Re: apache license 2.0 for consideration
Roy T. Fielding wrote: That is a very complex set of issues. First, the patent is not only licensed under the ASL2 -- it is actually licensed by the contributor to the ASF and any recipient of the ASF's software as part of their contribution. The Apache License makes the recipient aware of that license and the condition of reciprocity (as accepted by our community of contributors) under which that license was granted. The ASF does not have the right to sublicense any such patent; we merely pass the license along from the contributors. except that the 2.0 apache licence can, by design, be used unaltered by other entities than the asf -- so does the above continue to make sense if 'ASF' is replaced by any other entity? i think it does. let's not lose sight of the fact that the licensor applying the licence to its work may be something other than the asf. -- #kenP-)} Ken Coar, Sanagendamgagwedweinini http://Ken.Coar.Org/ Author, developer, opinionist http://Apache-Server.Com/ Millennium hand and shrimp! -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Eben Moglen wrote: On Tuesday, 24 February 2004, Arnoud Engelfriet wrote: I'm not even sure the license still exists if you take out the Contribution I made (embodying my patented method) and put it in some other work. In that case there would no mystery about the FSF position. If that's the right interpretation of the patent grant, then ASL2 isn't a free software license at all. The ASL2 says in section 3 that each Contributor gives everyone a patent license to make, use [etc] the Work. This is just like the GPL: the license is just for the work in question. Licensed are those patent claims that are necessarily infringed by [the] Contribution(s) alone or by combination of their Contribution(s) with the Work to which such Contribution(s) was submitted. So, if I have a patent on web server software arranged to perform method X and I contribute code to do method X to Apache, everyone using Apache has a license to do method X with my code and Apache. Now, someone takes my contributed code and combines it with another web server (or in a completely different program). This is allowed under the copyright license. But my patent license was _only_ to make, use [etc] ... *the Work*. The definition of Work in the ASL2 seems quite limited. The separate definition of Derivative Work and the constant use of the Work or Derivative Works thereof in the license makes it clear to me that Work and Derivative Work are different things. So apparently Derivative Works are not covered by the patent license! Perhaps it would be better to have contributors grant a license to make, use, sell etc. their _Contribution_ and combinations of their Contribution with the Work or Derivative Works thereof. Arnoud -- Arnoud Engelfriet, Dutch patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Roy T. Fielding wrote: On a side note, since software patent law is applied to the method of something and not to the particular expression, a patent license for doing that something remains in force regardless of the software that is later used to do it. The license is from the owner of the method to the legal entity using that method. Correct, although a patent license grant may very well be limited to one particular application. For example, I could license you under my patent to practice a method using only the software I provide to you. In other words, it is a blanket permission -- once you have the permission, you can use whatever tool you like (even one not derived from the ASL2 work) up until the permission is revoked. The ASL2 grants in clause 3 a patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer the Work. If I Contribute software for one of my patented methods to ASL2- licensed code, you have the right to use *that code* to practice my patented method. You do not have permission to use your own software to practice my patented method. In that case you are not using the Work. If a company sues for infringement on the basis of a patent being included in XY, where XY consists of X (non-infringing) and Y (infringing), then that will be brought up by the defense and the company will have to claim Y infringes as well (or drop the case entirely). As such, there is no need for the patent license to talk about derivative works. Nor would it be safe to do so, since derivative work is a concept of copyright law, not patent law. I'm not even sure the license still exists if you take out the Contribution I made (embodying my patented method) and put it in some other work. Arnoud -- Arnoud Engelfriet, Dutch patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Arnoud Engelfriet scripsit: I'm not even sure the license still exists if you take out the Contribution I made (embodying my patented method) and put it in some other work. It's hard to say, certainly. But consider this case: I have patented a gear, and I give you a patent license to make use of this gear. Now you build a machine one part of which is the gear, and I sue, claiming that you were licensed only to use the gear, not the machine of which the gear is a part. Surely you would reply that the very essence of a gear is to be used as a part of something. And likewise with a software library. IANAL, TINLA. -- He made the Legislature meet at one-horse John Cowan tank-towns out in the alfalfa belt, so that [EMAIL PROTECTED] hardly nobody could get there and most of http://www.reutershealth.com the leaders would stay home and let him go http://www.ccil.org/~cowan to work and do things as he pleased.--Mencken, Declaration of Independence -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
On Tuesday, 24 February 2004, Arnoud Engelfriet wrote: If a company sues for infringement on the basis of a patent being included in XY, where XY consists of X (non-infringing) and Y (infringing), then that will be brought up by the defense and the company will have to claim Y infringes as well (or drop the case entirely). As such, there is no need for the patent license to talk about derivative works. Nor would it be safe to do so, since derivative work is a concept of copyright law, not patent law. I'm not even sure the license still exists if you take out the Contribution I made (embodying my patented method) and put it in some other work. In that case there would no mystery about the FSF position. If that's the right interpretation of the patent grant, then ASL2 isn't a free software license at all. E -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Roy, Since you forwarded me the beginning of your list's thread on ASL2 and GPL2 I have been preparing the analysis you quite rightly sought. I regret that the statement at gnu.org was, and is, inadequate to explain FSF's concerns; you have correctly inferred that I did not write it. Today's statement by the Apache Foundation raises a new question for me, about which I had assumed that careful reading of your license would provide an answer, but which I answered for myself--after such a reading--differently than the Apache Foundation statement. I hope we can clarify both our views quickly, to avoid the public row that seems (unnecessarily so far as I can see) to be developing. For this purpose, I need to ask the question raised by today's statement: A developer, X, adds GPL'd code to Apache, and distributes the combination. The combined code, including the GPL'd code itself, practices the teaching of a patent, P, licensed under ASL2. A user, Y, asserts a defensive patent claim of infringement by Apache. Is the license to practice patent P in the GPL'd code added to Apache by X withdrawn or in force? Is the license as to the ASL code combined with the GPL code withdrawn or in force? I have been assuming, on the basis of the license text, which seemed clear to me, that the answer is withdrawn/withdrawn. Your statement of today asserting GPL compatibility suggests that the answer must be in force/in force. Can you help? Thanks and best regards. Eben -- Eben Moglen voice: 212-854-8382 Professor of Lawfax: 212-854-7946 moglen@ Columbia Law School, 435 West 116th Street, NYC 10027 columbia.edu General Counsel, Free Software Foundation http://moglen.law.columbia.edu -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Eben Moglen scripsit: A developer, X, adds GPL'd code to Apache, and distributes the combination. The combined code, including the GPL'd code itself, practices the teaching of a patent, P, licensed under ASL2. A user, Y, asserts a defensive patent claim of infringement by Apache. Is the license to practice patent P in the GPL'd code added to Apache by X withdrawn or in force? Is the license as to the ASL code combined with the GPL code withdrawn or in force? I have been assuming, on the basis of the license text, which seemed clear to me, that the answer is withdrawn/withdrawn. Your statement of today asserting GPL compatibility suggests that the answer must be in force/in force. Can you help? I would point out that ASL2's clause 3 does not mention derivative works at all: it provides a patent license only for the Work, not for anyu Derivative Works licensed (under the terms of clause 4) under a different license. Since the Academic Free License 2.0 (http://www.opensource.org/licenses/afl-2.0.php) uses essentially the same language as the ASL 2.0, it would be useful if the FSF could re-evaluate its position on the AFL as well. The other objection, to the trademark clause, seems moot given the FSF's acceptance of the extremely similar trademarks clause of the ASL2. I ask as a friend of Larry Rosen's and as a developer of AFL-licensed software. -- A mosquito cried out in his pain, John Cowan A chemist has poisoned my brain! http://www.ccil.org/~cowan The cause of his sorrow http://www.reutershealth.com Was para-dichloro- [EMAIL PROTECTED] Diphenyltrichloroethane.(aka DDT) -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Hello Eben, Thanks for responding -- I do find your messages to be a great help in understanding license issues. On Monday, February 23, 2004, at 10:58 AM, Eben Moglen wrote: Since you forwarded me the beginning of your list's thread on ASL2 and GPL2 I have been preparing the analysis you quite rightly sought. I regret that the statement at gnu.org was, and is, inadequate to explain FSF's concerns; you have correctly inferred that I did not write it. Today's statement by the Apache Foundation raises a new question for me, about which I had assumed that careful reading of your license would provide an answer, but which I answered for myself--after such a reading--differently than the Apache Foundation statement. I'll assume you are referring to the statement at http://www.apache.org/licenses/GPL-compatibility.html I hope we can clarify both our views quickly, to avoid the public row that seems (unnecessarily so far as I can see) to be developing. For this purpose, I need to ask the question raised by today's statement: A developer, X, adds GPL'd code to Apache, and distributes the combination. The combined code, including the GPL'd code itself, practices the teaching of a patent, P, licensed under ASL2. A user, Y, asserts a defensive patent claim of infringement by Apache. Is the license to practice patent P in the GPL'd code added to Apache by X withdrawn or in force? Is the license as to the ASL code combined with the GPL code withdrawn or in force? I have been assuming, on the basis of the license text, which seemed clear to me, that the answer is withdrawn/withdrawn. Your statement of today asserting GPL compatibility suggests that the answer must be in force/in force. Can you help? That is a very complex set of issues. First, the patent is not only licensed under the ASL2 -- it is actually licensed by the contributor to the ASF and any recipient of the ASF's software as part of their contribution. The Apache License makes the recipient aware of that license and the condition of reciprocity (as accepted by our community of contributors) under which that license was granted. The ASF does not have the right to sublicense any such patent; we merely pass the license along from the contributors. Therefore, assuming that the patent was infringed by a contribution made by the owner of that patent to an ASL2 product, the license remains in force regardless of what other code/license terms are added to make up derivative works. The contributor's patent license is withdrawn from Y upon the claim of infringement (upon Y claiming that Y owns a patent that has not been licensed to Apache). Thus, your interpretation is mostly correct in that the answer to both of your questions is withdrawn from Y. A second issue, however, is that the GPL does not allow redistribution of the combination if a patent does exist and is not available free to everyone. Again, the ASL2 has no impact here, since a separate license to the patent may exist outside of the ASL2's required license. For example, if a company says this patent may be freely used within any product distributed under the GPL or the Apache License, then a user would still retain rights to use under the GPL even if the Apache License termination clause is triggered. That is because the termination is passive -- it only matters if the user has no rights aside from those given by the ASL2. So, there may be a case where a given software product may be incompatible with the GPL, whether or not it is licensed under the ASL2, simply because an enforced patent exists within that product and has not been licensed (as might be the case for patented technology that was independently invented for Apache, or for which the inventor failed to reference the existing prior art). I hope you agree that, in such a case, it is the product that is incompatible with the GPL and not the generic licensing terms. It is the GPL's own restrictions on redistributing works covered by an enforced patent that cause the combination to be non-distributable, and that is true even if the patent was infringed by the GPL part of the code (e.g., combining two independent products into one product may cause a new infringement even if neither one infringed separately). As we said in our statement, the ASF does not knowingly distribute patented technology unless it is licensed free for everyone (not just under the ASL2). The purpose of the patent license in ASL2 is to protect our users against deliberate contributions of patented technology that are later enforced against the end-user. Since nobody knows when a patent is infringed until the patent is approved and enforced, usually years after it was introduced, our contributors have agreed to work together on a royalty-free basis for anything they might contribute, provided that the people benefiting from that broad license are also willing to license their patents in the same work (or at least agree not to enforce them
Re: apache license 2.0 for consideration
I would point out that ASL2's clause 3 does not mention derivative works at all: it provides a patent license only for the Work, not for anyu Derivative Works licensed (under the terms of clause 4) under a different license. On a side note, since software patent law is applied to the method of something and not to the particular expression, a patent license for doing that something remains in force regardless of the software that is later used to do it. The license is from the owner of the method to the legal entity using that method. In other words, it is a blanket permission -- once you have the permission, you can use whatever tool you like (even one not derived from the ASL2 work) up until the permission is revoked. If a company sues for infringement on the basis of a patent being included in XY, where XY consists of X (non-infringing) and Y (infringing), then that will be brought up by the defense and the company will have to claim Y infringes as well (or drop the case entirely). As such, there is no need for the patent license to talk about derivative works. Nor would it be safe to do so, since derivative work is a concept of copyright law, not patent law. IANAL, so I'm not sure if that is codified somewhere or simply the collective experience of those I've talked to. Roy -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Alexander Terekhov scripsit: Are you saying that your license allows GPL-forking? I think that it does allow things like distribution of GPL'd patches... but the resulting/originating derivative works would fall under multiple licenses -- the GPL for modifications and the ASL for all the remaining portions from the original work. Well, no. There are three cases: A. Patches are not an original work of authorship (because they are too trivial): the resulting derivative work can be licensed in any way that's compatible with the original. B. Patches are an original work of authorship: B-1. The original work's license is GPL-compatible: the resulting derivative work is licensed under the GPL. B-2. The original work's license is not GPL-compatible: the patches cannot be applied, or at least the result cannot be distributed. (with respect to reciprocation) licenses like the OSL and the CPL (and also the GPL) say that patches must be distributed under the original license even if, apart from referencing, they do NOT contain anything related to (copied from) the original work. This is only so if patches are themselves derivative works of the original. IMHO (IANAL, TINLA) they are not, and any parts of the original that appear inside a patch (a unified diff, e.g.) constitutes fair use and/or de minimis use of the original. The stock example is that the _C Answer Book_ is not a derivative work of K R, even though it just contains answers to the questions in K R, and therefore is meaningless without it. -- Her he asked if O'Hare Doctor tidings sent from far John Cowan coast and she with grameful sigh him answered that www.ccil.org/~cowan O'Hare Doctor in heaven was. Sad was the man that word www.reutershealth.com to hear that him so heavied in bowels ruthful. All [EMAIL PROTECTED] she there told him, ruing death for friend so young, algate sore unwilling God's rightwiseness to withsay. _Ulysses_, Oxen -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Are you saying that your license allows GPL-forking? No, I am saying that the Apache License says: You may add Your own copyright statement to Your modifications and may provide additional or different license terms and conditions for use, reproduction, or distribution of Your modifications, or for any such Derivative Works as a whole, provided Your use, reproduction, and distribution of the Work otherwise complies with the conditions stated in this License. Note that our license applies to the Work, not the copyright on the derivative work, and thus any GPL code combined with Apache License code will remain under the GPL. Roy -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Why, then, is the MIT license compatible with the GPL? Because the MIT license is silent about patents; in and of itself, it can't do anything to require you to breach the GPL's licensing terms. (It may be that the word use provides an implied patent license.) A specific MIT-licensed program may be GPL-incompatible, but MIT-licensed programs as a class are not, because they don't impose any requirements incompatible with the GPL's. Neither does the Apache License. Please read the terms carefully: 3. Grant of Patent License. Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable (except as stated in this section) patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer the Work, where such license applies only to those patent claims licensable by such Contributor that are necessarily infringed by their Contribution(s) alone or by combination of their Contribution(s) with the Work to which such Contribution(s) was submitted. If You institute patent litigation against any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Work or a Contribution incorporated within the Work constitutes direct or contributory patent infringement, then any patent licenses granted to You under this License for that Work shall terminate as of the date such litigation is filed. Under no circumstance does this add a restriction that does not already exist within the GPL. The GPL prohibits distribution of a work that is covered by non-free patents. The Apache License says that any patent licenses granted to you by virtue of it being contributed to Apache go away if you claim there exists a non-free patent in the work. Section 4: You may add Your own copyright statement to Your modifications and may provide additional or different license terms and conditions for use, reproduction, or distribution of Your modifications, or for any such Derivative Works as a whole, provided Your use, reproduction, and distribution of the Work otherwise complies with the conditions stated in this License. In other words, any GPL code that is combined with Apache License code remains under the GPL, which is the sole requirement for compatibility. I have not seen Eben's analysis of the approved version of the Apache License, so I have no idea why the FSF site claims they are incompatible. It is simply claimed on the site that it adds an additional restriction, as if nobody bothered to read the terms. Cheers, Roy T. Fieldinghttp://roy.gbiv.com/ Chief Scientist, Day Software http://www.day.com/ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Because the MIT license is a blanket grant of permission, almost without restriction: That is completely irrelevant. Unlike copyright, a patent does not move along with the work. The patent may be owned by a completely separate company of which the author is totally unaware at the time of distribution. Aside from that, there are no implied licenses to third parties. Just because a person has received software from one company, indicating that the company wished that person to use the software (an implied grant), does not imply that the person can give the same software to a third person and the implied license along with it. The GPL has a stronger restriction here in order to prevent the first person from receiving the software as GPL in the first place; no such restriction is true of any other license. If you argue that such a restriction is necessary for compatibility with the GPL, then no license is compatible with the GPL (including itself, because the patent may be approved after the initial GPL distributions). The point I am making is that the Apache License does not have any additional restrictions over the GPL. A derivative work that is distributed under only the GPL license does satisfy every single condition or restriction in the Apache License. If a patent does show up and a lawsuit ensues, then for those people the Apache License becomes no more permissive than the GPL. Roy -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
On Thu, 2004-02-19 at 16:10, Roy T. Fielding wrote: Because the MIT license is a blanket grant of permission, almost without restriction: That is completely irrelevant. Unlike copyright, a patent does not move along with the work. I may not be following your meaning here. Assuming that a particular section of code incorporating a patent is distributed under the MIT license, and that same particular section exists in a derivative work of that code distributed under the GPL license, I would assume that in most cases that same section of code still incorporates the patent. I'm probably missing something here. (I am ignoring the case where the same code can infringe or not infringe a patent depending on how it's used or what it's used for. I'm guessing that's outside the scope of the argument.) The patent may be owned by a completely separate company of which the author is totally unaware at the time of distribution. That is true. For simplicity I intended to only refer to copyrights and patents that the contributor had rights to license, although I admit I didn't word that in my previous message as clearly as I should have: And I'm assuming of course that the licensor/contributors have full copyrights and patent rights to the code. Aside from that, there are no implied licenses to third parties. Just because a person has received software from one company, indicating that the company wished that person to use the software (an implied grant), does not imply that the person can give the same software to a third person and the implied license along with it. Absolutely incorrect: Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated ^^^ documentation files... ^^^ -- Mark Shewmaker [EMAIL PROTECTED] -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
On Thu, 2004-02-19 at 16:30, Roy T. Fielding wrote: The GPL prohibits distribution of a work that is covered by non-free patents. The Apache License says that any patent licenses granted to you by virtue of it being contributed to Apache go away if you claim there exists a non-free patent in the work. I think you're using the term non-free to mean two different things in two different sentences. Let me reword: :-) | The GPL prohibits distribution of a work that is | covered by patents not distributable under GPL terms. The Apache | License says that any patent | licenses granted to you by virtue of it being contributed to Apache | go away if you claim there exists a patent in the work that's not | distributable under Apache license terms. If patents not distributable under GPL terms == patents not distributable under Apache license terms, then I would agree that the Apache license doesn't add a restriction not already in the GPL. However, I claim that if there exist Apache-licensed patents that are not also GPL-licensed, then the Apache license is not universally GPL-compatible. I also claim that since the Apache license can retract Apache-patent-licenses for people making patent infringement claims, that that retraction would have to apply to people using Apache-GPL'd code. Then, since the retraction applies to someone using GPL'd code, it breaks GPL licensing for everyone using that GPL'd code. So, I would then worry that the Apache License's explicit don't-sue-or-you'll-lose-your-patent-license restriction would be a restriction above and beyond what the GPL by itself provides. (By the way, I think we're still making some progress here, but if we end up just repeating ourselves, I'm content to drop things and wait to hear from FSF for more info, or discuss this elsewhere. No sense going back and forth if we get to the point where we're merely repeating ourselves.) In other words, any GPL code that is combined with Apache License code remains under the GPL, which is the sole requirement for compatibility. If someone can pull back a patent license via the Apache license through an Apache-GPL'd work, then..it looks to me like Apache requirements are still holding, and that the work is not just limited by GPL requirements. -- Mark Shewmaker [EMAIL PROTECTED] -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Mark Shewmaker scripsit: I also claim that since the Apache license can retract Apache-patent-licenses for people making patent infringement claims, that that retraction would have to apply to people using Apache-GPL'd code. Then, since the retraction applies to someone using GPL'd code, it breaks GPL licensing for everyone using that GPL'd code. So, I would then worry that the Apache License's explicit don't-sue-or-you'll-lose-your-patent-license restriction would be a restriction above and beyond what the GPL by itself provides. I agree, and add: The GPL license crashes only if the patent actually exists and is enforceable. But a mere claim, however non-colorable, of a patent can destroy the Apache license for the claimant. (The same applies to the AFL, the OSL, and various other modern licenses.) -- John Cowan www.ccil.org/~cowan www.reutershealth.com [EMAIL PROTECTED] All isms should be wasms. --Abbie -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
I think you're using the term non-free to mean two different things in two different sentences. Nope. Let me reword: :-) | The GPL prohibits distribution of a work that is | covered by patents not distributable under GPL terms. The Apache | License says that any patent | licenses granted to you by virtue of it being contributed to Apache | go away if you claim there exists a patent in the work that's not | distributable under Apache license terms. If patents not distributable under GPL terms == patents not distributable under Apache license terms, then I would agree that the Apache license doesn't add a restriction not already in the GPL. Try software not distributable under GPL terms. Talking about distributing the patent doesn't make any sense, which is why you are getting confused over the permissions. However, I claim that if there exist Apache-licensed patents that are not also GPL-licensed, then the Apache license is not universally GPL-compatible. No, that is reversed logic. If there exist GPL-licensed software that would somehow be restricted by the Apache License restrictions, then they would not be compatible. I don't think it is reasonable to invent new definitions of compatibility. I also claim that since the Apache license can retract Apache-patent-licenses for people making patent infringement claims, that that retraction would have to apply to people using Apache-GPL'd code. Then, since the retraction applies to someone using GPL'd code, it breaks GPL licensing for everyone using that GPL'd code. As it would if the code was entirely GPL'd. The GPL says that you cannot redistribute software that is covered by a patent wherein the patent is not licensed free for everyone. The Apache License says that if you claim the Apache software contains something that is not licensed free for everyone (i.e., specifically, you accuse someone of infringing your patent which implies that your patent is not free for everyone), then you can't use the Apache License as a defense against your own infringements. In contrast, the GPL says you cannot redistribute the software at all. Therefore, the Apache License is less restrictive than the GPL. Furthermore, since the Apache License's patent grant does not apply to the GPL'd code in any way (only the Work which is licensed under the Apache License) and has no impact on distribution whatsoever, whether or not such licenses disappear has no influence on the code covered by the GPL. In other words, any GPL code that is combined with Apache License code remains under the GPL, which is the sole requirement for compatibility. If someone can pull back a patent license via the Apache license through an Apache-GPL'd work, then..it looks to me like Apache requirements are still holding, and that the work is not just limited by GPL requirements. GPL is a copyright license. It cannot prevent the possibility of a patent, and does not try to. It merely states that, if a patent is being enforced, then the code cannot be distributed as GPL. That is true regardless of what license covers the original code that was later combined with the GPL work. The Apache License does not state, or even imply, that a non-GPL-compliant patent exists on the code. Nor does a GPL Derived Work have to be free of any patents (such is impossible if you look at what has been patented in the US); it merely has to be free of patents that restrict distribution to anything other than GPL terms. If such a patent exists, the GPL's own restrictions override any patent license granted by the Apache License -- whether or not you are the one suing someone is not relevant to the GPL. Roy -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Mark Shewmaker scripsit: So now Person_C is in the position of having Program_C that seemed to have been properly distributed to him under the GPL, but which he can no longer use because his rights to Patent_A have been revoked. That's equivalent to the case where Program_C requires Patent_Q held by Person_Q for which Person_C has no license. Program_C is not usable by Person_C although properly distributed to him. The fact that Person_C's license to Patent_A is being revoked because of a patent-defense clause is interesting, but I don't see how it's specially relevant. (IANAL, TINLA, as usual.) Person_B is also stuck--he can't distribute Program_B under the GPL anymore to anyone, because he's not allowed to distribute it to Person_C due to a lack of a patent license for Patent_A. Sure he can distribute it to Person_C; Person_C just can't use what he gets. The GPL doesn't restrict me from distributing to you because you can't (due to some legal disability) use the program, any more than I am restricted from distributing CP/M programs to you under the GPL just because you don't have a Z80 machine to run them on. In fact, looking at this from before Person_C starts a lawsuit, you can think of things from the point of view that in order for Person_[BC] to (continue to) have GPL-rights to Program_[ABC], Person_C is required to refrain from suing Person_A for patent infringement. The GPL rights of Person_[BC] are not altered, merely the use right of Person_C to Program_C. So would the requirement-not-to-sue be an additional requirement that would be counter to the GPL in the first place? No, or at least not on this line of reasoning. -- John Cowan www.ccil.org/~cowan www.reutershealth.com [EMAIL PROTECTED] All isms should be wasms. --Abbie -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
On Wed, 2004-02-18 at 10:49, [EMAIL PROTECTED] wrote: Mark Shewmaker scripsit: Person_B is also stuck--he can't distribute Program_B under the GPL anymore to anyone, because he's not allowed to distribute it to Person_C due to a lack of a patent license for Patent_A. Sure he can distribute it to Person_C; Person_C just can't use what he gets. The GPL doesn't restrict me from distributing to you because you can't (due to some legal disability) use the program, The GPL does restrict you from distributing to me if I can't further redistribute the program. Again, from the GPLv2: For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program. The GPL rights of Person_[BC] are not altered, merely the use right of Person_C to Program_C. From the Apache License Version 2.0: ...patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer the Work ^^ If the use right really only came from the Apache license, (which looks to be the case, as Person_A never distributed their patents in GPL code in the example), then I would think the rights of Person_[BC] to distribute code containing that patent would also be gone. (I called those GPL rights, because the distribution rights were rights the GPL seemed to be giving you, (even though it really had no power to do so in all cases.) In retrospect I shouldn't have made up a term like that without at least also listing a made-up definition. :-) ) (As an aside, I hate to think I probably contributed to a slashdotting of this topic. My intent was to merely point out that I considered there to be an unintentional incompatibility with the Apache License Version 2.0 and GPLv2. Apache folks, I apologize if I have triggered flames on the subject coming to you on this. In my mind at the very worst it's a fixable problem one way or another, and not at all flame-worthy material.) -- Mark Shewmaker [EMAIL PROTECTED] -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
On Wednesday, February 18, 2004, at 03:22 PM, Mark Shewmaker wrote: On Tue, 2004-02-17 at 20:20, Roy T. Fielding wrote: No, the patent (if there was one) would be an additional restriction on the GPL. The Apache License itself is not the patent and does not restrict the GPL any more than the GPL would have been restricted by the patent absent the Apache License. Let me make a more limited, but far more convoluted, claim: Code incorporating patents (when the code and contributors' patents are licensed solely under the Apache License Version 2.0) cannot be (safely?) incorporated into a derivative work distributed under GPLv2, because any recipient who receives a copy of such a derivative work is required to refrain from alleging infringement of any of his own patents incorporated into the original Apache-License-only code. (That requirement is more strict than the GPL's requirement to license the patent for GPL use.) Allow me to make a less convoluted translation: Code incorporating patents, when the code and contributors' patents are licensed solely under the MIT license, cannot be incorporated into a derivative work distributed under GPLv2, because any recipient who receives a copy of such a derivative work has no rights to use any of the patents incorporated into the original MIT code. Why, then, is the MIT license compatible with the GPL? Roy -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Roy T. Fielding scripsit: Code incorporating patents, when the code and contributors' patents are licensed solely under the MIT license, cannot be incorporated into a derivative work distributed under GPLv2, because any recipient who receives a copy of such a derivative work has no rights to use any of the patents incorporated into the original MIT code. Why, then, is the MIT license compatible with the GPL? Because the MIT license is silent about patents; in and of itself, it can't do anything to require you to breach the GPL's licensing terms. (It may be that the word use provides an implied patent license.) A specific MIT-licensed program may be GPL-incompatible, but MIT-licensed programs as a class are not, because they don't impose any requirements incompatible with the GPL's. -- Where the wombat has walked,John Cowan [EMAIL PROTECTED] it will inevitably walk again. http://www.ccil.org/~cowan -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
On Wed, 2004-02-18 at 20:01, Roy T. Fielding wrote: Allow me to make a less convoluted translation: Code incorporating patents, when the code and contributors' patents are licensed solely under the MIT license, cannot be incorporated into a derivative work distributed under GPLv2, because any recipient who receives a copy of such a derivative work has no rights to use any of the patents incorporated into the original MIT code. Why, then, is the MIT license compatible with the GPL? Reason 1 of 2: -- Because the MIT license is a blanket grant of permission, almost without restriction: | ...Permission is hereby granted, free of charge, to any person | obtaining a copy of this software and associated documentation files | (the Software), to deal in the Software without restriction, | including without limitation the rights to use, copy, modify, merge, | publish, distribute, sublicense, and/or sell copies of the Software, | and to permit persons to whom the Software is furnished to do so, | subject to... It out and out grants you permission (!!!) ... without restriction ... including without limitation ... [all this stuff] ... subject to ... [hardly anything]. It doesn't just grant you a copyright license, but grants you out and out permission. (I just noticed that in rereading the license for this email--shocking! IANAL, so I may be very confused here--can a Real Lawyer comment on this? I was quite surprised to notice the apparent breadth of the grant.) So the fact that the grant doesn't mention patents doesn't make me think patents aren't included, just as the fact that the grant doesn't mention copyrights doesn't make me think a copyright license isn't included. (Also note that the license even uses those magic patentspeak words make, use, and sell.) So, I would *assume* that a contributor submitting software initially under just the MIT license, (where the code incorporates patents they own), is in effect allowing their incorporated patents and incorporated copyrights to be used anywhere with hardly any limits or conditions. It would make no sense to me for someone to hand the world a piece of code containing his patents and say I hereby give anyone permission to do anything with this, without restriction and without limitation, and later come back and say that he wasn't licensing the incorporated copyrights or patents. So it looks to me that an MIT-licensed patent, (to use sloppy wording), is in effect also automatically compatibly GPL-licensed, (to continue the sloppiness), as well as being available for proprietary works. So I don't think your less convoluted translation holds, as the MIT license couldn't cause a patent to be licensed solely under itself. (Now, granted, IANAL, and IANAPL, so I may be totally off-base here. And I'm assuming of course that the licensor/contributors have full copyrights and patent rights to the code, there are no disputes on ownership, and that sort of thing.) Reason 2 of 2: -- However, even I'm totally wrong in the above, and the MIT license *doesn't* implicitly include a patent license when the owner of a patent distributes code under that license, there's still nothing in the MIT license that *retracts* any [patent] grants under any conditions in which the GPL would not also retract/disallow its granted rights. (For instance, lying about author attributions breaks both licenses.) So, the MIT license isn't adding any restrictions not already in the GPL, meaning the MIT license is GPL-compatible. (This is a simpler claim than the above; IANAL still applies of course, and I'm still assuming there aren't weird corner cases going on with IP ownership disputes and such.) (Side issue: I just read John Cowan's very similar but impressively far more concise response. I've got to learn to write more concisely!) -- Mark Shewmaker [EMAIL PROTECTED] -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Russell Nelson said on Mon, Feb 16, 2004 at 05:12:21PM -0500,: If nobody else reviews this license, then the license approval snip comply with the OSD (cough, cough). But still, could somebody else take a gander at this? This license was discussed on [EMAIL PROTECTED], and I had seen quite a few regulars on this and debian-legal there; and in one mail, Eben Moglen of FSF wrote:- quote FSF notes that section 5 is the only element of ASL 2.0 that is incompatible with version 2 of the GNU General Public License. FSF continues to believe that the achievement of compatibility between ASL and GPL would be of enormous benefit to the community of free software developers, allowing merger of valuable code bases currently separated by license incompatibilities. FSF is pleased to note the convergence implied by the ASL 2.0 draft. FSF will make efforts, in the development, discussion, and adoption of GPL 3 to further the process of convergence, by carefully considering the Apache Foundation's approach to the patent defense problem. For this reason, we consider the distinction between the approaches contained in the first and second sentences of section 5 to be particularly significant. /quote Sec. 5 referred to by Prof. Moglen was Sec 5 of the original draft as proposed by the Apache Foundation. This seems to have been renumbered as section 3 in the final license. Finally, on January 24th, Roy Fielding of the Apache Foundation stated on the same list:- quote They(*) are compatible. Whether or not they are considered compatible by the FSF is an opinion only they can make, but given that a derivative work consisting of both Apache Licensed code and GPL code can be distributed under the GPL (according to *our* opinion), there really isn't anything to be discussed. /quote Guess that settles the matter. I am not on a `always on' network, so cannot search out the archives of [EMAIL PROTECTED] for exact links to the above messages; the messages are archived by me though. (*) The ASL and GNU GPL. -- +~+ Mahesh T. Pai, LL.M., 'NANDINI', S. R. M. Road, Ernakulam, Cochin-682018, Kerala, India. http://paivakil.port5.com +~+ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
On Tue, 17 Feb 2004, Mahesh T. Pai wrote: Russell Nelson said on Mon, Feb 16, 2004 at 05:12:21PM -0500,: If nobody else reviews this license, then the license approval snip comply with the OSD (cough, cough). But still, could somebody else take a gander at this? This license was discussed on [EMAIL PROTECTED], and I had seen quite a few regulars on this and debian-legal there; and in one mail, Eben Moglen of FSF wrote:- quote FSF notes that section 5 is the only element of ASL 2.0 that is incompatible with version 2 of the GNU General Public License. FSF continues to believe that the achievement of compatibility between ASL and GPL would be of enormous benefit to the community of free software developers, allowing merger of valuable code bases currently separated by license incompatibilities. FSF is pleased to note the convergence implied by the ASL 2.0 draft. FSF will make efforts, in the development, discussion, and adoption of GPL 3 to further the process of convergence, by carefully considering the Apache Foundation's approach to the patent defense problem. For this reason, we consider the distinction between the approaches contained in the first and second sentences of section 5 to be particularly significant. /quote Sec. 5 referred to by Prof. Moglen was Sec 5 of the original draft as proposed by the Apache Foundation. This seems to have been renumbered as section 3 in the final license. Also, the second sentence referred to above by Eben in the older draft was the broader one that applied to any patent action taken against any open source software product. It was narrowed, in the draft that was eventually officially approved, to only cover patent actions regarding *the licensed software itself*, narrowing the scope but being much more acceptable. Finally, on January 24th, Roy Fielding of the Apache Foundation stated on the same list:- quote They(*) are compatible. Whether or not they are considered compatible by the FSF is an opinion only they can make, but given that a derivative work consisting of both Apache Licensed code and GPL code can be distributed under the GPL (according to *our* opinion), there really isn't anything to be discussed. /quote Guess that settles the matter. Well, Russ's matter is conformance with the OSD, not the GPL. Nothing came up in our own drafting and discussion of the ASL that suggested something beyond the OSD's constraints. The same basic contract is there - use our code for whatever purpose you want, just give us credit, don't call it Apache if it's your work, and caveat emptor. Brian -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Russell Nelson wrote: If nobody else reviews this license, then the license approval committee will have to work without your input. As we're only human, we might make a mistake, and approve an Apache license which didn't comply with the OSD (cough, cough). But still, could somebody else take a gander at this? When it was a draft, there was some discussion of the patent clauses. They have since been reworded to be more in line with other, OSI-approved licenses. I do wonder about 5. Submission of Contributions. Unless You explicitly state otherwise, any Contribution intentionally submitted for inclusion in the Work by You to the Licensor shall be under the terms and conditions of this License, without any additional terms or conditions. Notwithstanding the above, nothing herein shall supersede or modify the terms of any separate license agreement you may have executed with Licensor regarding such Contributions. Can you do that in a copyright license? It seems harmless, since it only applies if you intentionally submit something to the Licensor, but I've never seen it before in an open source license. For the rest, it reads like the Apache license 1.1 rewritten by a lawyer. :-) Arnoud -- Arnoud Engelfriet, Dutch patent attorney - Speaking only for myself Patents, copyright and IPR explained for techies: http://www.iusmentis.com/ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Arnoud Engelfriet wrote: I do wonder about 5. Submission of Contributions. Unless You explicitly state otherwise, any Contribution intentionally submitted for inclusion in the Work by You to the Licensor shall be under the terms and conditions of this License, without any additional terms or conditions. Notwithstanding the above, nothing herein shall supersede or modify the terms of any separate license agreement you may have executed with Licensor regarding such Contributions. Can you do that in a copyright license? 'copyright licence'? what's that? this clause is intended to cover all submissions to the licensor without having to explicitly execute any other document. for example, most (hopefully soon to be all) of the asf's committers have submitted CLAs (contributor licence agreements) which basically describe the terms of their submissions. however, stuff that comes in through the mailing lists or issue trackers in the form of patches *isn't* covered by a cla. this clause addresses that. -- #kenP-)} Ken Coar, Sanagendamgagwedweinini http://Golux.Com/coar/ Author, developer, opinionist http://Apache-Server.Com/ Millennium hand and shrimp! -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
Rodent of Unusual Size writes: i don't think anyone has submitted it yet. the apache software foundation approved version 2.0 of its licence, and would like to submit it for osi approval. it's online at http://www.apache.org/licenses/LICENSE-2.0 and i'm attaching the text version to this message. it is our belief that this new licence is just as osi-compliant as the 1.1 version, and is more clearly compatible with the gpl to boot. +1 -- The man that wanders far[EMAIL PROTECTED] from the walking tree http://www.reutershealth.com --first line of a non-existent poem by: John Cowan -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: apache license 2.0 for consideration
On Tuesday, February 17, 2004, at 04:04 PM, Mark Shewmaker wrote: On Sun, 2004-02-08 at 14:19, Rodent of Unusual Size wrote: it is our belief that this new licence is just as osi-compliant as the 1.1 version, and is more clearly compatible with the gpl to boot. Is the patent grant section GPL compatible? Yes. From the Apache License, Version 2.0: | If You institute patent litigation against any entity (including a | cross-claim or counterclaim in a lawsuit) alleging that the Work or a | Contribution incorporated within the Work constitutes direct or | contributory patent infringement, then any patent licenses granted to | You under this License for that Work shall terminate as of the date | such litigation is filed. From the GPLv2: | For example, if a patent license would not permit royalty-free | redistribution of the Program by all those who receive copies directly | or indirectly through you, then the only way you could satisfy both it | and this License would be to refrain entirely from distribution of the | Program. To me it looks like the patent grant is an additional restriction beyond the bare GPL when a program licensed under the Apache License Version 2.0 is then distributed under the GPL. No, the patent (if there was one) would be an additional restriction on the GPL. The Apache License itself is not the patent and does not restrict the GPL any more than the GPL would have been restricted by the patent absent the Apache License. Roy -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
apache license 2.0 for consideration
i don't think anyone has submitted it yet. the apache software foundation approved version 2.0 of its licence, and would like to submit it for osi approval. it's online at http://www.apache.org/licenses/LICENSE-2.0 and i'm attaching the text version to this message. it is our belief that this new licence is just as osi-compliant as the 1.1 version, and is more clearly compatible with the gpl to boot. -- #kenP-)} Ken Coar, Sanagendamgagwedweinini http://Golux.Com/coar/ Author, developer, opinionist http://Apache-Server.Com/ Millennium hand and shrimp! Apache License Version 2.0, January 2004 http://www.apache.org/licenses/ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION 1. Definitions. License shall mean the terms and conditions for use, reproduction, and distribution as defined by Sections 1 through 9 of this document. Licensor shall mean the copyright owner or entity authorized by the copyright owner that is granting the License. Legal Entity shall mean the union of the acting entity and all other entities that control, are controlled by, or are under common control with that entity. For the purposes of this definition, control means (i) the power, direct or indirect, to cause the direction or management of such entity, whether by contract or otherwise, or (ii) ownership of fifty percent (50%) or more of the outstanding shares, or (iii) beneficial ownership of such entity. You (or Your) shall mean an individual or Legal Entity exercising permissions granted by this License. Source form shall mean the preferred form for making modifications, including but not limited to software source code, documentation source, and configuration files. Object form shall mean any form resulting from mechanical transformation or translation of a Source form, including but not limited to compiled object code, generated documentation, and conversions to other media types. Work shall mean the work of authorship, whether in Source or Object form, made available under the License, as indicated by a copyright notice that is included in or attached to the work (an example is provided in the Appendix below). Derivative Works shall mean any work, whether in Source or Object form, that is based on (or derived from) the Work and for which the editorial revisions, annotations, elaborations, or other modifications represent, as a whole, an original work of authorship. For the purposes of this License, Derivative Works shall not include works that remain separable from, or merely link (or bind by name) to the interfaces of, the Work and Derivative Works thereof. Contribution shall mean any work of authorship, including the original version of the Work and any modifications or additions to that Work or Derivative Works thereof, that is intentionally submitted to Licensor for inclusion in the Work by the copyright owner or by an individual or Legal Entity authorized to submit on behalf of the copyright owner. For the purposes of this definition, submitted means any form of electronic, verbal, or written communication sent to the Licensor or its representatives, including but not limited to communication on electronic mailing lists, source code control systems, and issue tracking systems that are managed by, or on behalf of, the Licensor for the purpose of discussing and improving the Work, but excluding communication that is conspicuously marked or otherwise designated in writing by the copyright owner as Not a Contribution. Contributor shall mean Licensor and any individual or Legal Entity on behalf of whom a Contribution has been received by Licensor and subsequently incorporated within the Work. 2. Grant of Copyright License. Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable copyright license to reproduce, prepare Derivative Works of, publicly display, publicly perform, sublicense, and distribute the Work and such Derivative Works in Source or Object form. 3. Grant of Patent License. Subject to the terms and conditions of this License, each Contributor hereby grants to You a perpetual, worldwide, non-exclusive, no-charge, royalty-free, irrevocable (except as stated in this section) patent license to make, have made, use, offer to sell, sell, import, and otherwise transfer the Work, where such license applies only to those patent claims licensable by such Contributor that are necessarily infringed by their Contribution(s) alone or by