Re: Proposed: Debian's Five Freedoms for Free Works
Scripsit Adam Warner [EMAIL PROTECTED] On Tue, 2003-06-17 at 02:54, Henning Makholm wrote: It must be possible for me to enjoy the freedoms without communicating with anybody else but those whom I voluntarily decide to distribute the software to. Why should I have to communicate with anyone, even in a situation of copyleft code distribution? Well, distribution is a form of communication. It would be too much to require that I should be able to distribute software to someone without sending them data (i.e. communication). And how does your proposal proscribe the limits of information disclosure when one does voluntarily decide to distribute the software? It says that the only party I can be required to disclose anything (such as the source) to is the one to whom I distribute. So communication isn't the root issue. The root issue is the extent of one's information disclosure obligation (at the point of distribution). No. If that were the root issue, we wouldn't have problems with you must send patches upstream licenses - they do not require me to disclose anything to upstream that I don't have to disclose to the recipient I choose. On the contrary the root issue is exactly that such a license tries to dictate *whom* I must disclose this information to. -- Henning Makholm I didn't even know you *could* kill chocolate ice-cream!
Re: Proposed: Debian's Five Freedoms for Free Works
On Mon, Jun 16, 2003 at 08:50:59PM +0300, Anton Zinoviev wrote: But otherwise let us talk about guidelines rather that about definitions. You seem to be unaware of past discussions on this subject. Could I trouble you to catch up with the debian-legal archives since (at least) March of this year? -- G. Branden Robinson|Optimists believe we live in the Debian GNU/Linux |best of all possible worlds. [EMAIL PROTECTED] |Pessimists are afraid the optimists http://people.debian.org/~branden/ |are right about that. pgpRBjSMjdtmY.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
Scripsit Adam Warner [EMAIL PROTECTED] A guideline of privacy could be read as a positive obligation that DFSG-free software licences protect against information disclosure. How about, instead of information disclosure, to speak about forced communication or something like that? As I see it, the freedoms we are speaking about here could all be described by It must be possible for me to enjoy the freedoms without communicating with anybody else but those whom I voluntarily decide to distribute the software to. This would essentially equal the desert island / dissident test, but without the fancy narrative. -- Henning Makholm Slip den panserraket og læg dig på jorden med ansigtet nedad!
Re: Proposed: Debian's Five Freedoms for Free Works
On 13.VI.2003 at 13:06 Branden Robinson wrote: On Fri, Jun 13, 2003 at 03:29:03PM +0300, Anton Zinoviev wrote: I'd like to mention here that FSF talks about free software and free documentation and not about free works. Well, they're the Free *Software* Foundation. Presumably, they care first and foremost about software. The same with Debian: we are (still) a software distribution. It is questionable whether we have to require these freedoms from works that are not software, nor documentation. What's questionable about it? This sort of rhetoric is FUD. My statement was that it is not our job to require freedoms from works that will never be part of Debian. For our Debian distribution the difference is not much important as we distribute only software and documentation. Not true, unless you define software and documentation so broadly that you open yourself up to exactly the same sort of ridicule that I received for calling everything in Debian main effectively software. Yes, I defined software and documentation so broadly. :) Things like the music files for frozen-bubble are neither software nor documentaiton. OK. If a music file is a part of some software, then we expect to be able to modify it. On the other hand if I am not allowed to modify some work such as music or picture, then I will miss this freedom only in case I want to use it as part of some software. Thats why I think that these 4+1 freedoms are essentially software freedoms. You are thinking only about what the software can do, and not about what the *license* might do. I can not think about a license that provides me with the first 4 freedoms, but not with the fifth. Any hints? Maybe this depends on how one interprets the first 4 freedoms? I don't think it makes sense to treat the FSF as some sort of unstable psychotic who's likely to go off and start shooting people if his every whim is not sated. It wasn't my intention to persuade this. We are free to discuss any definitions of what is free software/documentation/work. I just wanted point out that FSF always tries to make its philosophical ideas clear. If Debian differs in something, then they will make their best to educate our FS community why Debian is not right. If we try to clarify/change the definition of free software in collaboration with FSF -- that is wonderful. But otherwise let us talk about guidelines rather that about definitions. Anton Zinoviev
Re: Proposed: Debian's Five Freedoms for Free Works
4) The freedom to change the Work for any purpose[1], to distribute one's changes, and to distribute the Work in modified form. Access to the form of the work which is preferred for making modifications, if applicable, is a precondition for this. OK, so there's lots of argument about preferred form. How about a more negative definition: Deliberately obfusticated or encrypted forms and program-generated forms are *not* preferred forms for making modifications.
Re: Proposed: Debian's Five Freedoms for Free Works
On Sunday, Jun 15, 2003, at 12:45 US/Eastern, Nathanael Nerode wrote: Deliberately obfusticated or encrypted forms and program-generated forms are *not* preferred forms for making modifications. Program-generated forms can become the preferred form. Its certainly possible to use something like glade, for example, to generate skeleton code then fill it in. Or, more weirdly, people do hack up configure, flex/bison output, etc.
Re: Proposed: Debian's Five Freedoms for Free Works
On Fri, Jun 13, 2003 at 12:57:13AM -0400, Greg Pomerantz wrote: Branden Robinson [EMAIL PROTECTED] wrote: I would say that the controlling preference is that of the person who last modified the Work and distributed it in that modified form. Anyone downstream from that person would have to keep the source in that form and the binary together. I think one formulation that makes this a bit more explicit is this: 4) The freedom to change the Work for any purpose, to distribute one's changes, and to distribute the Work in modified form. Unrestricted access to the form of the work which is preferred by its author for making modifications, if applicable, is a precondition for this. (with the possible addition of the words or translator after author). This makes explicit the fact that there is no single preferred form. If you allow individual authors to define their own preferred version, you solve problems like this -- In trying to specify, this phrasing overlooks other important preferred forms. Someone may make changes to an existing piece of software which involve rearranging the APIs. Is this a translation? I'm not sure it is; but it may change the source sufficiently that the original author would prefer not to use that form as a basis for further modifications. The important nuance here is that the preferred form is the form that the last person to modify the code worked in when making those changes. This is unfortunately hard to specify precisely in a definition. Andrea writes a program in C. Marty rewrites Andrea's program in Fortran. Ulysses gets Marty's binary and asks for source -- Marty can send his Fortran source because it is the form Marty prefers for making modifications. Alternately, assume Marty lives on a proprietary island and only has access to proprietary programming languages. Marty should not be barred from translating Andrea's program into a proprietary language and distributing his modifications in that language (which is preferred because its the only thing he's got). I think any definition of preferred form needs to pass this test. In other words, I think that any definition of preferred form that requires an open or transparent format will be non-free. The same holds true for document formats of course. The person who aims to prepare a derivative work should have the option of using whatever form she prefers, and should have no obligation beyond the distribution of modifications in her preferred form. I am not sure at this point the extent to which certain exceptions need to be in place in the case where the author has some vested interest in selling you a proprietary interpreter (in the extreme case, pay me $100 for the AES key you need to decrypt my preferred form). Any thoughts? I think there are two distinct issues here that need to be considered: whether a piece of software can be considered free if the only source available for it requires proprietary software to be useful, and whether a license can be considered free if it enforces redistribution of source in a form that is useful without proprietary tools. The first question seems to be the more important one to this discussion, since being able to use/compile/edit the software is more fundamental than being able to redistribute it in modified form. And I think the intuitive answer here is the correct one: for purposes of identifying whether a piece of software comes with enough source code to be considered free, we already have a separation in Debian between the main archive and the contrib archive. Software in contrib is *effectively* non-free, even though it's not *intrinsically* non-free; it's non-free by circumstance. Free Software is well out of its bootstrapping infancy, and being self-hosting is an important characteristic of Debian. I think this formulation of freedom in response to the first question leads naturally to an answer to the second question: yes, we recognize certain copyleft compromises as being free, and one of these is that a license can require the redistribution of source code under free terms. If source code written to a non-free interpreter is effectively non-free, then I think a license which doesn't recognize distribution of this sort of source code as fulfilling the source distribution requirements *is* free. I'm not sure such a license requirement is necessarily a good idea, and I'm not sure whether the GPL contains this requirement, but I do think such a requirement would not render a license non-free. -- Steve Langasek postmodern programmer pgpncN7roXdyY.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
On Sun, Jun 15, 2003 at 01:28:18PM -0500, Steve Langasek wrote: The first question seems to be the more important one to this discussion, since being able to use/compile/edit the software is more fundamental than being able to redistribute it in modified form. FWIW, I disagree with this prioritization. While some of the definitional freedoms may be dependent on others, none is less important than another. We do not have sufficient freedom if we are allowed to modify our freely-licensed software DVD players to ignore region coding, not pass the Macrovision signal, and skip FBI warnings commercials for other products manufactured or sold by the same company that produced a DVD in our posession, but forbidden from redistributing a free software DVD player that has these features available. -- G. Branden Robinson| To be is to do -- Plato Debian GNU/Linux | To do is to be -- Aristotle [EMAIL PROTECTED] | Do be do be do -- Sinatra http://people.debian.org/~branden/ | pgp44sh7EFzMI.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
On Sat, 2003-06-14 at 06:15, Branden Robinson wrote: On Fri, Jun 13, 2003 at 09:15:26AM -0400, Jeremy Hankins wrote: Branden Robinson [EMAIL PROTECTED] writes: I personally have advocated a fifth freedom: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. I think (though I'm not sure) that I agree with what you're trying to do, but I don't like using privacy as its basis[1]. Reasonable people can disagree, of course, but I think it's important to understand that privacy and the free flow of information are competing values, and the optimum is some point between either extreme that maximizes other social values. I don't find your observation objectionable; I have been wrestling with a sound philosophical basis for my instinctive feelings on this subject. As long as the goal is met, I'm not particularly enamored of grounding freedom 5 on the concept of privacy. It probably shouldn't be grounded explicitly on *any* political principle, since people can have differing value systems, yet agree on this particular point. Branden, perhaps the term information disclosure would better suit you/us than privacy? That is we propose a DFSG-free licence cannot mandate information disclosure of anything but the information forming a distributed and derived work. I agree with Jeremy that I don't like using the idea of privacy as a basis for the freedom. Not accepting licences that require mandatory information disclosure seems much more concrete and exactly encompass the consequences you set out in your initial post. Regards, Adam
Re: Proposed: Debian's Five Freedoms for Free Works
On Sat, Jun 14, 2003 at 03:50:24PM +1200, Adam Warner wrote: Branden, perhaps the term information disclosure would better suit you/us than privacy? Sure, if that's agreeable to others. That is we propose a DFSG-free licence cannot mandate information disclosure of anything but the information forming a distributed and derived work. Right. I agree with Jeremy that I don't like using the idea of privacy as a basis for the freedom. Not accepting licences that require mandatory information disclosure seems much more concrete and exactly encompass the consequences you set out in your initial post. Probably so, yes. -- G. Branden Robinson| Debian GNU/Linux | Ab abusu ad usum non valet [EMAIL PROTECTED] | consequentia. http://people.debian.org/~branden/ | pgpdw2wmREOFq.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
On Fri, Jun 13, 2003 at 05:02:27PM -0400, Gregory K.Johnson wrote: Branden Robinson [EMAIL PROTECTED] writes: I personally have advocated a fifth freedom: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. I need to work on the wording of this fifth freedom a bit to make it clear that it is fair for a person to whom Free Software is distributed to demand access to the source code, including the source code to any changes or improvements made by the person from whom one is receiving the software. The point is that my usage of your Free Software does not entitle you to access to or any rights in my improvements to your software unless I distribute the Software back to you specifically. I added a phrase to the end that I think addresses these concerns: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others, except as these changes or Works are willfully shared. My views are better expressed by Jeremy Hankins's mail; as noted in my original message, my thinking on point 5) wasn't as developed as it could have been. Mr. Hankings very nicely pulled together some principles I already had in mind but hadn't thought to ground freedom 5) on. As others have pointed out, making a Debian list of free software freedoms, with one freedom more than GNU's list, would be a bit confusing and a bit untidy, but hardly enough so to rule out the idea. I think that the most compelling reason to add privacy to the list is Let's just call it freedom five and not privacy. Some folks may feel, with some legitimacy, that it's not really privacy that's being protected, though privacy advocates will probably have little to complain about. to clarify that Debian holds the right to make private modifications or anonymous public modifications to software in as high a regard as the other freedoms; I consider as simply a handy fringe benefit that the statement would also condemn so-called spyware. As others have pointed out, spyware can be viewed as simply flawed -- and the other freedoms facilitate fixing the flaws. And privacy as a human right is, I think, too political and nonspecific to the project for an official stance to be warranted. IMO there's nothing wrong with spyware per se, unless you *define* spyware to be that which surveilles people's actions without their knowledge and/or consent. In that case software itself isn't spyware and cannot be -- rather, spyware is a manner of deploying and administering software. To Branden: Do you have plans to try to add such a document to the Web site's statements of policy? Not yet. I would prefer this proposal to percolate for a while longer and see if I get flamed first. What procedural hurdles stand in the way of doing so? (I'm new to Debian, so please forgive my lack sophistication about the project's inner workings.) If the proposal is broadly seen as a good idea, very little. If it turns out to be opposed by the DPL, the Debian Webmasters, or certain other prominent folks, it may require a General Resolution. According to clause 4.1.5 of the Constitution, the Developers can issue nontechnical policy documents and statements with a simple majority under the General Resolution process. My proposed document would not do anything to invalidate or supersede the Debian Social Contract or Free Software Guidelines, so people who feel that those documents enjoy a special, currently extra-Constitutial protected status should not have any objection on those grounds. But before we go that route I'd like to see if this document generates further discussion. Then the document needs to be updated to reflect the feedback-based consensus. It's not suitable in its current state. -- G. Branden Robinson|I just wanted to see what it looked Debian GNU/Linux |like in a spotlight. [EMAIL PROTECTED] |-- Jim Morrison http://people.debian.org/~branden/ | pgptQKiM7aGu7.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
Dylan Thurston [EMAIL PROTECTED] wrote: One clear difference is that the FSF finds the FDL license to be free on their terms [...] To my knowledge, the FSF have never claimed the FDL meets their definition of free software. Can you show otherwise, please? -- MJR/slef My Opinion Only and possibly not of any group I know. http://mjr.towers.org.uk/ jabber://[EMAIL PROTECTED] Creative copyleft computing services via http://www.ttllp.co.uk/ Thought: Changeset algebra is really difficult.
Re: Proposed: Debian's Five Freedoms for Free Works
In article [EMAIL PROTECTED], Adam Warner wrote: Branden, perhaps the term information disclosure would better suit you/us than privacy? That is we propose a DFSG-free licence cannot mandate information disclosure of anything but the information forming a distributed and derived work. But surely privacy is exactly when you have when information disclosure is not forced on you? Could you please elaborate on the difference? Peace, Dylan
Re: Proposed: Debian's Five Freedoms for Free Works [humor]
Hi Am Fre, 2003-06-13 um 23.30 schrieb Anthony DeRobertis: On Friday, Jun 13, 2003, at 04:57 US/Eastern, Joachim Breitner wrote: Unrestricted access to all not-common elements to produce the final product is a precondition for this. [...] Humans (non-common: the order of the 4 bases on the DNA string) :-) Hmmm... sounds like you're required to distribute yourself with a work to a make it free software. Oooh, can I clone a Linus from the kernel sources? Or are they not free? ;-) well, these were separate examples, and Linux is not a component needed to build the kernel. But if of course the all-around-apache-group has a subproject perfecthacker.apache.org, that provides the DNA of the perfect Free Software Hacker, this project would be free if it distributes the DNA sequence. I just wanted to show the universalism of my definition. Joachim -- Joachim Breitner e-Mail: [EMAIL PROTECTED] | Homepage: http://www.joachim-breitner.de JID: [EMAIL PROTECTED] | GPG-Keyid: 4743206C | ICQ#: 74513189 Geekcode: GCS/IT/S d-- s++:- a--- C++ UL+++ P+++ !E W+++ N-- !W O? M?+ V? PS++ PE PGP++ t? 5? X- R+ tv- b++ DI+ D+ G e+* h! z? Bitte senden Sie mir keine Word- oder PowerPoint-Anhänge. Siehe http://www.fsf.org/philosophy/no-word-attachments.de.html signature.asc Description: Dies ist ein digital signierter Nachrichtenteil
Re: Proposed: Debian's Five Freedoms for Free Works
On Sun, 2003-06-15 at 06:05, Dylan Thurston wrote: In article [EMAIL PROTECTED], Adam Warner wrote: Branden, perhaps the term information disclosure would better suit you/us than privacy? That is we propose a DFSG-free licence cannot mandate information disclosure of anything but the information forming a distributed and derived work. But surely privacy is exactly when you have when information disclosure is not forced on you? Could you please elaborate on the difference? A guideline of privacy could be read as a positive obligation that DFSG-free software licences protect against information disclosure. My concern is that a guideline expressed in this way could impact upon no discrimination against fields of endeavour. No mandatory information disclosure is expressed as a negative. A licence can't mandate information disclosure beyond the source for a distributed and derived work. No obligation to protect privacy arises. To my mind privacy is too likely to tie the information test to the licence (e.g. is it good policy that a licence can demand my credit card number?). A prohibition against information disclosure ties the test to each distributed and derived work (e.g. if an author uses his credit card number as a key to unlock what would otherwise be copylefted information then the credit card number must be disclosed). Branden's fifth freedom can be read as a right because it is written in a way that mirrors the FSF's four fundamental freedoms of (as Branden writes) freedoms intended to apply to non-public-domain works in general. That is it is most relevant as a copyleft principle. When the principle is applied to DFSG-free software which also distributes public domain and permissively licensed software it has to be recast as a negative. That's what I have done by suggesting a test of no mandatory information disclosure beyond the information provided by the source of a distributed and derived work. Though I would go further to suggest that a copyleft software licence that mandated certain privacy protections (where all derived works had to be distributed under the same licence) would not be DFSG-free because it could discriminate against persons, groups or fields of endeavour. Consider a copyleft licence mandating that any software distributed under the licence cannot collect email addresses as a privacy protection against spam. We arrive at such thorny issues because privacy is given the opportunity to exist as a fundamental free software freedom. I can immediately think of two potential implications to recognising that a DFSG-free licence can't mandate information disclosure beyond the source for a distributed and derived work: * The APSL becomes clearly non-DFSG free (but it already clearly discriminates against fields of endeavour when determining whether something is considered internal or personal use): http://opensource.org/licenses/apsl.php 1.4 Deploy means to use, sublicense or distribute Covered Code other than for Your internal research and development (RD) and/or Personal Use, and includes without limitation, any and all internal use or distribution of Covered Code within Your business or organization except for RD use and/or Personal Use, as well as direct or indirect sublicensing or distribution of Covered Code by You to any third party in any form or manner. 2.2 You may use, reproduce, display, perform, modify and Deploy Covered Code, provided that in each instance: ... (c) You must make Source Code of all Your Deployed Modifications publicly available under the terms of this License, including the license grants set forth in Section 3 below, for as long as you Deploy the Covered Code or twelve (12) months from the date of initial Deployment, whichever is longer. You should preferably distribute the Source Code of Your Deployed Modifications electronically (e.g. download from a web site); and * Debian could be more reluctant to accept that any licences that attempt to close the so-called ASP (Application Service Provider) loophole are DFSG-free. With a no mandatory information disclosure guideline a licence could not force source disclosure without distribution. Though I wonder how far the issue of distribution of a derived work could already extend under existing copyleft licences. Say a program that also includes artwork is distributed under the GPL and one modifies the program, allows access to it via a website ASP model and also modifies the artwork and displays it on the website. As one is distributing a derivative of the artwork via the website could that also mean one is required to supply the source to the entire GPLed work that also includes the modified program? Regards, Adam
Re: Proposed: Debian's Five Freedoms for Free Works
On Fri, Jun 13, 2003 at 01:15:38AM +0200, Joachim Breitner wrote: Am Don, 2003-06-12 um 23.21 schrieb Branden Robinson: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. Isn't that effectively this lonely island test? I know it as the Desert Island test, but I think we're thinking of the same thing, yes. Since if it would be required to disclose any information, the lonely islander would not be able to use it legally. Right. And if I got it right, then the lonely island test has been applied to all Debian software (or at least to those in doubt), so one can actually hope that every piece of Debian software and data already confirms to your 5th Requirement for Freedom. Well, more or less. As we've learned over the years, there are sometimes surprises lurking in very old packages that have been in Debian for a long time. Debian as a Project is far, far more careful about licenses now than we were 9+ years ago. -- G. Branden Robinson| To stay young requires unceasing Debian GNU/Linux | cultivation of the ability to [EMAIL PROTECTED] | unlearn old falsehoods. http://people.debian.org/~branden/ | -- Robert Heinlein pgptNmBs5Lw0u.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
On Fri, Jun 13, 2003 at 04:01:54AM +0200, Joachim Breitner wrote: Not sure: Technically, for example, you can modify a program in any possible way just by having access to the assembler code that the compiler generates out of the closed sources, but this would be far too difficult to be realistic. That is why specifically the preferred form has to be available. But a clearer definition would be great, of course. See my reply to Mr. Suffield. I'm not sure that an exhaustive definition of Free Works that attempts to hit every corner case is the best approach to defining Free Works. -- G. Branden Robinson|Optimists believe we live in the Debian GNU/Linux |best of all possible worlds. [EMAIL PROTECTED] |Pessimists are afraid the optimists http://people.debian.org/~branden/ |are right about that. pgpKH9mzIYt2T.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
On Fri, Jun 13, 2003 at 01:10:23AM +0100, Andrew Suffield wrote: On Thu, Jun 12, 2003 at 04:21:35PM -0500, Branden Robinson wrote: 4) The freedom to change the Work for any purpose[1], to distribute one's changes, and to distribute the Work in modified form. Access to the form of the work which is preferred for making modifications, if applicable, is a precondition for this. I find the second sentence here to be prejudicial and inaccurate. Mostly it leads to debates over what the preferred form for modification is, much like we've had debates over what source code is. Well, the concept is borrowed straight from the GNU GPL... Firstly, it deals with preferences. The problem here is that different people have different preferences, and it is not inconceivable that they might prefer different forms for modification. Take a document as an example; do you prefer latex source, or a word document? Given your answer, would you contend that everybody shares this preference?[0] I would say that the controlling preference is that of the person who last modified the Work and distributed it in that modified form. Anyone downstream from that person would have to keep the source in that form and the binary together. Unfortunately I can see an easy way to abuse this: Malicious proprieteer A takes a Free Work and modifies its Source extensively. A then distributes the modified Work and Source to complicit agent B, who converts the Source into a less useful format and makes a trivial change. Agent B then distributes the modified Work along with the Source in the hobbled form he can -- with some legitimacy -- claim to be his preferred form for modifying the work. We could alter our requirements such that any form ever used for preferential modification be transmitted downstream to recipients, but there are pathological cases there. (If some loon writes something in INTERCAL and another person translates it to C, should the original INTERCAL source be inflicted on the whole world?) It may be that it is impossible to nail down the appropriate balance in exact language. Even a Definition of Free Software will likely not give one the mathematical certainty that the Open Source Initiative appears to seek. Secondly, it implicitly states, through use of the definite article, that there is only one such form. This is needlessly confusing, not to mention often wrong. Yes, you seem to have been thinking along the same lines as I was in my second pathological case above. I contemplated a few ways to rephrase it, but whenever I tried, I found myself arriving back at the first sentence again[1]. As such, I think it'd be best to remove the second one outright; the freedom is already adequetely described by the first. *Any* form which allows you to modify the work for any purpose, is good enough. You may be right. The Debian Project can apply tests (like the Desert Island or Chinese Dissident tests) to Works and their licenses that may not be appropriate for encapsulation in a definition of Free Works. In other words, it may not be *necessary* to ensure that a definition of Free Software obviously prohibits distribution of .i or .s files in lieu of commented source code. If institutions like Debian, the FSF, and OSI call bullshit on such practices, it may be enough. We should only attempt to play lawyer's games up to a point (that being the point where the costs outweigh the benefits). -- G. Branden Robinson|To Republicans, limited government Debian GNU/Linux |means not assisting people they [EMAIL PROTECTED] |would sooner see shoveled into mass http://people.debian.org/~branden/ |graves. -- Kenneth R. Kahn pgp01tFC7L85z.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
On Thu, Jun 12, 2003 at 08:54:17PM -0400, David B Harris wrote: On Thu, 12 Jun 2003 16:21:35 -0500 Branden Robinson [EMAIL PROTECTED] wrote: Comments? One thing I don't think that's entirely clear is about the labelling of your changes. The GPL specifies that you must put a notice in a given file detailing the date and nature of the changes. Such may or may not be considered part of the copyright notice, and I'd like to see point #3 amended to say that the license may require notices of modification within the source material should it be redistributed. Or however you want to work it in :) I thought of this when writing my little essay, believe it or not, but left the point unaddressed. Since you wasted no time in picking up the point, I'll go ahead and utter my heresy: I do not regard works licensed under the GNU GPL as purely free, either under the FSF's definition of Free Software, or my derived definition of Free Works. The GNU GPL makes certain, fairly limited, compromises with pure freedom[1] to achieve its ends. RMS has said stuff like this for years, so maybe it's not so heretical. Without (in my opinion) placing much emphasis on the point, I think he feels this way about the GNU FDL as well. He knows full well that it takes freedoms away, even in ways that the GNU GPL does not. While I'm uttering heresies I'll also say that I think clause one of our Social Contract is poorly worded[2]. It steers people towards the concept of 100% purity when what it really means is that we try to ensure that each and every package in our distribution satisfies our guidelines for Free Software. I think the 100% business can mislead people into thinking that each package is or can be some sort of paragon of virtuous freedom. [1] Yes, yes, some BSD bigots have been saying for years that the GNU GPL actually takes freedom away because it takes away one's freedom to charge a million bucks for a binary and keep the source code secret. I find this observation about as astute as the one that says libertarians and (most) anarchists don't really advocate freedom because they don't think people should be left free to coerce and defraud each other. Given that concept of freedom, the observation is perfectly true, perfectly accurate, and perfectly useless. Maybe BSD bigots are actually solipsists, and that is why they keep kicking people out of their Core Teams ;-). [2] Actually, I went on record three years ago or more as saying that other parts of the Social Contract are poorly worded as well, so there. If you're going to bash me, you're late to the party. -- G. Branden Robinson| Debian GNU/Linux | Extra territorium jus dicenti [EMAIL PROTECTED] | impune non paretur. http://people.debian.org/~branden/ | pgpfShtfwGeiN.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
Branden Robinson [EMAIL PROTECTED] wrote: I would say that the controlling preference is that of the person who last modified the Work and distributed it in that modified form. Anyone downstream from that person would have to keep the source in that form and the binary together. I think one formulation that makes this a bit more explicit is this: 4) The freedom to change the Work for any purpose, to distribute one's changes, and to distribute the Work in modified form. Unrestricted access to the form of the work which is preferred by its author for making modifications, if applicable, is a precondition for this. (with the possible addition of the words or translator after author). This makes explicit the fact that there is no single preferred form. If you allow individual authors to define their own preferred version, you solve problems like this -- Andrea writes a program in C. Marty rewrites Andrea's program in Fortran. Ulysses gets Marty's binary and asks for source -- Marty can send his Fortran source because it is the form Marty prefers for making modifications. Alternately, assume Marty lives on a proprietary island and only has access to proprietary programming languages. Marty should not be barred from translating Andrea's program into a proprietary language and distributing his modifications in that language (which is preferred because its the only thing he's got). I think any definition of preferred form needs to pass this test. In other words, I think that any definition of preferred form that requires an open or transparent format will be non-free. The same holds true for document formats of course. The person who aims to prepare a derivative work should have the option of using whatever form she prefers, and should have no obligation beyond the distribution of modifications in her preferred form. I am not sure at this point the extent to which certain exceptions need to be in place in the case where the author has some vested interest in selling you a proprietary interpreter (in the extreme case, pay me $100 for the AES key you need to decrypt my preferred form). Any thoughts? Greg
Re: Proposed: Debian's Five Freedoms for Free Works
Hi, Am Fre, 2003-06-13 um 05.41 schrieb Andrew Suffield: Not sure: Technically, for example, you can modify a program in any possible way just by having access to the assembler code that the compiler generates out of the closed sources, but this would be far too difficult to be realistic. That is why specifically the preferred form has to be available. But a clearer definition would be great, of course. Suppose the author is one of those nutcases that *likes* writing assembly code. Under a requirement such as you describe, all the code he wrote would be non-free, since nobody else wants to work in that form. If you try and clarify enough to make this case free, you find yourself with a null statement. Now, let's take it one step further. I postulate that there are numerous packages in the archive which are so poorly written, that modifying them for a range of useful purposes (including fixing some bugs), is too difficult to be realistic; assume this is true for a moment. Are they therefore to be considered non-free? What about this one (attention, not legal english, but I hope it is at least correct english): Unrestricted access to all not-common elements to produce the final product is a precondition for this. This would require to publish the code, the Makefiles, any unpublic compiler patches, maybe some UML files that are needed, while elemtents like make, gcc and similar do not have to be distributed by the author. What is common and what not is of course not defined, but as Branden already said, it will probably not be able to nail it down completely. This definition works also with Documents (non-common: The LaTeX file, common: the letter document LaTex style), Hardware (non-common: the blue prints; common the lithography machine and the silicon), Humans (non-common: the order of the 4 bases on the DNA string) :-) Joachim Breitner -- Joachim Breitner e-Mail: [EMAIL PROTECTED] | Homepage: http://www.joachim-breitner.de JID: [EMAIL PROTECTED] | GPG-Keyid: 4743206C | ICQ#: 74513189 Geekcode: GCS/IT/S d-- s++:- a--- C++ UL+++ P+++ !E W+++ N-- !W O? M?+ V? PS++ PE PGP++ t? 5? X- R+ tv- b++ DI+ D+ G e+* h! z? Bitte senden Sie mir keine Word- oder PowerPoint-Anhänge. Siehe http://www.fsf.org/philosophy/no-word-attachments.de.html signature.asc Description: Dies ist ein digital signierter Nachrichtenteil
Re: Proposed: Debian's Five Freedoms for Free Works
On Thu, Jun 12, 2003 at 11:00:51PM -0500, Branden Robinson wrote: On Fri, Jun 13, 2003 at 01:10:23AM +0100, Andrew Suffield wrote: On Thu, Jun 12, 2003 at 04:21:35PM -0500, Branden Robinson wrote: 4) The freedom to change the Work for any purpose[1], to distribute one's changes, and to distribute the Work in modified form. Access to the form of the work which is preferred for making modifications, if applicable, is a precondition for this. I find the second sentence here to be prejudicial and inaccurate. Mostly it leads to debates over what the preferred form for modification is, much like we've had debates over what source code is. Well, the concept is borrowed straight from the GNU GPL... Firstly, it deals with preferences. The problem here is that different people have different preferences, and it is not inconceivable that they might prefer different forms for modification. Take a document as an example; do you prefer latex source, or a word document? Given your answer, would you contend that everybody shares this preference?[0] I would say that the controlling preference is that of the person who last modified the Work and distributed it in that modified form. Anyone downstream from that person would have to keep the source in that form and the binary together. Unfortunately I can see an easy way to abuse this: Malicious proprieteer A takes a Free Work and modifies its Source extensively. A then distributes the modified Work and Source to complicit agent B, who converts the Source into a less useful format and makes a trivial change. Agent B then distributes the modified Work along with the Source in the hobbled form he can -- with some legitimacy -- claim to be his preferred form for modifying the work. Hmmm... Wouldn't distributing the modified Free Work, even if it's only distributed to B, require A to make available the modified Free Work to third parties? Then one could start from there, and utterly disregard Bs obfuscated version. I'm pretty sure that is the case with the GPL, not sure about other licenses. Cheers, Nick -- x--x |I maintain a Zero-Tolerance policy for| | Zero-Tolerance policy maintainers. | |--| | Nicolas Kratz [EMAIL PROTECTED] [EMAIL PROTECTED] | x--x pgpyXxajgHGdO.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
In article [EMAIL PROTECTED], Branden Robinson wrote: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. ... The point is that my usage of your Free Software does not entitle you to access to or any rights in my improvements to your software unless I distribute the Software back to you specifically. So do you intend to exclude the MPL with this? Peace, Dylan
Re: Proposed: Debian's Five Freedoms for Free Works
On 12.VI.2003 at 16:21 Branden Robinson wrote: The Free Software Foundation promulgates, and the Debian Project generally accepts, four essential freedoms as defining Free Software. The following is an enumeration of freedoms intended to apply to non-public-domain works in general. I'd like to mention here that FSF talks about free software and free documentation and not about free works. It is questionable whether we have to require these freedoms from works that are not software, nor documentation. For our Debian distribution the difference is not much important as we distribute only software and documentation. Nevertheless there is a great philosophical difference. Do we start promoting philosophical ideas that are not directly related to our own work? I doubt there will be any benefit if we start doing this. 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. Yes, we should require this from all software in Debian. But I don't think it is desirable to add this to the definition of free software (and free documentation) because of this: * If some software spies me and its license disallows me to remove the bad code in it, then this won't be a free software, because I am not allowed to modify and improve it. * If the software spies me, but its license permits me to remove the bad code, then this will be bad software, but free software. I am allowed to improve it. * Debian has its guidelines rather than its definition of free software. I don't think we have any interest in more confrontation with FSF and initiate a third movement in our community (as OSI already did). Some people in FSF are too sensitive about definitions. Anton Zinoviev
Re: Proposed: Debian's Five Freedoms for Free Works
1) The freedom to use the Work for any purpose. 2) The freedom adapt the Work to one's needs. Access to the form of the ^to work which is preferred for making modifications (for software, the source code), if applicable, is a precondition for this. 3) The freedom to redistribute copies of the Work. 4) The freedom to change the Work for any purpose[1], to distribute one's changes, and to distribute the Work in modified form. Access to the form of the work which is preferred for making modifications, if applicable, is a precondition for this. What's the difference between change ... for any purpose (#4) and adapt ... to one's needs (#2)? If they mean the same thing then one of them is superfluous. It they mean different things then the difference should be made clearer. Looking at the FSF original, I see that #2 is the freedom to study how the program works ... and adapt I think you should restore the word 'study', since that seems to be the essence of the second FSF freedom. Also, access to source is a precondition for studying a program, but not necessary for adapting a program. I can adapt Microsoft Windows to my needs without the source code. By the way, it would be better if you preserved the FSF's numbering for these freedoms: FSF numbers them from zero through three. The fifth freedom you add is addressed on the FSF's page, even though it is not enumerated as one of the Four Freedoms: You should also have the freedom to make modifications and use them privately in your own work or play, without even mentioning that they exist. So this raises the question: Why not just reference the FSF page? If you have quibbles with the FSF definition, you could submit patches to the FSF. Thanks -- Thomas
Re: Proposed: Debian's Five Freedoms for Free Works
* Anton Zinoviev ([EMAIL PROTECTED]) wrote: On 12.VI.2003 at 16:21 Branden Robinson wrote: The Free Software Foundation promulgates, and the Debian Project generally accepts, four essential freedoms as defining Free Software. The following is an enumeration of freedoms intended to apply to non-public-domain works in general. I'd like to mention here that FSF talks about free software and free documentation and not about free works. It is questionable whether we have to require these freedoms from works that are not software, nor documentation. For our Debian distribution the difference is not much important as we distribute only software and documentation. Nevertheless there is a great philosophical difference. Do we start promoting philosophical ideas that are not directly related to our own work? I doubt there will be any benefit if we start doing this. as the world progresses more and more into the realm of 'multimedia' (hehe) we will see, i think, that debian will be distributing works not describable as software or documentation: image files, music, animations, etc, etc, etc. requiring that any such works be free is a very large step to take, and one that i feel is worth taking. i've recently been reading http://www.dreamsongs.com/MobSoftware.html, which has several interesting points on similar lines (but is mostly OT for this discussion). iain -- wh33, y1p33 3tc. If sharing a thing in no way diminishes it, it is not rightly owned if it is not shared. -St. Augustine
Re: Proposed: Debian's Five Freedoms for Free Works
Branden Robinson [EMAIL PROTECTED] writes: I personally have advocated a fifth freedom: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. I think (though I'm not sure) that I agree with what you're trying to do, but I don't like using privacy as its basis[1]. Reasonable people can disagree, of course, but I think it's important to understand that privacy and the free flow of information are competing values, and the optimum is some point between either extreme that maximizes other social values. To give a concrete example, accurate attribution of changes (e.g., a changelog) is a good thing because is strengthens the social structures that keep Free Software working, yet it's clearly a limit on privacy, albeit relatively minor. I'd say that there are two ideas implicit in the desert island test, and neither of them are really about privacy: * The only two parties involved in the exchange of Free Software are the distributor and the distributee. Requirements that arbitrary third parties play a role can not be met if that third party isn't on the desert island. I think of this as symmetry. * That the distributee isn't required to give any sort of consideration to the distributor. If the distributee finds the software on a note in a bottle (or perhaps on a computer in the ship that crashed on the island) he should have all the necessary rights to the software. I think of this as not requiring consideration. I think both of these are somewhat controversial, though. [1] Much as I dislike the guy, I find I agree with Brin in that making privacy into a fundamental right is likely to be more damaging than helpful to society. -- Jeremy Hankins [EMAIL PROTECTED] PGP fingerprint: 748F 4D16 538E 75D6 8333 9E10 D212 B5ED 37D0 0A03
Re: Proposed: Debian's Five Freedoms for Free Works
* Jeremy Hankins ([EMAIL PROTECTED]) wrote: Branden Robinson [EMAIL PROTECTED] writes: I personally have advocated a fifth freedom: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. I think (though I'm not sure) that I agree with what you're trying to do, but I don't like using privacy as its basis[1]. Reasonable people can disagree, of course, but I think it's important to understand that privacy and the free flow of information are competing values, and the optimum is some point between either extreme that maximizes other social values. To give a concrete example, accurate attribution of changes (e.g., a changelog) is a good thing because is strengthens the social structures that keep Free Software working, yet it's clearly a limit on privacy, albeit relatively minor. it's not a limit on privacy, because it only applies when an individual chooses to distribute a derived work. iain -- wh33, y1p33 3tc. If sharing a thing in no way diminishes it, it is not rightly owned if it is not shared. -St. Augustine
Re: Proposed: Debian's Five Freedoms for Free Works
In article [EMAIL PROTECTED], Thomas Hood wrote: 1) The freedom to use the Work for any purpose. 2) The freedom adapt the Work to one's needs. Access to the form of the ^to work which is preferred for making modifications (for software, the source code), if applicable, is a precondition for this. 3) The freedom to redistribute copies of the Work. 4) The freedom to change the Work for any purpose[1], to distribute one's changes, and to distribute the Work in modified form. Access to the form of the work which is preferred for making modifications, if applicable, is a precondition for this. What's the difference between change ... for any purpose (#4) and adapt ... to one's needs (#2)? If they mean the same thing then one of them is superfluous. It they mean different things then the difference should be made clearer. One clear difference is that the FSF finds the FDL license to be free on their terms, since it can be adapted to fit any substantive modifications to the program (adapt to one's needs), while Branden (and many others from Debian) would reject the FDL, since there is text which is unmodifiable and unremovable (it cannot be changed for any purpose). I think this follows from the wording difference; I don't think it needs any clarification. (And I like the change.) On the other hand, it will certainly get confusing to have two slightly different but nearly identical lists of freedoms around. But maybe that's OK. Peace, Dylan, NADD
Re: Proposed: Debian's Five Freedoms for Free Works
* Jeremy Hankins ([EMAIL PROTECTED]) wrote: Having to include a changelog entry describing my modifications and (at minimum) that the original author didn't make the change is quite a bit different from simply giving some code to a friend w/o telling whether I even modified the code. One is giving a minimal amount of information to a single person I may trust, the other gives quite a bit more info to anyone who ultimately receives a copy of the code. Neither is really what I'd call a serious invasion of privacy, but it's a matter of degree, not kind. I don't really think it's relevant to privacy at all - you're putting something out, and there are requirements an what you must distribute with it. you could quite easily fake the changelog, make it up afterwards etc. after all, the difference is fairly trivial - running 'diff' provides a minimal changelog - it's more a question of politeness, letting people know what/why/how has been altered. iain -- wh33, y1p33 3tc. If sharing a thing in no way diminishes it, it is not rightly owned if it is not shared. -St. Augustine
Re: Proposed: Debian's Five Freedoms for Free Works
Nicolas Kratz said: Hmmm... Wouldn't distributing the modified Free Work, even if it's only distributed to B, require A to make available the modified Free Work to third parties? Then one could start from there, and utterly disregard Bs obfuscated version. I'm pretty sure that is the case with the GPL, not sure about other licenses. With the GPL, no, but keeping the sources secret takes a bit of care. A gives the GPL binaries to B. At the same time, A asks B would you like source with that? (offering to hand B a CD with the source, but not making a written offer that B could pass on) Now, if B does not have the sources, B can not distribute the binaries under the terms of the GPL. In order to (re-)distribute the binaries, B would have had to take the source CD, and make it available to distributees. If this is the only way that A has distributed the binaries, then A has no obligation to distribute the sources to anyone ever again, even if B distributes the binaries all over the internet (and A can sue B if this happens, for violating the GPL) --Joe
Re: Proposed: Debian's Five Freedoms for Free Works
I was mildly confused with Branden's response to my message, and I've been asked by two other people privately what the conclusion of the debate was, so I'll just summarise quickly here the discussion Branden and myself had on IRC. I checked with Branden, and he's perfectly happy with the summary below. Regardless of whether he agrees or disagrees with a change log requirement, it's not appropriate to the essay. The essay is not intended to go into specific applications and restrictions, but is rather an attempt to broadly define our freedoms such that we may apply its principles when making day-to-day decisions about licensing. See http://lists.debian.org/debian-legal/2003/debian-legal-200306/msg00111.html , specifically: It may be that it is impossible to nail down the appropriate balance in exact language. Even a Definition of Free Software will likely not give one the mathematical certainty that the Open Source Initiative appears to seek. snip We should only attempt to play lawyer's games up to a point (that being the point where the costs outweigh the benefits). Now that I understand the goal of the essay, I certainly agree with him :) Regardless of whether the specific application of the principles involved gives us a yea or a nay on the change log issue, it's not appropriate for the essay. However, some people (myself included) wanted to ensure that we understood Branden's position with respect to change log requirements, so I asked him and he kindly explained. Basically, he is not averse to a change log requirement - we both agree that it is a reasonable restriction of the public domain so to speak (the public domain being the most truly free form a work can be released as, the user being able to do absolutely anything they want with no obligations, responsibilities, or restrictions). Its utility (providing an audit trail, a history, and ensuring that those complying with the license don't release something under somebody else's name with all the problems involved in that) outweighs the fact that it's an added responsibility. However, he feels that the text within the GPL which deals with this issue is out of date - it requires the nature of the changes and the dates of the changes to be documented within the changed source file itself. This is very rarely done, and it has been replaced with the common practice of keeping a ChangeLog file of some form. It also doesn't take into account the revision control situation, where if you distribute a source tree via a revision control system, those change logs are typically available from the revision control system and don't necessarily need to be included in the downloaded data itself. I certainly agree there as well, and would feel a more generic paragraph would be more useful. Obviously the current one is worse than useless, since so many people aren't in compliance with the license. pgpKuF37TwAem.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
Greg Pomerantz said: Branden Robinson [EMAIL PROTECTED] wrote: I would say that the controlling preference is that of the person who last modified the Work and distributed it in that modified form. Anyone downstream from that person would have to keep the source in that form and the binary together. I think one formulation that makes this a bit more explicit is this: 4) The freedom to change the Work for any purpose, to distribute one's changes, and to distribute the Work in modified form. Unrestricted access to the form of the work which is preferred by its author for making modifications, if applicable, is a precondition for this. A problem with this is to define who is the author. If it's the original author, and he prefers InterCal, then nobody else can make changes except in Intercal. (with the possible addition of the words or translator after author). This makes explicit the fact that there is no single preferred form. If you allow individual authors to define their own preferred version, you solve problems like this -- If, on the other hand, you define it as the last person who touched it, then all it takes is an employee from EvilCorp(TM) to testify that they prefer to work with C source code that is obfuscated. Or someone to claim that they prefer to keep thier files encrypted. (snip) I think any definition of preferred form needs to pass this test. In other words, I think that any definition of preferred form that requires an open or transparent format will be non-free. The same holds true for document formats of course. The person who aims to prepare a derivative work should have the option of using whatever form she prefers, and should have no obligation beyond the distribution of modifications in her preferred form. I am not sure at this point the extent to which certain exceptions need to be in place in the case where the author has some vested interest in selling you a proprietary interpreter (in the extreme case, pay me $100 for the AES key you need to decrypt my preferred form). Any thoughts? The word preferred has problems that it is subjective, and it is not clear whose preference is relevant. open and transparent have problems in their definition. Would suitable or generally accepted as suitable be a good substitution? (All non-common source files must be distributed in a form [generally accepted as] suitable for modification) It would allow Free software to be written in C# or a language with a non-free toolchain. It would allow sound files to be distributed as either .au or .ogg, or even .mp3[0]. It would allow documents written with MS Word to be distributed as .doc files, though. --Joe [0] You can't distribute .mp3 files under the GPL because you can't distribute the mp3 encoder under the terms of the GPL, and the encoder is required to build the .mp3.
Re: Proposed: Debian's Five Freedoms for Free Works
On Fri, Jun 13, 2003 at 10:32:40AM +, Dylan Thurston wrote: In article [EMAIL PROTECTED], Branden Robinson wrote: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. ... The point is that my usage of your Free Software does not entitle you to access to or any rights in my improvements to your software unless I distribute the Software back to you specifically. So do you intend to exclude the MPL with this? If I were dictator of Debian's licensing policy, yes. However, I am but one voice among many. I think the MPL is an instructive case, given how Mozilla's own licensing has evolved. -- G. Branden Robinson| There's nothing an agnostic can't Debian GNU/Linux | do if he doesn't know whether he [EMAIL PROTECTED] | believes in it or not. http://people.debian.org/~branden/ | -- Graham Chapman pgpP1XYr26lwo.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
On Fri, Jun 13, 2003 at 03:29:03PM +0300, Anton Zinoviev wrote: I'd like to mention here that FSF talks about free software and free documentation and not about free works. Well, they're the Free *Software* Foundation. Presumably, they care first and foremost about software. It is questionable whether we have to require these freedoms from works that are not software, nor documentation. What's questionable about it? This sort of rhetoric is FUD. For our Debian distribution the difference is not much important as we distribute only software and documentation. Not true, unless you define software and documentation so broadly that you open yourself up to exactly the same sort of ridicule that I received for calling everything in Debian main effectively software. Things like the music files for frozen-bubble are neither software nor documentaiton. Nevertheless there is a great philosophical difference. This is an assertion for which you have provided no foundation. Do we start promoting philosophical ideas that are not directly related to our own work? I think that's more a question for those who seek to have us promulgate philosophical ideas of any stripe, e.g., RMS. I doubt there will be any benefit if we start doing this. One might object to the GNU FDL on precisely those grounds, and more importantly to specific manuals carrying the GNU Manifesto. 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. Yes, we should require this from all software in Debian. But I don't think it is desirable to add this to the definition of free software (and free documentation) because of this: * If some software spies me and its license disallows me to remove the bad code in it, then this won't be a free software, because I am not allowed to modify and improve it. * If the software spies me, but its license permits me to remove the bad code, then this will be bad software, but free software. I am allowed to improve it. You are thinking only about what the software can do, and not about what the *license* might do. * Debian has its guidelines rather than its definition of free software. I don't think we have any interest in more confrontation with FSF and initiate a third movement in our community (as OSI already did). Some people in FSF are too sensitive about definitions. I don't think it makes sense to treat the FSF as some sort of unstable psychotic who's likely to go off and start shooting people if his every whim is not sated. I expect the FSF to conduct itself rationally and ethically, and I cannot think of a situation where they have not done so. It is neither necessarily irrational nor necessarily unethical to disagree with the Debian Project about anything. I personally refuse to live in fear of what the FSF or RMS might think or say. I think the Debian Project would be well advised to share this perspective. -- G. Branden Robinson| That's the saving grace of humor: Debian GNU/Linux | if you fail, no one is laughing at [EMAIL PROTECTED] | you. http://people.debian.org/~branden/ | -- A. Whitney Brown pgpdQZKPNit9n.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
On Fri, Jun 13, 2003 at 09:15:26AM -0400, Jeremy Hankins wrote: Branden Robinson [EMAIL PROTECTED] writes: I personally have advocated a fifth freedom: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. I think (though I'm not sure) that I agree with what you're trying to do, but I don't like using privacy as its basis[1]. Reasonable people can disagree, of course, but I think it's important to understand that privacy and the free flow of information are competing values, and the optimum is some point between either extreme that maximizes other social values. I don't find your observation objectionable; I have been wrestling with a sound philosophical basis for my instinctive feelings on this subject. As long as the goal is met, I'm not particularly enamored of grounding freedom 5 on the concept of privacy. It probably shouldn't be grounded explicitly on *any* political principle, since people can have differing value systems, yet agree on this particular point. I do not seek to establish an orthodox politcal theory for the Debian Project, just a broad definition of freedom for the things we distribute, and upon which we can fall back when the DFSG fails to anticipate a problematic licensing issue. To give a concrete example, accurate attribution of changes (e.g., a changelog) is a good thing because is strengthens the social structures that keep Free Software working, yet it's clearly a limit on privacy, albeit relatively minor. Yes, David B. Harris raised the same point. I'd say that there are two ideas implicit in the desert island test, and neither of them are really about privacy: * The only two parties involved in the exchange of Free Software are the distributor and the distributee. Requirements that arbitrary third parties play a role can not be met if that third party isn't on the desert island. I think of this as symmetry. Yup. * That the distributee isn't required to give any sort of consideration to the distributor. If the distributee finds the software on a note in a bottle (or perhaps on a computer in the ship that crashed on the island) he should have all the necessary rights to the software. I think of this as not requiring consideration. Yes; I've been yammering for a couple of years at least about adding a No Consideration in Exchange for Rights Under this License clause to the DFSG. I think both of these are somewhat controversial, though. I hope not. Your points seem quite cogent to me, and, I think, reflect my personal instincts pretty clearly without dropping the politically explosive term privacy into the equation. With luck that's true for many other Debian developers as well. Thanks for elucidating this issue. [1] Much as I dislike the guy, I find I agree with Brin in that making privacy into a fundamental right is likely to be more damaging than helpful to society. I haven't read Brin, so I'll leave this remark to others. :) -- G. Branden Robinson| Debian GNU/Linux | If ignorance is bliss, [EMAIL PROTECTED] | is omniscience hell? http://people.debian.org/~branden/ | pgpHdi1eQ77gt.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
Nicolas Kratz [EMAIL PROTECTED] writes: On Thu, Jun 12, 2003 at 11:00:51PM -0500, Branden Robinson wrote: [...] I would say that the controlling preference is that of the person who last modified the Work and distributed it in that modified form. Anyone downstream from that person would have to keep the source in that form and the binary together. Unfortunately I can see an easy way to abuse this: Malicious proprieteer A takes a Free Work and modifies its Source extensively. A then distributes the modified Work and Source to complicit agent B, who converts the Source into a less useful format and makes a trivial change. Agent B then distributes the modified Work along with the Source in the hobbled form he can -- with some legitimacy -- claim to be his preferred form for modifying the work. Hmmm... Wouldn't distributing the modified Free Work, even if it's only distributed to B, require A to make available the modified Free Work to third parties? Then one could start from there, and utterly disregard Bs obfuscated version. I'm pretty sure that is the case with the GPL, not sure about other licenses. B would have irrevocable permission to transfer its rights to anyone it pleased (which is just what A wanted), but A would not necessarily have any obligation to do any further distributing, or, even if it did continue, to distribute to any particular person. The one exception is the theme of section 3 of the GPL: if A distributes a binary, A must either distribute the source alongside or must include an offer to distribute the source to any third party who requests it within 3 years; presumably others would find out about the offer from B if B decided to redistribute the binary non-commerically. But B needn't disclose this offer; B could intentionally make itself ineligible to transfer A's offer by conducting its own distribution commercially; and, even more simply, A could just give B the source in the first place. -- Gregory K. Johnson http://gkj.gregorykjohnson.com/
Re: Proposed: Debian's Five Freedoms for Free Works
David B Harris [EMAIL PROTECTED] writes: One thing I don't think that's entirely clear is about the labelling of your changes. The GPL specifies that you must put a notice in a given file detailing the date and nature of the changes. Such may or may not be considered part of the copyright notice, and I'd like to see point #3 amended to say that the license may require notices of modification within the source material should it be redistributed. Or however you want to work it in :) When I read You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change [GPL section 2(a)], I notice a conspicuous lack of any requirements of elaborate record keeping. I think that the required notice would look something like this: // Foomatic v1.4 - a command-line utility for doing a particular thing // Copyright 2002 Sum-Wun Elss // // Changed 5/10/2003 - 5/25/2003 by Gregory K. Johnson // Changes Copyright 2003 Gregory K. Johnson // // This program is free software; you can redistribute it and/or // [and so on] There's no explicit requirement that the change notice include an itemized or line-by-line record of changes or even that it describe the nature of the changes. From a privacy standpoint, there is not even a requirement that the author identify himself. The change notice could just as easily include these lines: // Changed 5/10/2003 - 5/25/2003 by Anonymous Hacker // Changes hereby placed into the public domain or // Changed 5/10/2003 - 5/25/2003 by Anonymous Hacker // Copyright hereby assigned to Sum-Wun Elss I think the GPL preamble explains the FSF's decision: If the software is modified by someone else and passed on, we want its recipients to know that what they have is not the original, so that any problems introduced by others will not reflect on the original authors' reputations. -- Gregory K. Johnson http://gkj.gregorykjohnson.com/
Re: Proposed: Debian's Five Freedoms for Free Works
Branden Robinson [EMAIL PROTECTED] writes: I personally have advocated a fifth freedom: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. I need to work on the wording of this fifth freedom a bit to make it clear that it is fair for a person to whom Free Software is distributed to demand access to the source code, including the source code to any changes or improvements made by the person from whom one is receiving the software. The point is that my usage of your Free Software does not entitle you to access to or any rights in my improvements to your software unless I distribute the Software back to you specifically. I added a phrase to the end that I think addresses these concerns: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others, except as these changes or Works are willfully shared. To those who would argue that this is redundant because it is implicit that the right to privacy can be waived: It is necessary to clarify, as Branden Robinson suggested above, that the right to privacy may be de facto abridged by engaging in distribution. Otherwise licenses like the GPL, which can compel distribution of source to *everyone* who requests it if you commerically distribute a binary-only copy to *anyone*, would come into conflict with the definition. But, as Branden also said, its important not to open the barn door to required universal distribution under normal circumstances. To those who would argue that the wording isn't blunt enough: In Merriam-Webster's Collegiate Dictionary, the applicable definition of the conjuction as is: in accordance with what or the way in which. Thus, the end of the fifth freedom could be paraphrased except in accordance with the way in which these changes and Works are willfully shared. In the same dictionary, willful means this: done deliberately: intentional. Note that by willfully distributing a GPL-licensed program one agrees to the compelled-distribution clause because it is, to use my proposed wording, in accordance with the way in which these changes and works are willfully shared. Similarly, one agrees to whatever requirements to offer source code an acceptable license imposes. To those who would argue the addition is tacky or obtrusive: I don't think it's either of those things, and it's certainly more attractive than potential alternatives that speak of waiving [the] freedom or complying with requirements or refusing to distribute. Moreover it has the three qualities that are important in a document like a defintion of free software: it is short, it is directly comprehended, and it is firm in meaning. As others have pointed out, making a Debian list of free software freedoms, with one freedom more than GNU's list, would be a bit confusing and a bit untidy, but hardly enough so to rule out the idea. I think that the most compelling reason to add privacy to the list is to clarify that Debian holds the right to make private modifications or anonymous public modifications to software in as high a regard as the other freedoms; I consider as simply a handy fringe benefit that the statement would also condemn so-called spyware. As others have pointed out, spyware can be viewed as simply flawed -- and the other freedoms facilitate fixing the flaws. And privacy as a human right is, I think, too political and nonspecific to the project for an official stance to be warranted. To Branden: Do you have plans to try to add such a document to the Web site's statements of policy? What procedural hurdles stand in the way of doing so? (I'm new to Debian, so please forgive my lack sophistication about the project's inner workings.) -- Gregory K. Johnson http://gkj.gregorykjohnson.com/
Re: Proposed: Debian's Five Freedoms for Free Works
On Thursday, Jun 12, 2003, at 20:10 US/Eastern, Andrew Suffield wrote: I contemplated a few ways to rephrase it, but whenever I tried, I found myself arriving back at the first sentence again[1]. As such, I think it'd be best to remove the second one outright; the freedom is already adequetely described by the first. *Any* form which allows you to modify the work for any purpose, is good enough. No it isn't. Otherwise compiled object code would be good enough. Or the output of dvips. Both can certainly be modified (done it myself), but a work under the MIT X11 license that comes in binary form only isn't free, IMO (unless it was hand-coded in a hex editor...)
Re: Proposed: Debian's Five Freedoms for Free Works
On Thursday, Jun 12, 2003, at 22:01 US/Eastern, Joachim Breitner wrote: Not sure: Technically, for example, you can modify a program in any possible way just by having access to the assembler code that the compiler generates out of the closed sources, but this would be far too difficult to be realistic. We need to consider that some works were written in machine code using a hex editor. For example, I believe Tom Pitman originally coded CompileIt! in a hex editor. I think what we're going at is that its not free to make it hard to modify something by withholding certain forms of it. If I distribute a postscript file and withhold the Τεχ[0] source I generated it from, that's not free. If I write my document in PostScript, and distribute that[1], that is free. I think preferred form is thus very much the right term. However, it has to be the preferred form of the last major author. I say of the last major author because I think its free if someone takes my hand-crafted LaTeX, imports it to Lyx, considerably modifies it there, and then distributes the .lyx file. I think it's free if he runs dvips on it and considerably edits the generated PostScript, and distributes that. I don't think it's free if he just changes one word. [0] That's tau epsilon chi for the Unicode challenged. [1] Assuming all the other criteria are met.
Re: Proposed: Debian's Five Freedoms for Free Works [humor]
On Friday, Jun 13, 2003, at 04:57 US/Eastern, Joachim Breitner wrote: Unrestricted access to all not-common elements to produce the final product is a precondition for this. [...] Humans (non-common: the order of the 4 bases on the DNA string) :-) Hmmm... sounds like you're required to distribute yourself with a work to a make it free software. Oooh, can I clone a Linus from the kernel sources? Or are they not free? ;-)
Re: Proposed: Debian's Five Freedoms for Free Works
In article [EMAIL PROTECTED], Gregory K.Johnson wrote: ... But B needn't disclose this offer; B could intentionally make itself ineligible to transfer A's offer by conducting its own distribution commercially; ... I'm not sure what you're getting at, but under the terms of the GPL, B is not allowed to distribute object code at all without fulfilling one of the conditions of clause 3. The one (minor) bug (in which some recipient of object code might not have full access to the source) is if A gives B the object code, with a written offer; B waits for nearly 3 years, and then passes on the object code together with the offer. But I have no idea what this is relevant to. Peace, Dylan
Proposed: Debian's Five Freedoms for Free Works
[Originally this was going to be a reply to the Lego Mindstorms SDK question, but it turned into an essay. Oh well. :) ] As Richard M. Stallman of the Free Software Foundation has been saying for twenty years or more, the Free in Free Software refers to freedom, not price. A great deal of Free Software is available to the general public free of charge, but that is not the essential characteristic of Free Software. The Free Software Foundation promulgates, and the Debian Project generally accepts, four essential freedoms as defining Free Software. The following is an enumeration of freedoms intended to apply to non-public-domain works in general. 1) The freedom to use the Work for any purpose. 2) The freedom adapt the Work to one's needs. Access to the form of the work which is preferred for making modifications (for software, the source code), if applicable, is a precondition for this. 3) The freedom to redistribute copies of the Work. 4) The freedom to change the Work for any purpose[1], to distribute one's changes, and to distribute the Work in modified form. Access to the form of the work which is preferred for making modifications, if applicable, is a precondition for this. (You can read more about the Free Software Foundation's definition of Free Software at: URL:http://www.gnu.org/philosophy/free-sw.html. You will note that my wording differs slightly from the Free Software Foundation's. This is deliberate.) I personally have advocated a fifth freedom: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. I need to work on the wording of this fifth freedom a bit to make it clear that it is fair for a person to whom Free Software is distributed to demand access to the source code, including the source code to any changes or improvements made by the person from whom one is receiving the software. The point is that my usage of your Free Software does not entitle you to access to or any rights in my improvements to your software unless I distribute the Software back to you specifically. Other consequences of my proposed fifth freedom are that a Free Software licensor has no right to insist that a person to whom software is distributed disclose any more information about him- or herself than is strictly necessary for processing of the transaction. For example, a Free Software licensor cannot insists that a distributee disclose his credit rating (or compel a grant of permission to find out, by running a credit check), that a work of Free Software retain code that scans the contents of one's hard drive and reports on its findings to the author of the software, to a third party, or even to the user him- or herself, or that a Free book must retain a foreword which calls for the extermination of Lendu people. Comments? [1] Except the eradication of legal notices necessary to communicate and preserve the legal status of the software. This means applicable copyright notices, license terms, references to license terms, warranty disclaimers, and so forth. Freedom four *does* include the freedom to remove or change such material where it is incorrect or inapplicable, and add correct and applicable material of this nature. -- G. Branden Robinson|It is the responsibility of Debian GNU/Linux |intellectuals to tell the truth and [EMAIL PROTECTED] |expose lies. http://people.debian.org/~branden/ |-- Noam Chomsky pgpN15CzqnU4s.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
Branden said: snip Comments? Well, I love it. :-) --Nathanael
Re: Proposed: Debian's Five Freedoms for Free Works
Hi, Am Don, 2003-06-12 um 23.21 schrieb Branden Robinson: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. Isn't that effectively this lonely island test? Since if it would be required to disclose any information, the lonely islander would not be able to use it legally. And if I got it right, then the lonely island test has been applied to all Debian software (or at least to those in doubt), so one can actually hope that every piece of Debian software and data already confirms to your 5th Requirement for Freedom. Besides that, I fully support that proposal, since I value privacy very high Joachim Breitner Debian Developer to be :-) -- Joachim Breitner e-Mail: [EMAIL PROTECTED] | Homepage: http://www.joachim-breitner.de JID: [EMAIL PROTECTED] | GPG-Keyid: 4743206C | ICQ#: 74513189 Geekcode: GCS/IT/S d-- s++:- a--- C++ UL+++ P+++ !E W+++ N-- !W O? M?+ V? PS++ PE PGP++ t? 5? X- R+ tv- b++ DI+ D+ G e+* h! z? Bitte senden Sie mir keine Word- oder PowerPoint-Anhänge. Siehe http://www.fsf.org/philosophy/no-word-attachments.de.html signature.asc Description: Dies ist ein digital signierter Nachrichtenteil
Re: Proposed: Debian's Five Freedoms for Free Works
On Thu, Jun 12, 2003 at 04:21:35PM -0500, Branden Robinson wrote: 4) The freedom to change the Work for any purpose[1], to distribute one's changes, and to distribute the Work in modified form. Access to the form of the work which is preferred for making modifications, if applicable, is a precondition for this. I find the second sentence here to be prejudicial and inaccurate. Mostly it leads to debates over what the preferred form for modification is, much like we've had debates over what source code is. Firstly, it deals with preferences. The problem here is that different people have different preferences, and it is not inconceivable that they might prefer different forms for modification. Take a document as an example; do you prefer latex source, or a word document? Given your answer, would you contend that everybody shares this preference?[0] Secondly, it implicitly states, through use of the definite article, that there is only one such form. This is needlessly confusing, not to mention often wrong. I contemplated a few ways to rephrase it, but whenever I tried, I found myself arriving back at the first sentence again[1]. As such, I think it'd be best to remove the second one outright; the freedom is already adequetely described by the first. *Any* form which allows you to modify the work for any purpose, is good enough. [0] I can make many more arguments along these lines; in the name of brevity, I will refrain from doing so at this time. [1] Access to any form of the work that allows you to change it for any... oh, damn -- .''`. ** Debian GNU/Linux ** | Andrew Suffield : :' : http://www.debian.org/ | Dept. of Computing, `. `' | Imperial College, `- -- | London, UK pgpX9WJdnvxmr.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
On 13 Jun 2003 01:15:38 +0200 Joachim Breitner [EMAIL PROTECTED] wrote: 5) The freedom to retain privacy in one's person, effects, and data, including, but not limited to, all Works in one's possession and one's own changes to Works written by others. Isn't that effectively this lonely island test? Since if it would be required to disclose any information, the lonely islander would not be able to use it legally. And if I got it right, then the lonely island test has been applied to all Debian software (or at least to those in doubt), so one can actually hope that every piece of Debian software and data already confirms to your 5th Requirement for Freedom. Besides that, I fully support that proposal, since I value privacy very high Correct. What Branden is saying (among other things) is that the license should not require you to return changes to the author; thus any such license passes the lonely island test pgpuCqs2sidLo.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
On Thu, 12 Jun 2003 16:21:35 -0500 Branden Robinson [EMAIL PROTECTED] wrote: Comments? One thing I don't think that's entirely clear is about the labelling of your changes. The GPL specifies that you must put a notice in a given file detailing the date and nature of the changes. Such may or may not be considered part of the copyright notice, and I'd like to see point #3 amended to say that the license may require notices of modification within the source material should it be redistributed. Or however you want to work it in :) pgpwKQbyXlBM1.pgp Description: PGP signature
Re: Proposed: Debian's Five Freedoms for Free Works
On Fri, 13 Jun 2003 01:10:23 +0100 Andrew Suffield [EMAIL PROTECTED] wrote: On Thu, Jun 12, 2003 at 04:21:35PM -0500, Branden Robinson wrote: 4) The freedom to change the Work for any purpose[1], to distribute one's changes, and to distribute the Work in modified form. Access to the form of the work which is preferred for making modifications, if applicable, is a precondition for this. I contemplated a few ways to rephrase it, but whenever I tried, I found myself arriving back at the first sentence again[1]. As such, I think it'd be best to remove the second one outright; the freedom is already adequetely described by the first. *Any* form which allows you to modify the work for any purpose, is good enough. There are all sorts of tools out there that patch binaries, most of which may be nefarious; however, it does allow you to modify the work for any purpose. It's just obscenely difficult to do so for any but the most trivial of changes. How about: 4) The freedome to change the Work for any purpose, to distribute one's changes, and to distribute the Work in modified form. Access to the form of the Work in which the original author uses for making changes (if applicable) is a precondition for this. That'd get all realistic definitions of source, and will stop people from saying I want it in LaTeX, I don't care if you use plaintext files. Some rephrasing I think is still necessary; I don't like the original in there, but without it somebody might think the author is the person who's requesting the source. pgpxyQooyi15N.pgp Description: PGP signature