Re: Source Distribution License
Alexander's point is not exactly correct, but I think the main point was on target; namely, in addressing questions concerning the copyrightability for software, the object code is not likely to be treated differently than the source code. In some cases, the distinction between object code and source gets pretty fuzzy and, in my opinion, treating the two differently would dislodge whatever is left of the logic in our copyright jurisprudence that applies to software. Having said that, Alexander's mistake appears to be his reference to the copyright office. While the copyright law may treat source code and object code similarly, the copyright office does not. Instead, the copyright office may accept a filing for registration of software in the form of object code only under what is called the rule of doubt. Ostensibly, the rule of doubt means that courts give even less deference to the copyright office's finding of copyrightability than the court would acknowledge, if it were assumed that the copyright office actually read the source code sample submitted with the copyright registration application. The application of the rule of doubt should mean that the party alleging copyright authorship on the basis of a work in object code has a heavier burden of proof than a software developer who files for copyright registration using source code. Rod Rod Dixon Open Source Software Law Blog: http://opensource.cyberspaces.org : Mahesh T. Pai wrote: : [...] : General consensus is that binaries are modified/derived versions of : sources. : : AFAIK, The U.S. copyright office doesn't agree (the copyright : office regards the source code and object code as equivalent : for purposes of registration). : : regards, : alexander. : : -- : 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: Source Distribution License
Rod Dixon, J.D., LL.M. wrote: [...] Having said that, Alexander's mistake appears to be ... My mistake was the omission of reference (and context) to the source of my comment. http://www.digital-law-online.com/lpdi1.0/treatise26.html (VI.B. Source Code and Object Code) quote Even though source code and object code are distinct, it is still useful to maintain the concept that the source code and the object code are just different forms of the same copyrighted work. The Copyright Office regards the source code and object code as equivalent for purposes of registration. In fact, it generally requires a deposit of at least a portion of the source code (generally the first and last 25 pages - see their Circular 61) and questionsany registration that includes only object code. Where an applicant is unable or unwilling to deposit source code, he/she must state in writing that the work as deposited in object code contains copyrightable authorship. The Office will send a letter stating that registration has been made under its rule of doubt and warning that it has not determined the existence of copyrightable authorship. ... /quote regards, alexander. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Source Distribution License
I propose a source-only distribution license that restricts commercial use of binaries. I know it's on the border of the open source idea but if it catches on we might see software written for commercial use as open source software. The restrictions it creates only affects commercial users and endusers who are afraid of compilers or just want a binary distribution out of convenience. In both cases I'm in favor of the idea that the creator of the source code can earn a bit of money from commercial re-use. Bernhard html head titleSDL, Source Distribution License 1.0/title /head body center h1SDL, Source Distribution License 1.0/h1 p iThe following is a SDL license template. To generate your own license, change the values of OWNER, ORGANIZATION and YEAR from their original values as given here, and substitute your own./i /center p font size=+1 lt;OWNERgt; = Bernhard Fastenrathbr lt;ORGANIZATIONgt; = FAR LLC, Floridabr lt;YEARgt; = 2004br /font h2Copyright Notice/h2 Copyright (c) lt;YEARgt;, lt;OWNERgt; All rights reserved. h2Source Code License/h2 Permission is hereby granted, free of charge, to any person obtaining a copy of this source code and associated documentation files (the Source Code), to deal in the Source Code without restriction, including the rights to use, copy, modify, merge, publish, distribute, and/or sell copies of the Source Code, and to permit persons to whom the Source Code is furnished to do so, subject to the following conditions: h3Distribution Obligations/h3 Redistributions of the Source Code must retain the above copyright notice and include a full copy of this license agreement. p Neither the name of the lt;ORGANIZATIONgt; nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission. h3Distribution of Executable Versions/h3 Redistribution in any other form than human readable source code in compressed or uncompressed form (Binary Distribution) is not covered by this license. p A Binary Distribution derived from Source Code you received under this license should be coverered in a separate license agreement with the owner of the Source Code. If no such license agreement exists you are not permitted to make a Binary Distribution of any software derived from the Source Code. h2Disclaimer/h2 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS AS IS AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. /body /html
Re: Source Distribution License
Bernhard Fastenrath said on Sat, Mar 13, 2004 at 12:08:23PM -0500,: html head titleSDL, Source Distribution License 1.0/title /head Posting in html are not a good idea. h3Distribution of Executable Versions/h3 Redistribution in any other form than human readable source code in compressed or uncompressed form (Binary Distribution) is not covered by this license. p General consensus is that binaries are modified/derived versions of sources. So, this will fail OSD. -- +~+ Mahesh T. Pai, LL.M., 'NANDINI', S. R. M. Road, Ernakulam, Cochin-682018, Kerala, India. http://paivakil.port5.com +~+ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Source Distribution License
Mahesh T. Pai wrote: [...] General consensus is that binaries are modified/derived versions of sources. AFAIK, The U.S. copyright office doesn't agree (the copyright office regards the source code and object code as equivalent for purposes of registration). regards, alexander. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Quoting Russell Nelson ([EMAIL PROTECTED]): They were also wrong. Oh, we *can* stretch the definition, but inventing requirements out of whole cloth is an invitation to a party -- party to a lawsuit, that is. I understand (thanks to Lawrence) the reason why this could create problems maintaining the certification mark under trademark law. But I must admit I can't for the life of me think of any legal theory whereby denying certification would be a tort. Perhaps you could elaborate? -- Is it not the beauty of an asynchronous form of discussion that one can go and make cups of tea, floss the cat, fluff the geraniums, open the kitchen window and scream out it with operatic force, volume, and decorum, and then return to the vexed glowing letters calmer of mind and soul? -- The Cube, forum3000.org -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Legal soundness comes to open source distribution
On Wed, 14 Aug 2002, Russell Nelson wrote: I like mine (well duh!) because it explicitly says that all is fair in love, war, and software use and modification except for a few things. That's also its weakness because the list needs to be right; no more and no less. Actually, it's alright if initially it's too restrictive - you can always add to your list of exceptions over time, but removing exceptions would be politically tough (while doable) since it would invalidate previously valid licenses. Brian -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Legal soundness comes to open source distribution
Brian Behlendorf writes: On Wed, 14 Aug 2002, Russell Nelson wrote: I like mine (well duh!) because it explicitly says that all is fair in love, war, and software use and modification except for a few things. That's also its weakness because the list needs to be right; no more and no less. Actually, it's alright if initially it's too restrictive - you can always add to your list of exceptions over time, but removing exceptions would be politically tough (while doable) since it would invalidate previously valid licenses. Quite true. Better to err on the side of not giving much freedom to restrict, since we can always give them freedom later. -- -russ nelson http://russnelson.com | Crynwr sells support for free software | PGPok | businesses persuade 521 Pleasant Valley Rd. | +1 315 268 1925 voice | governments coerce Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Tuesday 13 August 2002 10:43 pm, Lawrence E. Rosen wrote: Whatever else open source licenses do, they do not explicitly make a licensee the owner of a copy. To transfer ownership requires a contract; a mere license won't do. What about the gift of a copy of the software, as in a download at the author's invitation? The transferal of property may require a contract, but that contract can be as informal as here, take this. Regardless of this confusing point, why does this make click-wrap problematic? If the user has the legal right to use or install the software, then the exercise of that right cannot be used to indicate license acceptance. Anything else would be legal blackmail. Which in fact is how I view many of these newfangled EULA's. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Larry's comment sums up my point quite well when he states: [snip] Whatever else open source licenses do, they do not explicitly make a licensee the owner of a copy. The implications of the licensee not being an owner of the copy of software he/she has possession of go directly to Bernstein's point. At any rate, in the context of open source licensing, Bernstein's argument requires understanding how section 117 relates to section 109 with respect to the status of the end-user/licensee. The matter is not pertinent to this discussion, but someone raised the issue. [snip] Regardless of this confusing point, why does this make click-wrap problematic? That's a good question. In my previous post, I attempted to summarize the arguments presented because I had the same reaction. I think a few people had a near-visceral reaction to the very idea of click-wrap and the contractual-open-source-license. Even so, I have repeated that click-wrap is but one way to show an indicium of mutual assent. Although in unusual circumstances there may be practical difficulties implementing a click-response for user input, the opposition to the concept of mutual assent seems over-blown. If dialog boxes are too confusing, there are other ways to achieve the same result. I can imagine some strategic advantage denominating an open source whatever as a copyright license rather than a contract, but I am befuddled by the opposition to what ostensibly are simple steps to decrease the likelihood of a successful challenge to the validity of the license. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Quoting Russell Nelson ([EMAIL PROTECTED]): Oh, it's *always* had to be changed. Anybody could insert restrictions on use into a license and ask us to approve it. Since the OSD says nothing about a license not being allowed to have restrictions on use, we would have to approve the license. Or they'll turn you into a newt? Russ, on other occasions, I believe you've been among those reminding people that the OSD isn't a black-box algorithm into which you plug candidate licences to determine whether they pass or fail. Whether I recall correctly or not, those were words of wisdom. I would expect that, if someone proposed a licence that satisfied OSD formalisms but denied rights to software usage, the Board's reaction would be Nice try. And its reaction to allegations that it has to approve such a licence would be No. Doesn't that solve the problem? There will probably always be clever licence provisions to attempt subversion of the OSD's intent, no matter how many of them get patched. It would save a lot of time and energy to fall back on the rule of reason -- and the right of usage is obviously necessary to and implied by the existing OSD terms. -- Is it not the beauty of an asynchronous form of discussion that one can go and make cups of tea, floss the cat, fluff the geraniums, open the kitchen window and scream out it with operatic force, volume, and decorum, and then return to the vexed glowing letters calmer of mind and soul? -- The Cube, forum3000.org -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Lawrence E. Rosen scripsit: Whatever else open source licenses do, they do not explicitly make a licensee the owner of a copy. To transfer ownership requires a contract; a mere license won't do. That seems farfetched to me. If I set out a table with cookies on it by the side of the road and a sign saying HELP YOURSELF!, it seems to me that by taking a cookie you become owner of the cookie and can do what you like with it. Similarly, if I put a printer (chained to a tree) by the side of the road, with a sign saying PUSH HERE TO GET A FREE COPY OF MY BOOK, then it seems to me that those who push the button and get a copy are owners of that copy and can do what they like with it. In neither case is there any mention whatsoever of ownership. I grant that in most cases ownership is transferred by means of contract, but I don't see how a contract can be required. -- John Cowan [EMAIL PROTECTED] http://www.ccil.org/~cowan One time I called in to the central system and started working on a big thick 'sed' and 'awk' heavy duty data bashing script. One of the geologists came by, looked over my shoulder and said 'Oh, that happens to me too. Try hanging up and phoning in again.' --Beverly Erlebacher -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Legal soundness comes to open source distribution
Lawrence E. Rosen writes: Several people, including Bruce Perens, Russ Nelson, myself, and most recently David Johnson, have suggested wording for such an OSD provision. None of those versions has caused the others on this list to stand up and cheer. Particularly Bruce's, which he never actually got around to submitting. I like David's, because it's such a shot across the bows. Unfortunately, his suggestion says nothing about modification restrictions, such as the GPL's, or BitKeeper's. Presumably people like being restrained from making or required to make certain modifications to GPL'ed software (because they relate to copyright notices), and dislike BitKeeper's restriction on removing logging code. Larry's is nice because it completely punts on the issue. What you could do, you can do. I like mine (well duh!) because it explicitly says that all is fair in love, war, and software use and modification except for a few things. That's also its weakness because the list needs to be right; no more and no less. -- -russ nelson http://russnelson.com | Crynwr sells support for free software | PGPok | businesses persuade 521 Pleasant Valley Rd. | +1 315 268 1925 voice | governments coerce Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
John Cowan wrote: Lawrence E. Rosen scripsit: Whatever else open source licenses do, they do not explicitly make a licensee the owner of a copy. To transfer ownership requires a contract; a mere license won't do. That seems farfetched to me. . . . In neither case is there any mention whatsoever of ownership. I grant that in most cases ownership is transferred by means of contract, but I don't see how a contract can be required. I look at it somewhat different. It's not that you NEED a contract to transfer ownership - its that if you transfer ownership in exchange for consideration (usually money), you HAVE a contract.Open source transactions involving payment of money are all agreements. The gift issue is different - I think you can transfer ownership in a gift transaction without a contract. Here are what I see as the options for what form an open source transaction/license can take: 1) SIMPLE sale (no additional terms, just transfer of ownerhship of a copy for money) or gift of a copy, PLUS a non-contractual permission notice which lets you copy, modify, distribute, etc. The notice license cannot be enforced by the user, only the copyright owner. 2) Sale PLUS license - user gets title to the copy, plus agrees to a license that allows copying, modification, etc. The user can enforce the contractual license. This could legally include use restrictions, restrictions on transfer, etc. As I said previously, its not common in a sale but it can be done. 3) PURE license, user does not get title, but gets a license to use, copy, modify, etc. This may have use restrictions, etc., or not. Traditional open source (GPL, BSD) follows the first. Proprietary follows the third. There is nothing inherently evil about PURE licenses. If you reserve title, but give the user all the rights they would have in a sale, plus the right to copy, etc., where is the harm? I'm not sure why you would want to do this, but you could. Is the choice of form central or relevant to the OSD? The previous acceptance by OSI of licenses that are clearly agreements would suggest that the answer is no.So perhaps we should go back to deciding what terms are allowable under the OSD, rather than the form of the transaction. Having said that, the simple sale/gift plus permission notice (option 1 above) versus making the transfer subject to a more complex agreement (sale plus license (option 2) or pure license (option 3)) IS A CRITICAL ISSUE because there is so little understanding or acknowledgement of option 1. If open source wants this form of transaction to work, it needs to do a better PR job. Just being clear about the proper characterization of the transaction would help immensely. And the consequences of requiring distributors (who may be unlawyered programmers) to enter into complex contracts with users needs to be further studied. Carol -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Rod Dixon wrote: I want to summarize what we have discussed on click-wrap because the issue is significant from the standpoint of the legal standing of open source licenses, and so I can include proposed responses in our research project on the OSD. It is my understanding that the issue initially involved the approval of a license, not a change to the OSD. The discussion of click-wrap then considered whether the fact that adding indicia of mutual assent to website agreements like open source licenses (e.g., a mouse click from the user) might have adverse implications for the position that open source licenses are non-contractual licenses. There was also some discussion concerning whether click-wrap conditions imposed on downstream or sub-licensees is practical (it may be difficult to implement). Finally, some raised the question whether the click-wrap condition is doomed to failure in cases where distribution is packaged with multiple programs carrying distinct licenses. Is this a fair summary? Rod This is very helpful. I would add to this list of issues the consequences of requiring distributors/programmers to enter into complex agreements which you need a lawyer to fully comprehend. Carol -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Carol A. Kunze scripsit: Traditional open source (GPL, BSD) follows the first. Proprietary follows the third. There is nothing inherently evil about PURE licenses. If you reserve title, but give the user all the rights they would have in a sale, plus the right to copy, etc., where is the harm? I'm not sure why you would want to do this, but you could. Proprietary licenses normally do *not* give you all the rights of an owner: far from it. If I (not being a licensed reseller) sell you a computer with lots of proprietary software on it, you get no use rights to that software. Ownership of libre software of course goes with the hard disk. -- John Cowan[EMAIL PROTECTED] At times of peril or dubitation, http://www.ccil.org/~cowan Perform swift circular ambulation,http://www.reutershealth.com With loud and high-pitched ululation. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Wednesday 14 August 2002 01:23 am, Rick Moen wrote: There will probably always be clever licence provisions to attempt subversion of the OSD's intent, no matter how many of them get patched. It would save a lot of time and energy to fall back on the rule of reason -- and the right of usage is obviously necessary to and implied by the existing OSD terms. I agree. The history of the legal profession has shown that a significant number of lawyers are willing and able to fold, spindle and mutilate even the most straightforward language. No matter how precise the OSD becomes, there will always be attempts to subvert it. The US Bill of Rights has some small amount of protection in the form of original intent. Perhaps the OSI could also consider original intent during its deliberations? -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Wednesday 14 August 2002 07:20 am, Russell Nelson wrote: I like David's, because it's such a shot across the bows. Unfortunately, his suggestion says nothing about modification restrictions, such as the GPL's, or BitKeeper's. That is because I wanted to limit the clause to what the user can do in the absence of a license. We already know that they need a license of some form before they can modify the software, but they shouldn't need one to use the software. In addition, modification is sufficiently addressed elsewhere in the OSD. p.s. I also wanted to ensure the right of first sale, right of criticism, and all that, but an enumeration of all the rights a user should possess would be extremely lengthy. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Quoting Lawrence E. Rosen ([EMAIL PROTECTED]): Of course, that makes it even more important for the OSD to be precise, and for the OSI board to be rigorous and not arbitrary in its review of licenses. That's another reason why I don't like Rick Moen's suggestion that OSI merely apply the rule of reason to its license approval process. I have informed the OSI board that, under trademark law, they cannot be arbitrary and capricious and still retain ownership of the certification mark. I was unaware of that concern, when I made the suggestion. (In fact, I hadn't known it involved any sort of legal question.) Perhaps you, or one of OSI's other attorney friends, have some notion how much discretion trademark caselaw suggests is permissible. -- Is it not the beauty of an asynchronous form of discussion that one can go and make cups of tea, floss the cat, fluff the geraniums, open the kitchen window and scream out it with operatic force, volume, and decorum, and then return to the vexed glowing letters calmer of mind and soul? -- The Cube, forum3000.org -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Rick Moen writes: Quoting Russell Nelson ([EMAIL PROTECTED]): Oh, it's *always* had to be changed. Anybody could insert restrictions on use into a license and ask us to approve it. Since the OSD says nothing about a license not being allowed to have restrictions on use, we would have to approve the license. Or they'll turn you into a newt? Russ, on other occasions, I believe you've been among those reminding people that the OSD isn't a black-box algorithm into which you plug candidate licences to determine whether they pass or fail. Whether I recall correctly or not, those were words of wisdom. They were also wrong. Oh, we *can* stretch the definition, but inventing requirements out of whole cloth is an invitation to a party -- party to a lawsuit, that is. -- -russ nelson http://russnelson.com | Crynwr sells support for free software | PGPok | businesses persuade 521 Pleasant Valley Rd. | +1 315 268 1925 voice | governments coerce Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
I kept my own email short because I knew there were other people, better qualified to speak on this. Rod, thanks for stepping forward. You presented the facts more thoroughly than I could. By the way, although you say you disagree with me, I don't think I disagree with you. I'm not sure where that leaves us. My issue with Bernstein is that he presents his opinion as if it were historical fact. This is dangerous for the unsuspecting reader. One part of his opinion is that Microsoft's end user license agreements (and, by extension, all software license agreements) are not enforceable; that you can simply ignore the license terms and do whatever you want with the software. That part of his argument really doesn't hold water for me. From: Rod Dixon [EMAIL PROTECTED] Last point: I do not think anyone made the argument that open source licenses that are contracts are not enforceable. _ Join the worlds largest e-mail service with MSN Hotmail. http://www.hotmail.com -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Russell Nelson wrote: [ Catching up on mail from ten days ago ] Carol A. Kunze writes: Here is the theoretical difference between proprietary and traditional (GPL, BSD) free software. With the former the user agrees to a license and does not get title to the copy of the program. Without agreeing to the license (and the use restrictions in it), the user has no legal right to use the copy of the software that they possess but do not own. Basically, its a license transaction where the user has no ownership in the copy of the software they possess. My understanding is that, if you have legally acquired a copy of the software, you have the right to run it. http://cr.yp.to/softwarelaw.html Absent a contract otherwise, a user can do anything they want to their copy, including use it, modify it, give it away, or resell it to someone else. Berstein says - In the United States, once you own a copy of a program, you can back it up, compile it, run it, and even modify it as necessary, without permission from the copyright holder. See 17 USC 117. You have to OWN the copy. When I say that in a proprietary license the licensor reserves title to the copy, I am saying the licensor takes the view that the user does not OWN the copy. The payment that is made is for a license to USE the software. So copyright rules that apply to OWNERS of copies do not apply if the copy of the software is still owned by the licensor and merely licensed to the user. This is the legal theory under which a proprietary licensor operates. I am not saying I think it should be this way, I am just explaining the legal foundation for this approach. It's the difference between buying and renting a house. If you buy a house you can do what you want with it, if you rent it you only get the rights your lease give you. So why form a contract, then? To get a warranty disclaimer. To get the recipient to agree that they lose their patent grant if they sue for patent infringement. If we can get those things without a contract, that would be a perfect world. The question here is whether we should amend the Open Source Definition so that it is clear whether click-wrap licenses are allowable or not. We could go either way, but we want to hear from you first. Your opinions solicited, and engaged! OSI has already blessed licenses which are intended to be agreements or contracts (see IBM license), so I'm confused about what the point is here.And why OSI definition would have to change. Am I missing something? They're not enforcable, at least as I understand it. I'm afraid they are. I still don't understand why there is this discussion about clickwrap when OSI has already OKed licenses which are contracts. Clickwrap vs. shrinkwrap is just a question of how the license is executed - that is, what mechanism the parties use to agree to it. Does the user click an I agree button online (clickwrap), or get a copy of the license in the box with the right to return if they don't like the terms (shrinkwrap). I still do not understand why the OSI definition would have to change. Why is the requirement for clickwrap any different from those licenses which OSI has blessed and which in fact are intended to be agreements? Can someone clue me in here? Carol -- -russ nelsonhttp://russnelson.com | Crynwr sells support for free software | PGPok | businesses persuade 521 Pleasant Valley Rd. | +1 315 268 1925 voice | governments coerce Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | -- 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: Legal soundness comes to open source distribution
Carol A. Kunze writes: Berstein says - In the United States, once you own a copy of a program, you can back it up, compile it, run it, and even modify it as necessary, without permission from the copyright holder. See 17 USC 117. You have to OWN the copy. When I say that in a proprietary license the licensor reserves title to the copy, I am saying the licensor takes the view that the user does not OWN the copy. That's an interesting view, given that they take money for it. The payment that is made is for a license to USE the software. So copyright rules that apply to OWNERS of copies do not apply if the copy of the software is still owned by the licensor and merely licensed to the user. This is the legal theory under which a proprietary licensor operates. Anybody can make up a theory for something. I have a theory that says that perpetual motion works. I can't get any physicists to sign onto that theory, though. I hope that no judge is willing to sign onto this legal theory. I am not saying I think it should be this way, I am just explaining the legal foundation for this approach. It's the difference between buying and renting a house. If you buy a house you can do what you want with it, if you rent it you only get the rights your lease give you. If I rent a house, the landlord wants it back. If I buy a piece of software, the putative owner of the copy of the software doesn't care if I destroy the copy. Pretty careless owner! They're not enforcable, at least as I understand it. I'm afraid they are. Click-wrap licenses? They're more enforcable than shrink-wrap licenses (which are unenforcable as I understand it)? I still don't understand why there is this discussion about clickwrap when OSI has already OKed licenses which are contracts. The discussion is over whether we should be ambiguous about whether the contract has been executed, or whether we should allow a license to require an acceptance ritual. I still do not understand why the OSI definition would have to change. Oh, it's *always* had to be changed. Anybody could insert restrictions on use into a license and ask us to approve it. Since the OSD says nothing about a license not being allowed to have restrictions on use, we would have to approve the license. -- -russ nelson http://russnelson.com | Crynwr sells support for free software | PGPok | businesses persuade 521 Pleasant Valley Rd. | +1 315 268 1925 voice | governments coerce Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Tuesday 13 August 2002 08:30 pm, Carol A. Kunze wrote: You have to OWN the copy. When I say that in a proprietary license the licensor reserves title to the copy, I am saying the licensor takes the view that the user does not OWN the copy. ... If you buy a house you can do what you want with it, if you rent it you only get the rights your lease give you. This is where the big disconnect occurs between the user and the manufacturer/licensor. When I rent a house, I KNOW that I am renting a house. But with software I have no clue. I have undergone every single motion of purchasing a product, obtained a sales receipt that itemizes a copy fo the software, yet I do not own it. Moreover, I don't even know this fact until the first time I try to use it. I am of the archaic and jurassic opinion that law that cannot be understood by the average layman is bad law. When the average consumer thinks they are buying a copy of Windows when they are not, because the law says they haven't, then the law is an accomplice to fraud. Skipping back to the middle of the last paragraph... The payment that is made is for a license to USE the software. From where I sit, it seems like the user is purchasing the right to VIEW the license. Only when they actually view the license and subsequently agree to it, do they gain the right to use the software. I still do not understand why the OSI definition would have to change. Why is the requirement for clickwrap any different from those licenses which OSI has blessed and which in fact are intended to be agreements? Can someone clue me in here? The main issue in my mind is not the simple click-wrap. That already exists in several forms for several Open Source products. Instead, the real issue (to me) is whether an Open Source license can require derivative works or downstream distribution to also use click-wrap. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Tuesday 13 August 2002 08:52 pm, Russell Nelson wrote: Oh, it's *always* had to be changed. Anybody could insert restrictions on use into a license and ask us to approve it. Since the OSD says nothing about a license not being allowed to have restrictions on use, we would have to approve the license. Please, please, please, please guarantee the right of users to use Open Source Software! Please. Everything else in the OSD is meaningless without it. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
David Johnson writes: On Tuesday 13 August 2002 08:52 pm, Russell Nelson wrote: Oh, it's *always* had to be changed. Anybody could insert restrictions on use into a license and ask us to approve it. Since the OSD says nothing about a license not being allowed to have restrictions on use, we would have to approve the license. Please, please, please, please guarantee the right of users to use Open Source Software! Please. Everything else in the OSD is meaningless without it. Yup, which is why the DFSG is so pathetic as a legal definition. I laugh at Bruce Perens when he says that we've changed the OSD too far away from the DFSG. We haven't changed it enough! But anyway, feel free to propose language. I've had my shot, and been shot down. -- -russ nelson http://russnelson.com | Crynwr sells support for free software | PGPok | businesses persuade 521 Pleasant Valley Rd. | +1 315 268 1925 voice | governments coerce Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
David Johnson wrote: I still do not understand why the OSI definition would have to change. Why is the requirement for clickwrap any different from those licenses which OSI has blessed and which in fact are intended to be agreements? Can someone clue me in here? The main issue in my mind is not the simple click-wrap. That already exists in several forms for several Open Source products. Instead, the real issue (to me) is whether an Open Source license can require derivative works or downstream distribution to also use click-wrap.-- David Johnson ___ http://www.usermode.org pgp public key on website Oh. Now I get it. Thanks. I'd have to ponder this more, but off the top of my head, it seems to me that requiring a distributor to enter into a license contract with the user means that the distributor should have a lawyer advising them. No offense to my own profession, but I don't like that idea. I am much more comfortable have an unlawyered developer use a non-contractual permission notice accompanying a sale or a free transfer of title, than entering into a complex legal agreement. In any event, I am going to have to go back and reread the approved licenses to see which ones require entering into an agreement and the extent to which downsteam distributors are required to do the same. Carol -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Tuesday 13 August 2002 09:12 pm, Russell Nelson wrote: But anyway, feel free to propose language. I've had my shot, and been shot down. I'll number this one zero for traditional reasons: 0) The possessor of a copy of the software must not be required to enter into or become party to any agreement or license in order to make use of the software, including the preparation of the software for subsequent use. Any problems with this? -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Tuesday 13 August 2002 09:37 pm, Carol A. Kunze wrote: In any event, I am going to have to go back and reread the approved licenses to see which ones require entering into an agreement and the extent to which downsteam distributors are required to do the same. Since distribution is an exclusive right of the author, it is perfectly reasonable for a distributor to enter into an agreement before distributing the software. It seems to be that clauses like by distributing this software you agree to... are not a problem. But the use of the software is not an exclusive right of the author. That's why click-wrap is problematic. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
I want to summarize what we have discussed on click-wrap because the issue is significant from the standpoint of the legal standing of open source licenses, and so I can include proposed responses in our research project on the OSD. It is my understanding that the issue initially involved the approval of a license, not a change to the OSD. The discussion of click-wrap then considered whether the fact that adding indicia of mutual assent to website agreements like open source licenses (e.g., a mouse click from the user) might have adverse implications for the position that open source licenses are non-contractual licenses. There was also some discussion concerning whether click-wrap conditions imposed on downstream or sub-licensees is practical (it may be difficult to implement). Finally, some raised the question whether the click-wrap condition is doomed to failure in cases where distribution is packaged with multiple programs carrying distinct licenses. Is this a fair summary? Rod - Original Message - From: David Johnson [EMAIL PROTECTED] To: Carol A. Kunze [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Sent: Wednesday, August 14, 2002 12:00 AM Subject: Re: Legal soundness comes to open source distribution On Tuesday 13 August 2002 08:30 pm, Carol A. Kunze wrote: You have to OWN the copy. When I say that in a proprietary license the licensor reserves title to the copy, I am saying the licensor takes the view that the user does not OWN the copy. ... If you buy a house you can do what you want with it, if you rent it you only get the rights your lease give you. This is where the big disconnect occurs between the user and the manufacturer/licensor. When I rent a house, I KNOW that I am renting a house. But with software I have no clue. I have undergone every single motion of purchasing a product, obtained a sales receipt that itemizes a copy fo the software, yet I do not own it. Moreover, I don't even know this fact until the first time I try to use it. I am of the archaic and jurassic opinion that law that cannot be understood by the average layman is bad law. When the average consumer thinks they are buying a copy of Windows when they are not, because the law says they haven't, then the law is an accomplice to fraud. Skipping back to the middle of the last paragraph... The payment that is made is for a license to USE the software. From where I sit, it seems like the user is purchasing the right to VIEW the license. Only when they actually view the license and subsequently agree to it, do they gain the right to use the software. I still do not understand why the OSI definition would have to change. Why is the requirement for clickwrap any different from those licenses which OSI has blessed and which in fact are intended to be agreements? Can someone clue me in here? The main issue in my mind is not the simple click-wrap. That already exists in several forms for several Open Source products. Instead, the real issue (to me) is whether an Open Source license can require derivative works or downstream distribution to also use click-wrap. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3 -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Legal soundness comes to open source distribution
But the use of the software is not an exclusive right of the author. That's why click-wrap is problematic. I understood the point that Rod Dixon was making is that section 117(a) of the Copyright Act applies, by its own words, to owners of a copy as distinguished from licensees. If that distinction matters, then section 117 does not authorize a *licensee* to make a copy of the software in memory in order to *use* the software; it only authorizes an *owner of a copy* to do so. Whatever else open source licenses do, they do not explicitly make a licensee the owner of a copy. To transfer ownership requires a contract; a mere license won't do. I note that section 117(b) states this differently. It allows the making of a copy by the owner or lessee of a machine as long as that machine lawfully contains an authorized copy of the computer program. The conclusion I draw from Rod's point is that the copyright act cannot be relied on as an authorization for use of software, at least to the extent that use requires the making of a copy in memory. Rod, am I reading you correctly? Regardless of this confusing point, why does this make click-wrap problematic? /Larry Rosen -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Legal soundness comes to open source distribution
I think you *almost* have it right, Rod. There are two different issues at stake here. First, Bruce Perens and others have been concerned for some time that the OSD does not address the right to use software; it is silent on that point. OSI has been asked in the past to approve licenses that impose obligations on licensees, ranging from the infamous advertising clause of the BSD that was withdrawn several years ago by the author of the license, to licenses that require users to notify the licensor before use, to licenses that seek to implement a click-wrap mechanism that requires a user (not just a distributor) to acknowledge the formation of a contract. The fear is that these licenses are just scratching the surface. What should the OSI board do when someone proposes a license that makes truly onerous demands on users? I have made it clear to the board that, because we have a certification mark, OSI can disapprove of a license only if it does not comply with the OSD; smell tests are not allowed to enter into the decision. So we all agree that we want to prevent smelly licenses that burden users, but we haven't defined smell. Several people, including Bruce Perens, Russ Nelson, myself, and most recently David Johnson, have suggested wording for such an OSD provision. None of those versions has caused the others on this list to stand up and cheer. The second issue deals directly with click-wrap. Perhaps it should be broadened to include shrink-wrap, for both forms of acknowledging the formation of a contract are potentially important for open source software. For example, Linux is available for download as well as in a box sold in stores. Some licensors and some distributors want to avoid any ambiguity about contract formation. They wish to follow procedures to obtain the assent of licensees so that, in the unlikely event that there is litigation, they can at least argue coherently that the licensee explicitly assented to the license. In response to this concern, and in light of a specific proposed license that wanted to *obligate* users and distributors to implement click-wrap procedures, I proposed a Click-Wrap Notice. I did not suggest it be made mandatory, but I did ask that it be posted on the OSI website as a standard way that people who want to can implement click-wrap. Please remember also that I called it a notice not a contract. If there is to be a contract, it must be the license itself, not some brief notice that purports only to highlight some important aspects of the license. The responses have been, to say the least, overwhelming. I wish that the responses had been more illuminating, however. Good questions have been raised about whether a click-wrap notice helps or hurts open source, whether it is legally necessary and, if so, in what circumstances and for which licenses, and whether it conflicts with the desireable goal listed above to avoid burdens on users. That's where you came in, Rod. /Larry Rosen -Original Message- From: Rod Dixon [mailto:[EMAIL PROTECTED]] Sent: Tuesday, August 13, 2002 10:37 PM To: David Johnson; Carol A. Kunze Cc: [EMAIL PROTECTED] Subject: Re: Legal soundness comes to open source distribution I want to summarize what we have discussed on click-wrap because the issue is significant from the standpoint of the legal standing of open source licenses, and so I can include proposed responses in our research project on the OSD. It is my understanding that the issue initially involved the approval of a license, not a change to the OSD. The discussion of click-wrap then considered whether the fact that adding indicia of mutual assent to website agreements like open source licenses (e.g., a mouse click from the user) might have adverse implications for the position that open source licenses are non-contractual licenses. There was also some discussion concerning whether click-wrap conditions imposed on downstream or sub-licensees is practical (it may be difficult to implement). Finally, some raised the question whether the click-wrap condition is doomed to failure in cases where distribution is packaged with multiple programs carrying distinct licenses. Is this a fair summary? Rod - Original Message - From: David Johnson [EMAIL PROTECTED] To: Carol A. Kunze [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Sent: Wednesday, August 14, 2002 12:00 AM Subject: Re: Legal soundness comes to open source distribution On Tuesday 13 August 2002 08:30 pm, Carol A. Kunze wrote: You have to OWN the copy. When I say that in a proprietary license the licensor reserves title to the copy, I am saying the licensor takes the view that the user does not OWN the copy. ... If you buy a house you can do what you want with it, if you rent it you only get the rights your lease give you. This is where the big disconnect occurs between the user
Re: Legal soundness comes to open source distribution
[ Catching up on mail from ten days ago ] Carol A. Kunze writes: Here is the theoretical difference between proprietary and traditional (GPL, BSD) free software. With the former the user agrees to a license and does not get title to the copy of the program. Without agreeing to the license (and the use restrictions in it), the user has no legal right to use the copy of the software that they possess but do not own. Basically, its a license transaction where the user has no ownership in the copy of the software they possess. My understanding is that, if you have legally acquired a copy of the software, you have the right to run it. http://cr.yp.to/softwarelaw.html Absent a contract otherwise, a user can do anything they want to their copy, including use it, modify it, give it away, or resell it to someone else. So why form a contract, then? To get a warranty disclaimer. To get the recipient to agree that they lose their patent grant if they sue for patent infringement. If we can get those things without a contract, that would be a perfect world. The question here is whether we should amend the Open Source Definition so that it is clear whether click-wrap licenses are allowable or not. We could go either way, but we want to hear from you first. Your opinions solicited, and engaged! OSI has already blessed licenses which are intended to be agreements or contracts (see IBM license), so I'm confused about what the point is here.And why OSI definition would have to change. Am I missing something? They're not enforcable, at least as I understand it. -- -russ nelson http://russnelson.com | Crynwr sells support for free software | PGPok | businesses persuade 521 Pleasant Valley Rd. | +1 315 268 1925 voice | governments coerce Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
I thought that section 117 was about the right to crack a program's copy protection (if necessary) in order to make a legitimate backup copy. Well, that's an oversimplification, but I think it's closer to the truth than Mr. Bernstein's argument. It goes to show that you shouldn't believe every opinion that you read on the Internet. (Follow the references back to the source; the quotes under patches both seem to be taken out of context. If you read them in their intended context you might find that they don't support Mr. Bernstein's opinion nearly as well.) Bruce - Original Message - From: Russell Nelson [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Monday, August 12, 2002 6:59 PM Subject: Re: Legal soundness comes to open source distribution [ Catching up on mail from ten days ago ] Carol A. Kunze writes: Here is the theoretical difference between proprietary and traditional (GPL, BSD) free software. With the former the user agrees to a license and does not get title to the copy of the program. Without agreeing to the license (and the use restrictions in it), the user has no legal right to use the copy of the software that they possess but do not own. Basically, its a license transaction where the user has no ownership in the copy of the software they possess. My understanding is that, if you have legally acquired a copy of the software, you have the right to run it. http://cr.yp.to/softwarelaw.html Absent a contract otherwise, a user can do anything they want to their copy, including use it, modify it, give it away, or resell it to someone else. So why form a contract, then? To get a warranty disclaimer. To get the recipient to agree that they lose their patent grant if they sue for patent infringement. If we can get those things without a contract, that would be a perfect world. The question here is whether we should amend the Open Source Definition so that it is clear whether click-wrap licenses are allowable or not. We could go either way, but we want to hear from you first. Your opinions solicited, and engaged! OSI has already blessed licenses which are intended to be agreements or contracts (see IBM license), so I'm confused about what the point is here.And why OSI definition would have to change. Am I missing something? They're not enforcable, at least as I understand it. -- -russ nelson http://russnelson.com | Crynwr sells support for free software | PGPok | businesses persuade 521 Pleasant Valley Rd. | +1 315 268 1925 voice | governments coerce Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | -- 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: Legal soundness comes to open source distribution
Russell Nelson wrote: ... it looks like a license without click-wrap is weaker at protecting your rights. By definition, Open Source *licenses* permit anybody to re-distribute without any explicit permission from the author. As has already been pointed out, if the user does not accept the (open source) license, he would be governed by the statute, and cannot re-distribute or modify my work. Other licenses attempt to restrict the user from exercising even the small rights available to him under the statutory law. So what do we really mean by my rights under an open source license? Under the Open Source definition, with respect to law of copyright, I permit everyone to re-distribute my work and retain only a small bundle of rights remains with me, called the moral rights. (example - US Code Title 17 Sec 106A). Every thing else excepted by open source licenses do not fall within the realm of copy right and come within the scope of product liability law or law relating to sale of goods. Issues relating to freedoms under the law of copy right apart, what is effectively protected by the software license are not my rights, but my liabilities under the laws relating to product liability and sale of goods. The time is coming when you won't be able to distribute software unless you have presented the license to the user and their assent is necessary to access the software. Even free software. What the law and the courts really say, when they insist on manufacturers producers and dealers ( and other similar entitles) informing the consumer and obtaining their assent about the warrant and its clauses, is that the user / consumer should be aware of the disclaimer, (that is, if the law permits such disclaimers at all). The primary (but often unspoken) reason for imposition of product liability is there is no way the user could find out why a particular product functioned the way it does. Even if he could, various laws relating to intellectual property prevented the user from making modifications to the product. Obviously, this is not the case with Open Source Software. The source code is out there, and it is for the user to access it, and find out whether the software will do what he wants it to do. Or else, I will do it for him, of course, for a fee. No court will impose a liability on creators of software who have no inkling about the use to which the software is being put to, unless either (a) consideration passes between the user and the creator or (b) the doctrine of injurious reliance is attracted (I suffered a loss because I trusted your promised donation and incurred an expense). Even in cases where exceptions to the principles of consideration or privity apply, liability would not be fastened unless the person sought to be held liable knew about the risks to which the user is exposing to himself. In other words, if the consumer intends to hold me liable for use of my software, (a) he should pay me for using it - the principle of consideration (c) he should obtain the software from me - the principle of privity. If he wants to hold me responsible on the basis of injurious reliance, there should be some representation by me to him regarding capabilities of the software. Of course, when law prohibits exclusionary clauses in contracts, nothing - neither click wrap, nor a signature in ink on a paper - can exclude my liability. The question here is whether we should amend the Open Source Definition so that it is clear whether click-wrap licenses are allowable or not. We could go either way, but we want to hear from you first. Your opinions solicited, and engaged! What should be debated is not whether Open Sourced software should be covered by click-wrap or not, but whether open source software should be covered by product liability or not. I feel that the answer is no. Except in countries where statutory provisions exist prohibiting contracts with exclusionary clauses, chances of an author being fastened with product liability for his open sourced work are very remote. What is really necessary is a campaign to take Open Source Software outside the scope of (compulsory) statutory product liability. Mahesh T Pai. Want to sell your car? advertise on Yahoo Autos Classifieds. It's Free!! visit http://in.autos.yahoo.com -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Bruce Perens wrote: http://www.nysd.uscourts.gov/courtweb/pdf/D02NYSC/01-07482.PDF . This is the kind of case (the facts disclosed by the case - not the decision in the legal sense) which arises coz. you claim to provide the user with one thing, and take away something else without telling him. No amount of disclaimers will save you from liability in such a situation. When you tell the user that he is getting a word processor, while in fact the program sends you copies of files created by the program, you are going to be faced with such situations. There is no use trying to shield yourselves with some warranty disclaimer, and that the user accepted the disclaimer is no excuse. This is quite different from a situation when you are providing something, let us say, which is capable of handling only ascii files, and call it a word processor while the user is looking for a Unicode aware program, a click through license *and* access to source code *might* help; especially if he (the user) does not tell you what he is looking for. Mahesh T Pai. Want to sell your car? advertise on Yahoo Autos Classifieds. It's Free!! visit http://in.autos.yahoo.com -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Sunday 04 August 2002 12:18 am, Mahesh T Pai wrote: What is really necessary is a campaign to take Open Source Software outside the scope of (compulsory) statutory product liability. I would hesitate to limit liability on the basis of Open Sourcedness. Rather, I would base it on the commercial nature of software. Non-commercial software should be outside the scope of product liability, but software obtained commercially, Open Source or otherwise, would not. My rationale is that an insurance burden upon non-commercial developers is so great that it would stop Open Source development outright, since 99% of Open Source software is non-commercial at the point of its creation. One should not be liable for good deeds, gifts or helping out the community. Commercial distributors, on the other hand, are at the minimum making the implicit claim that the software is merchantable, suitable as a product, and safe, unless otherwise stated. Of course, the real solution is to fix the mess that liability has become. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Fri, 2 Aug 2002, Russell Nelson wrote: From what various legal scholars tell me, a non-contractual license (such as the GPL) cannot cause you to give up your warranty rights. Is there a reference of some sort for this? It's about the only solid reason I see to need to go beyond copyright law. Is there any court precedent that suggests this? A case where someone was given something for free, with warranty disclaimed in a copyright license, and the court decided that warranty disclaimer was invalid? This is a pretty big delta to current understanding, so if a change as large as expanding the OSD to cover contracts is based upon this, we need more than hearsay. Are there any other reasons to consider allowing the OSD to cover contracts? My sense is that keeping it limited to copyright licenses has been key to its success to this point. Agreed. That's why I think we need to amend the OSD so that it clearly states that a license must not restrict use, modification, or redistribution of the software. The OSD, by applying to copyright licenses, already allows restrictions on redistribution. It'd be kinda toothless if it didn't... Brian -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Brian Behlendorf writes: On Fri, 2 Aug 2002, Russell Nelson wrote: From what various legal scholars tell me, a non-contractual license (such as the GPL) cannot cause you to give up your warranty rights. Is there a reference of some sort for this? It's about the only solid reason I see to need to go beyond copyright law. Is there any court precedent that suggests this? A case where someone was given something for free, with warranty disclaimed in a copyright license, and the court decided that warranty disclaimer was invalid? Well, see, that's the problem. You can protect critical variables with a spin-lock. If you fail to, your kernel crashes. I don't want the open source movement to crash. When the case above happens, Open Source will crash. It will be too late. Are there any other reasons to consider allowing the OSD to cover contracts? My sense is that keeping it limited to copyright licenses has been key to its success to this point. Mine, too. I'm not at all happy about having to go to click-wrap. It opens a whole can of worms. But I'm not happy about taking the risk of having nobody want to lose their house because they wrote free software. Small risk, maybe, but it has big consequences. Maybe click-wrap creates more problems than it solves? We need to ask the question rather than assuming the answer, as some would have us do. Agreed. That's why I think we need to amend the OSD so that it clearly states that a license must not restrict use, modification, or redistribution of the software. The OSD, by applying to copyright licenses, already allows restrictions on redistribution. It'd be kinda toothless if it didn't... Yes, yes. Not redistribution. But definitely use and modification. If you could put restrictions on modification, then BitKeeper is open source. And I don't think anybody besides Larry McVoy wants (wantED being a better term) that. -- -russ nelson http://russnelson.com | New Internet Acronym: Crynwr sells support for free software | PGPok | 521 Pleasant Valley Rd. | +1 315 268 1925 voice | IANAE Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | I Am Not An Economist -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Your questions actually raise many more issues than is apparent. The first critical hurdle we need to conquer is the confusion over whether open source licenses, which are presumed to be in compliance with the OSD, are properly denominated non-contractual licenses. I do not agree with that claim, but I do think that it is a claim that requires resolution. Although there may be strategic reasons for insisting that the GPL (or any other open source license) is merely a copyright license, there are consequences attached to the position. One way to consider this matter is to faithfully review the OSD to determine whether some of its articles exceed the boundaries of copyright law. If so, it may be unhelpful to ignore that fact in assessing whether an open source license is non-contractual. Regarding the question about giving up warranty rights, I am not familiar with any case explicitly on that point, but maybe someone else has more information on that matter. On the other hand, as is licensing is authorized under UCITA. In addition, the federal warranty law (Magnuson-Moss) only governs a WRITTEN warranty for consumer, mass-market goods, which, arguably, may include software distribution when the seller provides a written warranty. Generally, if a written warranty is provided, the seller cannot eliminate any implied warranty under the federal law; that condition might be what Brian is referring to by his reference to warranty rights. I believe the bottom line for open source licensing is that the federal law does not apply to as is licensing. You might conclude that as is licensing is not exactly consumer-friendly, but one might also view it as part of the trade-off for the freedom granted by the licensor. Rod Rod Dixon, J.D., LL.M. [EMAIL PROTECTED] http://www.cyberspaces.org/dixon/ My papers on the Social Science Research Network (SSRN) are available through the following url: http://papers.ssrn.com/author=240132 - Original Message - From: Brian Behlendorf [EMAIL PROTECTED] To: Russell Nelson [EMAIL PROTECTED] Cc: [EMAIL PROTECTED] Sent: Saturday, August 03, 2002 4:20 AM Subject: Re: Legal soundness comes to open source distribution On Fri, 2 Aug 2002, Russell Nelson wrote: From what various legal scholars tell me, a non-contractual license (such as the GPL) cannot cause you to give up your warranty rights. Is there a reference of some sort for this? It's about the only solid reason I see to need to go beyond copyright law. Is there any court precedent that suggests this? A case where someone was given something for free, with warranty disclaimed in a copyright license, and the court decided that warranty disclaimer was invalid? This is a pretty big delta to current understanding, so if a change as large as expanding the OSD to cover contracts is based upon this, we need more than hearsay. Are there any other reasons to consider allowing the OSD to cover contracts? My sense is that keeping it limited to copyright licenses has been key to its success to this point. Agreed. That's why I think we need to amend the OSD so that it clearly states that a license must not restrict use, modification, or redistribution of the software. The OSD, by applying to copyright licenses, already allows restrictions on redistribution. It'd be kinda toothless if it didn't... Brian -- 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: Legal soundness comes to open source distribution
Russell Nelson scripsit: If you could put restrictions on modification, then BitKeeper is open source. The GPL puts modest restrictions on modification, at least of interactive programs. All OS licenses, or nearly all, prevent you from modifying the copyright notices. -- One art / There is John Cowan [EMAIL PROTECTED] No less / No more http://www.reutershealth.com All things / To do http://www.ccil.org/~cowan With sparks / Galore -- Douglas Hofstadter -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Legal soundness comes to open source distribution
On Fri, 2 Aug 2002, Russell Nelson wrote: From what various legal scholars tell me, a non-contractual license (such as the GPL) cannot cause you to give up your warranty rights. On Sat, 3 Aug 2002, Brian Behlendorf wrote: Is there a reference of some sort for this? It's about the only solid reason I see to need to go beyond copyright law. Is there any court precedent that suggests this? A case where someone was given something for free, with warranty disclaimed in a copyright license, and the court decided that warranty disclaimer was invalid? This is a pretty big delta to current understanding, so if a change as large as expanding the OSD to cover contracts is based upon this, we need more than hearsay. Are there any other reasons to consider allowing the OSD to cover contracts? My sense is that keeping it limited to copyright licenses has been key to its success to this point. Agreed. That's why I think we need to amend the OSD so that it clearly states that a license must not restrict use, modification, or redistribution of the software. The OSD, by applying to copyright licenses, already allows restrictions on redistribution. It'd be kinda toothless if it didn't... I am baffled by everyone's confusion and philosophical rantings. Almost every license in OSI's approved list is much more than a copyright license. The MPL (and almost all similar licenses), for example, contains a patent grant that specifically applies to use and practice and it disclaims application of those patents to the combination of the Original Code with other software or devices. It contains a defensive suspension provision relating to patent litigation that applies to users of the software as much as to distributors. It contains a limitation of liability clause (not related to the warranty provision) that limits liability for damages of any sort. It contains restrictions on U.S. government rights. It contains a comprehensive set of miscellaneous provisions that govern contract interpretation, governing law, jurisdiction and venue, attorneys' fees, even a disclaimer of the United Nations Convention on Contracts (!) for the International Sale of Goods. What makes anyone think that this *CONTRACT* will be interpreted by the courts strictly under copyright law? Now, what if a distributor under the MPL insists that *users* of his software indicate clearly their assent to this contract? Should the OSD prevent such manifestations of assent? This entire discussion of click-wrap has gotten so politicized that you are confusing your own lawyers with your opinions. /Larry -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Is there a reference of some sort for this? It's the case at http://www.nysd.uscourts.gov/courtweb/pdf/D02NYSC/01-07482.PDF . IMO it's not all that germane to warranty disclaimer, and I'm not buying the chain of extrapolation that leads from this case to the conclusion that click-wrap might be necessary. It's about the only solid reason I see to need to go beyond copyright law. It's not about copyright law at all. The warranty obligation does not follow the copyright. It's about: 1. Is a simple warranty disclaimer that does not require agreement adequate? 2. How do you need to present the warranty disclaimer? 3. Do you really need a contract that other parties actually agree to in some way, for example by clicking yes? It's reasonably clear that you need one if you want someone else to indemnify you. It's not nearly so clear that you need one if you simply want to disclaim warranties. Agreed. That's why I think we need to amend the OSD so that it clearly states that a license must not restrict use, modification, or redistribution of the software. I agree that there should be no restrictions on use, modification, or distribution _other_than_those_ necessary to implement the goals of Open Source, such as disclaiming the warranty, preserving the copyright statement, mandating source distribution when the licensor chooses that option, and mandating transmission of the license to all parties. A simple no restrictions equates to public domain. Larry Rosen: I am baffled by everyone's confusion and philosophical rantings. That's distressing. This is your own community, or should be, since you claim to represent them. If they are confused, shouldn't you blame your presentation of the issue? If they are philosophical, and you didn't expect that, could it be that you've lost touch with them? So far, I see some significantly better alternatives than click-through. The very first should be a set of guidelines for distributions and other environments where free software is installed that would cause them to inform the user that: 1) There are licenses. 2) They disclaim warranties. 3) This is how you view the licenses. 4) This is how you look at the source code to perform your own due diligence. In the case of a distribution, most of them already do this at distribution install time. Debian does display a click-through warranty disclaimer when you install it. It also has a login message disclaiming warranties, but only on the text login. Obviously, this needs to be beefed up. In the case of package installers on something other than a Linux distribution, where we have less control of the enivronment, perhaps click-through is appropriate, but I still would oppose allowing it to be a license requirement. A license that requires it is going to cause us no end of trouble with the environments where we can deal with the problem more easily. Thanks Bruce -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Saturday 03 August 2002 06:02 am, Russell Nelson wrote: Maybe click-wrap creates more problems than it solves? We need to ask the question rather than assuming the answer, as some would have us do. If it will result in a divided rancorous community, dozens of new licenses that no one uses, and the perception in the general public that Open Source is not that different from closed source, then of course it will cause more problems that it will solve. One of the most basic tenets of Free and Open Software is that the right of the user to use, copy, distribute and modify the software is paramount. We assert the property rights of the user to his *copy* of the software. We do not tell the user what he can or cannot do with his copy. Once the software leaves our hands it is out of our control. Even the most restrictive clauses in OSS licenses are limited to ensuring that one user does not lay any property claim to the copy of another. Click-thru threatens to overturn this fundamental tenet. Regardless of what other effects it may have, it will severly damage the philosophical core of Open Source. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Legal soundness comes to open source distribution
Bruce, are you going to respond to any of my other comments besides my expression of bafflement? Or are you going to simply blame me for the confusion and lack of legal understanding on the part of *some* of the leaders of the open source community about whether licenses are contracts? I invite you to address directly my argument that the MPL (and similar licenses) is clearly, obviously, without question or doubt, a contract and not merely a copyright license. The case you cited, Specht, et al. v. Netscape, et al., does not deal with the issue of warranty. The decision addresses a preliminary matter, specifically whether a license that contains an arbitration clause can be enforced against licensees. That is a very important issue, for if arbitration can be compelled, the licensor may be able to dramatically reduce its costs and its risks of litigation before an uneducated jury. Many of my clients (licensors and licensees alike) demand an arbitration clause in their licenses for the simple reasons of cost avoidance and risk reduction. Arbitration is not available for copyright disputes; it is a contract provision. The courts have always ruled (most recently in a California Supreme Court case relating to employment contracts containing arbitration clauses) that arbitration provisions are favored and will be enforced against plaintiffs, but only if the plaintiff can be shown to have assented to the arbitration clause in a contract. I have myself litigated cases where the validity of an arbitration clause was the principal issue. Once that issue was resolved, the case settled because one or the other party wanted to minimize any further risk of loss. In Specht, the court first analyzed whether a contract was formed. This question was decided under state law even though the court had jurisdiction based on a federal question. Which state law to apply was itself a complicated issue, because the licensees were in multiple jurisdictions. The court decided to apply California contract law to that question. I believe that this means that a similar court, faced with a federal question of copyright infringement for which the defense of license is raised, would have to determine, as a preliminary matter, under state law, whether a contract (license) was formed, long before it addressed any copyright issues. That is how the Sun v. Microsoft case was resolved, by the way, although contract formation questions were not relevant to that decision. Once the court concluded that it was going to apply California contract law, it then proceeded to the key (for us) issue of whether the licensees assented to the contract through click-wrap procedures. The court held that Netscape's procedures were inadequate to establish assent through click-wrap and that a contract was therefore not formed. Therefore, the arbitration clause of the contract could not be enforced and Netscape could not compel arbitration. I've not spoken to Netscape's attorneys about this case but I presume there were high-level meetings about this in which the lawyers and website administrators were instructed to clean up their procedures to ensure click-wrap assent by licensees before their software could be downloaded. Here's another example of the importance of click-wrap. I have told Sun's attorneys that their previous procedures for obtaining click-wrap assent to their restrictive Java specification licenses were defective. (My arguments are similar to those raised by the defendants in Specht.) I therefore argue that Sun cannot restrict the open source community from implementing open source programs by attempting to enforce those license restrictions on Sun's specifications. As to your entirely separate issue, whether the OSD should contain a provision to prevent restrictions on use, I want to address that separately. When non-lawyers smash complex issues together in one email, they obviously get confused. /Larry -Original Message- From: Bruce Perens [mailto:[EMAIL PROTECTED]] Sent: Saturday, August 03, 2002 10:58 AM To: Lawrence E. Rosen Cc: 'Brian Behlendorf'; [EMAIL PROTECTED] Subject: Re: Legal soundness comes to open source distribution Is there a reference of some sort for this? It's the case at http://www.nysd.uscourts.gov/courtweb/pdf/D02NYSC/01-07482.PDF . IMO it's not all that germane to warranty disclaimer, and I'm not buying the chain of extrapolation that leads from this case to the conclusion that click-wrap might be necessary. It's about the only solid reason I see to need to go beyond copyright law. It's not about copyright law at all. The warranty obligation does not follow the copyright. It's about: 1. Is a simple warranty disclaimer that does not require agreement adequate? 2. How do you need to present the warranty disclaimer? 3. Do you really need a contract that other parties actually agree to in some way, for example by clicking
Re: Legal soundness comes to open source distribution
On Saturday 03 August 2002 09:25 am, Lawrence E. Rosen wrote: What makes anyone think that this *CONTRACT* will be interpreted by the courts strictly under copyright law? There are several reasons, but I'll go into just one: there is a significant number of laymen in the community that doubt the bona fides of click-thru contracts. Contracts of adhesion are not in question here, merely that subset known as shrink-wrap, click-thru and the like. The primary evidence that these contracts are not bona fide is that the licensor does not have the exclusive right to install and execute the software. The potential licensee already has rights to install and execute the software which they possess. Furthermore, one cannot make the exercise of a preexisting right the indication of assent. Here is an example: What if you purchased a movie DVD, inserted it into your DVD player, only to discover that a contract being displayed with the instructions to indicate agreement or non-agreement? Would such a contract be valid? What if you had to press a button saying I Agree in order to start your automobile? What if untwisting the cap off a bottle of beer indicated acceptance to hold the brewer blameless for any damages that might occur? What if breaking a paper seal printed with a contract were necessary in order to read of new book you just bought? Some Open Source licenses ARE contracts in certain situations. The GPL acts as a contract when you distribute the software, because the distribution of the software is an exclusive right of the author, and can certainly be used as an indication of assent. But any license that holds the exercise of preexisting rights to be assent is severely flawed. Now, what if a distributor under the MPL insists that *users* of his software indicate clearly their assent to this contract? Should the OSD prevent such manifestations of assent? A distributor can always wrap an OSS license with a form of click-thru. There are a few Open Source projects that do this, such as the Qt as obtained directly from Trolltech. But the larger question is whether an OSS license can mandate the use of click-thru. I have no problems with typing yes when I build Qt. But I would have serious problems if Trolltech required me to get a yes response from everyone I subsequently distributed the software to. My opinion is that the OSD can allow the use of click-thru, but that they must not allow the mandate of click-thru. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Sat, Aug 03, 2002 at 12:17:10PM -0700, Lawrence E. Rosen wrote: Bruce, are you going to respond to any of my other comments besides my expression of bafflement? Sure, no problem. Or are you going to simply blame me for the confusion and lack of legal understanding on the part of *some* of the leaders of the open source community about whether licenses are contracts? That is Brian Behlendorf of Collab.net you are talking about. His company offers training on Open Source licensing. HP buys it. If you are not getting through to Brian, backing up and starting again would be advised, because you are surely losing the rest of the audience. I invite you to address directly my argument that the MPL (and similar licenses) is clearly, obviously, without question or doubt, a contract and not merely a copyright license. Oh, I considered this so obvious that it wasn't necessary for me to comment upon it, and certainly I would not have disputed it. But it is peripheral to the issue of a warranty _disclaimer_, which like a copyright permission, does _not_ necessarily have to be in the form of a contract. The decision addresses a preliminary matter, specifically whether a license that contains an arbitration clause can be enforced against licensees. There are many license terms that I believe would require a contract. _Indemnification_ is one that is germane to this argument. Choice of venue and arbitration probably require a contract too. But I'm not convinced that a simple disclaimer of warranty requires a contract. Many of my clients (licensors and licensees alike) demand an arbitration clause in their licenses for the simple reasons of cost avoidance and risk reduction. Were I writing a proprietary software license, I would certainly ask for indemnification, choice of venue, an arbitration clause, and anything else that would be likely to hurt the other guy, and I would ask for them to be expressed in the most forceful possible way - I might even require internet registration so that I had confirmation that the licensee had agreed. After all, that sort of license is entirely one-sided - it's written for the copyright holder and nobody else. If I am able to express those terms at all when pursuing Open Source, I may not be able to express them with the greatest possible force, because they place an undue burden on the other participants, and are not likely to be accepted. This is simply the difference between a vendor-customer relationship and a partnership with a community. Thanks Bruce -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
I guess I am unsure of why there is such strong opposition to a clickwrap licensing requirement. The Netscape-Smart-download case follows the prevailing legal climate; namely, the licensor increases the risks of losing a legal challenge to the license (either under the enforcement of a license provision or the formation of the entire agreement) if the licensor does not carefully ensure that proof of mutual assent can be shown. Regarding Bruce's three questions: there are at least two federal laws that might be relevant to this question: Magnusson-Moss and E-SIGN, and there are likely to be nearly 50 state laws and 2 uniform codes relied upon by courts. In other words, I do think the correct answer to the first question is going to be yes. In response to question #1, I would ask another question: aside from ease on the license drafter, why would you want to impose terms (a disclaimer is still a license term, albiet a negation) under conditions that make it unclear to both parties whether the terms have been agreed to? This seems to run counter to the purpose of drafting terms. Questions 2 and 3 appear to be answered, in part, by the Netscape-Smart-Dowload opinion. I do not agree with all of the court's points (footnote 10 seems particularly distressing), but I think the court adopts the prevailing approach by characterizing the netscape license at issue as browser-wrap lacking manifestations of mutual assent. One final word of caution on this matter: once the OSI board resolves the approach to take on clickwrap, whether a particular warranty disclaimer will be enforced may depend upon a patchwork of state and federal consumer protection laws for mass market, open source licenses, which is likely to mean that some disclaimers may not be enforced even if they are enforceable. Rod Is there a reference of some sort for this? It's the case at http://www.nysd.uscourts.gov/courtweb/pdf/D02NYSC/01-07482.PDF . IMO it's not all that germane to warranty disclaimer, and I'm not buying the chain of extrapolation that leads from this case to the conclusion that click-wrap might be necessary. It's about the only solid reason I see to need to go beyond copyright law. It's not about copyright law at all. The warranty obligation does not follow the copyright. It's about: 1. Is a simple warranty disclaimer that does not require agreement adequate? 2. How do you need to present the warranty disclaimer? 3. Do you really need a contract that other parties actually agree to in some way, for example by clicking yes? It's reasonably clear that you need one if you want someone else to indemnify you. It's not nearly so clear that you need one if you simply want to disclaim warranties. Agreed. That's why I think we need to amend the OSD so that it clearly states that a license must not restrict use, modification, or redistribution of the software. I agree that there should be no restrictions on use, modification, or distribution _other_than_those_ necessary to implement the goals of Open Source, such as disclaiming the warranty, preserving the copyright statement, mandating source distribution when the licensor chooses that option, and mandating transmission of the license to all parties. A simple no restrictions equates to public domain. Larry Rosen: I am baffled by everyone's confusion and philosophical rantings. That's distressing. This is your own community, or should be, since you claim to represent them. If they are confused, shouldn't you blame your presentation of the issue? If they are philosophical, and you didn't expect that, could it be that you've lost touch with them? So far, I see some significantly better alternatives than click-through. The very first should be a set of guidelines for distributions and other environments where free software is installed that would cause them to inform the user that: 1) There are licenses. 2) They disclaim warranties. 3) This is how you view the licenses. 4) This is how you look at the source code to perform your own due diligence. In the case of a distribution, most of them already do this at distribution install time. Debian does display a click-through warranty disclaimer when you install it. It also has a login message disclaiming warranties, but only on the text login. Obviously, this needs to be beefed up. In the case of package installers on something other than a Linux distribution, where we have less control of the enivronment, perhaps click-through is appropriate, but I still would oppose allowing it to be a license requirement. A license that requires it is going to cause us no end of trouble with the environments where we can deal with the problem more easily. Thanks Bruce -- 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: Legal soundness comes to open source distribution
On Saturday 03 August 2002 01:11 pm, Rod Dixon wrote: The Netscape-Smart-download case follows the prevailing legal climate; namely, the licensor increases the risks of losing a legal challenge to the license (either under the enforcement of a license provision or the formation of the entire agreement) if the licensor does not carefully ensure that proof of mutual assent can be shown. I'm wondering what challenges an Open Source developer faces without a click-thru license. What risks does he face if it is ruled that the user did not assent to the license? For a proprietary developer, the risks involve critical reviews of the software being published, someone reverse engineering the software and discovering how it works, and the sale of used software under the first sale rule. But as Open Source developers we don't care about that stuff. For an Open Source developer, the risk is ONLY in regards to getting sued for damages. We don't want to get sued so we include warranty and liability disclaimers in our licenses. But it has not been made clear to me that mutual assent is necessary to disclaim warranty, or whether mutual assent is sufficient to remove liability. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Bruce Perens: 1. Is a simple warranty disclaimer that does not require agreement adequate? From: Rod Dixon [EMAIL PROTECTED] I do think the correct answer to the first question is going to be yes. In response to question #1, I would ask another question: aside from ease on the license drafter, why would you want to impose terms (a disclaimer is still a license term, albiet a negation) under conditions that make it unclear to both parties whether the terms have been agreed to? This is mostly an issue of practicality - and practicality is what drives many OSD questions. Debian, for example, has some 8000 packages, and a typical system will have 1000 to 3000 of them, some people install the whole kitchen sink which is probably around 6000 packages once package conflicts are resolved. The packages are produced by some 800 different package maintainers who are not employees of Debian and are not under the orders of any corporation. Of course there are many different owners for the software that is packaged. It's not clear that Debian is the warrantor, rather than the package maintainers and the copyright holders. There are at least 100 variations on the licenses, both different license versions and different entities offering the same licenses. If even one one-hundredth of the packages required click-wrap, it would not be practical to present them all. Imagine clicking through 30 licenses during an install. There would be no reasonable expectation that the installer had actually read the text of all of those licenses, which defeats the purpose of click-wrap. The same issue comes up in other venues, such as download sites, and applies to all other distributions, Red Hat, and so on, although most distributions are smaller than Debian and may have employees doing the packaging. The practical alternative is to present _once_ that there are licenses, that they in general disclaim warranties and that thus you should have no expectation of warranty, where you can find them, and the fact that since you have source you can perform your own due diligence. This seems to run counter to the purpose of drafting terms. Only if you are taking a vendor-centric view. Vendor-centric licenses are drawn with maximum possible terms to protect the vendor. Open Source licenses are drawn to protect the vendor as much as possible while still being practical and fair to redistribute and deploy throughout a broad community of users and derivative developers who are not motivated to accept an odious license. That means that we deliberately make some things easy - for example the act of copying and redistributing a software distribution, and installing and using that distribution. We may reduce the software producer's capability to defend themselves, by a reasonable amount, in order to achieve those goals. Thanks Bruce -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Your points have answered a couple of questions. If we look at this issue narrowly, it makes sense to say that clickwrap should not be a mandatory requirement of the OSD, but could be approved as appropriate for an open source licensor. The point being that there is nothing extraordinary about clickwrap/click-through method itself that sustains mutual assent; rather, there are many ways to accomplish this task that are more appealing as a practical matter. In that light, there are a number of ways to disclose a warranty disclaimer in a manner that best ensures that the end-user receives notice/consent. It is difficult to frame the warranty disclaimer issue abstractly and independently of the license because one walks from one potential quagmire to another despite the fact that a specific instance is probably a great deal less complicated. The bottom line is very close to how the discussion sees to have begun. If an open source licensor distributes software via a website, the license/warranty disclaimer/contract/ should make its way to the potential licensee in a manner that the netscape license in the smart-download case did not. Click-through dialog boxes seem to offer a level of assurance that a court might agree with the licensor that mutual assent is indicated. Clickwrap is not the only way to show mutual assent. More practical measures are certainly possible so I would agree that we should not get too affixed to clickwrap when less budernsome, but equally effective measures can be adopted. Should the OSD mandate a clickwrap measure? I agree with those who say no, but I would not undermine the importance of mutual assent when it is relevant. License drafters should be aware of the importance that contract formation rules have on the enforceability of the license regardless (and independent) of the terms. Is mutual consent relevant for warranty disclaimers only? I think this is a difficult question in the context of software licensing, but viewing the matter simply as a generic issue, my answer is: since an AS IS disclaimer is ostensibly not a promise of anykind, the effectiveness of the AS IS notice is likely to be controlled by consumer protection laws, rather than a genuine issue of copyright licensing (i.e. copyright law or contract law). Anyone with a consumer protection law background? I have made no further comment on the philosophical issues since they seem to raise the stakes of disunity more than the legal issues. Rod - Original Message - From: Bruce Perens [EMAIL PROTECTED] To: [EMAIL PROTECTED]; [EMAIL PROTECTED]; [EMAIL PROTECTED] Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Saturday, August 03, 2002 5:20 PM Subject: Re: Legal soundness comes to open source distribution Bruce Perens: 1. Is a simple warranty disclaimer that does not require agreement adequate? From: Rod Dixon [EMAIL PROTECTED] I do think the correct answer to the first question is going to be yes. In response to question #1, I would ask another question: aside from ease on the license drafter, why would you want to impose terms (a disclaimer is still a license term, albiet a negation) under conditions that make it unclear to both parties whether the terms have been agreed to? This is mostly an issue of practicality - and practicality is what drives many OSD questions. Debian, for example, has some 8000 packages, and a typical system will have 1000 to 3000 of them, some people install the whole kitchen sink which is probably around 6000 packages once package conflicts are resolved. The packages are produced by some 800 different package maintainers who are not employees of Debian and are not under the orders of any corporation. Of course there are many different owners for the software that is packaged. It's not clear that Debian is the warrantor, rather than the package maintainers and the copyright holders. There are at least 100 variations on the licenses, both different license versions and different entities offering the same licenses. If even one one-hundredth of the packages required click-wrap, it would not be practical to present them all. Imagine clicking through 30 licenses during an install. There would be no reasonable expectation that the installer had actually read the text of all of those licenses, which defeats the purpose of click-wrap. The same issue comes up in other venues, such as download sites, and applies to all other distributions, Red Hat, and so on, although most distributions are smaller than Debian and may have employees doing the packaging. The practical alternative is to present _once_ that there are licenses, that they in general disclaim warranties and that thus you should have no expectation of warranty, where you can find them, and the fact that since you have source you can perform your own due diligence. This seems to run counter to the purpose of drafting terms. Only if you are taking a vendor-centric
Re: Legal soundness comes to open source distribution
From: Rod Dixon [EMAIL PROTECTED] it makes sense to say that clickwrap should not be a mandatory requirement of the OSD, but could be approved as appropriate for an open source licensor. I'd better clear this up. There was no proposal for click-wrap to be a a mandiatory requirement of the _OSD_. The question was whether or not the OSD should allow a license that requires click-wrap. I mantain that it's not appropriate for the OSD to allow it. Thanks Bruce -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
David Johnson writes: Click-thru threatens to overturn this fundamental tenet. Regardless of what other effects it may have, it will severly damage the philosophical core of Open Source. I share your fear, and brought it to the board at the last meeting. Allowing contractural licenses really means modifying the OSD so that it *clearly* does not allow restrictions on use or modification. -- -russ nelson http://russnelson.com | New Internet Acronym: Crynwr sells support for free software | PGPok | 521 Pleasant Valley Rd. | +1 315 268 1925 voice | IANAE Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | I Am Not An Economist -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Legal soundness comes to open source distribution
Lawrence E. Rosen writes: The MPL (and almost all similar licenses), for example, contains a patent grant that specifically applies to use and practice and it disclaims application of those patents to the combination of the Original Code with other software or devices. But that, by itself, doesn't form a contract. Where is the consideration? Instead, it's just giving permission. I can say You can walk across my land, but that's all you can do without forming a contract, because it's my property. You're not giving up anything. It contains a defensive suspension provision relating to patent litigation that applies to users of the software as much as to distributors. I can put anything I want in my license. The question is whether it's enforcible in a context where there is no assent and no consideration. Is a contract formed in such a case? What makes anyone think that this *CONTRACT* will be interpreted by the courts strictly under copyright law? Contract? Or contract-wannabee? -- -russ nelson http://russnelson.com | New Internet Acronym: Crynwr sells support for free software | PGPok | 521 Pleasant Valley Rd. | +1 315 268 1925 voice | IANAE Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | I Am Not An Economist -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
John Cowan writes: Russell Nelson scripsit: If you could put restrictions on modification, then BitKeeper is open source. The GPL puts modest restrictions on modification, at least of interactive programs. Indeed. One has to wonder whether the GPL should be an approved license. Okay, so obviously we have to accept the GPL. How are we then to distinguish between Richard's good restrictions on modifications from Larry's bad restrictions on modifications? Maybe we need a GFPL, the General Free Public License, which allows all modifications but which is otherwise the GPL? I haven't looked at it in detail yet, but perhaps that is exactly what Larry's OSL is. All OS licenses, or nearly all, prevent you from modifying the copyright notices. That's a freedom you don't have regardless of what the license says. -- -russ nelson http://russnelson.com | New Internet Acronym: Crynwr sells support for free software | PGPok | 521 Pleasant Valley Rd. | +1 315 268 1925 voice | IANAE Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | I Am Not An Economist -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On 2002.08.01 23:18 Russell Nelson wrote: At the July OSI board meeting last week, we approved the Academic Free License (think MIT/BSD/X11/Apache with a patent grant) and we sent four licenses back for reconsideration. As someone who has submitted a license (the Bento Poetic License), is there any way for me to find out whether the license was sent back for reconsideration, rejected outright, or not even considered? Michael St. Hippolyte -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Michael St . Hippolyte writes: On 2002.08.01 23:18 Russell Nelson wrote: At the July OSI board meeting last week, we approved the Academic Free License (think MIT/BSD/X11/Apache with a patent grant) and we sent four licenses back for reconsideration. As someone who has submitted a license (the Bento Poetic License), is there any way for me to find out whether the license was sent back for reconsideration, rejected outright, or not even considered? Sorry, I hadn't yet gotten around to it. Yes, your license is being sent back for reconsideration. The license is approvable, but we are concerned that you may be acting under a misunderstanding of trademark law and as such the license may not serve your purpose. The board recommends that you consider obtaining legal advice before resubmitting this license. -- -russ nelson http://russnelson.com | New Internet Acronym: Crynwr sells support for free software | PGPok | 521 Pleasant Valley Rd. | +1 315 268 1925 voice | IANAE Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | I Am Not An Economist -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Russell Nelson wrote: The time is coming when you won't be able to distribute software unless you have presented the license to the user and their assent is necessary to access the software. Even free software. Our industry is maturing and we need to be more legally careful and rigorous. I think its early to state that free software needs a license to which the user agrees. Here is the theoretical difference between proprietary and traditional (GPL, BSD) free software. With the former the user agrees to a license and does not get title to the copy of the program. Without agreeing to the license (and the use restrictions in it), the user has no legal right to use the copy of the software that they possess but do not own.. Basically, its a license transaction where the user has no ownership in the copy of the software they possess. With free software, the user gets title to a copy of the software, which gives them the right to use it without a license. Basically, its a sale. The GPL then grants the owner of the copy additional permissions. These permissions are not intended to create contractual obligations on either party. However, they do give the user who exercises rights of the copyright owner in accordance with the permission, an estoppel defense against copyright infringement. Which is all the user needs. I think for open source and free software purposes, the non-contractual permission notice is far superior.However, it is a problem that the nature of the GPL (as a non-contractual permission notice) is not really clear. Most lawyers think it is supposed to be an agreement. It would help to be more upfront about it. The question here is whether we should amend the Open Source Definition so that it is clear whether click-wrap licenses are allowable or not. We could go either way, but we want to hear from you first. Your opinions solicited, and engaged! OSI has already blessed licenses which are intended to be agreements or contracts (see IBM license), so I'm confused about what the point is here.And why OSI definition would have to change. Am I missing something? Carol -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On 2 Aug 2002, Russell Nelson wrote: The question here is whether we should amend the Open Source Definition so that it is clear whether click-wrap licenses are allowable or not. We could go either way, but we want to hear from you first. Your opinions solicited, and engaged! I see a practical issue - if I install Debian from CD and fire up Mozilla, I don't want to have to go through ten dozen different dialog boxes with nearly inscrutable license terms listed in a small scrolling textbox I then am asked whether I accept or not before I can continue. Why so many? In going from bare hardware to loading the OS to browsing a web site, I'm likely to need to run applications and libraries written by many different groups of developers, each potentially with their own agreement, and each agreement potentially having some OSI-conformant-but-really-silly clauses, like you may not utter the word 'pancreas' while using our software. Even the BSD advertising clause is less of a potential annoyance than this could be. Maybe I'm taking this into reductio ad absurdum, but it's awful nice to know right now that there are no conditions on use with open source software, only conditions upon redistribution. Philosophically, I don't like the idea of someone being restricted in what they can do once they have the software in their hands. But then again, I have a bias towards minimalism anyways. Brian -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Fri, Aug 02, 2002 at 09:44:23AM -0700, Brian Behlendorf wrote: agreement potentially having some OSI-conformant-but-really-silly clauses, like you may not utter the word 'pancreas' while using our software. Even the BSD advertising clause is less of a potential annoyance than this could be. And does this use restriction trickle down to any derivatives of the software? If I use pieces of 10 packages, each with their own use restriction, am I going to end up having to use the resultant software only between the hours of 10 and 2, in the dark, without disclosing results, not uttering the word pancreas... Use licenses scare me. -drew -- M. Drew Streib [EMAIL PROTECTED] Independent Rambler, Software/Standards/Freedom/Law -- http://dtype.org/ msg05144/pgp0.pgp Description: PGP signature
RE: Legal soundness comes to open source distribution
Concerning use licenses: I understand your reluctance to use software that contains onerous use restrictions. If a license prohibits you from uttering the word pancreas, or any of the other silly restrictions noted in both Brian's and Drew's emails below, then don't use the license or software it licenses. Simply because a license is open source doesn't mean that we like the license terms or are willing to license it under those terms. It seems to me *unreasonable* to require, through some vague OSD provision that some of you want, that all open source licenses be *reasonable*. After all, some major players even consider the GPL to be unreasonable. /Larry -Original Message- From: M. Drew Streib [mailto:[EMAIL PROTECTED]] Sent: Friday, August 02, 2002 9:49 AM To: Brian Behlendorf Cc: Russell Nelson; [EMAIL PROTECTED] Subject: Re: Legal soundness comes to open source distribution On Fri, Aug 02, 2002 at 09:44:23AM -0700, Brian Behlendorf wrote: agreement potentially having some OSI-conformant-but-really-silly clauses, like you may not utter the word 'pancreas' while using our software. Even the BSD advertising clause is less of a potential annoyance than this could be. And does this use restriction trickle down to any derivatives of the software? If I use pieces of 10 packages, each with their own use restriction, am I going to end up having to use the resultant software only between the hours of 10 and 2, in the dark, without disclosing results, not uttering the word pancreas... Use licenses scare me. -drew -- M. Drew Streib [EMAIL PROTECTED] Independent Rambler, Software/Standards/Freedom/Law -- http://dtype.org/ -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Several packages of GPL'ed software for Win 32 come click wrapped. eg:- Bloodshed C++ from www.bloodshed.net and audacity. (any body want more examples?) If you do not click the accept button, the installation aborts. Mahesh T Pai. Russell Nelson wrote: The time is coming when you won't be able to distribute software unless you have presented the license to the user and their assent is necessary to access the software. Even free software. Our industry is maturing and we need to be more legally careful and rigorous. The question here is whether we should amend the Open Source Definition so that it is clear whether click-wrap licenses are allowable or not. We could go either way, but we want to hear from you first. Your opinions solicited, and engaged! Want to sell your car? advertise on Yahoo Autos Classifieds. It's Free!! visit http://in.autos.yahoo.com -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Pretty large amount of s/w is distributed in CDs, especially the open source variety. ( redistribution under the same license terms is one of the rights under the OSD ). In such case, the user would have acquired the media, (eg:- the CD coming with a magazine) and may or may not be aware of the contents. The contents of the same CD can often be distributed under different licenses. Here, the issue of becoming aware of the license terms can arise when the user wants to install the software. In such a situation, only use wrap will work. Therefore, I do not think that we have any option, except to recognize and accept use-wrap. Mahesh T Pai. David Johnson wrote: The first is where the license is presented during installation or first usage. The second is where the license is presented before one can acquire the software. I'll refer to the first as use-wrap and the second as download-wrap to avoid confusion. I have few problems with download-wrap if the only way to aquire the software is to click I agree. The user has no rights with regards to software which they do not possess. The problem is with use-wrap. By the time the user sees the license terms, they have already aquired the right to install and use the software, particularly so if they have aquired the software through a commercial transaction. If the license merely grants additional rights to the user, then use-wrap is no great problem. But if it lessens any rights already possessed by the user, then use-wrap is a serious wrong. I would have no problems with an Open Source license that mandates the use of download-wrap. But the mandate of use-wrap should never be part of an Open Source license. Just because the heathens do it doesn't mean we should as well. Want to sell your car? advertise on Yahoo Autos Classifieds. It's Free!! visit http://in.autos.yahoo.com -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Fri, Aug 02, 2002 at 10:31:36AM -0700, Lawrence E. Rosen wrote: Simply because a license is open source doesn't mean that we like the license terms or are willing to license it under those terms. It seems to me *unreasonable* to require, through some vague OSD provision that A better example: A benchmark suite is licensed under an OSI license, with the use provision that you cannot publish results with the open source version of the suite. You may copy it, redistribute it, use it internally, etc, but one of the most commercially useful features (a published result) is not allowed by a use license. Of course, a commercially licensed version is also available which allows you to publish results. A company can very easily cripple software via a use license. Should that package call itself open source? -drew -- M. Drew Streib [EMAIL PROTECTED] Independent Rambler, Software/Standards/Freedom/Law -- http://dtype.org/ msg05149/pgp0.pgp Description: PGP signature
RE: Legal soundness comes to open source distribution
M. Drew Streib wrote: [mailto:[EMAIL PROTECTED]] A better example: A benchmark suite is licensed under an OSI license, with the use provision that you cannot publish results with the open source version of the suite. You may copy it, redistribute it, use it internally, etc, but one of the most commercially useful features (a published result) is not allowed by a use license. Of course, a commercially licensed version is also available which allows you to publish results. A company can very easily cripple software via a use license. Should that package call itself open source? Then how about a provision of the OSD that reads something like the following: An open source license cannot restrict any fair use rights that would be available for a copyrighted work in the absence of a license. That certainly would prevent someone from restricting the publication of benchmark results, or using the work for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research. 17 U.S.C. §107. /Larry -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
begin Lawrence E. Rosen quotation: Then how about a provision of the OSD that reads something like the following: An open source license cannot restrict any fair use rights that would be available for a copyrighted work in the absence of a license. And which country's definition of fair use are you planning on adopting? -- Jack Valenti is to the American film viewer and the American public as the Boston strangler is to the woman home alone. -- http://cryptome.org/hrcw-hear.htm(search for Boston) -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
David Johnson (me) wrote: I would have no problems with an Open Source license that mandates the use of download-wrap. One day later and I have come to my senses. Let me rephrase my comment... I might not have too serious of a problem with an OSS license that mandates distributors to display the license before the user receives it, but I reserve the right to see the clause(s) in question before I commit myself. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
On Friday 02 August 2002 10:12 am, Mahesh T Pai wrote: In such case, the user would have acquired the media, (eg:- the CD coming with a magazine) and may or may not be aware of the contents. The contents of the same CD can often be distributed under different licenses. Here, the issue of becoming aware of the license terms can arise when the user wants to install the software. In such a situation, only use wrap will work. The result of such licenses will be that distributors will refuse to include such software in their collections. One such presentation of license and demand for acceptance might be palatable, maybe even two. But when the installation of Linux or a BSD involves several, possibly dozens, of such demands, the user is going to very upset. What if any of these packages became dependencies? Failure to agree to one of them could result in an unusable system! -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
Brian Behlendorf writes: I see a practical issue - if I install Debian from CD and fire up Mozilla, I don't want to have to go through ten dozen different dialog boxes with I don't like it any more than you do. You're being asked to agree to give up the right to any warranty. From what various legal scholars tell me, a non-contractual license (such as the GPL) cannot cause you to give up your warranty rights. Only a contract can do that. Yes, the law *is* an ass, and needs to be changed. Until then, a warranty disclaimer needs agreement. Philosophically, I don't like the idea of someone being restricted in what they can do once they have the software in their hands. Agreed. That's why I think we need to amend the OSD so that it clearly states that a license must not restrict use, modification, or redistribution of the software. -- -russ nelson http://russnelson.com | New Internet Acronym: Crynwr sells support for free software | PGPok | 521 Pleasant Valley Rd. | +1 315 268 1925 voice | IANAE Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | I Am Not An Economist -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
M. Drew Streib writes: Use licenses scare me. They scare me too. That's why I think we need to change the OSD. -- -russ nelson http://russnelson.com | New Internet Acronym: Crynwr sells support for free software | PGPok | 521 Pleasant Valley Rd. | +1 315 268 1925 voice | IANAE Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | I Am Not An Economist -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
RE: Legal soundness comes to open source distribution
Lawrence E. Rosen writes: Then how about a provision of the OSD that reads something like the following: An open source license cannot restrict any fair use rights that would be available for a copyrighted work in the absence of a license. That certainly would prevent someone from restricting the publication of benchmark results, or using the work for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research. 17 U.S.C. §107. What about adding 17 U.S.C. §117 rights as well? That would ensure that anybody who got a copy of the software would be able to use it, back it up, and restore it if necessary. -- -russ nelson http://russnelson.com | New Internet Acronym: Crynwr sells support for free software | PGPok | 521 Pleasant Valley Rd. | +1 315 268 1925 voice | IANAE Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | I Am Not An Economist -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Legal soundness comes to open source distribution
At the July OSI board meeting last week, we approved the Academic Free License (think MIT/BSD/X11/Apache with a patent grant) and we sent four licenses back for reconsideration. Here's the hitch: we were asked to approve a license which includes a requirement for click-wrap. The submittor had already been asked if that requirement was a necessity. She said yes, because of various legal precedents. We consulted a few people and yes, it looks like a license without click-wrap is weaker at protecting your rights. So, folks, the lawyers are coming. The time is coming when you won't be able to distribute software unless you have presented the license to the user and their assent is necessary to access the software. Even free software. Our industry is maturing and we need to be more legally careful and rigorous. The question here is whether we should amend the Open Source Definition so that it is clear whether click-wrap licenses are allowable or not. We could go either way, but we want to hear from you first. Your opinions solicited, and engaged! -- -russ nelson http://russnelson.com | New Internet Acronym: Crynwr sells support for free software | PGPok | 521 Pleasant Valley Rd. | +1 315 268 1925 voice | IANAE Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | I Am Not An Economist -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
My response is yes. In fact, the OSD recommendations I am developing as part of the OSD Model Code proposal will include a suggestion on which article and what language might be best to accomplish this. I am hoping to post the complete proposal during the fall semester. - Rod Rod Dixon, J.D., LL.M. Visiting Assistant Professor of Law Rutgers University School of Law - Camden [EMAIL PROTECTED] http://www.cyberspaces.org/dixon/ My papers on the Social Science Research Network (SSRN) are available through the following url: http://papers.ssrn.com/author=240132 - Original Message - From: Russell Nelson [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Thursday, August 01, 2002 11:18 PM Subject: Legal soundness comes to open source distribution At the July OSI board meeting last week, we approved the Academic Free License (think MIT/BSD/X11/Apache with a patent grant) and we sent four licenses back for reconsideration. Here's the hitch: we were asked to approve a license which includes a requirement for click-wrap. The submittor had already been asked if that requirement was a necessity. She said yes, because of various legal precedents. We consulted a few people and yes, it looks like a license without click-wrap is weaker at protecting your rights. So, folks, the lawyers are coming. The time is coming when you won't be able to distribute software unless you have presented the license to the user and their assent is necessary to access the software. Even free software. Our industry is maturing and we need to be more legally careful and rigorous. The question here is whether we should amend the Open Source Definition so that it is clear whether click-wrap licenses are allowable or not. We could go either way, but we want to hear from you first. Your opinions solicited, and engaged! -- -russ nelson http://russnelson.com | New Internet Acronym: Crynwr sells support for free software | PGPok | 521 Pleasant Valley Rd. | +1 315 268 1925 voice | IANAE Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | I Am Not An Economist -- 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: Legal soundness comes to open source distribution
On Thursday 01 August 2002 08:18 pm, Russell Nelson wrote: The submittor had already been asked if that requirement was a necessity. She said yes, because of various legal precedents. We consulted a few people and yes, it looks like a license without click-wrap is weaker at protecting your rights. So, folks, the lawyers are coming. Does that mean we should get to working cleaning out our flintlocks :-) Seriously, the problem here is the term click-wrap. There are two types of license presentation in use today that are referred to by this term. The first is where the license is presented during installation or first usage. The second is where the license is presented before one can aquire the software. I'll refer to the first as use-wrap and the second as download-wrap to avoid confusion. I have few problems with download-wrap if the only way to aquire the software is to click I agree. The user has no rights with regards to software which they do not possess. The problem is with use-wrap. By the time the user sees the license terms, they have already aquired the right to install and use the software, particularly so if they have aquired the software through a commercial transaction. If the license merely grants additional rights to the user, then use-wrap is no great problem. But if it lessens any rights already possessed by the user, then use-wrap is a serious wrong. I would have no problems with an Open Source license that mandates the use of download-wrap. But the mandate of use-wrap should never be part of an Open Source license. Just because the heathens do it doesn't mean we should as well. The time is coming when you won't be able to distribute software unless you have presented the license to the user and their assent is necessary to access the software. Even free software. Our industry is maturing and we need to be more legally careful and rigorous. First, this sounds like download-wrap, so the problem is not great. However, I still doubt that it is going to be necessary for most Open Source Software. The only rights the user will have to modify, distribute and copy the software must come from the license, and since those activities are normally the only activities regulated by OSS licenses, it does not matter if the user sees the license or not. The only potential problem is with the presentation of the warranty disclaimer. By all means, commercial software should be presenting the disclaimer to the user, whether by download-wrap or use-wrap. But a lack of merchantability disclaimer for non-merchanted software is not, in my non-lawyerly opinion, much of a problem. Besides which, I'm pretty certain that the primary purpose of proprietary click-wrap licenses is not to disclaim warranty. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Legal soundness comes to open source distribution
I agree with David that click-wrap (or click-through or web-wrap...) generally denotes what he describes as download-wrap licenses. Leaving aside the matter of use-wrap licensing, courts seem to viewing click-wrap licensing in two forms: the passive license and the active license. What is at issue, often, is electronic contract formation. In other words, did the user manifest consent to the terms of the license? In this regard, passive licenses or click-wraps that do not require the user to actually click a dialog box, slect something on a pull-down menu, or some other user interface input demonstrating I agree are less likely to be viewed as licenses that offer clear evidence that a user may have consented to the terms of the license. In my opinion, it is far better that this potential problem is fixed or avoided than ignored. Consequently, I think the OSD should include something on the click-wrap issue in its next iteration. With regard to the comment about whether it matters if the user sees the license, I suppose in a strictly technical sense it is true that manifestation of consent does not necessarily mean that licensors must prove that users must actually have seen the license to be bound or denominated a licensee, but a movement that prides itself on being generally consumer-friendly might be inclined to adopt a practice that more likely than not ensures that potential licensees read the copyright license to which they are bound. Therefore, I would suggest that in response to the prevailing legal climate and as a policy matter, the click-wrap issue is important enough to be considered in the next version of the OSD. - Rod [EMAIL PROTECTED] http://www.cyberspaces.org/dixon/ My papers on the Social Science Research Network (SSRN) are available through the following url: http://papers.ssrn.com/author=240132 - Original Message - From: David Johnson [EMAIL PROTECTED] To: Russell Nelson [EMAIL PROTECTED]; [EMAIL PROTECTED] Sent: Friday, August 02, 2002 12:49 AM Subject: Re: Legal soundness comes to open source distribution On Thursday 01 August 2002 08:18 pm, Russell Nelson wrote: The submittor had already been asked if that requirement was a necessity. She said yes, because of various legal precedents. We consulted a few people and yes, it looks like a license without click-wrap is weaker at protecting your rights. So, folks, the lawyers are coming. Does that mean we should get to working cleaning out our flintlocks :-) Seriously, the problem here is the term click-wrap. There are two types of license presentation in use today that are referred to by this term. The first is where the license is presented during installation or first usage. The second is where the license is presented before one can aquire the software. I'll refer to the first as use-wrap and the second as download-wrap to avoid confusion. I have few problems with download-wrap if the only way to aquire the software is to click I agree. The user has no rights with regards to software which they do not possess. The problem is with use-wrap. By the time the user sees the license terms, they have already aquired the right to install and use the software, particularly so if they have aquired the software through a commercial transaction. If the license merely grants additional rights to the user, then use-wrap is no great problem. But if it lessens any rights already possessed by the user, then use-wrap is a serious wrong. I would have no problems with an Open Source license that mandates the use of download-wrap. But the mandate of use-wrap should never be part of an Open Source license. Just because the heathens do it doesn't mean we should as well. The time is coming when you won't be able to distribute software unless you have presented the license to the user and their assent is necessary to access the software. Even free software. Our industry is maturing and we need to be more legally careful and rigorous. First, this sounds like download-wrap, so the problem is not great. However, I still doubt that it is going to be necessary for most Open Source Software. The only rights the user will have to modify, distribute and copy the software must come from the license, and since those activities are normally the only activities regulated by OSS licenses, it does not matter if the user sees the license or not. The only potential problem is with the presentation of the warranty disclaimer. By all means, commercial software should be presenting the disclaimer to the user, whether by download-wrap or use-wrap. But a lack of merchantability disclaimer for non-merchanted software is not, in my non-lawyerly opinion, much of a problem. Besides which, I'm pretty certain that the primary purpose of proprietary click-wrap licenses is not to disclaim warranty. -- David Johnson ___ http://www.usermode.org pgp public key on website -- license
Re: Section 2 source distribution terms (was Re: GPL vs APSL (was:YAPL is bad))
Thorsten Glaser writes: This breaks things which do not only consider code being licensed under $any_license but any kind of work (be it code, documentation, books etc.) OSI certifies software, not documentation, as open source. There are various documentation licenses out there. We're not decided on whether we want to certify documentation. -- -russ nelson [EMAIL PROTECTED] http://russnelson.com Crynwr sells support for free software | PGPok | It's a crime, not an act 521 Pleasant Valley Rd. | +1 315 268 1925 voice | of war. For my take, see: Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | http://quaker.org/crime.html -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Section 2 source distribution terms (was Re: GPL vs APSL (was: YAPL is bad))
Karsten M. Self writes: - Ensure that sources are distributable. Not only distributable, but also available. Sigh. Last time I sat down to rewrite #2, I ended up concluding that we really need to have *two* OSD's: one describing source code, and another describing the distribution of a special type of derived work created through mechanical means from source code which is also not only licensed under an Open Source license, but which is also actually available. Phwew! Started this posting with only three periods left in the type tray, and I managed to complete it using just those three! -- -russ nelson [EMAIL PROTECTED] http://russnelson.com Crynwr sells support for free software | PGPok | It's a crime, not an act 521 Pleasant Valley Rd. | +1 315 268 1925 voice | of war. For my take, see: Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | http://quaker.org/crime.html -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Section 2 source distribution terms (was Re: GPL vs APSL (was:YAPL is bad))
Dixitur de Russell Nelson respondebo ad: (...) Good. Close. Better than my previous attempt. What do you think of this: 2. Source Code The license applies to source code. A compiled executable is considered a derived work. Such an executable is only open source if its source code is also open source. When a compiled executable is not distributed with source code, there must be a well publicized means of obtaining the source code for no more than a reasonable reproduction cost -- preferably, downloading via the Internet without charge or access restrictions. The source code so offered must be in the preferred form in which a programmer would modify the program. Deliberately obfuscated source code does not qualify. Intermediate forms such as the output of a preprocessor or translator are not allowed. This breaks things which do not only consider code being licensed under $any_license but any kind of work (be it code, documentation, books etc.) which is the form I prefer to write. I usually put work and not code under X.Net (formerly MIT/BSD), because I feel that this is more concerning the distribution as whole and does not center/focus on the code. In my eyes, for most simple work the documentation is lot more an effort than the actual code. I might be wrong, and I definitively am wrong on things such as the Linux kernel, as any larger project, but for smaller projects this seems ok. Of course, a big problem with the OSD is that it talks about legal requirements, and yet was not touched by a lawyer before being cast into stone. Any kind of extensive rewrite probably ought to be done by people with actual experience with the law, as opposed to dilettantes like you and I. And me. Ok, but OSI is a nice attempt. -mirabilos -- Redistribution via AOL or the Microsoft network prohibited! According to billg's Win32 GetMessage() API, the return value of type BOOL can be one of {nonzero|zero|-1} and thus BOOL is tristate. -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Section 2 source distribution terms (was Re: GPL vs APSL (was: YAPL is bad))
It's my understanding that OSI is trying to come up with a plan to review the OSD. I may be presenting a proposal to larry soon to help them in that effort. Even so, I think the lawyers could benefit from the input of the developers. I would not abandon the project. Rod Original Message- From: Russell Nelson [EMAIL PROTECTED] To: [EMAIL PROTECTED] [EMAIL PROTECTED] Cc: Bcc: Subj: Re: Section 2 source distribution terms (was Re: GPL vs APSL (was: YAPL is bad)) Type: IPM.Note Sent: Wednesday, September 26, 2001 12:55 AM Karsten M. Self writes: Proposed language: 2. Source Code The license most provide for distribution in source code as well as compiled form. Where some form of a product is not distributed with source code, there must be a well publicized means of obtaining the source code for no more than a reasonable reproduction cost -- preferably, downloading via the Internet without charge or access restrictions. The source code so offered must be in the preferred form in which a programmer would modify the program. Deliberately obfuscated source code does not qualify. Intermediate forms such as the output of a preprocessor or translator are not allowed. For licenses in which distribution without source is allowed, an OSD Qualifying Distribution shall be defined as an offering of the software, under qualifying license terms, with source or an offer of source as described in this paragraph. Good. Close. Better than my previous attempt. What do you think of this: 2. Source Code The license applies to source code. A compiled executable is considered a derived work. Such an executable is only open source if its source code is also open source. When a compiled executable is not distributed with source code, there must be a well publicized means of obtaining the source code for no more than a reasonable reproduction cost -- preferably, downloading via the Internet without charge or access restrictions. The source code so offered must be in the preferred form in which a programmer would modify the program. Deliberately obfuscated source code does not qualify. Intermediate forms such as the output of a preprocessor or translator are not allowed. Of course, a big problem with the OSD is that it talks about legal requirements, and yet was not touched by a lawyer before being cast into stone. Any kind of extensive rewrite probably ought to be done by people with actual experience with the law, as opposed to dilettantes like you and I. -- -russ nelson [EMAIL PROTECTED] http://russnelson.com Crynwr sells support for free software | PGPok | It's a crime, not an act 521 Pleasant Valley Rd. | +1 315 268 1925 voice | of war. For my take, see: Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | http://quaker.org/crime.html -- 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: Section 2 source distribution terms (was Re: GPL vs APSL (was: YAPL is bad))
Rod Dixon, J.D., LL.M. writes: It's my understanding that OSI is trying to come up with a plan to review the OSD. From my message that you quoted?? No, no plan, but instead more like dreams. Larry has told us more than once that the OSD looks to a lawyer like a computer program written by a lawyer looks to a programmer (no offense intended to those extremely smart people who are proficient at both). While it may work, it doesn't reflect best current practice. But maybe your offer will help turn that dream into a reality? I'm not trying to talk you out of it, but instead to appraise you of the current state. -- -russ nelson [EMAIL PROTECTED] http://russnelson.com Crynwr sells support for free software | PGPok | It's a crime, not an act 521 Pleasant Valley Rd. | +1 315 268 1925 voice | of war. For my take, see: Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | http://quaker.org/crime.html -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Section 2 source distribution terms (was Re: GPL vs APSL (was:YAPL is bad))
On Wed, 26 Sep 2001, Russell Nelson wrote: Of course, a big problem with the OSD is that it talks about legal requirements, and yet was not touched by a lawyer before being cast into stone. Any kind of extensive rewrite probably ought to be done by people with actual experience with the law, as opposed to dilettantes like you and I. IMO, either rewrite it trying to be more clear, or rewrite it trying to cover all of the bases legally. Again, IMO, writing it to be more clear, and relying upon the common sense of the OSI, seems reasonable. -- Matthew Weigel Research Systems Programmer [EMAIL PROTECTED] ne [EMAIL PROTECTED] -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Section 2 source distribution terms (was Re: GPL vs APSL (was: YAPL is bad))
Karsten M. Self writes: Proposed language: 2. Source Code The license most provide for distribution in source code as well as compiled form. Where some form of a product is not distributed with source code, there must be a well publicized means of obtaining the source code for no more than a reasonable reproduction cost -- preferably, downloading via the Internet without charge or access restrictions. The source code so offered must be in the preferred form in which a programmer would modify the program. Deliberately obfuscated source code does not qualify. Intermediate forms such as the output of a preprocessor or translator are not allowed. For licenses in which distribution without source is allowed, an OSD Qualifying Distribution shall be defined as an offering of the software, under qualifying license terms, with source or an offer of source as described in this paragraph. Good. Close. Better than my previous attempt. What do you think of this: 2. Source Code The license applies to source code. A compiled executable is considered a derived work. Such an executable is only open source if its source code is also open source. When a compiled executable is not distributed with source code, there must be a well publicized means of obtaining the source code for no more than a reasonable reproduction cost -- preferably, downloading via the Internet without charge or access restrictions. The source code so offered must be in the preferred form in which a programmer would modify the program. Deliberately obfuscated source code does not qualify. Intermediate forms such as the output of a preprocessor or translator are not allowed. Of course, a big problem with the OSD is that it talks about legal requirements, and yet was not touched by a lawyer before being cast into stone. Any kind of extensive rewrite probably ought to be done by people with actual experience with the law, as opposed to dilettantes like you and I. -- -russ nelson [EMAIL PROTECTED] http://russnelson.com Crynwr sells support for free software | PGPok | It's a crime, not an act 521 Pleasant Valley Rd. | +1 315 268 1925 voice | of war. For my take, see: Potsdam, NY 13676-3213 | +1 315 268 9201 FAX | http://quaker.org/crime.html -- license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3
Re: Source Distribution
On Tue, Sep 05, 2000 at 01:37:47AM +, Ben DeLong wrote: Hi all, I have a question. I am building a Linux appliance based completely on open source software. In putting this appliance together I have not added nor modified a single line of code in any of the open source applications that I'm using. As this is an appliance and storage space is limited, there is no room to include the source within the appliance. Given that all the code in this appliance is freely available from a variety of locations on the net, do I have to go to the trouble of providing a source CD for all the open source programs that I have included in it? Cf: Section 3 of the GNU GPL, which spells out source distribution obligations. You must distribute sources, or a promise to provide same at no more than cost of physically performing the source distribution. If and only if you are considering a non-commercial use of GPL'd code can you rely on pointing to existing distribution points. Distribution need not be on the appliance itself. -- Karsten M. Self [EMAIL PROTECTED] http://www.netcom.com/~kmself Evangelist, Opensales, Inc.http://www.opensales.org What part of "Gestalt" don't you understand? Debian GNU/Linux rocks! http://gestalt-system.sourceforge.net/K5: http://www.kuro5hin.org GPG fingerprint: F932 8B25 5FDD 2528 D595 DC61 3847 889F 55F2 B9B0 PGP signature