Alexander Terekhov wrote: > > [... "derived work" (i.e. "derivative work" under GNU law) ...] > > I suppose that id "lrosen" belongs to http://www.rosenlaw.com/rosen.htm. > > Nice to see both Hollaar and Rosen commenting GNU legal nonsense > version three. (Note that the GPLv2 contains the same GNU definition of > "derivative work".)
Interesting things are going on over there at gplv3.fsf.org. http://gplv3.fsf.org/comments/rt/readsay.html?Query=%20Creator%20=%20'lrosen'%20%20AND%20'CF.NoteUrl'%20LIKE%20'gplv3-draft-1'%20&Order=DESC&OrderBy=id&Rows= It appears that Rosen was (?is?) on the Committee A. He identified a bunch of issues (including his comments and a bunch of comments made by others) and claimed them for the Committee A. Now, just a few days later someone "fontana" downgraded and removed all that stuff from docket for Committee A. I suppose that "fontana" belongs to Moglen's underling at SFLC Richard Fontana. http://www.softwarefreedom.org/team.html I knew that gplv3 process was destined to deliver first class circus... and it turns out to be just stunning. ;-) regards, alexander. ------ Comment 641: Incompatibility and attorney's fees Regarding the text: your terms may add limited kinds of additional requirements on your added parts In section: gpl3.licensecompat.p1.s1 Submitted by: lrosen on 2006-01-23 at 21:08 EST Comment noted by lrosen on 2006-01-23 at 21:08 EST: Eben reminded me to ask about this: Does the fact that another license has an attorney's fees provision automatically make it incompatible with GPLv3? I would much rather that GPLv3 were merely incompatible with "licenses that contradict the terms of this GPLv3 license for this work," and leave out all the other list of "limited kinds of additional requirements" below. Otherwise, we'll be arguing about various kinds of license compatibility provisions forever. Claimed for Committee A by lrosen on 2006-01-23 at 21:09 EST Identified as an Issue by lrosen on 2006-01-23 at 21:09 EST Issue downgraded, moved back to Inbox by fontana on 2006-01-25 at 17:24 EST Remove from docket for Committee A by fontana on 2006-01-25 at 17:24 EST Comment 639: Contemplated by whom? Regarding the text: reasonably contemplated use of the covered work In section: gpl3.licensingpatents.p0.s2 Submitted by: lrosen on 2006-01-23 at 20:13 EST Comment noted by lrosen on 2006-01-23 at 20:13 EST: What is the reach of the "reasonably contemplated" provision? Contemplated by whom? As of what date? Must the contemplation be written somewhere? Most patent licenses are limited to claims that are "necessarily infringed by" or "embodied in" the software as delivered. Anything more is risky for holders of large and diverse patent portfolios. Claimed for Committee A by lrosen on 2006-01-23 at 20:17 EST Identified as an Issue by lrosen on 2006-01-23 at 20:17 EST Issue downgraded, moved back to Inbox by fontana on 2006-01-25 at 17:24 EST Remove from docket for Committee A by fontana on 2006-01-25 at 17:24 EST Comment 638: True, but so what? Regarding the text: Not a Contract. In section: notacontract.0.0 Submitted by: lrosen on 2006-01-23 at 18:58 EST Comment noted by lrosen on 2006-01-23 at 18:58 EST: NONE of the free and open source licenses are contracts without the external formalities of offer, acceptance and consideration, and the GPLv3 IS a contract when those external formalities are undertaken--despite what the GPLv3 license says. The fact that some licenses memorialize the contract formation externalities doesn't actually make them contracts; it is the formalities themselves that do. Each licensor decides for him/herself whether to require formalities; the license author cannot decide that on anyone else's behalf. There are many advantages to forming a contract, including the opportunity for the licensor to seek contract remedies such as specific performance. There are no disadvantes to forming a contract with the GPLv3, because the license terms are still enforceable under either contract law or copyright law. If a licensor seeks to enforce the GPLv3 under copyright law rather than contract law, the license then is merely a potential defense to a claim of infringement. Contract law is the same, except more alternative remedies exist for licensors besides statutory or actual damages, or injunction. Why isn't that something good we should encourage, rather than discourage them with factually inaccurate phrases such as "Not a Contract"? Claimed for Committee A by lrosen on 2006-01-23 at 20:16 EST Identified as an Issue by lrosen on 2006-01-23 at 20:16 EST Issue downgraded, moved back to Inbox by fontana on 2006-01-25 at 17:24 EST Remove from docket for Committee A by fontana on 2006-01-25 at 17:24 EST Child comment of 638: Comment 259: Should this be removed? Regarding the text: Therefore, by modifying or propagating the Program (or any covered work), you indicate your acceptance of this License to do so, and all its terms and conditions. In section: gpl3.notacontract.p0.s5 Submitted by: joe on 2006-01-17 at 02:25 EST Comment noted by joe on 2006-01-17 at 02:25 EST: Rational: (1) It is probably unnecessary. (2) It could be interpreted by some as turning the GNU GPL into a (unilateral?) contract (yes, even though it is under "NOT A CONTRACT" in big bold letters). If this sentence (and others) is merely supposed to explain copyright law or what is a conclusion of the rest of the license, maybe such sentences should be highlighted in a way that makes it clear that they are non-normative (like the preamble) on the basis that it is helpful to have explanations for the layman without the danger that they be misinterpreted as part of the licensing terms or add to the complexity of those terms for readers or users of the GNU GPL. Comment noted by verdyph on 2006-01-17 at 15:32 EST: unilateral contracts are illegal in many countries. The obligations and rights from one party MUST be reasonnably be balanced by obligations and rightsbythe other party. As the GPL gives obligations and rights to the licencee, the full exemption of obligations in the GPL is illegal. In my opinion, this is the WEAKEST aspect of the GPL, meaning that it insufficiently lawfully protects the rights of authors and users face to other threats like patent claims. Please enforce the GPL so that it becomes a legal contract with the complete protection of laws and international treaties (notably at WIPO, WTO, and the related international treaties like Conventions of Berne, Paris and Washington), which can be easily defended face to a court (at least in all OECD countries and in the European Union, including the Internation Court of Human Rights, and also used as a valid licence to cover the texts and software applications used in governmental and international institutions for the public interest). Made a sub-comment of ticket #638 by lrosen on 2006-01-23 at 20:46 EST Claimed for Committee A by lrosen on 2006-01-23 at 20:46 EST Child comment of 638: Issue 394: Internationalization: copyright licenses are contracts in some jurisdictions Regarding the text: Not a Contract In section: notacontract.0.0 Submitted by: oliva on 2006-01-17 at 16:37 EST Comment noted by oliva on 2006-01-17 at 16:37 EST: A Brazilian lawyer once told me that a copyright license is indeed a contract in Brazil. It might be a specific kind of unilateral contract whose name escapes me now, but it is a contract nevertheless. Comment noted by neroden on 2006-01-18 at 18:20 EST: Change the name to "Acceptance by Actions" Made a sub-comment of ticket #638 by lrosen on 2006-01-23 at 20:44 EST Claimed for Committee A by lrosen on 2006-01-23 at 20:44 EST Identified as an Issue by lrosen on 2006-01-23 at 20:44 EST Child comment of 638: Comment 463: Not a contract in the U.S. - but others? Regarding the text: Not a Contract. In section: notacontract.0 Submitted by: dwheeler1 on 2006-01-17 at 23:12 EST Comment noted by dwheeler1 on 2006-01-17 at 23:12 EST: My understanding is that the GPL is _NOT_ a contract in the U.S. --- it's a license. But in other countries, I understand that copyright law is considered a subset of contract law. Is there a way to keep this concept, yet internationalize it? If nothing else, rename this to "This is a license, not a contract"; otherwise, in countries where copyrights ARE contracts, "not a contract" seems self-contradictory. Comment noted by pde on 2006-01-23 at 14:42 EST: As I understand it, the GPL is not a contract, but an *offer* of a contract. Once accepted, in the U.S., it is a contract. The thing that makes the GPLv3 draft unusual amongst contracts is that litigation following an irreconcilable dispute would most likely be for copyright infringement, not breach of contract. The case would be fought as breach of contract only if both parties preferred it that way or if unusual circumstances forced at least one party onto that ground. GPLv2, with automatic termination, was even more strongly biased in favour of dispute settlement under copyright law. Made a sub-comment of ticket #638 by lrosen on 2006-01-23 at 20:45 EST Claimed for Committee A by lrosen on 2006-01-23 at 20:45 EST Child comment of 638: Comment 528: This heading does not describe the content of the section Regarding the text: Not a Contract. In section: notacontract.0.0 Submitted by: stone on 2006-01-18 at 23:04 EST Comment noted by stone on 2006-01-18 at 23:04 EST: Unless the text of the section expressly states that this License is not a contract, I don't see the point of implicitly making this claim in the section header. Perhaps `Acceptance as a Condition of Modification or Propagation' would be better. Comment noted by terjebr on 2006-01-19 at 03:34 EST: I think "Acceptance of Linsence" is better. Anyway, I agree that "Not a Contract" is a bad heading. Either it is, or it is not, and that cannot then be changed by a heading. Even in some countries copyright law is so that it must be regarded as a contract, or it is not valid. Then in the US it is best if it is not regarded as a contract. So it is best for all parties if the License itself do not try to address the issue. Comment noted by pde on 2006-01-23 at 14:23 EST: Spelling aside, I think terjebr has a point :). Made a sub-comment of ticket #638 by lrosen on 2006-01-23 at 20:45 EST Claimed for Committee A by lrosen on 2006-01-23 at 20:45 EST Child comment of 638: Comment 628: Consideration Regarding the text: Not a Contract In section: notacontract.0.0 Submitted by: weel on 2006-01-22 at 21:09 EST Comment noted by weel on 2006-01-22 at 21:09 EST: Seeing as the GPL is not a contract, wouldn't it, at least in common law jurisdictions, be revokable? If so, are there any known workarounds? I understand that there was at some point the idea that you might sell a license to the FSF for some small amount of consideration, thereby giving the FSF the rights stated in the GPL, but irrevokably so. The FSF presumably can be trusted not to revoke licenses. Is such a scheme made harder by the fact that the license explicitly states that it is not a contract? [Note: I submitted something like this earlier, but somehow it doesn't show up.] Made a sub-comment of ticket #638 by lrosen on 2006-01-23 at 20:55 EST Claimed for Committee A by lrosen on 2006-01-23 at 20:55 EST Comment 637: Derivative works Regarding the text: that is to say, a work containing the Program or a portion of it, either modified or unmodified. In section: gpl3.definitions.p0.s2 Submitted by: lrosen on 2006-01-23 at 18:39 EST Comment noted by lrosen on 2006-01-23 at 18:39 EST: The statement beginning "that is to say..." is not an accurate description of "derivative works" under US copyright law. If you want the copyleft provisions of GPLv3 to apply to "collective works" then you should say so explicitly, rather than use language reminiscent of the definition of collective works when trying to describe derivative works. The current draft, in this respect, is both ambiguous and potentially very misleading. See 17 USC 101. Claimed for Committee A by lrosen on 2006-01-23 at 20:17 EST Identified as an Issue by lrosen on 2006-01-23 at 20:17 EST Issue downgraded, moved back to Inbox by fontana on 2006-01-25 at 17:24 EST Remove from docket for Committee A by fontana on 2006-01-25 at 17:24 EST Child comment of 637: Issue 372: still unclear Regarding the text: that is to say, a work containing the Program In section: gpl3.definitions.p0.s2 Submitted by: manuell on 2006-01-17 at 12:43 EST Comment noted by manuell on 2006-01-17 at 12:43 EST: In short, a linux distro contains a GPL'ed kernel but is not a 'derivative' of that kernel. Why keep the 'that is to say...' text ? (confusion beetween 'derivative' and 'compilation') Comment noted by neroden on 2006-01-18 at 17:54 EST: Yes. Drop this, it confuses people. Comment noted by mhoyles on 2006-01-19 at 11:33 EST: Count me among the confused. I don't know if this is intended to broaden the meaning of derivative work (to include collections of software) or if it is mearly a (misleading?) explaination of the legal term derivative work. It might be better to seperate explainations, expressions of intent, etc. from the definitions to avoid ambiguity. Made a sub-comment of ticket #637 by lrosen on 2006-01-23 at 20:39 EST Claimed for Committee A by lrosen on 2006-01-23 at 20:39 EST Identified as an Issue by lrosen on 2006-01-23 at 20:39 EST Child comment of 637: Issue 466: Can we define 'derivative work' more clearly? Regarding the text: derivative or collective works based on the Program. In section: gpl3.distribmod.p5.s1 Submitted by: tungsten on 2006-01-17 at 23:27 EST Comment noted by tungsten on 2006-01-17 at 23:27 EST: I know the term 'derivative work' is a somewhat gray area of the law in the US at least, but I think this license should define "derivative work" in a precise way, so that a programmer reading this license can decide whether his code counts as a derivative work. Right now he has to ask a lawyer for an opinion (which of course is not definitive), or go to court to settle the matter. Comment noted by swillden on 2006-01-18 at 15:18 EST: IANAL, but I don't think it's useful for the GPL to clarify "derivative work". The legal definition is fuzzy, so the GPL can't clarify it without providing a somewhat different definition... one that is either weaker or stronger in a given case. I don't think the GPL wants to weaken the meaning of "derivative work". If anything, it would better serve the purpose of Free Software to strengthen it a little, closing some common escape hatches (dynamic linking, etc.). But the GPL can't actually strengthen the definition of derivative work, because if a programmer's work is not derivative per the law, then the programmer doesn't need the GPL and any definitions it provides are moot. So if there's no legal force behind any clarification that strengthens the definition, and if we don't want a clarification that weakens the definition, it seems better to omit the clarification. Made a sub-comment of ticket #637 by lrosen on 2006-01-23 at 20:41 EST Claimed for Committee A by lrosen on 2006-01-23 at 20:41 EST Identified as an Issue by lrosen on 2006-01-23 at 20:41 EST ------ _______________________________________________ Gnu-misc-discuss mailing list Gnu-misc-discuss@gnu.org http://lists.gnu.org/mailman/listinfo/gnu-misc-discuss
