RE: OSL 2.0 and linking of libraries
Lawrence E. Rosen wrote, in part: You don't need the clarification. Simply linking a program against a library or loading machine readable code compiled from source code doesn't create a derivative work of software. Huh!!!? Clarification or recent citation please? As far as I understand it, when moduleA + moduleB = statically linked executable executable IS a derivative work of both moduleA and moduleB. And if sourceA is compiled or translated to create moduleA, then moduleA IS a derivative work of sourceA, (but this is a slightly different question.) Dynamic linking is further different, and your statement may be accurate under that circumstance. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: OSL 2.0 and linking of libraries
Read this and try to extrapolate it to software and static linking [dynamic linking aside for a moment]: http://www.law.cornell.edu/copyright/cases/125_F3d_580.htm If I were defending, and my attorney tried to cite only that one as defense for a software license/copyright violation, I wouldn't feel too confident. See also http://lwn.net/2001/1108. (License agreements and first sale doctrine) Here's the ruling: http://tinyurl.com/3c2n2 (Does a denial of injunctive relief carry the same weight of precedence as the outcome of a trial?) If the ruling at that URL (Softman v Adobe) extends to all software licenses and EULA's, and the Lee v A.R.T. case applies to compiled works, then I have serious doubts that most provisions of most of the OSI-approved licenses mean anything at all for most users of software. It would mean that if one is simply going to compile, link, and/or use software (and not edit the source, or distribute copies) then licenses (open or closed-source) can be ignored if you obtained a lawful copy. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: OSL 2.0 and linking of libraries
Here we have collections which unambiguously are collections: the question about statically linked software is precisely whether or not it is a collection. I think someone must successfully argue that it is only a collection (and does not meet the definition of derivative work:) from http://www4.law.cornell.edu/uscode/17/101.html A ''derivative work'' is a work based upon one or more preexisting works, such as a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgment, condensation, or any other form in which a work may be recast, transformed, or adapted. A work consisting of editorial revisions, annotations, elaborations, or other modifications which, as a whole, represent an original work of authorship, is a ''derivative work''. My take on this definition is that most statically linked programs include a relocation table and symbol tables which are annotations of the source code. These annotations are not particularly original, but if you declare that your statically linked program is not an original work of authorship, then it is not protected under copyright law. Now that may be OK to the open source community, but it will be, um, difficult position for a proprietary software producer to hold. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: OSL 2.0 and linking of libraries
Roy T. Fielding [EMAIL PROTECTED] wrote, in part: My take on this definition is that most statically linked programs include a relocation table and symbol tables which are annotations of the source code. These annotations are not particularly original, but if you declare that your statically linked program is not an original work of authorship, then it is not protected under copyright law. That only means it is not separately protected under copyright. The owner of the source code copyright retains control over all copying of the work, including copies that involve mechanical transformation and later copying of that transformation. You forgot 17 USC 117. See comments below... Otherwise, copyright would not be maintained when a modern digital copier is used to copy an art print, which is clearly not the case. Likewise, printing a document using PostScript does not create a separately copyrighted document even though the raw script looks nothing like the WYSIWYG editor window. I agree with Larry Rosen's view, but confusion in regards to this issue is so rampant that failure to state it explicitly in the license leads to an endless stream of FAQs. I don't think appealing to analogy of art prints is a good strategy, because software is special because of the fair uses permitted by 17 USC 117. To summarize... 1. The Softman v Adobe opinion referenced in this thread presented a court which very strongly disagreed that purchased software was licensed, even when the copyright holder claimed it was only licensed not sold. 2. The opinion on static linking, advocated here by Mr. Rosen and others, is that no derivative work is created by mere compilation and linking. 3. Then we have 17 USC 117, which states: (a) Making of Additional Copy or Adaptation by Owner of Copy. - Notwithstanding the provisions of section 106, it is not an infringement for the owner of a copy of a computer program to make or authorize the making of another copy or adaptation of that computer program provided: (1) that such a new copy or adaptation is created as an essential step in the utilization of the computer program in conjunction with a machine and that it is used in no other manner, or [snip] So if all the above holds, if I become a lawful owner of a copy of licensed software without assenting to the license, I can ignore the license terms if all I want to do is compile, install, and use the software unmodified, and the software is not encumbered by patents. (Only if there were patents involved, or I were to desire to create a derivative work, or make copies of the original or derivative work, I would need to agree to the open source license.) But let's say I didn't want to do anything more than compile and use itDoes that also mean I can ignore clauses in the license about limitations of liability? The Termination for Patent Action in OSL (and others) doesn't apply either then? Specification of Venue is irrelevant to me too? The GPL can easily be circumvented in practice too by automated package download and build systems. It makes the GPL not much stronger than the LGPL. I think this discussion is incredibly important to open source software. If the above end up being true, companies using dual-licensing strategies to generate revenue will not be able to rely on that revenue model. Wow. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: For Approval: Open Source Software Alliance License
Sean Chittenden [EMAIL PROTECTED] wrote, in part: Why does everyone insist that they're protecting my interests by likening a piece of BSD code that goes closed source as a bad thing or as if it's not what I want? That is precisely what I want people to be able to do! That's a smart business for reusing someone else's wheel design, kinda like a dated patent. If you are a small software shop, you sell services. So, what is the biggest threat to that line of work? Commoditization and customer lock-in to big-vendor proprietary solutions. GPL tilts the playing field against that. (I am not insisting you adopt GPL, just explaining one reason why GPL helps the small shop and IS a rational decision.) And yes, if you are selling a software library, we can see why BSD-licensed code is more sellable than GPL-licensed code. The market is better. But you haven't made your point as to why a BSD+anti-copyleft is more sellable, unless your customer is MicroSoft, or other company hoping to segment the world of open source software into as many incompatible islands as possible. The GPL is like the perpetual patent though, it never expires and becomes usable to other businesses. *shudder* It only took you two paragraphs to break your promise to maintain the distinction between closed source and commercial. GPL software is extremely usable in business.[1] If you can't keep your promise on the distinction, don't post any more. Forrest [1] Examples abound, but one that comes to mind is an business who would never dream of releasing the code I write for them under the GPL or any OSI certified license, but very happily uses GnuPG and the Windows drag-n-drop program I gave them to send me proprietary data via email. They didn't want to open up their VPN to me and PKZIP encryption isn't very good. Very cool. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Updated license - please comment
b) Accompany it with a written offer, valid for at least three tears, to give any third party, at no charge, a complete machine-readable copy of the corresponding source code, to be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software Well, are you sure you proofread it? Three tears? :-) I found that and stopped reading it. But I don't know that clause 2 d is open source. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Updated license - please comment
Mark Rafn [EMAIL PROTECTED] wrote, in part: It doesn't even seem close to me. Let me know if I'm insane, or reading it wrong, but I can't see how such a restriction can be considered open source. I know they're straight from the LGPL, but they are irrelevant there because the LGPL is a pure superset of the GPL (see LGPL section 3), unlike the license under discussion. Yes, this indicates that I think the LGPL without section 3 would be non-open-source. I agree with you. I have difficulty understanding 2d. It seems complicated. In private email, Christophe Dupre summarized the intention this way: 2 d is there to make sure that the library remains usefull by itself, and does not become a wrapper for proprietary tools. I don't know if 2 d meets that intention or not. It is hard to understand 2d. But I have my doubts that the intention is compatible with Open Source, (depending on how we define Proprietary) -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: OSD Model Code -- Article 1 (Free Distribution)
From: Rod Dixon, J.D., LL.M. [EMAIL PROTECTED] To:Forrest J. Cavalier III [EMAIL PROTECTED], [EMAIL PROTECTED] Subject: Re: OSD Model Code -- Article 1 (Free Distribution) Do you mean clause 5 of version 2.0 of the Artistic License? If so, would you agree that the proposed change, either your suggestion or Larry's, would avoid the problem caused by the current Art. 1 of the OSD or do you think there is still a problem with clause 5? http://www.opensource.org/licenses/artistic-license.php clause #5 reads: 5. You may charge a reasonable copying fee for any distribution of this Package. You may charge any fee you choose for support of this Package. You may not charge a fee for this Package itself. However, you may distribute this Package in aggregate with other (possibly commercial) programs as part of a larger (possibly commercial) software distribution provided that you do not advertise this Package as a product of your own. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: OSD Model Code -- Article 1 (Free Distribution)
With my rewording, there's also no need for the confusing term aggregate software distribution. We only need to rely on the definition of the term copies in the Copyright Act. 17 USC 101. I like the clarity of Larry's , but I think the clumsy wording of OSD #1 was to permit the Artistic License clause #5 to qualify. Please read the Artistic License clause #5 and see if the new proposed wording will continue to treat that license the same way. Forrest -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: OSD Model Code -- Article 1 (Free Distribution)
From: Russell Nelson [EMAIL PROTECTED] To:[EMAIL PROTECTED] Subject: Re: OSD Model Code -- Article 1 (Free Distribution) I think you're doing more than clarifying. I think you're introducing additional restrictions on a licensor -- and that's good! What if somebody tried to restrict a licensee from selling or giving away the software as a stand-alone package?? Take a look at the Artistic License clause #5. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: time frame between request for approval and acknowledge
Russell Nelson [EMAIL PROTECTED] wrote in [EMAIL PROTECTED] John Cowan writes: With respect, Russ, that's bassackwards. Collectively if not individually, the members of the list have far more free man-hours than you do. You should pass submissions straight on to the list and let one or more of us shoot them down if they are obvious losers. Okay, unless list members have an objection, that's how I'll handle submissions in the future. I wondered why the list has only been seeing submissions of high quality licenses. We used to more commonly see all kinds of OSD violations in the submitted licenses. (I thought the world was getting smarter about open source. Ah well.) I agree the list members could function as first-line review for licenses, AS LONG AS THEY REMEMBER THIS: We are ambassadors for open source. We want more open source. And those proposing licenses are newbies, not converts. Be gentle. I think Russ started being first-line contact back when clearly bad licenses were very likely to get flamed. That became a big turn-off to the suits bringing licenses to license-discuss. We don't even need someone volunteering that I would never use such a license. It is adequate to explain the violation if you think there is one, and point them to the rationale at www.opensource.org. If they want to follow-up, then that is the time for continuing a further detailed discussion. Forrest -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
OSL 1.1 treatment of documentation
3) Grant of Source Code License. The term Source Code means the preferred form of the Original Work for making modifications to it and all available documentation describing how to access and modify the Original Work. access is not well-defined. Is your intent to compel book publishers to give away the text of their books written for users of the Work? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: OSL 1.1 treatment of documentation
Lawrence E. Rosen [EMAIL PROTECTED] wrote in part: 3) Grant of Source Code License. The term Source Code means the preferred form of the Original Work for making modifications to it and all available documentation describing how to access and modify the Original Work. access is not well-defined. Is your intent to compel book publishers to give away the text of their books written for users of the Work? No, and I don't think the word access conveys that meaning. accessed appears in OSL 1.1 paragraph 5. And it seems that use is different than access. Can you explain what you meant in each paragraph? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: OSL 1.1 treatment of documentation
Lawrence E. Rosen [EMAIL PROTECTED] wrote in part: Then, Forrest's question: what about a book that isn't a derivative work? Could contract law and some technically inept judge compell the book publisher to release the book's source code (DocBook / TeX / whatever) under OSL? Not if it ain't a Derivative Work, I'd say. How does not being Derivative Work matter? Isn't the OSL a contract and not a copyright license? Derivative Work doesn't even appear in OSL 1.1 paragraph 3. Consider a case when the software is meant to be deployed as a network service. Users use the software by accessing it via a network. Seems to me by the OSL paragraph 3, a book publisher is compelled to include a machine-readable copy of their book if they provide a copy of the Original work. For access and modify the Original Work did you mean obtain? locate? identify? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
OSL 1.1 treatment of derivative works
Larry wrote, in part: Not if it ain't a Derivative Work, I'd say. Does OSL 1.1 1(c) with paragraph 3 require distribution of derivative works or not? Paragraph 3 mentions Original Work, not Original and Derivative Works. So it seems not. - Shouldn't the or in 1(c) be and? - -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: A few here may have an opinion on this
I looked up MSFT's latest financial release (operating results for Sept 02 quarter, Oct 17). They had $7.4B revenue and $2.7B net income. EBIT was $4B and they paid $1.3B in taxes. In the Sept 01 quarter (a really bad quarter) EBIT was $1.9B and they still paid $600M in taxes. Do you still think that proprietary software is not a great business model? You missed my feeble attempt at sarcasm with my MicroSoft commentBut can anyone provide a quick update for 2002 of the numbers and accounting techniques explained in the following from fall 2000? http://www.sfgate.com/cgi-bin/article.cgi?file=/chronicle/archive/2000/10/09/MN3707.DTL -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Copyright
modifications, and distribute my own version, can I remove all instances of RedHat ? 2. If I am allowed to, to what extent? The OSD allows licenses to prohibit that. Some jurisdictions may prohibit it. As John Cowan noted separately, trademark law can require the removal of trademarks. But a copyright notice is not a trademark. But the answers does the law and license permit it and should it be done are probably different. Of course no one should do that. I think there are the obvious ethical reasons. I think there is an important practical consideration: By removing all traces of the authorship, you will appear to be the responsible party, and not a mere re-distributor of the work. That may result in legal liability you didn't count on. You must be sure to communicate all license terms and warranty disclaimers, and who is offering the license and making the disclaimers. Doing anything else is fraud (usually illegal) or creates liability you don't want to assume. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Warranty
Sure. Take a look at http://linuxjournal.com/article.php?sid=6155. So, how does someone with sense or conscience redistribute software and offer a warranty of non-infringement for software they acquired or is a combined work? Relying on a cascade of breach of warranty lawsuits back through the distribution chain doesn't sound like a reasonable business risk. Is there an insurance industry around this? (And if there were, is it as risk averse as the underwriters for all other lines of insurance have become in the last 13 months?) I think the best to reasonably expect from a licensor of software at reasonable, or little. or no cost, is: For materials exclusively authored by Provider, Provider warrants that the listed licenses apply to You. Provider warrants that all licenses and disclaimers that Provider is aware apply to You as of this date are as follows: [list here] And providing the second part of that warranty is a stretch for most OSS distributors. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: discuss: OCLC Office of Research Open Source License
This is how combined work is defined... A Combined Work results from combining and integrating all or parts of the Program with other code. A Combined Work may be thought of as having multiple parents or being result of multiple lines of code development. Is this definition precise enough to exclude agregate works? As written, it seems that Agregate works are a subset of Combined works. This indicates to me that Section D restrictions, because they would apply to CD complilations for example, are not OSD compliant. As another note, I personally do not consider object code non-modifiable. In fact, back in the day when assemblers were slow, simple flaws in code could be hand-patched by replacing the defective code with a jump instruction out to unused memory, and hand-assembling the code to perform the correct operation and jump back. This hand-patching would save about 25 minutes of time for each bug, with the added BIG advantage that the 180 page assembler output would remain usable (because subroutine addresses would remain the same.) -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Open Source Click-Wrap Notice
[EMAIL PROTECTED] wrote, in part: Your answer added nothing to the discussion. Please give some legal argument why a single click-wrap won't bind the licensee to all relevant licenses. How do you form a contract without presenting the terms? Is there a way to review the terms without clicking? Is such vague language sufficient to incorporate all the terms (of those possibly 800 licenses) by reference? Seems against common sense to me. Further, any consciencious, thinking user would never click to agree to something like that. I know I wouldn't. It is just too unpalatable. I would want to agree individually, not in bulk. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: copyleft lite?
Seems to me that it is not GPL compatible. The GPL expressly prohibits combining with licenses that have more restrictive terms than the GPL. without fee is a restriction the GPL does not include. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Approval request for BXAPL
This is a very complicated license. Thanks for providing the remarks and annotations. Very nice. After a quick read, I think that it should not be OSI approved, for numerous reasons, some follow. Because the license is so complicated, it is not clear to me that addressing the following points would be adequate to make it OSD compliant. I think the two fundamental issues are it very much seems to be a EULA, trying to function under copyright law, and EULA's are hard to get through OSD. and the Rationale a) is very difficult to set forth legally. Any license which attempts to make such distinctions must get extra scrutiny. After all, the license ends up defining Programming Tool as Any Software or portion of such Software that is declared as being a programming tool by the Copyright Holder. Such declaration must be located in the Copyright Notice. which I think is ripe for abuse and inconsistency in itself. The definition of User is too broad. It allows any Distributor to force someone to be a User simply by sending them a copy. The Source code definition includes this statement, Source files or members that contain obfuscated source do not count as Source Code. Obsfucated is not well-defined. I've see a lot of legitimate source files that appear to be obsfucated. Other OSI-approved licenses have definitions of Source. Also, as written, I think this definition includes compilers and linkers (and more! run-time ld? ) as Source code. Paragraph 6 says, in part: No right is granted to the trademark(s) of any Contributor even if such marks are included in the Software. The names of Contributors or any of their products may not be used in any way without prior written permission from the pertinent Contributor. Derivatives and/or Dependent Software may not be named after the Software, nor may they be given a name that might be confused with the name of any Contributor or any Contributor's products and/or trademarks. Remarks Since Contributor is defined as to include any Distributor it is fundamentally impossible to know the set of Contributors. Without knowing that set, it is impossible to know what names you are not permitted to use. The even if such marks are included is a problem when you also require (in a separate paragraph) verbatim distribution of the software. I read that as when there is any trademark in the software, you are not permitted to distribute. --- Paragraph 10 claims that all items which make use of ... the original or modified versions of the Software are Derivatives. That is plainly wrong. md5sum will make use of the Software to compute the MD5 sum, and by Paragraph 10, md5sum is a Derivative of the Software. --- Paragraph 12 uses the term Distributor and Contributor in a manner inconsistent with the definition in the Item 2. --- Paragraph 12.2 is unsatisfiable because the set of Contributors is unknown. It is against the OSD because as I read it, any new official contributed modification will invalidate existing versions. This constitutes the need for a separate license: always checking some official site that there are no new contributions. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Approval request for BXAPL
Steve Lhomme [EMAIL PROTECTED] wrote, in part: A Contributor can be (or not) a Distributor. A Distributor can be (or not) a Contributor. That's what the definitions say. The definition (at General #2) is as follows, and is formatted thusly: Contributor: Any Distributor and/or any User supplying any Modifications and/or Derivatives and/or Dependent Software in any form to the Copyright Holder, but not to any User(s). That makes it clear that every Distributor is a Contributor -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Approval request for BXAPL
[Discussion of Paragraph 6] The even if such marks are included is a problem when you also require (in a separate paragraph) verbatim distribution of the software. I read that as when there is any trademark in the software, you are not permitted to distribute. -- In my opinion that would not constitute the use of such trademarks. Would this not be so under US law? One way of including marks is as icons and graphics. These may appear as output during the normal course of running the software. (splash screens, for example.) They may appear at page headers and footers of printed pages produced by the software. I'm not a lawyer, but under US law, trademarks have to be used as a description of something, a description which identifies the source. If there is a trademark in the software at all, then it must identify the source. If I am not that source, then I can't use the mark without separate agreement. Am I allowed to remove the marks by the BXPL? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Approval request for BXAPL
The definition of User is too broad. It allows any Distributor to force someone to be a User simply by sending them a copy. But does it arm any part of the license ? Or just a personal feeling ? 8.5 seems to have an effect for Users 15 may also. 16 also, but 16 is hard to follow. Realistically, I think no court will find you had obligations if someone sent you a source without request. But what if someone obtains the source for determination if it infringes a patent? Does the license allow them to inspect the software without obligating them to #16? How? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Approval request for BXAPL
Also, as written, I think this definition includes compilers and linkers (and more! run-time ld? ) as Source code. ld is not a Source file. The BXAPL says Source Code is ... and any other files or members needed to create the executables required to properly execute the Software What if special features of ld are used? What if special features of MSVC are used in the source code? gcc is really-high quality software now, but if you dig into any complicated piece of software at least 10 years old, you can probably locate well-commented code written a certain way to avoid triggering some gcc bug. Am I allowed to put a note on the source code which says: Only compiles with MSVC 6.0 and later.? (I think I know your intent, but that doesn't match how I read the license.) -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: UnitedLinux and open source
David Johnson wrote in part, in a message to [EMAIL PROTECTED] Here is the FSF's definition, which is remarkably similar to your own. *) The freedom to run the program, for any purpose. *) The freedom to study how the program works, and adapt it to your needs. *) The freedom to redistribute copies so you can help your neighbor. *) The freedom to improve the program, and release your improvements to the public, so that the whole community benefits. I don't see where the original Artistic License (or any OSI approved license including the APSL) fails to meet this definition. I think a careful reading of the QPL shows it is not a Free Software License, (even though the FSF web pages say it is) for two reasons: 1. QPL 6c compels you to disclose source to the original developer. I think this is against FSF freedom to associate (or not). 2. QPL 6a is worded such that EVERY distributor must provide at cost copies to ANYONE with a binary copy, regardless of who distributed that binary. This is against FSF freedom to charge a profit. (This is OK with the OSD, because the OSD allows licenses which prohibit a profit.) I have emailed the [EMAIL PROTECTED] questions about this a number of times over the last year, but have never received a reply. There are other licenses, (like Jabber?), which also prohibit profit when distributing copies. My opinion is that neither the FSF nor the OSI should approve such licenses. But they do. Forrest J. Cavalier III Mib Software -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Proposed License-RPL
3.4 Licensor reserves the right at any time to cease public release of the Licensed Software or, if released, to change fees and prices, features, specifications, capabilities, functions, licensing terms, release dates, general availability or other characteristics of the release. Title, ownership rights, and intellectual property rights in and to the Licensed Software shall remain in Licensor and/or its Contributors. Revocable/Temporary rights are not compatible with the OSD. (I think there are other clauses which are in conflict with the OSD, but that one seems the most serious after a quick read.) You can find some commentary on revocable rights in Bruce Perens' comments on the original APSL. Apple made modifications to create the current APSL. If you did not intend to be able to change license terms of already released software, then the paragraph could be reworded. - It would be helpful if you follow the OSI submission guidelines, (especially when you describe how this license is similar and differs from existing approved licenses.) It seems that there are many similiarities between the RPL and the APSL or the Jabber license. Are you aware of those licenses? -- Forrest J Cavalier III Get help selecting a license, or knowing OSD compatibility with LIDESC: http://mibsoftware.com/librock/lidesc/tags.htm -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Discuss: The Open-Source Milestone Application Framewor
Akil Franklin wrote: How could we reword clause 6 to make it less clumsy and to make it more palatable? As written, clause 6 obligates everyone who reuses the smallest piece of your software. 6. All advertising materials regarding products derived from this software and/or features derived from its use must include the following acknowledgement: There is much written on how the original BSD advertising clause reaches too far, but it does have a limitation All advertising materials mentioning features or use of this software must display... Is there a reason the BSD phrasing does not meet your goals? Is there a reason you cannot dismiss the clause entirely? Even though encouraging advocacy of products by users is a big reason that companies want to use open source, it is a challenge to do successfully. IMHO, if you try to grab too much in the license, you will be seen as too self-promoting, which will discourage adoption (possibly because users get suspicious of your motives: are you trying to help them, or help only yourself?) -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Discuss: The Open-Source Milestone Application Framewor
Akil Franklin [EMAIL PROTECTED] wrote: Article 7 is basically meant to ensure that the authors of the framework are notified when it is used (i.e. placed into a production system either modified or not). Again, the goal is the sharing of information. Perhaps the following would be better: 7. Milestone must be notified when this product is placed into a production system (either modified or not). Such notification can be made by visiting www.techMilestone.com and filling out the appropriate online form. If you have any questions about this process, please send an email to [EMAIL PROTECTED] Better? I don't think it is better. Notifying a specific entity requires that the entity exists. This may not be the case in 10 years, or next week (which is not an insult, just a statement of the risk that an adopter will recognize.) Secondly, requiring a specific method of notification requires that the method is available. This could allow the licensor to control who can use the method, which is a clear OSD violation. For community philosophy on this point, you can read what Bruce Perens et al. wrote about the APSL 1.0 clause 2.2(c) at. http://perens.com/Articles/APSL.html As you can see in the APSL version 1.2, clause 2.2(c) does not have the same problem. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Discuss: The Open-Source Milestone Application Framewor
Akil Franklin [EMAIL PROTECTED] wrote: Forrest J. Cavalier III [mailto:[EMAIL PROTECTED]] wrote: Notifying a specific entity requires that the entity exists. This may not be the case in 10 years, or next week (which is not an insult, just a statement of the risk that an adopter will recognize.) Secondly, requiring a specific method of notification requires that the method is available. This could allow the licensor to control who can use the method, which is a clear OSD violation. Excellent points. The main issue seems to be what happens when Milestone goes out of business or changes their URL? It seems to me that we can fix this issue by amending a few conditions to the clause. How about the following: [snip] Yes, better, but... What about the second issue? You could run a firewall which excludes access based on IP address you select, or other criteria (browsers?) You could reject email based on criteria you choose. The No Separate License clause of the OSD should be (and is) interpreted broadly. Requiring signing up on a web page is requiring a separate license. Even if you provide an outlet clause in the case of unavailability, it is not going to be OSD compliant. Even if such a clause would pass the OSD, what does unavailable to the global internet mean? If 1 site cannot reach it does that count as unavailable to the global internet? If unavailable to the global internet means that no sites can reach you, how does anyone know? To a specific observer, how is destination unreachable distinguished from a firewall which black-holes traffic from certain IPs? To a specific observer, how is no response distinguished from a Distributed denial of service against your servers for 15 days? What defense does someone have if they say they registered, but you say they did not? You can have legally binding agreements which happen over the web, but it isn't simple. I am starting to think that this license and these clauses would benefit from further review by legal counsel before it is resubmitted. Forrest -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: New FSB branding model
You should just be using URIs (or even URNs) for the license file. Don't use a hash because this can change if you license file changes (think whitespace). Just use urns... urn:gpl or urn:lgpl should be fine. That was considered. There is no central authority for assigning URNs. Many licenses are already available and freely distributable, and it is hoped that those are used instead of re-created. But in the LIDESC open architecture, anyone can create a new stamp and description file without going through a central authority. LIDESC stamps already include a urn-like item: the name of the license text file. Because there is also a hash code, LIDESC can search a list of directories for a file which matches in name AND hash code. So even if there is a collision in file name, LIDESC still identifies the right license. As for whitespace, \r is removed when calculating the hash code, but there are some contrived cases where differences in whitespace have legal significance. Consider indenting whitespace when there is a phrase like You must place the following notice... followed by an indented copyright notice, and then a warranty disclaimer, not indented. Did I fully reply to the issues you raise? Forrest -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Discuss: The Open-Source Milestone Application Framework So
6. All advertising materials regarding products derived from this software and/or features derived from its use must include the following acknowledgement: Is that worded as you want it? There is some clumsiness there and I am not sure what features derived from its use really means or is different from products derived from this software You know that this clause reaches much farther than the orginal BSD advertising clause. There are probably some who will not adopt the software because of it. 7. Milestone must be notified of the redistribution and use of this product. Such notification can be made by visiting www.techMilestone.com and filling out the appropriate online form. If you have any questions about this process, please send an email to [EMAIL PROTECTED] I seem to recall previously submitted licenses rejected for clauses similar to this. (On the grounds that this fails the no separate license requirement of the OSD.) Consider what happens when the company goes out of business, or the WWW is replaced by something better. I think a requirement to notify the public is OK, a requirement to notify a specific entity is not. 8. Milestone Technology Group, LLC is granted the right to use the name and logos of products derived from this software in communications related to the Milestone Application Framework. Yikes! That may not be against the OSD, but I think you are going to frighten away anyone who considers this software. Forrest -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Discuss: Submission of Sybase Open Watcom Public License v.
COMMENT: This section has been added because of last year's Specht v. Netscape decision, which threw doubt on the validity of what = has become an accepted industry standard - of trying to make terms and conditions applicable (particularly in the web site context) without ma= king it clear to the user that he/she is accepting a contract and being boun= d by its terms, and without requiring some affirmative manifestation of assent. Was that paragraph supposed to say something different? As written, it indicates that prior to Specht v Netscape there was some way to make contract terms and conditions applicable without making it clear to the parties. (and that it was an industry standard.) That's nonsense. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: source code button
Here's the big problem with this clause: I can ignore it with impunity! The clause is Affero GPL 2(d), which means it is effective only when you are preparing derivative works. This is how GPL clause 2 starts: 2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions: This is good enough to preserve freedom, since it prevents proprietary forks. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Affero GPL. Big loophole?
Steve Lhomme wrote in part. Because as the software is GPLed, you must supply the source code on request. Where does clause 2(d) state that? I want to discuss 2(d) specifically. Certainly the other clauses of the GPL will remain in force, but there are circumstances where only Clause 2(d) will apply. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Affero GPL. Badgeware?
Henri Poole [EMAIL PROTECTED] writes: I wouldn't interpret it that way. Were you involved in the creation of the text? Do you speak for Affero? That would be very helpful to know. This license is going to be used by others. (The intent and interpretation that Affero has is important, but not the only one that will matter.) The facility is defined previously as any user interacting with the Program was given the opportunity to request transmission to that user of the Program's complete source code There is no definition of facility. It does not specify that you cannot modify the facilty...but that you must offer an equivalent opportunity for all users interacting with I read the text has two separate requirements, separated by and you must not remove that facility and must offer My point is what purpose is served by having the must not remove? If you argue that the second is the necessary requirement, then why have both? According to the text in 2(d), if I completely replace an existing method with my own (maybe I copied a new file over top of an existing one), didn't I remove that facility, which the license prohibits? Have I merely modified it, even though not a single bit of it remains? It bothers me that the clause could be, but is not continue to provide an equivalent facility You seem to argue that was the intent. But if you change it to be that, the text becomes redundant, (since that is what the and must offer clause states.) It is my understanding that one way that courts interpret legal documents is that every clause must provide meaning. It is not safe to argue that clauses should be ignored as redundant when there is a way to interpret them in a way which provides non-redundant meaning. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Affero GPL. Big loophole?
Free Software Foundation Announces Support of the Affero General
Public License, the First Copyleft License for Web Services
http://www.fsf.org/press/2002-03-19-Affero.html
(NOTE: The FSF suggests comments to them. I CC'ed them, but
I'd prefer discussion in a forum. license-discuss seems
appropriate.)
Please, someone tell me how I am misreading the added clause
2(d).
Where are the teeth? In my reading, the essence of the interesting
part of clause 2(d) is
must offer an ... opportunity ... to request
immediate transmission ... of the complete source code.
I find it _extremely_ odd that it does not compel transmission.
(Other GPL terms are clear that you must supply source code. This
clause is different. Why?)
If compelling the transmission is the goal, why is the license
clause only compelling the offer of opportunity to request?
Is this on purpose? Why create a huge loophole when it could
have been written clearly?
As an exercise, could someone explain why the following response is
_not_ compliant with 2(d)?
int main(int argc,char **argv)
{
printf (We received your request for the complete source code.\n);
printf (BRAGPL 2(d) does not obligate us to supply it when responding.\n);
printf (BRHave a nice day!\n);
}
Here is the added clause 2(d) (reformatted line breaks,
anyone have a .txt copy?)
-
* d) If the Program as you received it is intended to interact with users
through a computer network and if, in the version you received, any user
interacting with the Program was given the opportunity to request
transmission to that user of the Program's complete source code, you
must not remove that facility from your modified version of the Program
or work based on the Program, and must offer an equivalent opportunity
for all users interacting with your Program through a computer network
to request immediate transmission by HTTP of the complete source code
of your modified version or other derivative work.
-
--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Affero GPL. Badgeware?
Free Software Foundation Announces Support of the Affero General Public License, the First Copyleft License for Web Services http://www.fsf.org/press/2002-03-19-Affero.html I think clause Affero GPL 2(d) provides for the propagation of badgeware, meaning that it obligates propagation of a brand or mark. The important bit is you must not remove that facility. I find that this 2(d) operates much differently than GPL 2(c). GPL clause 2(c) obligates providing a run-time notice under certain conditions. In other words, 2(c) does not specify that a _specific_ notice is to be preserved. In contrast, Affero 2(d) does. How does this work in practice? If there is a download graphic, I think the only way to comply is that you keep the graphic. If there is a paragraph with superfluous marketing, you must keep paragraph. This lets the originator insert branding and propaganda, and forces it to be propagated. This is unnecessary. I suggest that software freedom is preserved fully if you delete the part of 2(d) I placed in brackets below. I hold that that bracketed text is an unnecessary compromise of freedom, serves no useful purpose except to promote branding, and should be removed before this license gets the full support of the Free Software Movement. In making the deletion, some further adjustment may be necessary, so that the important text modified version of the Program or work based on the Program appears. (the funny word spacing comes directly from the HTML.) * d) If the Program as you received it is intended to interact with users through a computer network and if, in the version you received, any user interacting with the Program was given the opportunity to request transmission to that user of the Program's complete source code,you [must not remove that facility from your modified version of the Programor work based on the Program, and] must offer an equivalent opportunity for all users interacting with your Program through a computer network to request immediate transmission by HTTP of the complete source code of your modified version or other derivative work. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: OSD modification regarding what license can require of
From: Russell Nelson [EMAIL PROTECTED] To:[EMAIL PROTECTED], [EMAIL PROTECTED] Subject: Re: OSD modification regarding what license can require of The OSI-approved the W3C license which has a requirement of displaying a notice to users. If it's in the documentation, such a notice is displayed. It says nothing about that notice having to be displayed by the software itself, whereas the GPLv2 specifically says that the program must display the message. Yes, I want to make a slight correction. The W3C says that the notice must be viewable, it is not, as I wrote before, a requirement of displaying. The point of my message was to point out the phrasing viewable to users. (because I think it would impact ASPs.) -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: OSD modification regarding what license can require of
Russell Nelson [EMAIL PROTECTED] wrote: We have never removed OSI approval for any license, APSL 1.0 so I'm sure that the GPL is in no danger of not being an Open Source license. You are the tail trying to wag the dog, Russ. If the OSD can't accept the GPLv2, then the only plausible conclusion is that the OSD is wrong. Plain and simple, that. We can't make any conclusions about GPLv3, since we haven't seen the details. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: OSD modification regarding what license can require of user
Bruce wrote (in part) So, what if it turns out that the present GPL doens't hold up with regard to dynamic linking? Some future version of the GPL might have to place a constraint on the user regarding combination of works on the user's system that would, if it were distributed in that form, be considered a derived work. I think that should be allowed by the OSD. Did you somehow mean constraint on the publisher regarding combination of works on the user's system? Thus, there seem to be a few sorts of requirement on the user that I think _should_ be permitted by the OSD. All of them are intended to further the goals of Open Source or Free Software. freedom and requirement are direct opposites. Imposing requirements may further the goals of Open Source, (whatever they are, I thought they were freedom.) Imposing requirements on users is unlikely to further freedom, by definition. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: OSD modification regarding what license can require of
I just want to point out that there is one license already approved which has a public performance clause like Bruce gave as an example.. The OSI approved the APSL, with clauses 2.2c-d, which require publication of sources upon deployment. Can we use this concrete case to clarify the goal of this discussion? So Bruce, (correct me if I am wrong), your goal is OSD changes which better ensure user freedom, but still allow approval of the APSL (and as-yet-unwritten licenses with clauses like you mentioned.) Or is the goal instead to relax the OSD to allow approval of more licenses with clauses like you mentioned? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: OSD modification regarding what license can require of
David Johnson [EMAIL PROTECTED] wrote (in part) The only way around this is for the author to put his morality on the shelf and try to impose click-wrap licensing on me (bad), or to be honest and present me with a contract prior to my aquisition of the software (doubtful it would meet the OSD), or to make these terms a condition of the granted permissions. The latter might fit into the OSD somewhat, but it would still be a no-op regarding the ASP issue. David didn't make it explicit, (and I'm sure he knows...,) but it should be stated that the first two only way around this are clearly not options with software under OSD-compatible licenses. Click-wrap licensing doesn't work when recipients can modify the source code. They can simply disable the wrapper when they redistribute. Presented and accepted contracts don't work for a similar reason. (OSD #7 says all rights must apply with no separate license.) So that leaves only the third, which is make these terms a condition. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Discuss: BSD Protection License
OSD-related issues that I see 1. Someone already pointed out the OSD #1 issue. If the license doesn't explicitly permit selling copies, then copyright law reserves the right to the author. 2. Except possibly for the copyleft clause 4c, the license fails to state that the terms apply to everyone receiving a copy of the software. The OSD could word it better, but the license terms must attach to the program being distributed. Is there any OSI approved license which relies on a chain of licensees? Look at how OSI-approved licenses are worded and compare. They are grants and attached licenses. (Sometimes you have to read the definition of you or recipient.) The BSD Protection license defines you in a way that has a self-reference. There is no way to become a you unless you are a licensee, and there is no way to be a licensee unless you get it from Clause 4c. 3. The distribution term of clause 3c is obviously in conflict with the OSD. That clause should not be approved. The OSI approval process has rightly taken a dim view to license clauses which are OSD conflicts, even when there is an alternative clause. The BSD Protection License is two different licenses. If rewritten in the standard way of disjoint licenses, then the version with Clause 4 terms would seem pretty close to OSD compliant to me. - I think the ideas behind this license are somewhat novel. But the wordings leave me a lot of questions. 1. There is no definition of the terms open license and closed license. (If those terms are not useful in selecting either clause 3 or 4, then they should not be in the wording.) 2. I think as worded, you must make the election of Clause 3 or 4 as an organization, not on a work-by-work basis. For example, suppose I distribute two separate derivative works, and I choose the clause 3 for the first work. Later, I make a separate work and choose clause 4. I think that clause 4c says that I cannot continue distributing the other work under the terms of clause 3. 3. When read a certain way, the wording of the copyleft clause 4c is even more tainting than the GPL copyleft (ref GPL section 2). Such an interpretation of clause 4c doesn't seem to be in the spirit of the preamble, so I expect the wording can be adjusted to clarify. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Discuss: BSD Protection License
I didn't define Definitions, either. ducks I have no legal training, No legal training required for discussion here. And according to Larry, if you have legal training, there is some discussion you should not be doing here. :-) but I thought it would be clear that Modification and redistribution under open license and ... under closed license were simply section headings. Hmmm. Not clear to me. They are worded as selection criteria, in which case the selector must know the definitions. (Someone else also pointed out that the election could be made more clear. I still think it should just be a disjointed license.) Since Clause 3a and 4a are identical. 3b and 4b are similar, Did you consider rewriting so that 3a and 3b/4b are always required, and then have a must do one of the following clauses? (BTW, 4b is kind of confusing. The full details of a modification ARE the modification. Did you mean full description?) Err, which wording did you want to see adjusted? The GPL 2(b) is in context. The GPL is very carefully worded and concise which indicates to me that the GPL paragraphs after 2(c) are required in the license itself to influence the interpretation of the clauses in that section. The BSDPL does not have the same explanatory text, and 4c is in a slightly different context than GPL 2c. As I said, I understand from the preamble what you meant. But the preamble is not the license. The unintended interpretation of 4c will taint works reasonably considered independent. Here's a test. If I distribute source for software under BSDPL terms that consists of two source files, one BSDPL, and one that was originally under the MIT license. Can the recipient distribute the works separately, and use the MIT license for the second? The GPL is clear and allows it. The BSDPL is not clear. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: request for approval of APOSSL
APOSSL is a BSD style licence save for the following special points. * the name of the software should not include pronoic.org or Pronoic Ltd. That makes it like the Apache license, I think. * the software should be described as being pronoic unless you ask for permission to use the term pronoic. Yes. I see that clause 4 says that. in that case your request will be denied. Clause 4 doesn't say that. It says that you can get written permission by contacting [EMAIL PROTECTED] Before obtaining written permission also indicates that written permission is available. Read it: * 4. The term pronoic should be used to endorse and promote products derived *from this software before obtaining written permission. For written *permission, you must contact [EMAIL PROTECTED] But since you have just written you will never give permission, the license clause 4 is nonsense and is an offer in bad faith. As far as I am aware, having slightly off-beat goals is not a reason to deny an OSSL such accreditation. It is well recognized that the OSD is incomplete. I think it is proper that the OSI board rejects licenses with nonsense clauses or that are internally contradicting, even if they do not have an OSD conflict. I think it is proper that they pay no attention to licenses when the author says up front that he puts it forth in bad faith. Whine all you want about how it isn't fair that you have a quine'd license which is OSD comptible. I still recommend it is soundly rejected as a this statement is false license. Use the Apache license. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: request for approval of APOSSL
What does quine'd mean? http://www.ship.edu/~deensl/pgss/Day16/goedel.html (I admit I used the term loosely to describe a statement which can be read as a self-reference at more than one level that creats a contradiction.) Here is the response I would give you about OSI approval for your license. Your request will be rejected is your request will be rejected. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: request for approval of APOSSL
I wrote: Here is the response I would give you about OSI approval for your license. Your request will be rejected is your request will be rejected. I thought of another appropriate response... We will refuse when you ask is we will refuse when you ask. And another... We won't reject until you ask is we won't reject until you ask. (I don't speak for the OSI board, of course.) You've had your fun. Now move along. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: serious?
all fun aside, I am serious about APOSSL and believe I have reacted in a serious manner to all serious points made. Serious means more than simply not joking. John Cowan pointed out a major mistake in 1.0, which was totally the opposite of what you intended. If you were serious you would have found a major error by seriously reading what you wrote and discussing it with others before it got to license-discuss. If you were serious you would have considered using an attorney to help with the precise language to capture your goals, as well as making it OSD compatible. Clearly you are not that serious. But since you followed the standard submission procedure, I am willing to approach this anew, all joking aside, and report that I am convinced that Version 1.2 has a serious difference compared to your stated intent, and is in conflict with the OSD. Summary: --- Based on your other emails, you expected that using should was more lenient than using must in clause 4. You want clause 4 to be encouragement to use the term, but not a requirement. The problem is, that since clause 4 appears as a condition of license, the wording there must be interpreted as a condition. (And that alters how we read should.) Because of this condition, the license conflicts with the OSD no discrimination requirement for certain fields of endeavor. (Namely those which do not fit the definition of pronoic.) Exmplanation: First, let me establish that the field of endeavor I described is plausible. I get junk email ads for internet spy software. That leads me to believe that there is a field of endeavor which is the opposite of pronoic. Let's call the field of endeavor Software for the paranoid and to make people paranoid The license states there is a list of conditions. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: .. One of those conditions is .. * 4. The term pronoic should be used to endorse and promote products derived *from this software before obtaining written permission. For written *permission, you must contact [EMAIL PROTECTED] The first sentence of clause 4 must be interpreted as a condition to be met. When is this condition met? Clearly, a descriptive term should be used only when it is an accurate description. In the field of endeavor I gave as the example, pronoic as defined by the license is not an accurate description. It is false that the term should be used. Ergo, clause 4 is not satisfied, the license conditions are not met. Since that example field of endeavor gets no rights from this license, the license discriminates. You cannot get OSD approval for this license. (Alternately, one might argue that they just need to ask permission to get out of clause 4. That conflicts with the OSD no separate license requirement.) Conclusion: -- Maybe there is a bandage or touch up which will fix these problems by altering a word or two. (I doubt it, but I don't care to try.) As I wrote before, having an attorney write the license is not a requirement for OSI submission. But, in my opinion, (and I mean this in the nicest possible way), consulting an attorney about your case might be a good idea. Forrest -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: request for approval of APOSSL
[snip] We seek to spread our ideas, meme like, through both non-commercial and commercial channels We do not seek to restrict use of our software by anyone, and for the most part our licence is bog-standard OSS stuff, but we do have some weird demands on them should they do; like a deal with Satan, only funny and positive not at all evil [snip] only funny and not evil is matter of opinion This is not a Free software license because clause 4 requires promotion of derivatives I should be free to create a derivative and keep it totally private, which is not allowed by clause 4 I also think the OSI should not approve it Your text explanation of the clause 4 and 5 is not going to be part of the license Approval is based on what the license says, not what you say it says In my interpretation, Clause 5 is a clear OSD conflict pronoic is not a word, it is (appears to be) a name Distinctions in capitalization of Pronoic cannot be significant Therefore, the phrase: Products derived from this software will always be pronoic can only have one interpretation: that pronoic is a name, and that all derivatives must be named pronoic Then clause 5 goes on to say that Pronoic may not appear in the name without prior written permission That's a conflict with the OSD Your meme may be important to you, but I think you can find better ways to propagate it than compelling behavior with a license BTW, I am glad you looked through the archives (It would be nice if everyone would follow that procedure before submitting) But the license-submission procedure appears on wwwopensourceorg, and following it will streamline your submission -- license-discuss archive is at http://crynwrcom/cgi-bin/ezmlm-cgi?3
Re: request for approval of APOSSL
dave sag [EMAIL PROTECTED] wrote, in part Clause 4 does NOT require promition of derivatives at all. Should you never obtain written permission, you never need endorse anything. 4. The names Pronoic, or pronoic.org must be used to endorse and promote products derived from this software before obtaining written permission. You argue that the clause means must be used WHEN endorsing and promoting. My interpretation, on the other hand, is that as written it compels you to endorse and promote products. What interpretation do others make? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: request for approval of APOSSL
pronoic is a word (albeit a made up word) meaning the opposite of paranoic it is also a name, but so is apple, and netscape and apache they can use their name in their own licences Undefined words no place in legal documents If a made up word appears, or is offset in it will be assumed to be a name or a trademark Captilization differences in names are a pretty thin edge to hang an argument on in a court of law When read that way, the only way to comply is to get written permission The OSD conflict is that you cannot require written permission That is what I meant when I write there was conflict BTW, as I understand it, there are important reasons that you never want your trademarkable name to have a defined meaning I mention this because if you try to solve the license conflicst by defining pronoic and you are successful in propagating your meme, then I think you be unable to protect a trademark on the company name or software Forrest PS the name vs description distinction is amusing to anyone who knows Douglas Hofstadters writings You get a wink and a nodm from the geeks here But they don't belong in a legal document -- license-discuss archive is at http://crynwrcom/cgi-bin/ezmlm-cgi?3
Re: request for approval of APOSSL
the key here is the qualifying 'before obtaining written permission' should you NEVER obtain written permission you never need endorse anything Huh? How does a court of law distinguish someone who will never obtain permission from someone who has not yet decided to obtain permission? Before obtaining written permission is equivalent to without written permission I there any other consistent way to interpret it? There is no requirement to consult a lawyer about a license before submitting for approval But if you are trying to do something out of the ordinary (or clever), writing a license is not a simple task Legal professionals can provide important guidance I won't make further discussion until you resubmit Forrest -- license-discuss archive is at http://crynwrcom/cgi-bin/ezmlm-cgi?3
Re: Updated license from Edustructures
From: Steve Setzer [EMAIL PROTECTED] Date sent: Tue, 12 Feb 2002 12:42:12 -0700 To:[EMAIL PROTECTED] Subject: Updated license from Edustructures [snip] Oh, and per Karsten, I submitted this one in text form. If/when the approval process is done, I will generate a new HTML version conforming to OSD's requirements and submit that for the OSD web site. [snip] You may think you submitted in text form, but you did not. Text form ideally means Not MIME, with no transfer encoding. It at least means content-type text/plain. Here is what your message had. Content-Type: multipart/mixed; Content-Type: text/plain; Content-Transfer-Encoding: quoted-printable Content-Type: multipart/appledouble; Content-Type: application/applefile; Content-Transfer-Encoding: base64 Content-Type: application/octet-stream; Content-Transfer-Encoding: base64 Sorry, but I think most people ignored your message. (But it was during a RMS et al slugfest on fsb@crynwr, so maybe the encoding wasn't the reason.) But I urge you to resubmit it. I dug out the text of your submission. There are gratuitous changes in the copy of the GPL which appears. Most of them appear to be formatting, but some may not be. For example, They have used a word youmay which I do not understand. They changed a brief to an, year to in the section on how to apply these terms to your software. [Note to opensource.org webmaster - please fix the '' and '' tags in http://www.opensource.org/licenses/gpl-license.html when they should be lt; and gt;] I do not dismiss these changes as simple cut and paste errors. This discussion is about a legal document. Precision is essential, not optional. Plus, the GPL is copyrighted. Changes are not allowed. There are many wording changes to the Jabber section. Reading and analyzing those changes is going to take some time. Since the Jabber license is one of the most complex, and just on the edge of acceptable open source, in my opinion, I urge that the Edustructures license gets a full review, and not try to piggyback on the approval of Jabber. The changes they made are not minor. I recommend that until a text submission appears without the GPL changes, (one that makes it easier to compare to the existing licenses these are based on,) this license NOT BE APPROVED. Forrest -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Squeak License OSD-compliance
Sorry for the delay. I think most of us were distracted by a discussion on fsb@crynwr at the time. [snip] http://www.squeak.org/license.html Basically it's an X11/MIT flavoured license, but with a couple of added things that we are not sure about how they impact OSD-compatibility: - paragraph 5, You agree to idemnify and hold Apple harmless... - paragraph 6, Export restrictions Also, there's in paragraph 2 a sentence that restricts the bitmap fonts that Apple donated to the project; however this is not a big issue as there now exists jurisprudence that invalidates this clause. ...Then it should be easy to remove those sentences from pagagraph 2. Adding additional burdens as a condition of license is not playing fair. There are a number of clauses in this license, including paragraph 6 which I think make this a Not-Free software license. But we are discussing OSD approval, not FSF approval. Requiring compliance with U.S. laws, even in other jurisdictions is unnecessary. But the way paragraph 6 is worded, I think it obviously fails the OSD non-discrimination test. I quote that paragraph: 6. Export Law Assurances. You may not use or otherwise export or reexport the Apple Software except as authorized by United States law and the laws of the jurisdiction in which the Apple Software was obtained. In particular, but without limitation, the Apple Software may not be exported or reexported (i) into (or to a national or resident of) any U.S. embargoed country or (ii) to anyone on the U.S. Treasury Department's list of Specially Designated Nationals or the U.S. Department of Commerce's Table of Denial Orders. By using the Apple Software, you represent and warrant that you are not located in, under control of, or a national or resident of any such country or on any such list.p I am sorry that there are some previous licenses to work with here. Squeak is pretty well-known. But my recommendation: DEFINITELY DO NOT APPROVE. Forrest -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Apache License Question
The new program files that are added would only have the new Apache license. I have had the impression the new software Yazd would be distributed under the new Apache license. Therefore, the new license would be the license of the software, and there is no need to mention the old one on the site. You must comply with the old license if you are distributing anything which is derived from that software. I'd say at a minimum, the warranty disclaimer and copyrights of the old software must be propagated to the new downloaders. That is one purpose of the Apache 1.1 clause #2: 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. (And clause #1 generally preserves the notice in source forms.) The original software creators have complained that the license on the site should also have the old license too. I was wondering if this is true. You must comply with the old license terms. Depending on how you are distributing and what your phrase license on the site means, you could be OK or not. But I would point out that I dislike license surprises when I download. I dislike reading a license before downloading, and then finding a different license after downloading. If anyone could help me out with this I would very much appreciate it. Best Regards Aflatoon Aflatooni Yasna Inc. http://www.yasna.com/ -- 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: Advertising Clauses in Licenses
Russell Nelson [EMAIL PROTECTED] wrote (in part) There is much in the OSD which is insufficiently explicit. For example, we have maintained that there are no possible restrictions a license could put on users, because there is no possible mechanism one could use to constraint them, because no approved license can require that the user execute a license (OSD#7). I think it is a problem that execute a license is not defined. As far as I can tell, the GPL is indeed a license. And making a modified version, or distributing a copy, binds you to the license. Isn't that executing a license? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: The Simple Permissive License, v0.1
Tina Gasperson wrote: Everyone is permitted to use, modify, and redistribute this software, provided the above copyright notice and this permission notice are included with all copies, modifications, and redistributions. I think you need a warranty disclaimer. Here is the shortest free software license with warranty disclaimer that I have seen in existing use: Permission granted to use, copy, modify, distribute and sell so long as the above copyright and this permission statement are retained in all copies. THERE IS NO WARRANTY - USE AT YOUR OWN RISK. That is from Scott E. Lee's excellent sdncal library which allows computing calendar day numbers for gregorian, julian, jewish, etc. The copyright notice that the license refers to is simply: Copyright 1993-1995, Scott E. Lee, all rights reserved. The question I have about the above SDNCAL license is if it requires the notice embedded within a binary. Does anyone have an opinion on that? In other words, does the law treat a compiled version as a copy or a modified version? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: License for web application
What I want to accomplish is that if someone deploys a changed version of my application he'd be required to publish those changes (or at least send them to me and license me to use them in my free version), and that the visitors of the generated pages would have a way to identify the original author (me) even if the app was changed. The Mib Software LIDESC project identifies one existing license with something similar. Did you have a look at the wording in the Apple Public Source License Version 1.2 (clauses 2.2c-d) There is a copy at http://mibsoftware.com/librock/lidesc/apsl.txt If there are any other licenses analyzed with this requirement, they will be eventually listed here: http://www.mibsoftware.com/librock/lidesc/exlibrock_3MSRC_1C_DEPLOY.htm License selector and compatibility reporter: http://www.mibsoftware.com/librock/lidesc/ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Advertising Clauses in Licenses
John Cowan wrote: Lawrence E. Rosen scripsit: As for the GPL, where does it say that you can't distribute source via a website? As I read it, you must merely distribute source code on a medium customarily used for software interchange. I now get almost all of my software, including proprietary programs, through downloads. See http://www.gnu.org/licenses/gpl-faq.html#DistributeWithSourceOnInternet et seqq. To be clear, that QA covers the case when you initially distributed binaries only. The relevant part is GPL section 3. If you satisfy 3a, (distribute both binaries and source from a web site) then the GPL does not require satisfying 3b (distribute on a medium customarily used...) Further in the GPL, it states it plainly: If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code, even though third parties are not compelled to copy the source along with the object code. The GPL is clear and fair on this point. You are obligated to provide source code to those who received a binary version from you. The Q Public license 1.0 (QPL), for exmaple, says something different. QPL 4 b says you must give source code to all recipients of a binary, without any charges beyond the costs of data transfer (which precludes a profit.) I think that is unfair. Consider that I can decided to sell sources only at $50 profit. Company B takes my sources, distributes executables, and then goes out of business. Company C, who received binaries from B, now can come to me and demand sources without profit. That's not fair, and I don't see how it became OSI approved. Forrest. License analyser and compatibility checker: http://www.mibsoftware.com/librock/lidesc/ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
W3C and GPL compatibility (was Re: Discussion: ADPL)
Forrest J. Cavalier III wrote: There are two ways that I see trademark clauses written in licenses. As I read the GPL, trademark protection clauses as a condition of license are not GPL compatible. But trademark warning statements along the lines of Nothing in this license gives you don't seem to impose a further restriction, and therefore are GPL compatible, in my understanding. That's also the message I'm getting from the FSF at the moment (more specifically, from Bradley M. Kuhn). (Thanks for confirming that. FSF licensing answers must be swamped, based on delayed responses to items I submitted in the last month.) That FSF position might explain the W3C license being OK. The W3C is a pretty unique license, because the only conditions it seems to impose are propagating the license and notice (including the trademark statement.), It doesn't make compliance with the statement a condition for permission to use the software. That seems like a strategy which could be abused. Suppose I tacked a notice onto the W3C license: The copyright holders expect payment of $50.00 from you to name at address. Wouldn't that still be GPL compatible and OSD compliant, since it expresses an expectation of payment. It doesn't require it. I wonder what kind of revenue stream it would create. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Discussion: ADPL
(a) Except as required by the Acknowledgement section below, this license does not grant any rights to use the trademark ArsDigita, the ArsDigita logo or the terms ArsDigita Community System,ACS, or ArsDigita Corporation, even if such marks are included in the Original Code. Doesn't that say I must not redistribute the Original Code unless I remove the marks? -- Is it possible to receive trademark registration for three letters ACS ? If not, then the license should not be trying to restrict it. If protecting names is important, is there a reason that the wording similar to the Apache license was not adequate? -- [EMAIL PROTECTED]/ul II. strongAcknowledgement/strong ulAll advertising or promotional materials, download pages, documentation, and/or physical media mentioning features or use of this software, and any redistributions in whatever form, must display the following acknoledgement in a reasonably prominent location: This product includes software developed by ArsDigita Corporation for use in the ArsDigita Community System (ACS) (a href=http://www.arsdigita.com/acs/http://www.arsdigita.com/acs//a). /ul That takes the advertising clause a couple of steps too far. Physical media?! Does the notice really have to be printed on a 3.5 diskette, (not the label mind you, but the actual diskette!?) What about a hard drive?! The notice will take up an unreasonable amount of space, even on a CD. The notice has to be on every single page of (printed) documentation too?! What if I tell someone in email about how to use the program? Isn't that documentation? What if I write a message to license-discuss about the license? (Isn't this message mentioning use of the software?) Make sure you add their notice to this message before you store it. UCB removed the advertising clause from the original BSD license. (I question if there are ever valid reasons for an advertising clause) But if there are, why wasn't the more accepted wording from the original BSD used? What is it they are worried about? -- I noticed at least one spelling mistake. -- -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
License Analyzer
Semi-automated License analysis and compatibility reporter: http://www.mibsoftware.com/librock/lidesc/index.htm What it does: Based on answers to an HTML questionnaire, some conflicts with the GPL, the FSF definition of Free software, the OSD, and closed-source licensing are reported. The report also includes lists of previously analyzed example licenses which have similar clauses. Intended purposes of making this work public: - A license selector. Developers can get a list of licenses with attributes of interest, and choose an existing one. - A way to quickly identify known licenses with certain types of clauses. - Issues awareness tool. Those unfamiliar with Free/Open software concepts can identify problem clauses in a license, before approval submission for example. - Starting point for further study, and better communication when discussing/researching licenses. Further details are at the URL above. I welcome criticism and comments. On a related note, is there a list anywhere of licenses which failed to meet the OSD? The FSF commentary on failed licenses is very interesting. Such a list would be very useful to this project (and I expect it would be useful to visitors to opensource.org.) Does anyone have records of board votes and the license texts which failed? I want to add more examples to the analyzer. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
OSD fuzziness/omissions?
In analyzing licenses for the LIDESC project http://www.mibsoftware.com/librock/lidesc/index.htm there are a few types of clauses which in my mind do not meet the spirit of the OSD, but are not clearly rejected by it. Some of these came up by looking at which licenses pass the OSD, but are rejected by the FSF. (A pretty interesting point of study.) I know there is a move to tighten up the OSD. Here are some cases that might be discussed further. 1. Usage restrictions. The OSD relies on the guarantee of the right to run the software made by 17 USC for an owner of a copy. But if someone is not an owner of a copy, and just a license holder, or not under 17 USC, the OSD provides no help. The open source shareware licenses I proposed illustrate this. But there are other usage restrictions which have come up. For example: ``Disclosure of test results, except for the purposes of reporting a suspected problem in the execution of the tests, or claims of passing the tests are not permitted without the previous written consent of X/Open (The Open Group).'' 2. OSD says may not discriminate. But GPL 3c makes a distinction between commercial and non-commercial permissions. Under what interpretation of the OSD is that allowed? This impacts the wording LIDESC uses for the symbolic tag librock_USE_C_DISCRIM. 3. In Bruce's APSL 1.0 commentary, http://perens.com/Articles/APSL.html he says that APSL 1.0 is rejectable for a number of reasons which aren't clearly against the OSD in my mind, but certainly they are against the spirit. I reference the LIDESC symbolic tags below. For a discussion, read the APSL.html commentary. RIGHTS_TEMP: For those who do nothing to breach the license, are distribution, use, or other rights revocable, or temporary. (For example: time-limited or based on continued existence or non-existence of approval, relationship, law, entity, intellectual property litigation brought by others, etc.) 3NOTE*: Notification of a specific party is is required. 3MSRC*: Giving modified sources to a specific party is required. If someone could document where they are against the OSD, then I will add the OSD section numbers to the report when such licenses are reported in conflict. If they aren't covered by the present wording of the OSD, maybe they should be in a future version. 4. OSD #8 rejects tieing a license to a specific product. But what about the case where a license says something may NOT be used (or distributed) with a specific product? Over the years I have read licenses which had some of the above. But I did not collect them as examples. I am hoping such licenses will be contributed to the project as examples. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: OSD compliant shareware
Okay. But my point was that the copyright holder can grant portions of his rights under copyright without obtaining the signature of the recipient(s), while usage rights require a contract. Do you think the GPL creates a contract? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Version 2. Open source shareware
Karsten M. Self wrote: The attempt to key this to continued use rather than transfer has no standing under copyright as previously stated. Based on the discussion so far, everything seems to depend on whether the recipient has a license, or is an owner of a copy. I was hoping the OSD would reject the license, without needing such a long discussion. Guess not. Here is the improved version 2. The trick is to distribute the shareware as source only. In that case, in order to run it, it must be compiled, and (unless you are compiling and linking in memory), that creates a derivative work, a right reserved to the copyright owner. Then use the GPL method of forcing a license agreement without signing one. I don't think the limited evaluation period is an issue for OSD rejection or not. (I wrote the 30 day period into the license because that is common for shareware. This license version 2 doesn't have an evaluation period.) Here is version 2. (Clause 2 and 4 are the interesting parts. Parts are heavily borrowed from other licenses.) The one improvement needed is that someone could make a binary, pay the $20, and then users of that copy would not have to pay. (They accept the license only when they made a copy or derivative work.) There would have to be a copyleft clauses too, so that the license on the binary would be preserved. (I left that out for easier reading.) To head off OSD #7 objections, under this license running is not a right attached to the program. If you own the copy, that right is supplied by 17 USC 117. If you accept the license, that right is supplied by paying the $20 fee. - Copyright (c) year copyright holders 1. You may copy and distribute verbatim copies of the Program's source code as you receive it, in any medium, provided that you conspicuously and appropriately publish on each copy an appropriate copyright notice and disclaimer of warranty; keep intact all the notices that refer to this License and to the absence of any warranty, and agree and comply with the terms of this license. You may charge a fee for the physical act of transferring a copy, and you may at your option offer warranty protection in exchange for a fee. 2. You may compile this into an executable form, or modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above. 3. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance. 4. This software is licensed, not sold. If you wish to run the software after agreeing to this license, you must pay $20 to copyright holder. 5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for using, copying, distributing or modifying the Program or works based on it. THE SOFTWARE IS PROVIDED AS IS, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Version 2. Open source shareware
John Cowan wrote: Won't work (IANAL, TINLA), at least in the U.S.; any copies of computer programs that are needed in order to use the program are specifically non-infringing, by section 117. Normally, this refers to copying the binary form from disk to core, but it plainly would cover compiling if that's required. That is why the conclusions at MAI vs Peak were such suprising reading: you can have a copy, but not be an owner of that copy, and 17 USC 117 is for persons who own a copy. In MAI vs Peak, it was established that Peak's customers were licensees, not owners of a copy, and 17 USC 117 provided no permission. I am not a lawyer either, and I would not have known about MAI vs Peak unless it was brought up on this list. Should I interpret MAI vs Peak differently? -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Intel's proposed BSD + Patent License
Russ Nelson wrote: Forrest Tell me why you have to put the OSI's good name on this. The only way we can reject a license is to point to the OSD term which it violates. The license under discussion violates FSF Freedom 0, The freedom to run the program, for any purpose (freedom 0). If the OSD cannot discern it as unfree, the OSD must be amended to safeguard the right to use as well. It has been about a year since someone else tried to drive a poor license through this loophole. They wanted to require an end-user license fee for each copy. Since the OSD was silent on use, they figured it was OSS. Their license had other problems, and after a lot of flames and second and third attempts, they faded away (and I don't recall hearing anything about them at the last LWCE NYC that they expected to attend and announce their new whatever it was.) -- 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: Is inherited class a derivative work?
Angelo Scneider wrote: As I pointed out allready: linking to an API is not, I repeat: not a derived work. derived work is a legal term. You can not redefine it in your license. I didn't say I agreed with the FSF/RMS interpretation, I just mentioned what I remember it to be. One of the threads here on license discuss was from November 1999, where RMS posted into Re: Can Java code EVER be GPLd, at all? and a followup thread. (I seem to recall there were other threads on this topic, which also resulted in him posting, but I do not have URLs.) But you can go back and read the thread. Here is one of the RMS postings, (but do read the thread for important context.) http://www.crynwr.com/cgi-bin/ezmlm-cgi?3:mss:1225:199911:mjchloghelpieeomgjag -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Is inherited class a derivative work?
The discussion on this topic has been very interesting. I am unsure who posted the comment about the lawyers at FSF, but if that person could obtain clearance to post the complete explanation on why FSF has taken the position that the use of inheritance constitutes the creation of a derivative work, this might be extremely helpful for our discussion. If this is a reliable legal position, it might discourage use of the GNU GPL. Hence, this is a rather important matter. I think this fits with RMS and FSF previously published ideas (which were not from lawyers.) For years RMS and the FSF have the stance that if there is only one implementation of a library/API, and you write something which links to it, your work is a derivative work of that library. I expect the RMS/FSF reasoning is consistent in the case of inherited class. The library/API aspect is a topic discussed on license-discuss in the past (in a number of threads regarding the GPL and loopholes, if I recall.) This is the first I have heard that FSF lawyers had a hand in crafting a position on this particular topic. Since the FSF argument always seemed a little weak to me, I would like to read the lawyer-strengthened position also. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Is inherited class a derivative work?
Michael Beck [EMAIL PROTECTED] wrote: Derived class is a derivative work, because it is based on, or extends, the original class. Using would be instantiating an object from it - stand-alone, or as a part of another class (composition). There would be no adaptation of the existing class. According to Copyright law, a derivative work is a work based upon one or more preexisting works. It includes any form on which a work may be recast, transformed, or adapted. (17 USC Sec. 101). I think the issues that were raised in this thread are not totally tested or settled in case law. You might want to check the list archives for related discussion. I recall one thread had the subject Copyrightable APIs There are distinct types of works discussed in this thread, which I think that is causing some confusion. Instead of asking whether an inherited class is a derivative work, it may be more meaningful ask about one or more of the following items specifically. - The base class API (the set of functions and their parameter types.) - The derived class API - The base class implementation source - The base class executable version - The derived class implementation source - The derived class executable version -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Self-certification
These are good suggestions. They will be turned into reality with the
greatest alacrity if you make them changes yourself and submit them to
us.
Proposed consumer-centric index.html for www.opensource.org is
attached. Comments very much encouraged, on-list or in private.
Forrest
This message contains a file prepared for transmission using the
MIME BASE64 transfer encoding scheme. If you are using Pegasus
Mail or another MIME-compliant system, you should be able to extract
it from within your mailer. If you cannot, please ask your system
administrator for help.
File information ---
File: index.htm
Date: 3 Oct 2001, 22:53
Size: 5385 bytes.
Type: Unknown
index.htm
Re: Self-certification
This is the problem Russel Nelson and I are investigating in our discussion of section 2 of the OSD. Right. I didn't see you discuss that the wording for appplying the mark needs to be on the other web page, not just in OSD #2. (And maybe if the change was there, you would not even have to change OSD #2.) In any case, if OSD #2 gets reworded by your efforts, please see that the certification mark page is updated also. --- On a related point, in looking at the website (which I prefer over the initial version, if anyone remembers it), I am wondering about the emphasis and target audience. opensource.org has always been laid out as a website for producers. (I think it sort of invites you to write your own license, based on the ordering and wording of links.) But isn't the purpose of a mark to inform consumers, not producers? Perhaps the opensource.org certification mark page (or something like it) should be the home page. The link to the points of the OSD should be from within the context of the certification mark page only, (where it describes licenses which qualify.) I'm sure you would want to preserve some of the cheerleading text from the home page, placing it into the certification mark page. And you would want to split the certification mark page, which is far too long as it is. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
OSD #2 (was Re: GPL vs APSL (was: YAPL is bad))
Greg London [EMAIL PROTECTED] wrote (in part) It seems to me that the MIT does not meet item #2 of the OSD, then. The APSL goes above and beyond #2 requirements. But the MIT license seems to fall short. OSD #2 seems to be setting a clear minimum requirement that source code must be included with any distribution, or be made publicly available. OSD #2 is an oddball clause to be sure. All the other clauses describe The license, and #2 starts The program... (#7 starts differently too, but it describes the license as well.) This point has been touched before on this list, but I recall it was a secondary discussion. At the time no one proposed any changes in the wording of the requirements. In my reading (and yours too), it is possible to distribute software under an OSI certified license, and fail to meet OSD #2. That seems like a problem which should be discussed somewhere. What should be done about it? Is the OSI going to certifying distribution mechanisms as well as licenses? (Unlikely) It hardly seems likely that the BSD and MIT, (et al) licenses which don't guarantee downstream source are going to be decertified. Does OSD #2 need to be reworked? I hope that Bruce can comment on this point. -- Forrest J. Cavalier III, Mib Software Voice 570-992-8824 http://www.rocketaware.com/ has over 30,000 links to source, libraries, functions, applications, and documentation. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Quick Reference For Choosing a Free Software License
Rod Dixon, J.D., LL.M. [EMAIL PROTECTED] wrote: My main suggestion, however, was that some of the text - - the legal stuff - - should be re-written so it is clear that an opinion of the author is being expressed rather than a legal opinion being passed on by the author. I agree that being able to distinguish which is which will benefit the readers. But, do you think that readers will see it as the latter when there is no reference to case law and the author states that he is not a law professional? I suggested adding references. But from your point of view, that might make matters worse. (It is certainly possible to provide references to obsolete and superseded case law, appearing to offer rigorous, researched legal guidance, while being inaccurate.) I hope my point is taken as constructive. I would never argue that only lawyers should speak on legal matters; journalists and some politicians frequently do a fine job. The critical factor may be to stay within a context (...difficult as that might be in some contexts on the Internet...) where a reader is likely to conclude that they are reading commentary, opinion, or belief, but not legal advice. I won't dispute that only lawyers are qualified to give good legal advice. But I still don't know why you object to writing which appears to be legal advice. If there are inaccuracies in the page, let's see them corrected. Saying the page sounds too professional and accurate is strange criticism. Perhaps an admonition should have been directed to the readers, since problems only arise if the purpose of the web page is misconstrued by them. The author has already taken honest steps to minimize that possibility and I think that approaches the limit of what can be done, unless a blank web page is what you advocate.
RE: Quick Reference For Choosing a Free Software License
Rod wrote: Oh, my! I most defintely do not want to discourage you, and the chart is not a bad idea, but your commentary is written as if you ARE dispensing legal advice. Rod, you are a very frequent contributor to the list, and I don't want to discourage you. But... you have never criticized a non-lawyer for posting to license-discuss. What is different about a web page? It has a very clear, prominent statement that the author is not a lawyer and it is not legal advice (even though such statements are not required.) This is a great resource! I like the page as it is, legalistic wording and all. To make it an even better resource, can I suggest: - Provide links/references for the some people think or other opinions. This is what readers will need to track down case law and form clearer opinions. - For the community likes it item, use the counts of license types at sourceforge.net and/or freshmeat.net. GPL is the most popular, so you can remove the '?' for that one I think. - Add more licenses. (I wouldn't try to expand the table horizontally, just vertically.) Are you going to do only Free software licenses, not Open Source? - Put the Explanations of columns further up on the page, and set it as H1 - Change the Does not restrict You! to be Does not restrict the copyright holder. (Software users, not just authors, will be reading the page too.) I think the paragraph following is mostly ok, but the first sentence has the opposite meaning from what you intend. - The page isn't overly long right now, but if it grows, you might want to split some of the sections into separate HTML pages. Keep up the good work! Forrest J. Cavalier III, Mib Software Voice 570-992-8824 http://www.rocketaware.com/ has over 30,000 links to source, libraries, functions, applications, and documentation.
Re: X.Net, Inc. License
John Cowan [EMAIL PROTECTED] Matthew C. Weigel [EMAIL PROTECTED] My opinion is that MIT License with specified jurisdiction should be approved, as this seems like a valid concern. It should be noted for the record that such licenses are not GPL-compatible. When the jurisdictions are different, they are not compatible with each other either. That seems like a problem. Forrest J. Cavalier III, Mib Software Voice 570-992-8824 http://www.rocketaware.com/ has over 30,000 links to source, libraries, functions, applications, and documentation.
Re: command-line calls of GPL'd executables
phil hunt [EMAIL PROTECTED] wrote: Alice writes a program, aprog, which she licenses under the GPL. Bob writes another program, which invokes the aprog executable, using the POSIX system() call. Does Bob's program have to be released under a GPL-compatible license? (Assume for the sake of argument that there is no other program that does the same as aprog). What if aprog is, instead, licensed under the BSDL with advertising clause. Can be GPL his program? Understanding the copyright meaning of derivative work is essential to applying the GPL (and other licenses.) Derivative work cannot be defined in terms of linking or how software is executed. There are gray areas. (This is hard for many programmers to accept.) To answer the question you posed, you have to decide: Is using that system() call creating a new work based on aprog? The most likely answer is no. (But you can contrive examples and distributions where the answer is yes.) If you have a question about applying a particular license in reusing software, why not write to the authors and respect their interpretation? Forrest J. Cavalier III, Mib Software Voice 570-992-8824 http://www.rocketaware.com/ has over 30,000 links to source, libraries, functions, applications, and documentation.
Re: copyrightable APIs? (was RE: namespace protection compa
"Chloe Hoffman" [EMAIL PROTECTED] wrote pages./P PI think the real question is not whether an API is copyrightable but how an API is infringed and what is a derivative work of an API. You admit that some parts of the API would not qualify as original. Infringement would therefore depend on what portions of the API are "original enough" to qualify under copyright law. These APIs get published so that they can be used. Once the software is written that provides the API, it becomes a physical object which can be studied. It is pretty hard to argue that the facts of that physical object, including the exported interface, can never be expressed without infringing copyright. Forrest (P.S. To Chloe Hoffman: If you must use MIME, can you at least post in text/plain? Thanks!)
Re: Subscription/Service Fees - OSD Intent
Eric Jacobs [EMAIL PROTECTED] wrote: But whether or not shareware-with-source can be practically or legally enforced is not my main point. My main point is that OSD #7 cannot be sensibly construed as a criterion that a requirement-to-pay be waived for users to whom the software is redistributed, *without* also implying the waiver of other kinds of requirements, such as GNU GPL-style "viral" requirements. True, with respect to #7. But there is OSD #1. The license shall not restrict any party from selling or giving away the software as a component of an aggregate software distribution containing programs from several different sources. The license shall not require a royalty or other fee for such sale. A "royalty" is payment to the author. --- I spent some time discussing the IntraDAT viewpoint off-list in January. The stumbling block and misunderstanding were very simlar. Perhaps the OSD could be written to be more clear. There was also another point of contention. Apparently in German the idea of "right to do something" is not the same as "permission to do something." (It was very difficult to determine that it was this difference which was causing us to argue in circles.) I am (mostly) bi-lingual, but not German, and I never asked anyone else about it. He explained the difference using the example of a museum open to the public. Any member of the public has a "right" to enter the museum. But they still have to pay the admission fee. I explained that when the OSD used the term, it was in the sense of "license" and "permission." When you define "right" as "not otherwise restricted by law", as he did, the OSD reads very differently. That is how IntraDAT planned to charge usage fees without violating the OSD. --- Nowhere in the OSD is the right to use software required. Although I believe this to be permitted by copyright law, perhaps placing a mention in the OSD (or rationale) would also eliminate confusion. -- Forrest J. Cavalier III, Mib Software Voice 570-992-8824 http://www.rocketaware.com/ has over 30,000 links to source, libraries, functions, applications, and documentation.
Re: The Open Source Definition: 3. Derived Works
Christoph Steinbeck [EMAIL PROTECTED] wrote (in part) What is the point of letting them change the code and change the license to whatever they like (proprietory, e.g.). The other replies explained why the wording is "allow" instead of "require" in OSD point #3. They were accurate. I think you asked a deeper question. And a lot has been written on the subject if you dig around. (Lots of flame wars too, and I don't want to start one here.) In the big picture, - existing copyright law creates very broad restrictions. - A software license is permission to do something not already permitted by copyright law. (Some licenses are written as contracts, which would mean that copyright law isn't the starting place. The contract has all terms and conditions, and may even restrict something that was allowed under copyright laws. ) - Every license meeting the OSD must provide certain permissions. - The OSD allows other restrictions and permissions to be part of the license. (This is where the GPL and BSD-style licenses differ.) The way you phrased your question, it seems you already understand the freedom-ensuring benefits of the GPL. The BSD or MIT-style license provides the benefit of being able to re-use software in proprietary works as well as open source works. Some authors find that benefit more compelling than the benefits of the GPL. (If I say anything more, I'll be taken as a troll, I fear. Trust me, there has been a lot of discussion. Should be easy to find.) Forrest J. Cavalier III, Mib Software Voice 570-992-8824 http://www.rocketaware.com/ has over 30,000 links to source, libraries, functions, applications, and documentation.
Berkeley DB: (was RE: IPL as a burden)
"Ben Tilly" [EMAIL PROTECTED] It boils down to saying that you are free to use this software, with or without modificaiton, in any software for which source is available. You may not remove their copyright notice. Said notice includes contact information in case you want to negotiate a different license or purchase support for Berkeley DB. The license is at http://www.sleepycat.com/license.net. The license says *Redistributions in any form must be accompanied by information on *how to obtain complete source code for the DB software and any *accompanying software that uses the DB software You can derive "closed source" software from the Berkely DB software. If you choose to distribute, you must publish the source. (This is similar to the GPL.) Forrest J. Cavalier III, Mib Software Voice 570-992-8824 http://www.rocketaware.com/ has over 30,000 links to source, libraries, functions, applications, and documentation.
Re: What license to pick...
It is nice to see someone ask questions before they take a license and assume it does something it does not, and mis-use it. Does the GPL allow us (the toolkit creators) to ask a fee for commercial use of our toolkit? That is a question that can be interpreted in a few different ways which change the answer. You, as the toolkit creators, can distribute the software for a fee as you choose. You can even decide to distribute by charging only certain customers, and not others. A Free software license (...Free with a capital F here refers to freedom: the abilities of anyone possessing a legal copy of the source code...) means the authors have given freedom to use and distribute the source code. That freedom includes permission to redistribute under the license, with or without charging. Each Free software license is specific on what is and is not included in the freedoms it gives. It is hard to summarize the differences, you should read the licenses. You mentioned you were inclined to use the GPL. Under the GPL, you are prohibited from making additional use restrictions. - You may not require royalty payment for commercial (or any) use. (A royalty refers to a payment to the author when a copy is used. This is different than charging for transferring a copy.) - You may not require someone pay you when they transfer or use a copy. - You may not limit the fees someone may charge for making a copy. - You may not require that modifications be sent to you. (But the GPL does require that when a derivative work is distributed, the GPL'ed source code is included.) Other Free licenses have different requirements and restrictions. But to be a Free license, they may not prohibit use in a commercial, or certain environment. (That's what the non-discrimination clause of the Open Source Definition says.) When/if you start accepting modifications from the community, then you have to be careful that the contributions can be incorporated and licensed the same way. Everyone here strongly suggests you use an existing, approved license instead of writing your own. The difficult "leap of faith" for authors is to believe that the marketplace by nature (rather than through closed-license requirements) generally preserves the value in software. As long as your software fits into one of the business cases that Eric Raymond discussed for Open Source software, then you'll probably be happy using an Free/Open Source license. And if you don't match one of those business cases, then, yes, you will be giving away value. The open source license will ensure the value is preserved or increased (more people will benefit) but if you can't stay in business, then don't do it. Forrest J. Cavalier III, Mib Software Voice 570-992-8824 http://www.rocketaware.com/ has over 30,000 links to source, libraries, functions, applications, and documentation.
Re: Generic Simple License
Justin Wells wrote:
I'm not sure that such short disclaimers will work.
What is the smallest warranty disclaimer you have seen and think
would work?
Also, since you do not
require people to copy the license on to further works, you will get sued by
third parties who had no opportunity to read your disclaimers.
I don't understand how. OK, well I understand that anyone can
bring a lawsuit against anyone just by going through the motions,
but let's discuss liability.
Suppose a recipient ("second party") removes the license and
distributes the work (to a "third party.") The second party
must have provided some license to the third party, or the
third party lacks legal standing to use the work, but that is
not exactly important.)
How does such removal of the license text and disclaimer make the
authors any more liable for damages arising from third party use?
Related example situation:
Suppose someone takes a commercial shrink wrap package, and
emails a copy (illegal) to someone, who experiences some
type of loss as a result of using it.
Who is liable for the loss?
Related example situation:
If software is "released to the public domain" is there
any lingering liability?
Here are thoughts and motivation behind the license:
- I do not want to have a license propagation requirement since it
is a barrier to combining software written under various licenses.
Also, I want recipients to have the ability to create commercial
closed source, where the third-party does not have distribution
rights, as well as open source.
How can I be more liberal?
- After thinking about open source/free software off and on for
years, and trying to predict the future of open source, I think
branding will continue to be ultra-important. Trust of software
supplier is also important.
So although I don't generally like compliance to existing laws
to be a condition of licensing, (reference the Apple Open Source
patent revocation discussion) I think it provides small authors
a bit more (and necessary) leverage against abuses.
Thanks for writing. I hope to discuss these issues a bit more.
Forrest
Re: Generic Simple License
Because disclaiming implicit warranties and all liabilities has to be done explicitly and prominently. If the disclaimers are removed, then it is neither explicit nor prominent--so the disclaimers are probably unenforceable. So what changes if the second party removes the disclaimers despite the statement to propagate them? Same loss and lawsuit. How can the second party create more liability for the authors? What is the remedy for the authors? Sue the second party? I don't think so, but justice often doesn't follow "common sense." I know there are a handful of legal theories of liability, so I'd like to know more about the underpinnings here. -- And what about releasing it to the public domain? Where is the liability then? Authors would say "Here is a program in the public domain, it comes with no warranty." Same second party distribution. Same third party loss and lawsuit? Who's liable? -- I am not a lawyer either. But since the lawyers find a way to keep law in our lives, we non-lawyers are forced to discuss law.
Generic Simple License
The inspiration is the Apache license, with the advertising clauses removed. Can this be improved? Can it be made more simple? Can it be more generic? Forrest J. Cavalier III http://www.mibsoftware.com/ - This work is copyrighted by the authors. You may use and/or distribute the work and derivatives through these conditions: You inspect the work, which is provided AS IS, for suitability. THE AUTHORS DISCLAIM ALL WARRANTIES AND ARE NOT LIABLE FOR CONSEQUENCES ARISING FROM USE OR INABILITY TO USE THE WORK. and You do not misuse author trademarks or other intellectual property. The appropriate place to attribute our authorship is the source and not in the executable programs or other derivatives you create.
Re: The LaTeX Project Public License
You may not modify in any way a file of The Program that bears a legal notice forbidding modification of that file. This is counter to the whole idea of free software, and violates OSD #3. http://www.opensource.org/osd.html One reason that RMS started the free software movement in the first place is that he wanted to be able to modify a program to better suit his needs and he was denied that freedom. Either you grant freedom, or you do not. I don't see how a license which says "we sometimes grant you freedom, check the file" is much relevant to [EMAIL PROTECTED], or the free software community. - This license attempts to extend trademark-type protections to filenames, even though those filenames may be obvious and not meet any qualities of marks which deserve or need protection. - The license is probably incompatible with just about every other open source license. It is pretty complicated, so perhas that may just be my confusion. Forrest
Re: UCITA
(IANAL, but soon I'll need one if UCITA comes near me.) I believe RMS fears are justified. The UCITA language appears to be cleverly crafted. With respect to the implied warranty. UCITA creates implied warranty out of thin air EVEN IF there is no consideration (i.e. $$ to the provider) excerpt below.) It also sets _specific_ wording requirements in disclaimers. (Is there any thing else like this in law? I thought that the intent of contracts was more important than the words used...) - I am particularly outraged by the language creating the implied warranty in section 403 (and similar in 401.) "...a licensor that is a merchant with respect to computer programs of the kind warrants..." I read this as: if I sell _any_ software, I am a merchant, and must warrant that _all_ the software I license as being merchantable (per UCITA 403), AND noninfringing (per 401), EVEN IF I license some open source software at no cost. (If I don't sell software regularly, if it is my hobby, then there is no obligation.) That is a very unfair and chilling burden. RMS is right, I think. - One saving point for open source software appears to be this: 406 (d) If a licensee before entering into a contract has examined the information or the sample or model as fully as it desired or has refused to examine the information, there is no implied warranty with regard to defects that an examination ought in the circumstances to have revealed to the licensee. That "ought" is a very thin thread to hang a defense on though. And maybe in some cases you can add wording to licenses to limit damages (due to breach of implied warrantees) to the purchase cost. Too tired to read UCITA 803, and 803d seems to say that you can't always do limit. - As a software consultant who also distributes open source software at no cost, disclaiming warranty is absolutely essential. 406.b.1.a allows me to disclaim warranty under 403 with the standard language mentioning "merchantibility." The wording of this paragraph seems to indicate it can be done in the standard inclusion of the disclaimer with the source. But it is not clear that is sufficient. (RMS's objection too?) I don't see a way to disclaim the implied non-infringence warranty of 401. That's a bigger problem. If you want to see something really scary for consumers, read UCITA 107d: A person that uses an electronic agent that it has selected for making an authentication, performance, or agreement, including manifestation of assent, is bound by the operations of the electronic agent, even if no individual was aware of or reviewed the agent's operations or the results of the operations. WOW! But, oh, don't worry about that, I bet the lobbyists say. Because UCITA 206a allows you to petition the COURT for "appropriate relief if the operations resulted from fraud, electronic mistake, or the like." I think there is going to be some nice money to be made by trojaning software to do credit card deposits. Before UCITA, it is fraud and maybe conspiracy, and you could end up in jail. After UCITA, as long as the "person selected the electronic agent" (which you trojaned) then you are OK. Overcharge 1,000,000 card holders $2 each, each individual won't petition the court, and you retire with $2MM. Somebody please explain: 1. UCITA could have been a nice clarification of some sticky parts of contract law applied to computers. Contract law works quite nicely most of the time and is a heck of a lot easier to understand than UCITA. So can somebody give an example of a problem that UCITA solves? 2. Is there any benefit for consumers in UCITA? (Beyond avoiding the scary future the lobbyists predict that no one will be able to write and sell software in the future if UCITA isn't passed?) Again, IANAL. Forrest J. Cavalier III Mib Software
Re: How To Break The GPL
I would very much like to hear that there is a flaw in this logic. If so, where is it? In my understanding, Alice must not have used the GPL'ed software in her design and testing. It would be very hard to avoid this in practice. Claiming to have avoided it, and still distributing instructions for Bob to do it, should be regarded with grave suspicion. This is something which has been discussed before on this list, and elsewhere as well, I'd imagine. Forrest J. Cavalier III http://www.mibsoftware.com/
Re: How To Break The GPL
From: Mark Wells [EMAIL PROTECTED] On Fri, 3 Mar 2000, Forrest J. Cavalier III wrote: I would very much like to hear that there is a flaw in this logic. If so, where is it? In my understanding, Alice must not have used the GPL'ed software in her design and testing. It would be very hard to avoid this in practice. Claiming to have avoided it, and still distributing instructions for Bob to do it, should be regarded with grave suspicion. Alice didn't distribute any actual GPL'd software with her proprietary code. Says who? If she distributed a derivative work of GPL'ed software, then it must be GPL'ed. The question is whether or not Alice has a derivative work. Someone in this thread suggested that Alice could have used a common API and never used GPL software during development. That was NOT the theoretical example proposed. Are you saying that the GPL prohibits the _use_ of GPL software in development of proprietary software? Where did I say that? Alice can develop all the proprietary software she want with GPL'ed software. When she distributed it, it must be under the GPL. The GPL does not say derivatives MUST be published. It says they must be published under the GPL. Big difference. So I can't use, say, GCC to compile my non-GPL program? GCC has a special exception. If it did not have that exception, then anything compiled with gcc might be considered a derivative work.
Re: How To Break The GPL
From: John Cowan [EMAIL PROTECTED] Says who? If she distributed a derivative work of GPL'ed software, then it must be GPL'ed. The question is whether or not Alice has a derivative work. In my first scenario, Alice made a derivative work but didn't distribute it. She then distributed her own original work to Bob, who made another distributed work (identical to the first one) but didn't distribute it. I see where we have not met in understanding. Let's distinguish between "derivative work" (which I believe is a term in copyright law meaning a new work which incorporates material from a previous work), and "work based on the Program", which is a phrase used in the GPL. I admit to using the term "derivative work" sloppily in the text you have quoted. Let's start again Did Alice create a derivative work? Not in the definition I just gave. Did Alice create a work based on the GPL'ed program? She had no other way of writing and testing her program. Others must obtain the GPL'ed program to use hers. So I say yes. Now I believe what others have said I RMS and the FSF say the GPL says in this case: that Alice must GPL her work. But after another close reading of the GPL 2.0, I am not sure! The phrase "derivative work" and "work based on the Program" are used in several places, and if one of these terms applies to Alice's work, and the other doesn't, then it may indeed be a loophole! What do others think of the language in section 2, 3, and 5, and 6 in this case? I think section 6 could be reworded to clear the problem. From the GPL 2.0: (I added the carets below) 6. Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to ^^^ these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License. This should also include "any work based on the Program." Would you agree?
Re: Certification Process
I will not excuse the people who are supposed to be doing certifications. I am not one of them. I think the certification process should be outlined AND time limited. However, I do know that many of the people behind OSI were unavailable yesterday. There is an open source conference. Certification (and marks) are important. The whole point of the certification mark is to be able to make the concept of free software visble to END USERS, and allow them to "look for the union label." Like a recycled, or environment-friendly label. I asked about the certification trademark and its importance in bringing the message to end users during the QA at yesterday's thebazaar.org panel discussion. ESR responded it was still bogged down in procedural problems, but was coming. I hope soon, but no indication of time frame was given. As for the certification of new licenses, it is a need that the OSI people have agreed exists and should be served. You observe that those who have agreed there is a need for new certifications have not been serving everyone equally. I agree... but... That's life. Bruce is busy. We all are. He has to prioritize. I believe he has claimed that he isn't getting paid for these consultations. I don't know if it has changed. If he only has time to service a "Sun"-sized company that has a huge amount of software they want to open source or a small guy with a 2,000 line library, he will service the big company because the end result is more free software affecting more people sooner. Everyone except the conceited (and many lawyers fall into that Not-Invented-Here philosophy) agree that using an existing license is better than writing a new one. Pick one if at all possible. And if you are a small developer, you are better off using a known license almost no matter what. (You don't want an unfamiliar license to be the reason your stuff is not accepted. You will lose more value that way than you can ever protect by using a new license.) Also, it often goes without saying that if you haven't had a lawyer write (or at least read and approve) a license, and the license has major issues (like totally misusing some wrong assumptions about copyright laws) then don't expect much attention. No one has time to baby sit. Forrest J. Cavalier III
New license: Integrity Open Source License
Of the popular OSD compliant licenses, I like the MIT style license, but I wanted something that goes a little bit further in protecting the author's name and integrity of original source. (OSD #4 at http://opensource.org/osd.html) The OSD states that the license can restrict the distribution of modified works to be "patch files" but I have become convinced by discussions in the community that this is unworkable. So changing as little as possible, from the MIT license, I have the following. As a small developer/consultant, name and trademark recognition is very important, of course. That really ends up being the value I sell. The intent of the license is to allow distribution of unmodified software far and wide, but requires that modified versions be advertised and distributed using different names. If someone knows of a MIT-style license with a similar condition that already exists, please reply. I could not locate anything similar, and resorted to yet another open source license. With discussion and approval here, (and a better license name,) my hope is that many other developers will use the license when they wish to release open source software under a license that better protects name recognition. - - - - - - - - - - - - Copyright Notice - - - - - - - - - - - - Copyright 1999 Forrest J. Cavalier III MODIFICATION WARNING: This software may have been modified. See http://www.mibsoftware.com/libmib/ for documentation and up-to-date original copies, custom versions, support, or other licensing terms. Other licensing arrangements are made in writing. We would appreciate that you notify us of defect discoveries and enhancements so that others can benefit from improvements to this software. - - - - - - - - - - - - - License - - - - - - - - - - - - - - - - - Permission is hereby granted, to any person obtaining a copy of this software (the "Software"), to deal in the Software without restriction, including the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, subject to all the following conditions. 1. The text beginning with the copyright notice above, and including the four clauses of this license shall be included in full and unmodified in all copies of the source files or portions thereof. 2. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. 3. The author(s) grant you a non-exclusive right to use the author- given names and trademarks of the Software to publish and distribute only unmodified source files. Your rights to deal in the Software under this license are subject to the condition that you do not use nor associate trademarks, names, or reputation of the author(s) with modified derivative works except within the source files. 4. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Re: Hoping for something Useful
Your request is off topic for license-discuss. There are many free/open source WWW indexing programs in the rocketaware index: http://www.rocketaware.com/spec/infotool/www/robot/
RE: [openip] Re: rights and freedoms
From: "Robert M. Muench" [EMAIL PROTECTED] Answer: Because there is one US software patent issued, on average, every 20 minutes, and it would take several days of work to find prior art for each one and prepare a submission. But not all patents are related to the topic we are interested in. Of course it will be impossible to build up a shadow-PTO but defeating the most trivial patents should be possible. It's better to invalid some patents than non at all. I am not subscribed to [EMAIL PROTECTED] I am planning to improve my programmer's webliography in a way which may help. http://www.rocketaware.com/ is the complete programmer's webliography and index. It features more than 10,000 links of interest to software developers, including a special section listing more than 6000 open source applications and libraries. There are also FAQs and links to categories at dmoz, yahoo, and freshmeat. Lots of good stuff. At some point, I will cross-categorize and provide links to to the software patent categories at the IBM patent server (www.patents.ibm.com.) I was planning to do this anyway. If others think this is a good idea, then I will get it done sooner. I don't know a lot about challenging patents. Is Greg Arahonian (sp?) still around Usenet? Forrest J. Cavalier, Mib Software http://www.rocketaware.com/ Complete Programmer's Webliography 10,000 off-site links. 200 categories.
Re: Some general principles of naming
Richard Stallman [EMAIL PROTECTED], wrote (in part): If you think it is proper to use a name that gives credit to those who developed a system, but you think (as I do) that it is impractical to give credit in that way to all the contributors, I suggest making a list of them in order of decreasing priority. Then you can give credit to the first N of them, for whatever N you think is feasible. Despite having one of the largest word sets of all human languages, English seems to have enough ambiguity and double meaning. "Free" is almost hopelessly confusing. Richard, are you using the word "principal" to mean "first?" When I see that word, I generally take the meaning "most important." I think it is proper for those who are developing (present tense) and spreading (present tense) free software to use whatever name they like. Those who developed (past tense) or started (past tense) a project may or may not deserve to be named creditors. There are a lot of 80% solutions in the software world. Indeed, the future holds the promise of gathering pre-built components to create new applications. I respect those who provide the last 20% as being most important. Did Linus copy Linux in its entirety from the GNU project or did the GNU project simply inspire and provide some of the building blocks that the architect happened to use? The brick makers don't get to name the building. If you mean "most important" when you say "principal", then how do you consider the GNU project to be the most important contributor? Did the GNU project design and provide the architecture of Linux? Did the GNU project advise or support Linus? Would Linus have completed Linux even if the GNU project never existed? I am not questioning the importance of the GPL in the success of Linux. I am asking how the GNU project, not Linus, was most important developer of the Linux kernel. Forrest Cavalier Mib Software http://www.mibsoftware.com/
Re: GNU License for Hardware
Well, to explain all the reasons, the political and economic circumstances would need about 30 pages ... I though you where an american and you knew that, are you not? Regards, Angelo Your education seems to lack the realization that any telling of history is one of opinion and interpretation, not a dissection or autopsy. History is not science. Apparently, you have been given one interpretation of an extremely complex period of American history and incorrectly assumed it to be a statement of indisputable fact. Secondly, wars do not occur for reasons easily summarized in one email message, or even 30 pages. Wars require individuals willing to risk life. The reasons individuals will risk life are extremely varied, ranging from the practical "it pays well" to percepted "well-being of my family" or "well-being of society in which I and my family will live." Every combatant will have a slightly different reason for being a combatant, and have slightly different "rules of acceptable engagement." Any summary in an attempt to explain an aggregate will of a group to war against another, will be hopelessly approximate. Not that this has much to do with GNU license for hardware. Forrest
license counts
A caveat with taking freshmeat license counts Freshmeat.net doesn't list everything. Freshmeat does not list older applications and utilities, such as common system utilities or library functions. Even with just what is distributed with OpenBSD or FreeBSD, this is thousands of items. But for each such BSD item not listed, there is likely a GNU and Linux work-alike which is also not listed. So they might offset each other pretty well. Draw conclusions if you wish, the data is below. Forrest J. Cavalier III, Mib Software Voice 570-992-8824 The Reuse RKT: Efficient awareness for software reuse: Free WWW site lists over 6000 of the most popular open source libraries, functions, and applications. http://www.mibsoftware.com/reuse/ - Count of "license type" field of freshmeat.net 15-Oct-99 4-Mar-99 License type count ratio count ratio - 1 0.000 0 0.000 Eiffel Forum Freeware License 2 0.000 2 0.001 AFPL 2 0.000 0 0.000 Apache style 2 0.000 0 0.000 IBM Public License 9 0.002 2 0.001 QPL 14 0.003 7 0.003 source-available commercial 15 0.003 9 0.004 MIT 21 0.004 6 0.002 MPL 22 0.004 0 0.000 Artistic GPL 39 0.007 10 0.004 Free Trial 48 0.009 23 0.009 Shareware 59 0.011 82 0.032 unknown 78 0.014 26 0.010 Public Domain 82 0.015 39 0.015 commercial 117 0.022 56 0.022 Artistic 171 0.031 82 0.032 BSD type 180 0.033 74 0.029 free for non-commercial use 208 0.038 92 0.036 LGPL 211 0.039 177 0.070 free to use but restricted 226 0.042 104 0.041 OpenSource 241 0.044 106 0.042 Freeware 263 0.048 118 0.047 freely distributable 34190.630 15100.598 GPL
