Article on open-source licenses (and the OSI)
http://www.networkcomputing.com/1222/1222ws1.html A very good sum up of the current (blurry) situation on licenses. (especially for developpers getting lost) -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Article on open-source licenses (and the OSI)
On Tuesday 30 October 2001 05:54, you wrote: http://www.networkcomputing.com/1222/1222ws1.html A very good sum up of the current (blurry) situation on licenses. (especially for developpers getting lost) -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3 In good natured sarcasm, and speaking for myself.. This looks complicated to those who are too close to the situation. There are a million points of controversy or confusion to anything that's picked apart. Frankly, John Doe developer would have to be a glutton for punishment to even bother considering licensing their work under most of the approved licenses. And then he's asking for it. The license page does point out the classic licenses most commonly used for open-source software. How much of a clue do you have to buy someone? Just because a license has been given the nod as meeting the OSD doesn't mean that it _has to be_ considered for your particular work. It does mean IMHO: 1) go ahead and stick 'OSI Certified' on it, 2) I can see some program's funky license check it against the list know that whatever legalese is in there that it meets the OSD. C'mon, how many programs are going to go outside the top 6-7 licenses anyway? If more do then there will be more conversation on them everyone's understanding goes up on the finer points of that license. Perhaps there should be an Open Source Top 5 chart? 8^) Putting the pain in pain in the ass since., -- Steve Mallett | Stable, Open-Source Apps [EMAIL PROTECTED] | http://OSDir.org [EMAIL PROTECTED] http://open5ource.net/steve/ GPG, Voice, etc. Non-cooperation with evil is as much a duty as cooperation with good. -- Mohandas Gandhi -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Article on open-source licenses (and the OSI)
On Tue, 30 October 2001, Steve Lhomme wrote: http://www.networkcomputing.com/1222/1222ws1.html The part that looked promising was actually the tiny table in gif form, that is barely noticable on page two. http://img.cmpnet.com/nc/1222/graphics/1222ws1.gif The idea of being able to compare license to license in table form has its appeal. (whether that particular table achieves that or not, eh, maybe not) the problem would be to come up with the table categories such that they align with copyright law and real-world effects. copy, modify, distribute is all copyright law cares about. but there's distinctions within those categories. i.e. LGPL vs GPL: one lets you distribute outside code linked with the original code, and the other does not. I don't know what all the distinctions are between all the licenses to know if they can be easily broken down into a small, finite set of categories, put in table form with yes/no answers, and have such a table fully describe the effects of the license. It would be nice. The idea would be for a non-lawyer programmer to be able to look at the table and select the license that does what they want. it would be a lot easier to compare apples to apples if all the columns lined up nicely. Greg -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Intel's proposed BSD + Patent License
Is there anyone have samples of license including patent issues ? I am working on such problems for a consortium which will develop and distribute free software. The limitation of the patent license for any use of the software only on GPL licensed OS is very interesting, I think and can help patent owners to use opensource software. In fact, because of the exhaustion of IP rights, it is very dangerous to grant any right on a software because it may risk to imply a license of the patents covering this software. So it is very important to limit the scope of such license. what do you think about such limitations ? with thanks and regards, - Yann Dietrich IP/Licensing Legal Counsel France Telecom RD / VAT member of the editorial board of http://www.juriscom.net homepage : http://www.chez.com/ydietrich -Message d'origine- De : Russell Nelson [mailto:[EMAIL PROTECTED]] Envoye : mardi 30 octobre 2001 17:40 A : [EMAIL PROTECTED] Cc : Kolb, Doug; Stamnes, Michelle; Simon, David Objet : Intel's proposed BSD + Patent License [ Please review this license. If you do so promptly enough, we may be able to include it in tomorrow's board meeting. -russ ] After reviewing the following license with key open source developers, Intel is submitting a new license called the BSD + Patent License. This is a combination of the BSD license (the post July 22, 1999 version without the advertising clause), with patent license language derived from the Common Public License (CPL). Contributors to Linux or other GPL works may provide their independently created code under the BSD license (see the FSF's web site at http://www.fsf.org/licenses/license-list.html#ModifiedBSD). To address concerns that users of Intel's contributions under the BSD license risked patent infringement claims from Intel, this license submission expressly includes patent licensing language. Intel modified the BSD license in the following ways: 1. Intel made OPTIONAL the inclusion of a copyright notice (i.e., Redistributions of source code of the Software may retain the above copyright notice, this list of conditions and the following disclaimer). 2. Intel added certain definitions derived from the patent license in the Common Public License, and added a license grant under certain Intel patents to distribute Intel software contributions, alone or as incorporated in any operating system licensed under the GPL (version 2.0 or later). Please post the license to the license-discuss list under my name on behalf of Intel Corporation. Thank you for your consideration. Regards, Michelle V. Stamnes Intel Corporation The Intel BSD + Patent License Recipient has requested a license and Intel Corporation (Intel) is willing to grant a license for the software entitled _ (the Software) being provided by Intel Corporation. The following definitions apply to this License: Licensed Patents means patent claims licensable by Intel Corporation which are necessarily infringed by the use or sale of the Software alone or when combined with the operating system referred to below. Recipient means the party to whom Intel delivers this Software. Licensee means Recipient and those third parties that receive a license to any operating system available under the GNU Public License version 2.0 or later. Copyright (c) 1996-2001 Intel Corporation All rights reserved. The license is provided to Recipient and Recipient's Licensees under the following terms. Redistribution and use in source and binary forms of the Software, with or without modification, are permitted provided that the following conditions are met: Redistributions of source code of the Software may retain the above copyright notice, this list of conditions and the following disclaimer. Redistributions in binary form of the Software may reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. Neither the name of Intel Corporation nor the names of its contributors shall be used to endorse or promote products derived from this Software without specific prior written permission. Intel hereby grants Recipient and Licensees a non-exclusive, worldwide, royalty-free patent license under Licensed Patents to make, use, sell, offer to sell, import and otherwise transfer the Software, if any, in source code and object code form. This license shall include changes to the Software that are error corrections or other minor changes to the Software that do not add functionality or features when the Software is incorporated in any version of a operating system that has been distributed under the GNU General Public License 2.0 or later. This patent license shall apply to the combination of the Software and any operating system licensed
Re: Intel's proposed BSD + Patent License
Intel hereby grants Recipient and Licensees a non-exclusive, worldwide, royalty-free patent license under Licensed Patents to make, use, sell, offer to sell, import and otherwise transfer the Software, if any, in source code and object code form. This license shall include changes to the Software that are error corrections or other minor changes to the Software that do not add functionality or features when the Software is incorporated in any version of a operating system that has been distributed under the GNU General Public License 2.0 or later. This is odd. What is the software when linked into an operating system? Does the software include anything linked with it, and therefore, would a change to the operating system that adds features be considered a change to the software? If yes, then the software becomes pretty much useless--you can link it into an operating system only if you don't change the operating system. If no, than anyone can change the software any way they want anyway, merely by claiming the changes are changes to the operating system. Also, wouldn't it violate the fields of endeavor clause and specific to a product clause to only allow the software to be linked with operating systems? (I can see a way around this: claim that of course the user is allowed to link it with anything, it's just that it would violate patents, but it's not the license which says you're prohibited from violating patents, just an external law. The problems with this workaround are left as an exercise to the reader.) -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
Russell Nelson wrote: [ Please review this license. If you do so promptly enough, we may be able to include it in tomorrow's board meeting. -russ ] I have tried to turn around this response as fast as possible, so please forgive any defects of detail. Executive summary: this license is not Open Source because works governed by it cannot be freely used except on a limited class of operating systems, and is not truly open-source software even there. To address concerns that users of Intel's contributions under the BSD license risked patent infringement claims from Intel, this license submission expressly includes patent licensing language. So it does, but unfortunately not the right kind. What the BSD license gives away, the patent licensing language takes back. 2. Intel added certain definitions derived from the patent license in the Common Public License, and added a license grant under certain Intel patents to distribute Intel software contributions, alone or as incorporated in any operating system licensed under the GPL (version 2.0 or later). This provision carves out a safe harbor for Linux, but not for other free operating systems. It would be the height of irony if software licensed under a so-called BSD + Patent License was not usable by any BSD operating system, but such appears to be the case. Criterion 8 (License Must Not Be Specific To A Product) is violated, in substance if not to the letter; this license is in effect specific to Linux. This license shall include changes to the Software that are error corrections or other minor changes to the Software that do not add functionality or features when the Software is incorporated in any version of a operating system that has been distributed under the GNU General Public License 2.0 or later. And so even on Linux, the software can't be substantially reused or improved; only bugs can be fixed. The copyright license allows you to make such improved versions, but the patent license prevents you or anyone else from actually *using* them for anything more than research. In short, the hood is still welded shut, even though there is a maintenance port on the side of the car. Criterion 3 (Derived Works) goes by the board here, since the kind of derived work that can be made (or more accurately, used) is highly restricted. The patent license shall not apply to any other combinations which include the Software. Everything that is not permitted is forbidden. This license is in effect much like the pre-QPL Trolltech license, which was not Open Source because (inter alia) it forbade Open Source Qt programs for Windows and MacOS. Trolltech's own Qt ports to those operating systems are still encumbered, but nothing (except time and trouble) prevents someone from porting the Open Source Qt/X11 version to Windows and MacOS. If Intel's sole concern with the BSD license is that users may feel at risk from its patents, it should adopt the CPL, which well addresses those very concerns. It would still be possible for Intel to distribute a port of its software to unfree operating systems that was itself encumbered without interfering with Open Source. I am not a lawyer; this is not legal advice. -- Not to perambulate || John Cowan [EMAIL PROTECTED] the corridors || http://www.reutershealth.com during the hours of repose || http://www.ccil.org/~cowan in the boots of ascension. \\ Sign in Austrian ski-resort hotel -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Article on open-source licenses (and the OSI)
Karsten M. Self writes: The discussion of OSI certification is again unclear and IMO misleading, perhaps reflection a lack of clarity on the part of the OSI. Possibly. It certainly would seem to be a loophole that you could use the BSD license on binary code for which source is not available. We pondered doing something about this, then decided to give credit to the open source developer for having some brains. Without source code, how could anybody call it open source?? *If* we were going to do something about it, I would say that we only certify source code. Binaries are derivative works. The suggestion that the OSI focus on licensing conflicts is interesting, but, given experience, hopelessly optimistic. Education on the issues is one thing, actively tangling in disputes quite another, from an organizational, operational, and resources standpoint. Yes, license compatibility is a morass. If you thought Vietnam was bad, wait until you see Afghanistan. As Vizini said, Never get involved in a land war in Asia. -- -russ nelson [EMAIL PROTECTED] http://russnelson.com Crynwr sells support for free software | PGPok | Why are we still fighting 521 Pleasant Valley Rd. | +1 315 268 1925 voice | the war on drugs when there Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | is a real war to fight? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
John Cowan writes: Russell Nelson wrote: [ Please review this license. If you do so promptly enough, we may be able to include it in tomorrow's board meeting. -russ ] I have tried to turn around this response as fast as possible, so please forgive any defects of detail. Thank you. So it does, but unfortunately not the right kind. What the BSD license gives away, the patent licensing language takes back. Let me analogize: Let's say that I'm pointing a gun at you. Are you at any more disk if I let you know it's pointed at you? What if I tell you the conditions under which I will pull the trigger? Essentially, we are all of us completely and totally screwed by the patent system. If I invent something that you have put into your (unpublished -- at least as far as the patent system is concerned) code for decades, and patent it, I 0WN J00. Doesn't matter if you're IBM and I'm Joe Blow, or vice-versa even. Criterion 8 (License Must Not Be Specific To A Product) is violated, in substance if not to the letter; this license is in effect specific to Linux. On the other hand, if they left that patent license off, we would certify it. However, you would have less freedom to use such patented software because you don't have a license. Patents suck THS much (picture me opening my hands above my head to show you how bad patents are). In essence, every open source license which does not include a patent grant is just so much hot air. About the only thing we can reasonably do is be thankful that any patent rights are included, and approve the license based on the copyright permissions granted. Note that the Intel BSD+Patent License does not make copying dependent upon patent noninfringement. The patent grant is a separate term. And what about jurisdictions? Jurisdiction is even more significant in the patent field than in the copyright field. At least copyright lawyers have the Berne Convention. So in a jurisdiction where software cannot be patented, the patent grant is meaningless. -- -russ nelson [EMAIL PROTECTED] http://russnelson.com Crynwr sells support for free software | PGPok | Why are we still fighting 521 Pleasant Valley Rd. | +1 315 268 1925 voice | the war on drugs when there Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | is a real war to fight? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
On Tue, Oct 30, 2001 at 04:15:23PM -0500, Russell Nelson wrote: based on the copyright permissions granted. Note that the Intel BSD+Patent License does not make copying dependent upon patent noninfringement. The patent grant is a separate term. Could it be a separate document altogether? Step 1: Release under BSD license, not tied to any other terms. Step 2: Grant a patent license, under a separate document. It should be clear that step 1 is in no way reliant on step 2. I know that patents suck _thii...iis_ much, but I'm wondering if the way to solve this is to keep the patents away from the copyright. This also solves the we would approve it under BSD, but not with restrictions argument, since there is a full BSD 'normal' license grant. The real problem is that the BSD license doesn't necessarily infer a patent license grant, which in some minds makes it GPL incompatible to begin with. :/ (can of worms) -drew -- M. Drew Streib [EMAIL PROTECTED] | http://dtype.org/ FSG [EMAIL PROTECTED]| Linux International [EMAIL PROTECTED] freedb [EMAIL PROTECTED]| SourceForge [EMAIL PROTECTED] PGP signature
Re: Article on open-source licenses (and the OSI)
On Tue, 30 October 2001, Russell Nelson wrote: If you thought Vietnam was bad, wait until you see Afghanistan. As Vizini said, Never get involved in a land war in Asia. to give credit where credit is due, (which seems apropos for this list): The US has broken the second rule of war. That is, don't go fighting with your land army on the mainland of Asia. Rule One is don't march on Moscow. -- Field-Marshall Montgomery (in Montgomery of Alamein ?published? 1976) It's also attributed to Eisenhower speaking to JFK some time towards the beginning of Vietnam. who said it first, I am not certain. A search on the net yielded over a hundred hits. I could only find one reference to Eisenhower, none to Montgomery, and the rest were quoting The Princess Bride. I had to look it up at home. ah well. Greg -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
On Tue, 30 October 2001, Russell Nelson wrote: Essentially, we are all of us completely and totally screwed by the patent system. If I invent something that you have put into your (unpublished -- at least as far as the patent system is concerned) code for decades, and patent it, I 0WN J00. Doesn't matter if you're IBM and I'm Joe Blow, or vice-versa even. given: http://www.nolo.com/encyclopedia/faqs/pts/pct3.html#FAQ-294 =Patents must be novel (that is, it must be different from all =previous inventions in some important way). = =Patents must be nonobvious (a surprising and significant development) =to somebody who understands the technical field of the invention. I don't see how you could patent something that I've had in code for decades. It's neither nonobvious nor novel. Granted, software patents can be a pain (Some perl/tk widgets had to have functionality ripped out because they supported a patented image format) and, IMHO, stupid (the one-click patent from days gone by) but has the scenario you described actually happened? (i.e. decades old code getting patented out from under someone) Greg -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
On Tuesday 30 October 2001 02:06 pm, M. Drew Streib wrote: On Tue, Oct 30, 2001 at 04:15:23PM -0500, Russell Nelson wrote: based on the copyright permissions granted. Note that the Intel BSD+Patent License does not make copying dependent upon patent noninfringement. The patent grant is a separate term. Could it be a separate document altogether? Step 1: Release under BSD license, not tied to any other terms. Step 2: Grant a patent license, under a separate document. I heartily second this motion. There is no need for YAOSSL, especially one that is nearly identical to existing licenses (expect for the patent stuff). Having a Patent License for patent encumbered software is far better than no patent license for same, but the software cannot be Open Source in either case. The only software that could be considered Open Source under the proposed license would be software that was NOT encumbered by patents to begin with, making the Patent License redundant. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
On Tuesday 30 October 2001 08:40 am, Russell Nelson wrote: Intel hereby grants Recipient and Licensees a non-exclusive, worldwide, royalty-free patent license under Licensed Patents to make, use, sell, offer to sell, import and otherwise transfer the Software, if any, in source code and object code form. This license shall include changes to the Software that are error corrections or other minor changes to the Software that do not add functionality or features when the Software is incorporated in any version of a operating system that has been distributed under the GNU General Public License 2.0 or later. Now wait one cotton pickin' minute here! You mean to tell me I can't use this software on FreeBSD? Then why the *$#@! did you even start off with the BSD license! If the OSI certification mark only applies to licenses, then this one might, just might, squeak by if you serve enough booze at the board meeting. But if the mark applies to software, then it clearly fails. Software under this license and simultaneously encumbered by a patent fails the OSD on numerous points. The Patent License is better than no license at all, but then again, a kick in the pants is better than a kick in the head. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Intel's proposed BSD + Patent License
I'll jump in here, not to defend or explain the patent system (heaven forfend I even try that on this list!), but to point out that 35 U.S.C. sect. 273 offers some potential relief in the situation you describe. It's intended to provide a defense to people who were commercially using a patented business method more than a year prior to the patent's filing. It's a confusing and complicated law and has not seen wide use (yet). Laura A. Majerus Fenwick West LLP 2 Palo Alto Square Palo Alto, CA 94306 Phone: 650-858-7152 Fax:650-494-1417 http://www.fenwick.com -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] Sent: Tuesday, October 30, 2001 6:25 PM Cc: [EMAIL PROTECTED] Subject: Re: Intel's proposed BSD + Patent License On Tue, 30 October 2001, Russell Nelson wrote: Essentially, we are all of us completely and totally screwed by the patent system. If I invent something that you have put into your (unpublished -- at least as far as the patent system is concerned) code for decades, and patent it, I 0WN J00. Doesn't matter if you're IBM and I'm Joe Blow, or vice-versa even. given: http://www.nolo.com/encyclopedia/faqs/pts/pct3.html#FAQ-294 =Patents must be novel (that is, it must be different from all =previous inventions in some important way). = =Patents must be nonobvious (a surprising and significant development) =to somebody who understands the technical field of the invention. I don't see how you could patent something that I've had in code for decades. It's neither nonobvious nor novel. Granted, software patents can be a pain (Some perl/tk widgets had to have functionality ripped out because they supported a patented image format) and, IMHO, stupid (the one-click patent from days gone by) but has the scenario you described actually happened? (i.e. decades old code getting patented out from under someone) Greg -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3 -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
On Tuesday 30 October 2001 06:24 pm, [EMAIL PROTECTED] wrote: =Patents must be novel (that is, it must be different from all =previous inventions in some important way). = =Patents must be nonobvious (a surprising and significant development) =to somebody who understands the technical field of the invention. Unfortunately, the patent system is no longer being run by logic, common sense, or even the LAW. It's being run by lawyers. Despite our veneer of civilization, might still makes right, and the lawyers have a monopoly on the application of might. The law means whatever they say it means. (Yes, I'm cynical today) The current system is based on whoever filed first. It may not be the law, but it is the practice. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Can anyone say his or her software is open source?
On Friday 30 November 2001 03:31 pm, Tina Gasperson wrote: ACARA (http://www.openchannelsoftware.com/projects/ACARA) is a program originally developed by NASA. ACARA is now being handled by the Open Channel Foundation (http://www.openchannelsoftware.com). ACARA's license terms (http://www.openchannelsoftware.com/project/view_license.php?group_id=129l icense_id=20) violate at least three points of the Open Source definition (AFAIK, IANAL), yet the Open Channel Foundation claims all of the software it distributes is open source. Is this OK from a legal standpoint? It's legally okay to use the term Open Source Software without getting permission from anyone. But some uses of Open Source Software can be considered misrepresentation or fraud, activities that are illegal. The software they are selling is not Open Source Software, yet they claim that it is. That's fraud in my book. There's no trademark on the term wool carpet, yet if I sold you a wool carpet that was really acrylic, I would be in a world of legal hurt. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
Russell Nelson scripsit: Let me analogize: Let's say that I'm pointing a gun at you. Are you at any more disk if I let you know it's pointed at you? What if I tell you the conditions under which I will pull the trigger? If you bind yourself by contract NEVER to pull the trigger, I feel reasonably safe -- not totally safe, of course, but reasonably. That is what the CPL does. Criterion 8 (License Must Not Be Specific To A Product) is violated, in substance if not to the letter; this license is in effect specific to Linux. On the other hand, if they left that patent license off, we would certify it. Sure. But the patent license is restrictive, so you shouldn't certify. -- John Cowan http://www.ccil.org/~cowan [EMAIL PROTECTED] Please leave your values| Check your assumptions. In fact, at the front desk. | check your assumptions at the door. --sign in Paris hotel |--Miles Vorkosigan -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
John Cowan writes: Sure. But the patent license is restrictive, so you shouldn't certify. No, *not* having the patent license is restrictive. It's really weird that a license whose controversial terms grant people more freedoms would be clearly approvable if those terms were taken out. -- -russ nelson [EMAIL PROTECTED] http://russnelson.com Crynwr sells support for free software | PGPok | Why are we still fighting 521 Pleasant Valley Rd. | +1 315 268 1925 voice | the war on drugs when there Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | is a real war to fight? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
David Johnson writes: Now wait one cotton pickin' minute here! You mean to tell me I can't use this software on FreeBSD? Sure you can. GPL your copy of it. If the OSI certification mark only applies to licenses, then this one might, just might, squeak by if you serve enough booze at the board meeting. But if the mark applies to software, then it clearly fails. It applies to software (a certification mark can only be used on a good or service, and a license is neither such). We could approve it, saying By the way, this license cannot be used by patented software. Of course Intel wouldn't be able to use it for the purpose they desire, so there's little point in doing that. Software under this license and simultaneously encumbered by a patent fails the OSD on numerous points. Oh, it gets even more interesting than that. What if a third party owns patent rights to the software? It's *still* not freely usable by all. How is this situation made worse simply because some users have gotten a patent license? Those users still can't use it because of the third-party patent. -- -russ nelson [EMAIL PROTECTED] http://russnelson.com Crynwr sells support for free software | PGPok | Why are we still fighting 521 Pleasant Valley Rd. | +1 315 268 1925 voice | the war on drugs when there Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | is a real war to fight? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Can anyone say his or her software is open source?
On Fri, 30 Nov 2001, Tina Gasperson wrote: Does a license have to comply with the published requirements (http://www.opensource.org/docs/definition.html) in order for the distributor or creator of the software to call it open source? No. It is encouraged socially to 'help the end user' by using the Open Source Initiative's definition of 'open source software' in deciding whether to call one's own software such, but there is no legal impetus to do so. The Open Source Initiative owns the servicemark OSI Approved Open Source Software, and that is all. disclaimer: This is a possible NewsForge story; if you don't want to be quoted please say so in your reply. Feel free to quote me. Better yet, take a look in the archives for the instance (last week? earlier this week) of a company calling their software 'open source.' -- Matthew Weigel Research Systems Programmer [EMAIL PROTECTED] ne [EMAIL PROTECTED] -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Can anyone say his or her software is open source?
The question presented is actually a very good one in my opinion because it calls out a subtle complexity with discussing open source and understanding what is really meant by the various uses (and, possibly, misuses) of the term. I think a programmer may freely identify their software as closed source, open source, or friendly source as long as there is no trade mark issue and the intent is not to mislead consumers or create unfair competition among competitors. As far as I can tell, the term or phrase open source is generic, and often used as a marketing phrase in much the same manner as diet soda is used. I cannot imagine what the legal basis would be to bring a fraud claim on the use the term open source. On the other hand, I think the member organizations representing the open source community as well as vested individuals might have some obligation to keep the press informed of what their view is of what it means to develop and distribute open source/free software as that term is used in the open source community. In this respect, I see troubling uses of the term open source in the press when covering the open source community far more frequently than I notice free-riding software developers using the term as a marketing scheme that one might say is inappropriate, but hardly illegal. OSI's OSD is a different question entirely, but that is matter for OSI, although the issue is not whether someone violate[d] the OSD. There is no legal harm arising from violat[ing] a definition. More important, the OSD needs some helpful re-working, and it may not be unexpected that some open source projects have significantly diverged from the OSD. Rod Rod Dixon Visiting Assistant Professor of Law Rutgers University Law School - Camden www.cyberspaces.org [EMAIL PROTECTED] On Friday 30 November 2001 03:31 pm, Tina Gasperson wrote: ACARA (http://www.openchannelsoftware.com/projects/ACARA) is a program originally developed by NASA. ACARA is now being handled by the Open Channel Foundation (http://www.openchannelsoftware.com). ACARA's license terms (http://www.openchannelsoftware.com/project/view_license.php? group_id=1 29l icense_id=20) violate at least three points of the Open Source definition (AFAIK, IANAL), yet the Open Channel Foundation claims all of the software it distributes is open source. Is this OK from a legal standpoint? It's legally okay to use the term Open Source Software without getting permission from anyone. But some uses of Open Source Software can be considered misrepresentation or fraud, activities that are illegal. The software they are selling is not Open Source Software, yet they claim that it is. That's fraud in my book. There's no trademark on the term wool carpet, yet if I sold you a wool carpet that was really acrylic, I would be in a world of legal hurt. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3 -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
On Tuesday 30 October 2001 08:21 pm, Russell Nelson wrote: David Johnson writes: Now wait one cotton pickin' minute here! You mean to tell me I can't use this software on FreeBSD? Sure you can. GPL your copy of it. ??? GPL my copy of FreeBSD ??? Precisely how do I go about this? Replace all copies of the BSD license under /usr/src and recompile? And what precisely are the boundaries of the operating system? Does it include BIND and sendmail? XFree86? tcsh? What about third party kernel modules? What if I don't load those modules while I'm using the software? Is Intel going to have some click through agreement where you have to verify that you have correctly relicensed your operating system? Would redistributors be obligated to perform similar verification measures? Oh wait! You had me going for a while. Good joke. Hah hah. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Can anyone say his or her software is open source?
On Tuesday 30 October 2001 08:52 pm, Rod Dixon, J.D., LL.M. wrote: the term or phrase open source is generic, and often used as a marketing phrase in much the same manner as diet soda is used. I cannot imagine what the legal basis would be to bring a fraud claim on the use the term open source. But the term used wasn't open source. The term used was Open Source Software, including the capitalization. Open Source is being used as part of a proper name, and not as a generic adjective. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
David Johnson writes: On Tuesday 30 October 2001 08:21 pm, Russell Nelson wrote: David Johnson writes: Now wait one cotton pickin' minute here! You mean to tell me I can't use this software on FreeBSD? Sure you can. GPL your copy of it. ??? GPL my copy of FreeBSD ??? Precisely how do I go about this? Replace all copies of the BSD license under /usr/src and recompile? Why bother? Why not simply decide in your own head that, if you ever give away a copy of your FreeBSD, you'll do so under the GPL. Poof, instant GPL'ed operating system. One of the things we *don't* do at the OSI is only approve sensible licenses. We also approve the ones that aren't likely to accomplish the author's goals. As long as they comply with the OSD, we approve 'em. And when they don't, we don't. -- -russ nelson [EMAIL PROTECTED] http://russnelson.com Crynwr sells support for free software | PGPok | Why are we still fighting 521 Pleasant Valley Rd. | +1 315 268 1925 voice | the war on drugs when there Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | is a real war to fight? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
I think approving this sets a dangerous precedent. In order to approve this, the OSI has to take the view that well, we approve documents of any length, of any content, as long as the software license parts are OSD compliant. We ignore everything else in the document. Are you saying that if I take an OSI-approved license, insert a clause The members of the OSI board are unthinking robots! They had no choice in approving this license. I end up with something you must approve? (I am tired or I am sure that I could Godel and Quine my way to a really clever false = true insertion.) Any software license which restricts use to only publicly available GPL'ed OSs, (the way their patent license does), would obviously fail to meet the OSD. Tell me why you have to put the OSI's good name on this. There is no precedent for it. (The GPL may not play nice with other licenses, but it restricts copying, not use.) - This patent license shall apply to the combination of the Software and any operating system licensed under the GNU Public License version 2.0 or later if, at the time Intel provides the Software to Recipient, such addition of the Software to the then publicly available versions of such operating system available under the GNU Public License version 2.0 or later (whether in gold, beta or alpha form) causes such combination to be covered by the Licensed Patents. Is that a record for the longest sentence in an OSI-considered license? I read that 10 times before I understood it. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
On Tuesday 30 October 2001 09:07 pm, Russell Nelson wrote: ??? GPL my copy of FreeBSD ??? Precisely how do I go about this? Replace all copies of the BSD license under /usr/src and recompile? Why bother? Why not simply decide in your own head that, if you ever give away a copy of your FreeBSD, you'll do so under the GPL. Poof, instant GPL'ed operating system. Okay, we can play fun little metaphysical games with the BSD license, but it ignores an important point: the BSD+Patent License restricts what operating systems you may use it on. I can't use it on Solaris, QNX or IRIX. And I can't use it on some imaginary MPL or QPL licensed OS. This is a violation of clause 8. Wasn't the OpenMotif license reject for the same reason? -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
David Johnson writes: Okay, we can play fun little metaphysical games with the BSD license, but it ignores an important point: the BSD+Patent License restricts what operating systems you may use it on. No it doesn't, for several reasons: 1) You might live in a jurisdiction where software parents are not recognized. 2) The software you are using may not be patented. 3) The patent in question might not be valid. 4) You might have a license for the patent. 5) You might be able to negotiate a free license for your operating system. The Patent License does not say that you cannot run the software if your operating system is not GPL'ed. It says that you are granted a license if your operating system is GPL'ed. That's *all*. It's presence does not take anything away from anyone; therefore if the license is approvable without it, then it is approvable with it. -- -russ nelson [EMAIL PROTECTED] http://russnelson.com Crynwr sells support for free software | PGPok | Why are we still fighting 521 Pleasant Valley Rd. | +1 315 268 1925 voice | the war on drugs when there Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | is a real war to fight? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
On Tuesday 30 October 2001 10:07 pm, Russell Nelson wrote: The Patent License does not say that you cannot run the software if your operating system is not GPL'ed. It says that you are granted a license if your operating system is GPL'ed. That's *all*. And that's enough to disqualify it. The patent related rights granted may be specious and useless, but they are rights. The OSD says The rights attached to the program must not depend on the program's being part of a particular software distribution, and all parties to whom the program is redistributed should have the same rights as those that are granted in conjunction with the original software distribution. My interpretation is that the rights mentioned are all rights, not just those related to the software's opensourcedness. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Can anyone say his or her software is open source?
On Tue, 30 Oct 2001, David Johnson wrote: On Friday 30 November 2001 03:31 pm, Tina Gasperson wrote: ACARA (http://www.openchannelsoftware.com/projects/ACARA) is a program originally developed by NASA. ACARA is now being handled by the Open Channel Foundation (http://www.openchannelsoftware.com). ACARA's license terms (http://www.openchannelsoftware.com/project/view_license.php?group_id=129l icense_id=20) violate at least three points of the Open Source definition (AFAIK, IANAL), yet the Open Channel Foundation claims all of the software it distributes is open source. Is this OK from a legal standpoint? There's no trademark on the term wool carpet, yet if I sold you a wool carpet that was really acrylic, I would be in a world of legal hurt. AFAIK openchannelsoftware.com has several similar offerings which are not really Open Source e.g. their NASTRAN offering (also developed by NASA) Regards, -- martin // Martin Konold, Stauffenbergerstr. 107, 72074 Tuebingen, Germany // // Email: [EMAIL PROTECTED]// // KDE 2.2.1: It is real! // -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Article on open-source licenses (and the OSI)
Perhaps there should be an Open Source Top 5 chart? 8^) How do you mean? Most frequently used licenses? Yes. -- Steve Mallett | Stable, Open-Source Apps [EMAIL PROTECTED] | http://OSDir.org [EMAIL PROTECTED] http://open5ource.net/steve/ GPG, Voice, etc. Non-cooperation with evil is as much a duty as cooperation with good. -- Mohandas Gandhi -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
