Re: Can anyone say his or her software is open source?
Chris Gray wrote: You'll also see that Going To The Media (tm) was proposed and rejected as a first approach: softly softly did it. I'll give them a call this morning, cordially mention some of the points made, and see what the reaction is. -t -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
Abe Kornelis 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 ] -- This raises some questions. We recently had a lengthy discussion on the speed with which licenses are handled by the OSI board. It's good to see that speed is attempted, but it leaves me wondering when other license proposals (4 in wait, as far as I know) will be discussed on a board meeting. They're on the agenda also. But speed has its disadvantages. You sent your mail at 11:40 Eastern Standard Time. For us in europe that is 17:40. We don't meet for another 7 hours from now. It's obvious that this license will take a lot more discussion. I'm not going to put it on the agenda (although of course other board members may choose to introduce 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 Tue, 30 October 2001, David Johnson wrote: 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. The current system is based on whoever filed first. It may not be the law, but it is the practice. But if I code some software, register it with the copyright office, put a LGPL license on it, put it on the web, and I DON'T get a patent for it, then, YOU shouldn't be able to take my code and patent it, just because there's no prior patent art, doesn't mean it's novel. at work, we get a patent refresher every year or so. basically, we are not to tell our customers anything about future ASIC products without first passing it by our legal department for approval. apparently if we say too much, give away too much detail, we can lose a right to a possible patent. code registered with the copyright office decades ago should similarly disqualify patentability. 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. OK, time for a beer.;) Greg -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Intel's proposed BSD + Patent License
what is the policy of opensource about patents ? and such license in opensource license ? what is the way ? I think it is important to integrate patents issues in such license. In fact, the main risk for a patent holder is not about licensing business but to not exhaust its patent rights in a non delimited field. I think that a lot of patent holder will be agree to grant license for use in the opensource field but not in the traditional business field, that is the purpose of the limitation of Intel, i think. But, the problem with BSD, or in my case with LGPL, is that such software can be integrated in a traditionnal commercial software, and in this case, patent holder would like to have money. And not if the software is integrated in an opensource software .. ... finally, most of big companies will enter in opensource process if they have solutions to manage such issues, I think and it can be great for the developement of opensource software, no ??? 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 : Karsten M. Self [mailto:[EMAIL PROTECTED]] Envoye : mercredi 31 octobre 2001 09:45 A : [EMAIL PROTECTED] Objet : Re: Intel's proposed BSD + Patent License on Tue, Oct 30, 2001 at 09:32:40PM -0800, David Johnson ([EMAIL PROTECTED]) wrote: 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? I'd disqualify it on 6, but not 8. It's not specific to a product, but it *is* specific to a field of endeavor: OS applications. Peace. -- Karsten M. Self [EMAIL PROTECTED] http://kmself.home.netcom.com/ What part of Gestalt don't you understand? Home of the brave http://gestalt-system.sourceforge.net/ Land of the free Free Dmitry! Boycott Adobe! Repeal the DMCA! http://www.freesklyarov.org Geek for Hire http://kmself.home.netcom.com/resume.html -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Can anyone say his or her software is open source?
From: Matthew C. Weigel The Open Source Initiative owns the servicemark OSI Approved Open Source Software, and that is all. Not quite! The certification mark is OSI Certified and the goods are open source software. Thus the usage is: OSI Certified Open Source Software /Larry Rosen 650-216-1597 [EMAIL PROTECTED] www.rosenlaw.com -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
This is not legal advice. No client-attorney relationship is established. Speaking solely for myself. etc etc - Original Message - From: Russell Nelson [EMAIL PROTECTED] To: John Cowan [EMAIL PROTECTED] Cc: [EMAIL PROTECTED]; Kolb, Doug [EMAIL PROTECTED]; Stamnes, Michelle [EMAIL PROTECTED]; Simon, David [EMAIL PROTECTED] Sent: Tuesday, October 30, 2001 4:15 PM Subject: Re: Intel's proposed BSD + Patent License John Cowan writes: Russell Nelson wrote: [snip] 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. This statement can't be categorically made and misses nuances. For example, in probably every country in the world, there are prior user rights of some form i.e. someone using a patented invention before application for that invention can continue to use that invention.Unfortunately in the U.S. they are very narrow rights but in other countries they are quite broad. Further, prior use and prior invention can be a bar to or invalidate a patent in the U.S. and elsewhere - the devil is in the details. 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. I think you are assuming that the BSD grant does not include rights to patents. From what I recall, the BSD makes no reference to specific rights, whether copyright, patent or anything else for that matter (except it requires the inclusion of the copyright notice). 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 _ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
This is not legal advice. No lawyer-client relationship is established. Speaking only for myself. etc etc. From: [EMAIL PROTECTED] CC: [EMAIL PROTECTED] Subject: Re: Intel's proposed BSD + Patent License Date: 30 Oct 2001 18:24:32 -0800 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. If the invention embodied within the code has never become publicly known or used or has otherwise been kept secret, then it is possible. The patent system is all about encouraging disclosure. 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) Not sure about the circumstance regarding the decades old code. If the invention was properly made known or used in the public then a granted patent for that invention may be invalid. You have to remember that the patent system is not all knowing - it may not know of this decades old code unless it has been cited to the relevant patent office. If the invention was kept secret, then a patent for the invention may be valid. However a user of the invention before the grant of the patent may be able to continue to use the invention under prior use rights. Patent systems are different around the world so no general statement can be made. Further, each situation is very fact specific. Greg -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3 _ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
This is not legal advice. No lawyer-client relationship is established. Speaking for myself only. etc etc From: [EMAIL PROTECTED] CC: [EMAIL PROTECTED] Subject: Re: Intel's proposed BSD + Patent License Date: 31 Oct 2001 06:22:39 -0800 On Tue, 30 October 2001, David Johnson wrote: 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. The current system is based on whoever filed first. It may not be the law, but it is the practice. But if I code some software, register it with the copyright office, put a LGPL license on it, put it on the web, and I DON'T get a patent for it, then, YOU shouldn't be able to take my code and patent it, just because there's no prior patent art, doesn't mean it's novel. Someone may be able to get a patent but it may be invalid. The relevant patent office may not be aware of this code. Further, someone is not entitled to a patent for someone else's invention. The applicant must have invented the invention not have taken it. This issue is called derivation in patent law circles. Non-patent prior art is relevant to the examination of applications for patents and for invalidating granted patents. at work, we get a patent refresher every year or so. basically, we are not to tell our customers anything about future ASIC products without first passing it by our legal department for approval. apparently if we say too much, give away too much detail, we can lose a right to a possible patent. code registered with the copyright office decades ago should similarly disqualify patentability. It may but it depends on what is actually registered. In many cases a registration for software includes a redacted version of source code (to preserve trade secrets). Accordingly, the invalidating effect may be limited. 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. While I share your concerns about where the patent system is headed, I would note that the patent system is ultimately run by the various national governments. They can put an end to the patent system. Lawyers are ultimately just middlemen and dependent on the graces of the national governments. Further, lawyers typically act on behalf of clients. While lawyers certainly have a vested interest, they are dependent on benefactors (otherwise they are doing something else). So, I would suggest that the responsibilities of the government and interests of the patent system end-users not be lost. Just my biased two cents ;-) OK, time for a beer. ;) Greg -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3 _ Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
On Wednesday 31 October 2001 06:22 am, [EMAIL PROTECTED] wrote: But if I code some software, register it with the copyright office, put a LGPL license on it, put it on the web, and I DON'T get a patent for it, The key here is register it. I would also place a description of the software using general terms under the same registration. You want your software to be found when someone makes a patent search. And if it was missed (or ignored) during the search, you want legal documentation proving the prior art. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Response to comments on Intel's proposed BSD+Patent license
There seem to be a number of comments on the BSD+ Patent license we have proposed that claim that the license is not open because it only licenses a specific product; i.e., Linux. First, this is not true. The patent license that is extended is for ANY OS that is licensed under the GPL. It may be Linux or any other OS that is licensed under GPL. Second, and far more fundamental, all of the threads seem to agree: 1. BSD is a copyright only license. 2. BSD grants NO rights to patents. 3. BSD is an open license. It is not logical to say that a license that grants MORE rights than the BSD is not open. If you use the software in an OS licensed under GPL, you also get a patent license on the use of that software. For the sake of example, let's assume that instead of granting the additional value of a patent license from Intel, the proposed license said If you use the software in an OS that is licensed under the GPL, Intel will pay you $100. The license merely provides an incentive for a particular use, but does not prohibit other uses. Now, change the value to being a patent license. That does not change the fact that there is additional value; it is just value of a different form. How is that not an open license? Finally, under the proposed license, you can use the software in Solaris or any other proprietary OS or in any other piece of software (in addition to the GPL based OS's). You just don't have a patent license; so you are no worse off than with the BSD license. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Response to comments on Intel's proposed BSD+Patent license
On Wed, 31 Oct 2001, Stamnes, Michelle wrote: Finally, under the proposed license, you can use the software in Solaris or any other proprietary OS or in any other piece of software (in addition to the GPL based OS's). You just don't have a patent license; so you are no worse off than with the BSD license. I'm afraid the license is not entirely clear on this point: 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. Does that mean that the license *only* includes such changes to the software under *only* such operating systems? 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. It is not clear whether this patent license means the license to make 'bug-fixing/ modifications, or the license before that: 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. If it's the latter, then use is effectively restricted (through patent law, and not copyright law, but I don't think the OSD allows for such distinction). If it's the former, then the right to create derivative works is effectively restricted (again, through patent law and not copyright law). Please remember that the OSI certifies *software*, so - in my opinion - software distributed under this license, whose use or sale infringes upon patent claims licensable by Intel, restricts the user's ability to make derived works unacceptably, and discriminates against persons not using a GPL'd operating system. This means, IMO, that if software whose use or sale infringes upon patents is to be considered OSI Certified Open Source Software, the patent license must also support the OSD. Compare this license to the license a while back that restricted the ability of the user to modify some pay for this software routines: if the copyright holder of that software managed to get a patent for his pay for this software routines, and distributed it under this license, would it be considered OSI Certified Open Source Software? -- Matthew Weigel Research Systems Programmer [EMAIL PROTECTED] ne [EMAIL PROTECTED] -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
Russell Nelson wrote: Abe Kornelis 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 ] -- This raises some questions. We recently had a lengthy discussion on the speed with which licenses are handled by the OSI board. It's good to see that speed is attempted, but it leaves me wondering when other license proposals (4 in wait, as far as I know) will be discussed on a board meeting. They're on the agenda also. But speed has its disadvantages. You sent your mail at 11:40 Eastern Standard Time. For us in europe that is 17:40. We don't meet for another 7 hours from now. It's obvious that this license will take a lot more discussion. I'm not going to put it on the agenda (although of course other board members may choose to introduce it). -- Ok, that's fair enough. I do think it's important that anybody on the list gets a decent chance to react to any license for which approval is requested. Since this seems indeed to be the case, I apologize for my apparent overreaction to your first mail. Regards, Abe. -- Abe F. Kornelis, B.V. Bixoft Het Jaagpad 63, 3461 HA Linschoten The Netherlands phone: +31-6-22755401 To visit our website: either: http://www.bixoft.com or: http://www.bixoft.nl -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
