, and What makes software copyrightable anyway?). I've put together a 50-some page monograph that contains 'most everything I know about the subject (and about the process of construing the GPL, apart from the warranty disclaimer, about which I've learned more since); copies available for private circulation upon request. Excuse me for asking, but why is this monograph not freely available? Surely, as a non-lawyer, you have no hope of profiting from it, and having a succint, linkable statement of your arguments would do wonders for preventing such go-arounds as this one, and might even persuade laymen like me. I will examine your links in the future. The manner of linking doesn't make a difference in principle, and the statically linked case is no more a derived work than the dynamically linked case. Dynamic linking makes it abundantly clear that neither the source code nor the binary form of the program contains any more of the original's creative expression than is dictated by interoperability requirements, and may add an additional defense under 17 USC 117, but the only effect is to make it less likely that a district judge will rule incorrectly (IMHO, IANAL, TINLA). The reason that your analogy is incorrect is that the soundtrack is an essential part of the creative expression in the film. The relationship between program and library is not one of creative expression, it is one of engineering use. Net-SNMP, for instance, employs the OpenSSL programming interface in order to incorporate not the expressive content of OpenSSL but its functionality. That's not the relationship between a film and its soundtrack, it's the relationship between a film and the sprockets in the projector. Use is not my concern; distribution is. Certainly, a user may combine Net-SNMP and OpenSSL in any way they want; even the GPL allows that. But (again, statically, so as to be entirely clear) when somebody puts Net-SNMP and OpenSSL together (or, to be more precise, OpenSSL and the Perl extension libraries) into a single binary and distributes the result, does that person have any copyright obligations to the Perl, Net-SNMP, and/or OpenSSL authors? Furthermore, it's my sense that copyright holders have all kinds of discretionary power in restricting distribution of their works by others. In what way is this power restricted when it comes to the dynamic linking question? The FSF legal team are very much interested parties, and are on record as stating that the GPL is more useful as a tool for subverting the dominant paradigm of copyright than as an actual legal document. They have never, to my knowledge, cited any legal precedent more recent than the 1710 Statute of Anne (whose content they misrepresent) in support of the contentions in the FSF FAQ. I am no lawyer, but I will cheerfully walk you through the statutory, historical, and judicial support for my perspective in mind-numbing detail. Judge for yourself. Of course, the FSF has had a relatively successful run at spreading their meme throughout the legal community. The point was not that they are pure in some vague moral sense, but that they have well-established reputations and a long history of getting various portions of the legal community to see things their way. Thus, looking at things from an authority perspective, you can see the force behind their arguments. You are not forced to accept arguments from authority. There's a primary literature (appellate court decisions) that you can check for yourself (through FindLaw, for instance). You can verify that an argument's citations to the primary literature are honest and accurate representations of the thrust of the case. You can easily find similar cases that aren't cited in the argument, and see if they lead to similar conclusions, and verify which cases are frequently cited in later opinions. None of this takes a law degree, only some research skills and a passing familiarity with the legal lingo. Except that I have already done this. Arguably, I have not done it enough. But have you? How would I, or anyone else, know? Real legal degrees might be one way to vouch for the completeness of one's knowledge, but the FSF has those. Prestige in the legal community might be another, but the FSF has that, too. Success in negotiation? Also covered. Success in the courtroom? No one has that to any significant degree, and the lack of cases speaks to the effectiveness of negotiation, another point in the FSF's favor. Perhaps we could get Eben Moglen to duke it out with you here. If you can pull it off, more power to you. Until then, what can I rely on? -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of
Re: LGPL module linked with a GPL lib
On 7/27/05, Jeff Licquia [EMAIL PROTECTED] wrote: Does such compilation in itself give Debian any rights on its own, or is the compilation seen as non-copyrightable? The collective work (special case of compilation) that is a Debian CD is copyrightable. The copyright covers the creative expression that goes into selecting and arranging its contents. I would agree that hello.c is unencumbered with regards to its copyright. But the hello executable created through compilation (statically, to be clear) certainly has some kind of copyright obligation with regard to the libraries contained within it, doesn't it? Perhaps derived work is not the right term, but I find it thoroughly incredible that the copyright of the library is irrelevant with regards to the executable. The statically compiled executable contains a copy of (non-trivial parts of) the C library, and of course copying and distributing those bits requires license from the copyright holder. And, if so, then by the terms of the GPL, the whole of the executable must be itself covered by the terms of the GPL if any component part is. Nope. The GPL is written with passim references to works based on the Program, which is defined in Section 0 in terms of the copyright law meaning of derivative work. I'm not going to rehash the whole detailed parsing of the GPL language here (see the d-l archives or ask for a copy of my write-up), but to the extent that the drafters intended to forbid the (uncopyrightable) combination of GPL library X and library-using program Y, they failed to do so. (IANAL, TINLA.) Well, yes, in the case of glibc. In fact, one of the possible solutions to the KDE/Qt problem involved reimplementing Qt under the GPL, at which point the proprietary Qt would not be a prerequisite for KDE. The existence of an alternate implementation of a given published library interface certainly disposes of any question of whether a program is a derivative work of either implementation; but this is by no means necessary for the program to be non-infringing. So does this mean that libraries cannot impose restrictions on the programs that use them? I think far more groups than the free software community rely on this being the case; see the license agreement on any Microsoft SDK for an example. Whether or not that agreement purports to bind a developer in ways that copyright law does not, there are limits to what terms a court will permit in a contract of adhesion. The license terms might create a cause of action for breach of contract, but that's a very different animal from copyright infringement. If you know of a case more applicable than Lexmark v. Static Control and Specht v. Netscape, I'd be very interested to hear about it. It would have been good for someone to have made those arguments at the time. Why they did not do so is a mystery. (Of course, those arguments could have been made, and missed by me, in which case cites for those arguments would be worthwhile. Obviously, the cites were missed by far more people than me.) How many participants in the KDE/Qt brouhaha actually cited relevant case law? In any case, there's a perfectly good argument that for Debian to piss off the FSF is not a good idea whether or not they have a legal leg to stand on. I personally would be ashamed to lend my good name to their conduct in recent years, but YMMV. Cheers, - Michael
Re: LGPL module linked with a GPL lib
On 7/27/05, Jeff Licquia [EMAIL PROTECTED] wrote: Excuse me for asking, but why is this monograph not freely available? Surely, as a non-lawyer, you have no hope of profiting from it, and having a succint, linkable statement of your arguments would do wonders for preventing such go-arounds as this one, and might even persuade laymen like me. It's freely available by request, but not published, for several reasons. One, I don't want to risk detracting significantly from anyone's reputation in a way that isn't very firmly grounded in fact, and so I prefer to gather as much criticism as I can privately before presenting any conclusions in any form more official-looking than a post to a random discussion list (where counter-arguments by other people are only a click away). Two, a couple of lawyers and law students of my acquaintance have suggested (rather to my surprise) that it might be law journal material, and academic journals like to be the first actual publisher of a given paper. (There's no profit for me in that; if law journals operate anything like scientific journals, they'd probably charge me for the privilege; but it's kind of cool to have a publication in a peer-reviewed journal in any field, and if I ever want to go to law school it can't hurt.) Three, the paper (in LyX or PDF form) is inappropriately large for distribution using a mailing list archive's resources, and I haven't gotten around to resurrecting any of the random places where I have hosted content in the past. :-) Use is not my concern; distribution is. Certainly, a user may combine Net-SNMP and OpenSSL in any way they want; even the GPL allows that. But (again, statically, so as to be entirely clear) when somebody puts Net-SNMP and OpenSSL together (or, to be more precise, OpenSSL and the Perl extension libraries) into a single binary and distributes the result, does that person have any copyright obligations to the Perl, Net-SNMP, and/or OpenSSL authors? As far as I can tell (IANAL, TINLA), they have the same obligations that they would have if they shipped printouts of the three projects' source code in the same box. Copyright is about creative expression -- literal, human-readable text, images, and so forth, plus a limited extension towards the non-literal that is historically applied to translations to another human language, adaptations to another dramatic form, sequels using the same characters and mise en scene, etc. It's not about interlocking functionality or about the idea content of a work. A copyright license is of course a creature of contract law, despite the nonsense about copyright-based licenses that the FSF would have you believe. The GPL, in particular, is a perfectly valid offer of bilateral contract, and as such can condition the grant of license on whatever return performance is allowable in any other contract of adhesion under the law in a given jurisdiction (in the US, mostly state-level civil and commercial code). But its strident claims to divide the permitted from the forbidden exclusively using copyright criteria do have some legal effect -- the offeree is on solid ground when arguing that copyright law as judicially interpreted trumps erroneous paraphrases in the GPL text and (IMHO) misrepresentations on the FSF's web site. Furthermore, it's my sense that copyright holders have all kinds of discretionary power in restricting distribution of their works by others. In what way is this power restricted when it comes to the dynamic linking question? That discretionary power is channeled entirely through the terms of a license agreement (as modified by statutory overrides such as the 17 USC 203 termination provisions), and power over distribution is heavily curtailed by the doctrine of first sale. That's not of any great relevance to the GPL -- there are much stronger arguments against the hypothetical dynamic-linking ban based on the GPL text itself -- but it's useful background material. Of course, the FSF has had a relatively successful run at spreading their meme throughout the legal community. The point was not that they are pure in some vague moral sense, but that they have well-established reputations and a long history of getting various portions of the legal community to see things their way. Thus, looking at things from an authority perspective, you can see the force behind their arguments. The legal community has been pretty resistant to this meme, as far as I can tell. Programmers and journalists are another story. There's a delicious romantic storyline about Robin Stallman Hood and Friar Eben Tuck, and they have the advantage of what I think could fairly be called a monomaniacal dedication to spreading that storyline over the past 15 years or so. Although I have no personal knowledge on the financial side, it certainly looks to me like it has made them both rich men. Little snippets in the public record -- Jim Blandy's comment at
Re: LGPL module linked with a GPL lib
On Mon, Jul 25, 2005 at 09:17:25AM -0500, Jeff Licquia wrote: On Mon, 2005-07-25 at 11:59 +0200, Loïc Minier wrote: GStreamer's build process builds separate binaries for the various plugins, these are then dlopened when requested. I would personnally think that installing only Debian's GStreamer packages that are linked to LGPL libraries doesn't make your GStreamer installation / packages GPL (that is the build process has nothing to do with the resulting packages). I would even thing that installing GStreamer plugins packages which link to GPL libraries don't make your installation nor your running GStreamer applications GPL (that is only dlopening() something GPL makes the whole program in memory GPL, while it remains in memory). In a technical sense, you're right, in that each binary retains its separate copyright status. Most people, however, are concerned about the restrictions effectively placed on them more than about the specific status of any particular binary. I think it'd be a stretch to say that this prohibits proprietary gstreamer plugins, but I doubt you could *include* them in a gstreamer distribution. Third-party ones should still be okay. I see two ways in which this practically effects people using Debian. One, Debian could decide to package a plugin linking to a free but GPL-incompatible library, such as OpenSSL. Two, others might want to add a few proprietary plugins on top of Debian and distribute the result. So, I'd say that the former is probably prohibited and the latter is probably allowed. Incorporating proprietary plugins into Debian is somewhere around the borderline case, but I can't see that ever being an issue. This seems worth mentioning in the copyright file, even if the license itself doesn't change. Yeah, as far as the above goes. -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | `. `' | `- -- | signature.asc Description: Digital signature
Re: LGPL module linked with a GPL lib
Hi, On Mon, Jul 25, 2005, Jeff Licquia wrote: From the GPL: Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted... So the particular details of how things are distributed in memory while running aren't directly relevant. Modification and distribution are what matters, and it's clear from looking at the packages that GStreamer is distributed in Debian in conjunction with GPLed bits in a manner that's more than mere aggregation. I'm not sure to understand: you mean that since some LGPL GStreamer plugins are shipped in Debian along with GPL packages and they can play together means that the whole is GPLed? Would it be ok to have a copyright file along these lines: The source code for all plugins in the GStreamer Plugins source package is licensed under the LGPL, however some plugins are built with the help of header files from GPL libraries, and will be linked to GPL libraries when loaded in memory. Thus, using these plugins will switch their license to GPL, and you can only use them in applications with a license compatible with the GPL. You should have received a copy of the GPL and LGPL licenses ... Is a list of plugins necessary? I guess it's up to the interested person to check, nowadays it's relatively easy with tags and Debian's copyright files, and I don't want to maintain such a list. Bye, -- Loïc Minier [EMAIL PROTECTED] Come, your destiny awaits! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
** Loïc Minier :: Hi, On Mon, Jul 25, 2005, Jeff Licquia wrote: From the GPL: Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted... So the particular details of how things are distributed in memory while running aren't directly relevant. Modification and distribution are what matters, and it's clear from looking at the packages that GStreamer is distributed in Debian in conjunction with GPLed bits in a manner that's more than mere aggregation. I'm not sure to understand: you mean that since some LGPL GStreamer plugins are shipped in Debian along with GPL packages and they can play together means that the whole is GPLed? Would it be ok to have a copyright file along these lines: The source code for all plugins in the GStreamer Plugins source package is licensed under the LGPL, however some plugins are built with the help of header files from GPL libraries, and will be linked to GPL libraries when loaded in memory. Thus, using these plugins will switch their license to GPL, and you can only use them in applications with a license compatible with the GPL. You should have received a copy of the GPL and LGPL licenses ... Is a list of plugins necessary? I guess it's up to the interested person to check, nowadays it's relatively easy with tags and Debian's copyright files, and I don't want to maintain such a list. I find this discussion ultimately absurd. Debian is *not* distributing a derivative work. Debian does *not* distribute a work that includes both plugins/libraries. The fact that the things are (dynamically) linked at run time, especially combined with the fact that the plugins are opened with dlopen() and use stable API, is *more* than enough to lift any (inexistent IMHO) no-link requirement of the GPL. Please don't do that. -- HTH, Massa
Re: LGPL module linked with a GPL lib
Jeff Licquia writes: On Tue, 2005-07-26 at 11:14 -0300, Humberto Massa Guimarães wrote: I find this discussion ultimately absurd. Debian is *not* distributing a derivative work. Debian does *not* distribute a work that includes both plugins/libraries. The fact that the things are (dynamically) linked at run time, especially combined with the fact that the plugins are opened with dlopen() and use stable API, is *more* than enough to lift any (inexistent IMHO) no-link requirement of the GPL. I find most of this response confusing. First of all, it's clear that Debian *is* distributing a derived work based on GPLed libraries, called Debian GNU/Linux. The specific case in question may fall under the mere aggregation clause of the GPL, but then this is the point you should argue. US copyright law distinguishes between the classes of work called derivative works and compilations. The usual reading is that a compilation or collective work (which is a subset of compilation) is what the GPL means by mere aggregation, and a derivative work is covered by the stricter terms. The Berne Convention makes a similar distinction of a collection versus the works that comprise the collection. A compilation or collective work under US law is not necessarily a derivative work of any of its components. The GPL's use of derivative and derived is fuzzy in this sense, which is one reason the terms from copyright law are used more often than the GPL's terms. Michael Poole
Re: LGPL module linked with a GPL lib
On 7/26/05, Michael Poole [EMAIL PROTECTED] wrote: [snip] A compilation or collective work under US law is not necessarily a derivative work of any of its components. The GPL's use of derivative and derived is fuzzy in this sense, which is one reason the terms from copyright law are used more often than the GPL's terms. Almost -- a compilation or collective work is almost _never_ a derivative work of any of its components. The GPL drafter just plain got it wrong in Section 0, and the legal definition in 17 USC 101 (and its parallels in other Berne Convention countries) overrides the GPL's incorrect paraphrase. Extensively discussed on debian-legal in the last few months (disclaimer: only those few d-l participants with actual legal credentials seem to agree with me); you might start at http://lists.debian.org/debian-legal/2005/07/msg00336.html . Cheers, - Michael (IANAL, TINLA)
Re: LGPL module linked with a GPL lib
I wrote: ... only those few d-l participants with actual legal credentials seem to agree with me ... Er, that overreaches a bit in both directions; sorry. I'm more strident on the topic than the people with credentials are, and there are certainly other d-l regulars who question the FSF FAQ's stance on the matter. I just meant to say that I seem to be in a minority but it's a minority in which I'm comfortable. :-) Cheers, - Michael
Re: LGPL module linked with a GPL lib
Hi, I agree with most of what you said, except I'd like clarification on this part: On Sun, Jul 24, 2005, Jeff Licquia wrote: The copyright of the rest of GStreamer depends on how it's distributed. In Debian, it's clear that GStreamer is distributed with MAD support, which makes its effective license the GPL. However, someone interested in distributing proprietary plugins or apps for GStreamer (as part of a derivative, for example) could do so by removing the GPL plugins from the distribution before adding the non-free bits. This wouldn't even require a recompile to do. GStreamer's build process builds separate binaries for the various plugins, these are then dlopened when requested. I would personnally think that installing only Debian's GStreamer packages that are linked to LGPL libraries doesn't make your GStreamer installation / packages GPL (that is the build process has nothing to do with the resulting packages). I would even thing that installing GStreamer plugins packages which link to GPL libraries don't make your installation nor your running GStreamer applications GPL (that is only dlopening() something GPL makes the whole program in memory GPL, while it remains in memory). Is that correct? Bye, -- Loïc Minier [EMAIL PROTECTED] Come, your destiny awaits! -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: LGPL module linked with a GPL lib
On Mon, 2005-07-25 at 11:59 +0200, Loïc Minier wrote: GStreamer's build process builds separate binaries for the various plugins, these are then dlopened when requested. I would personnally think that installing only Debian's GStreamer packages that are linked to LGPL libraries doesn't make your GStreamer installation / packages GPL (that is the build process has nothing to do with the resulting packages). I would even thing that installing GStreamer plugins packages which link to GPL libraries don't make your installation nor your running GStreamer applications GPL (that is only dlopening() something GPL makes the whole program in memory GPL, while it remains in memory). In a technical sense, you're right, in that each binary retains its separate copyright status. Most people, however, are concerned about the restrictions effectively placed on them more than about the specific status of any particular binary. From the GPL: Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted... So the particular details of how things are distributed in memory while running aren't directly relevant. Modification and distribution are what matters, and it's clear from looking at the packages that GStreamer is distributed in Debian in conjunction with GPLed bits in a manner that's more than mere aggregation. I see two ways in which this practically effects people using Debian. One, Debian could decide to package a plugin linking to a free but GPL-incompatible library, such as OpenSSL. Two, others might want to add a few proprietary plugins on top of Debian and distribute the result. This seems worth mentioning in the copyright file, even if the license itself doesn't change.
Re: LGPL module linked with a GPL lib
On Sun, 2005-07-24 at 20:50 +0200, Loïc Minier wrote: The GStreamer suite ships a lot of plugins which are dlopened() when needed. Some of them link with GPL libraries. I received a bug report (#317129) to change the copyright files of libgstreamer0.8-0 and gstreamer0.8-mad to GPL. The upstream README mentions the situation, so I think I will mention it in the README.Debian with the next upload, but is the copyright supposed to reflect this? Does the whole distribution switch to GPL? I believe not, but would like a confirmation. The copyright of all the source code is independent of its dependencies. Thus, the license of the source is the LGPL, full stop. The copyright of the plugin binary itself is affected by dynamically linked libraries, since it's the combined work that's involved. I would expect the MAD plugin binary is effectively the GPL, since it cannot function without the MAD library. Still, technically, the binary has its own license, which is only required to be GPL-compatible (as the LGPL is). The copyright of the rest of GStreamer depends on how it's distributed. In Debian, it's clear that GStreamer is distributed with MAD support, which makes its effective license the GPL. However, someone interested in distributing proprietary plugins or apps for GStreamer (as part of a derivative, for example) could do so by removing the GPL plugins from the distribution before adding the non-free bits. This wouldn't even require a recompile to do. Unless, of course, Debian elects to exercise the upgrade-to-GPL clause in its particular copies of GStreamer. As I understand it, you're not asking about what's allowed, but what's required, so this isn't relevant. All this probably warrants mention in the copyright file, but overall I do not think it's accurate to say that the license, effective or otherwise, for GStreamer is the GPL. There are too many exceptions. IANAL, TINLA, etc.
