Re: The Show So Far
Anthony Towns aj@azure.humbug.org.au writes: then you have to say which restrictions you think are acceptible and which you think aren't. I've sketched out my method of analyzing such a question, but you haven't. Would you care to do so please? No, I'm sorry; I reserve my right to do so on a case-by-case basis. I can answer for the cases we already have though: I don't think the freedom to distribute binaries without source is a key freedom, nor do I think being able to make changes that you can keep private forever is a key freedom. I don't think removing those freedoms should be forbidden from free licenses. I do think both freedoms have value in some circumstances. I don't think it's particularly onerous working around not having those freedoms. I don't think hiding source code is in anyone's long term interests, either companies or activists. I've given a specific case. Can you articulate why a you must give me your tax return if possible, and costs are paid requirement is not acceptible? You've said it isn't, but it clearly meets the DFSG. So, now the question is: what principle would you invoke to explain why that's not an acceptible condition? I have explained what principle I would invoke... Have you?
Re: Dissident versus ASP
Anthony Towns aj@azure.humbug.org.au writes: On Mon, Mar 17, 2003 at 04:31:48PM -0800, Thomas Bushnell, BSG wrote: Anthony Towns aj@azure.humbug.org.au writes: The claim is that: Dissident test + Practical objections == Can't close the ASP loophole and, furthermore that that equality goes both ways. That is that the Dissident test is just another way of saying that the only ways you're allowed to close the ASP loophole are ones which are practically unreasonable. Sorry, when I say I agree with this claim, I mean the = direction. The = direction is surely false; for example, I formulated the dissident test long before the ASP loophole (of any sort) had ever been pointed out to me. Huh? I never thought about a number bigger than 10^128, therefore a^b * a^c = a^(b+c) is surely false for numbers that large You're not making any sense. Because it's logical equivalence, not numerical equality. That is: I agree that the dissident test + the practical objections imply that there is probably no way to close any of the things called the ASP loophole. I do not agree with the claim that the dissident test is just another way of saying that the only ways your allowed to close the ASP loophole are ones which are practically unreasonable. Rather, the dissident test arises out of a serious attempt to explain why certain kinds of conditions (including, for example, the tax return provision) are not acceptible in a free software license. I certainly did not original compose the dissident test because I was worried about the ASP loophole. I know that I mentioned this test on December 16, which is before I ever heard of the ASP loophole. Moreover, the = direction suggests that the only reason for the dissident test would be to keep the ASP loophole open. I don't think so; there are plenty of other reasons for caring about that test. And, there might well be licenses which fail the dissident test, but are not ASP loophole closing licenses.
[no subject]
CRACK HERE
Re: The Show So Far
On Mon, Mar 17, 2003 at 10:21:13PM -0800, Thomas Bushnell, BSG wrote: Anthony Towns aj@azure.humbug.org.au writes: No, I'm sorry; I reserve my right to do so on a case-by-case basis. I've given a specific case. Can you articulate why a you must give me your tax return if possible, and costs are paid requirement is not acceptible? Sure; it's a plainly stupid idea. No one's seriously advocating it, and it doesn't benefit anyone. Please at least come up with examples that are vaguely _plausible_. If you want that formulated as a principle, as though that makes it somehow better, I've already said: ] Sending your tax return, or your latest entries ] in your diary, or whatever, to someone random and sending your changes ] to some program to its author aren't comparable. One's never sensible ] or reasonable, the latter's a good thing that we'd want to encourage ] independent of whether it's required by the license. Which is to say: sending your tax return to someone when you change a program is not a reasonable thing to do. As such, it's not a reasonable thing for a license to require you to do. Cheers, aj -- Anthony Towns [EMAIL PROTECTED] http://azure.humbug.org.au/~aj/ I don't speak for anyone save myself. GPG signed mail preferred. ``Dear Anthony Towns: [...] Congratulations -- you are now certified as a Red Hat Certified Engineer!'' pgpJo15ZXHsfr.pgp Description: PGP signature
Re: Dissident versus ASP
On Mon, Mar 17, 2003 at 10:26:47PM -0800, Thomas Bushnell, BSG wrote: Dissident test + Practical objections == Can't close the ASP loophole You're not making any sense. Because it's logical equivalence, not numerical equality. a^b * a^c = a^(b+c) is true, but a^b * a^c = a^(b+c) isn't true? Or are the rules different for ideas that conform to Peano's axioms, than ideas that prefer modus ponens? I agree that the dissident test + the practical objections imply that there is probably no way to close any of the things called the ASP loophole. No, you don't agree because that's not what I'm saying. The whole point of this is to get rid of probablys and work out what is _actually the case_. I do not agree with the claim that the dissident test is just another way of saying that the only ways your allowed to close the ASP loophole are ones which are practically unreasonable. That's nice. Disprove it. Unfortunately it looks like you don't have enough of a handle on logical argument to even understand what that means. Yeesh. Rather, the dissident test arises [...] It doesn't matter how it arises. It matters what it *implies*. In particular, that it limits the ways of solving the ASP loophole through licensing to those that are technically objectionable. Moreover, the = direction suggests that the only reason for the dissident test would be to keep the ASP loophole open. Nonsense. There was _no_ value judgement implied, and the only conclusion to be drawn is that *dropping* the dissident test is one of exactly two ways of allowing people to *close* the ASP loophole. Cheers, aj -- Anthony Towns [EMAIL PROTECTED] http://azure.humbug.org.au/~aj/ I don't speak for anyone save myself. GPG signed mail preferred. ``Dear Anthony Towns: [...] Congratulations -- you are now certified as a Red Hat Certified Engineer!'' pgpSx8UhPiqB8.pgp Description: PGP signature
Re: Dissident versus ASP
Scripsit Jakob Bohm [EMAIL PROTECTED] Here is one hypothetical text (TINLA, IANAL, IANADD): If you offer to one or more parties the service of running the Program in exchange for a monetary fee or other significant consideration, and the act of running the Program is in essence the service charged for not merely a means of providing another service, the parties who have actually paid for that service receive the right to obtain a copy of The Program as follows. That's a use restriction and will never be free in my opinion. Use restrictions are non-free. Period. -- Henning MakholmKurt er den eneste jeg kender der er *dum* nok til at gå i *ring* på et jernbanespor.
Re: Dissident versus ASP
On Tue, Mar 18, 2003 at 10:34:35AM +1000, Anthony Towns wrote: Huh? It seems meaningless to me: if you employ some people to work on your program, you put them under NDA so that they agree not to disclose the source code; if you work with other groups, you do likewise to them. The license would forbid this, just like the GPL's clause 6 would. If necessary, you do the NDAing at arm's length, something like: A changes the program E employs B under a contract that they don't distribute the program or its source, etc E asks A to give B a copy of the program A gives B a copy of the program This would leave A free to distribute the modifications. E isn't covered by the program's license since he never has anything to do with it. I don't think it would be remotely reasonable or enforcable (or in line with giving people more free software) to somehow stop A from giving the program to B, or to stop B from being able to give copies to people who E approves of. If the program were GPLed, it _would_ stop B from being able to give copies to anyone unless the copies were unrestricted. This doesn't seem to be a problem in practice, though IIRC it collided with US export restrictions at some point. I think the end result of your scenario is that only B is restricted. Richard Braakman
Re: Dissident versus ASP
On Tue, Mar 18, 2003 at 12:06:12PM +0200, Richard Braakman wrote: If necessary, you do the NDAing at arm's length, something like: A changes the program E employs B under a contract that they don't distribute the program or its source, etc E asks A to give B a copy of the program A gives B a copy of the program This would leave A free to distribute the modifications. Sorry, it would leave A free to distribute the original program, but the point is that it would leave B with a copy of the program to work on (presumably under E's direction), but unable to give copies of it to anyone else whenever he likes. You might consider E to be employing both A and B (and thus that A is already under similar restrictions). If the program were GPLed, it _would_ stop B from being able to give copies to anyone unless the copies were unrestricted. No, the GPL's satisfied here: ] 6. Each time you redistribute the Program (or any work based on the ] Program), the recipient automatically receives a license from the ] original licensor to copy, distribute or modify the Program subject to ] these terms and conditions. You may not impose any further ] restrictions on the recipients' exercise of the rights granted herein. ] You are not responsible for enforcing compliance by third parties to ] this License. The only person redistributing the program is A. The person imposing further restrictions is E -- not You. I think the end result of your scenario is that only B is restricted. That's the point though: you can end up stopping all your employees from distributing the code through fairly standard employment contract terms, and you're done - the ASP loophole's open again, and you can make your modifications and stick them in your CGI directory. If you want to generalise it, think of: A = www.debian.org B, C, D = employees E = agent for employer Each employee can download from A happily; each is under NDA with E, and E never downloads anything so isn't bound by the license. B, C and D can share their changes quite happily if they pass them around if diff --ed form. This is the harder way of doing it, of course; the easiest way is to say that it's not the employees who own the copy of the program but the company, and that all the work the employees do is a work for hire and copyright is owned by the company; then there's no question of the employees being able to distribute copies -- they don't own the copy they have so they have no rights at all -- cf the various sourceless beta tests of GPLed software that people have gotten away with. Cheers, aj -- Anthony Towns [EMAIL PROTECTED] http://azure.humbug.org.au/~aj/ I don't speak for anyone save myself. GPG signed mail preferred. ``Dear Anthony Towns: [...] Congratulations -- you are now certified as a Red Hat Certified Engineer!'' pgppf4wJVVaeX.pgp Description: PGP signature
Re: the FSF's definition of Free Software and its value for Debian
On Mon, Mar 17, 2003 at 02:30:26PM +0100, Henning Makholm wrote: Scripsit Branden Robinson [EMAIL PROTECTED] On Sun, Mar 16, 2003 at 03:03:02PM +0100, Henning Makholm wrote: I think it is counterintuitive to read the directly or indirectly as a restrictive phrasing. On the contrary, it is meant to be inclusive, pointing out explicitly that the rights granted can *not* be restricted to *direct* recipients only. I don't see what's unclear, ambiguous, or inefficient about saying the recipient and all third parties. Sure. It just looked (to me; I may be dense) as if your objection was material rather than just related to a poor choice of words. In my opinion, clarity is a virtue. My objection becomes material with the first licensor who decides to resolve the license's ambiguous wording in an unexpected or undesirable manner. -- G. Branden Robinson|The basic test of freedom is Debian GNU/Linux |perhaps less in what we are free to [EMAIL PROTECTED] |do than in what we are free not to http://people.debian.org/~branden/ |do. -- Eric Hoffer pgpRYYxnk7Y6k.pgp Description: PGP signature
Re: the FSF's definition of Free Software and its value for Debian
On Mon, Mar 17, 2003 at 09:29:32AM -0500, Jeremy Hankins wrote: Branden Robinson [EMAIL PROTECTED] writes: I don't see what's unclear, ambiguous, or inefficient about saying the recipient and all third parties. ^ in posesion of the (modified) software, right? Otherwise it can sound like source must be available to everybody again. Yes, your correction seems apropos. So, my proposal for the wording of this clause is: Each time you distribute the Document (or any work based on the Document), you grant to the recipient and all third parties in possession of the Document the authority to gain access to the work by descrambling the work if it is scrambled, decrypting the work if it is encrypted, and otherwise avoiding, bypassing, removing, deactivating, or impairing any and all technological measures effectively controlling access to the work. -- G. Branden Robinson|Somebody once asked me if I thought Debian GNU/Linux |sex was dirty. I said, It is if [EMAIL PROTECTED] |you're doing it right. http://people.debian.org/~branden/ |-- Woody Allen pgp00J9A3DsQA.pgp Description: PGP signature
Re: QPL clause 3 is not DFSG-free
On Mon, Mar 17, 2003 at 02:33:47AM -0500, Branden Robinson wrote: Now, maybe the latter is what Trolltech *means*, but it's not what the license *says*. When we've got representatives of the FSF asserting that there is no fair use right to private modification because of the _Texaco_ case, I'd say the safe bet is to: 1) get clarification from the licensor; and 2) get the license clarified and ensure that it respects the fair use right to private modification. The distinction between these two, in case anyone doesn't understand, is that 1) refers to a specific copyright holder's interpretation of the license. 2) fixes most or all instances of the license to ensure that it is not creatively interpreted in a way that abrogates Fair Use rights. -- G. Branden Robinson| Debian GNU/Linux | If existence exists, [EMAIL PROTECTED] | why create a creator? http://people.debian.org/~branden/ | pgpnFAnmCxX3C.pgp Description: PGP signature
Re: Dissident versus ASP
Bernhard R. Link [EMAIL PROTECTED] writes: * Jeremy Hankins [EMAIL PROTECTED] [030317 17:31]: Folks who are providing an ASP-style service generally are going to have big web servers and lots of bandwidth anyway; I'm not convinced that distribution of source would be a significant burden for them. Note the generally. You are aware that you imply, that only such persons should have the possibility to do so? I'm saying that in the most typical case, these folks will have big servers big pipes. I'm certainly not saying that it's ideal that only folks with big servers and big pipes be able to provide ASP services. Am I misunderstanding you? I'm simply trying to divide the problem space up a bit. The other two scenarios described situations where ASP-providers didn't have big servers or big pipes. What if some small individual wants to start a server doing decss. Not sitting in the USA another place under the jurisdiction of the large media-industry, there might be nothing doable against him. If he has to expose the source, he would thereby importing a circumvention device to the USA, thus eliminating any chance to visit the US without beeing convicted. Fine, in this hypothetical if he's unable to provide the source to folks in the US, the license would not allow him to provide the service to folks in the US. Exactly analogous to someone trying to distribute deCSS binaries. Exactly what hurdle he would have to place in front of US users is up to him (and his lawyer). What I merely want to say by this, is that distribution is not only a technical burden, but more. (Running a slightly modified webserver that sends out his sourcecode in Germany may expose one to liability for example. With the current GPL one can look at the modifications some months later and test them a bit better to get rid of any problems with that) I'm certainly not familiar with German law (I'm not even really familiar with US law). But does this same liability apply if you make changes to a CVS repository? -- Jeremy Hankins [EMAIL PROTECTED] PGP fingerprint: 748F 4D16 538E 75D6 8333 9E10 D212 B5ED 37D0 0A03
Re: Dissident versus ASP
[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes: Jeremy Hankins [EMAIL PROTECTED] writes: Folks who are providing an ASP-style service generally are going to have big web servers and lots of bandwidth anyway; I'm not convinced that distribution of source would be a significant burden for them. But the proposals for closing the loophole apply not only to the big nasty people with the wide pipes, but also to the people, who, say, stripped out the source-downloading feature because they are over a satellite link to Antartica, and the changes were published anyhow in the lab back in Chicago. Yes. Which is why I didn't end my message with the bit you quoted, and went on to talk about such folks. And for what it's worth, I've never advocated the Affero bit (i.e., the quine-like functionality). I think that the only way the requisite flexibility can be maintained would be to describe the need (and permissible mechanisms) to provide source in the license text. -- Jeremy Hankins [EMAIL PROTECTED] PGP fingerprint: 748F 4D16 538E 75D6 8333 9E10 D212 B5ED 37D0 0A03
Re: The Show So Far
[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes: Jeremy Hankins [EMAIL PROTECTED] writes: * Software is a social artifact with significant social consequences, and therefore ought to be responsive to social pressures (i.e., not just individuals). [...] My favorite is the first, which is why I think freedoms should attach to use. I'm willing to take this disagreement as fundamental, though (which for the current purposes means we'll argue it out if we're ever sitting together over a beer, but probably not 'till then). So this is a different sort of argument, and calls for a different response. It's not about what makes a license a free software license, but more fundamentally, about how software ought to work. First of all, I must have radically misunderstood the message I was replying to above. I was simply pointing out that there are more ways than one to argue toward Free Software -- and certainly other ways than your artificiality of copyright argument. My attitude is that we should acknowledge that and not try too hard to settle on a particular grounding for Free Software. Since we are presuming free speech, and a broadly free-software consensus, we aren't going to tolerate laws that *prohibit* publication. So the only question is: should we have a law that *requires* publication? I'm not sure you're arguing to the point here. If you are, I'm misunderstanding you. How is this an argument for attaching freedoms to copies rather than use? -- Jeremy Hankins [EMAIL PROTECTED] PGP fingerprint: 748F 4D16 538E 75D6 8333 9E10 D212 B5ED 37D0 0A03
Re: The Show So Far
On Mon, Mar 17, 2003 at 07:19:48PM +1000, Anthony Towns wrote: On Sun, Mar 16, 2003 at 08:01:33PM -0800, Thomas Bushnell, BSG wrote: The fundamental premise of free software is that copyright is an artificial limitation on what I can do whit a piece of software, and that I should be able to modify it and copy it. I don't think so; the fundamental premise of free software is: * The freedom to run the program, for any purpose * The freedom to study how the program works, and adapt it to your needs * The freedom to redistribute copies so you can help your neighbor * The freedom to improve the program Historically, the only way you could access a program at all was to possess a copy of it, so it made sense to worry about how you could possess a copy, but not have those freedoms. These days, people use software they don't possess every day, in pretty significant ways. You guys are talking past each other. Copyright law traditionally restricts at least some of the activities you (Anthony) describe, and has done so since before the days of von Neumann. Media cartels are applying greater and greater pressure to ensure that copyright laws in fact restrict *all* of the activities you enumerated. Therefore, I cannot perceive your statement as anything but an elaboration of Thomas's, rather than the contradiction your posit it to be. -- G. Branden Robinson|A celibate clergy is an especially Debian GNU/Linux |good idea, because it tends to [EMAIL PROTECTED] |suppress any hereditary propensity http://people.debian.org/~branden/ |toward fanaticism.-- Carl Sagan pgpKtQeHrMzjt.pgp Description: PGP signature
Re: GPLv3 / Affero / RPSL
On Mon, Mar 17, 2003 at 09:44:42AM -0600, John Goerzen wrote: On Mon, Mar 17, 2003 at 02:48:52AM -0500, Branden Robinson wrote: As you well know, the role of spokesman for Debian was arrogated by Joseph Carter, who failed to, as I recall, accurately convey to TrollTech the concerns about the QPL being expressed at the time. In fact, so far as I know we have no record of his correspondence with them on this issue. No disputes there; however, it was Debian's responsibility to ensure that our spokesman -- official or not -- was accurately reflecting our concerns and communicating back do us. Failing that, Debian should have replaced him -- or at least had some other people participate in the discussions. So I don't think it's entirely a case of being able to pin it all on Joseph; the rest of us should have seen it coming and done something about it. (I include myself in that group) Agreed; I was not attempting to imply that one developer's soiling of the carpet absolved the rest of us from the responsibility of cleaning it up. But at the same time, the primary responsibility for the mess *does* lie with he who made it. :) They were big enough to admit that they erred with Qt's licensing initially, and they did not lose the respect of the community. Indeed, I think they did lose a lot of respect, and people continue to harbor ill will. I see that loss of respect as stemming from the original licensing decision, not the later one where they dual-licensed under the GPL. In one surprisingly candid post[1] from Miguel de Icaza in response to a Petrely article critizing Gnome that ran on Slashdot, Miguel stated: Those with long-term visions believe strongly that the foundation for building applications on Linux should be royalty free so Gnome is a good choice there. ... Some people want gnome because it makes sense license-wise (Red Hat and Sun seem to be concerned about *this* particular issue). [1] http://mail.gnome.org/archives/desktop-devel-list/2003-March/msg00026.html In reviewing this message, this seems to me an attack not so much on TrollTech in particular so much as an attack on the practice of licensing libraries under the GPL instead of the LGPL. At least, I cannot understand how Miguel is constraining his argument to TrollTech, except by way of a convenient example. There's nothing he says on this subject that can't be held to apply to any other case of a popular library that is licensed under the GPL but not the LGPL. I am not sure Debian should join in with that particular bit of disapprobation, therefore I suggest we disregard it. Now, obviously I don't agree that a GPL license is a problem or imposing royalties on anyone (where THAT came from is anyone's guess). As I said, he's contrasting the GPL with the LGPL. Miguel, if I understand him correctly, is asserting that it's a feature of GNOME that one can make closed, proprietary applications based on its toolkit without having to pay royalties. The same is not true of Qt. But if someone like Miguel misunderstands, I don't think that Troll has fully removed the stigma from their actions yet. I am unconvinced of this. I continue to think that the stigma arose from the original Qt licensing decision which motivated the GNOME project in the first place. The distinctions between the GPL and LGPL are far more subtle from a social standpoint, since they're both almost universally accepted as Free Software licenses. Basically, I think Miguel is now criticizing TrollTech on strategic, rather than moral, grounds.[1] I could be wrong about that, but if I am, then I perceive an ideological split between Miguel and the FSF, which views use of the LGPL as a compromised choice, not the preferred one. So why cannot we be big enough to admit that, if we have officially declared the QPL a DFSG-free license (a declaration of which I can find no evidence), that we erred, and that we recommend people follow TrollTech's example and dual-license their works under the QPL and GPL (with the footnote that many other DFSG-free licensing arrangements are possible, and they should feel free to contact us)? I must admit, I'm beginning to see it your way :-) Do we have any idea how much and what software is licensed under QPL only, with no dual-license provision? The only instance I know of at present is the Objective-CAML compiler. A few years ago the Zend engine (part of Zope) was QPL'ed but I don't know if that's still the case. I seem to remember some talk about PHP 4.x being QPLed as well. I do not know that dual-licensing with GPL has been rejected in any of these cases. [1] And, of course, is pointing out how GNOME's licensing makes it more suited as a foundation for proprietary apps. -- G. Branden Robinson| The Rehnquist Court has never Debian GNU/Linux | encountered a criminal statute it [EMAIL PROTECTED]
Re: The Show So Far
[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes: Jeremy Hankins [EMAIL PROTECTED] writes: But despite the above I do want to point out that the argument about the only thing stopping the possessor can easily (and, IMHO, more justifiably) be used against the GPL and in favor of BSD-style licensing. Simply s/possessor/possessor of source/ to see what I mean. No, and this is an important point. The BSD-license does not restrict modification to the possessor of source. If you have a binary, you can still change it. There is no restriction, and if you are clever enough, you can do it. So the situations are not parallel, and crucially so. How so? If anything, what you're saying here is a further argument against the need for a requirement to pass source along with binaries, since, according to you, we don't really need the source to make changes. -- Jeremy Hankins [EMAIL PROTECTED] PGP fingerprint: 748F 4D16 538E 75D6 8333 9E10 D212 B5ED 37D0 0A03
Re: The Show So Far
On Tue, Mar 18, 2003 at 08:28:37AM +1000, Anthony Towns wrote: On Mon, Mar 17, 2003 at 10:27:31AM -0800, Thomas Bushnell, BSG wrote: 1) The freedom to take away other poeple's freedom, and Number (1) is a real imposition, but not a real freedom. The freedom to XXX is not a real freedom. Look, I know it's fun to redefine words so that you can pretend whatever you're arguing against is a contradiction in terms, but it doesn't go anywhere. I take it you subscribe to a state of nature definition of freedom, then, in which the only true freedom belongs to a man who lives by himself on a planet? In any social scenario, the freedoms of one party are necessarily limited by the freedoms (or rights) of another. Rights and freedoms are always going to be in tension; it's the nature of the beast. Thomas is entitled to argue that the freedom to withhold source code from the recipeients of object code is as objectionable as the freedom to enslave or the freedom to kidnap. Whether he makes a compelling case for such an argument is another story, but it's not unsound on its face, as you have characterized it. -- G. Branden Robinson| There is no gravity in space. Debian GNU/Linux | Then how could astronauts walk [EMAIL PROTECTED] | around on the Moon? http://people.debian.org/~branden/ | Because they wore heavy boots. pgpHzytZ1RuIM.pgp Description: PGP signature
Re: The Show So Far
On Tue, Mar 18, 2003 at 05:21:55PM +1000, Anthony Towns wrote: Sure; it's a plainly stupid idea. No one's seriously advocating it, and it doesn't benefit anyone. Please at least come up with examples that are vaguely _plausible_. [...] Which is to say: sending your tax return to someone when you change a program is not a reasonable thing to do. As such, it's not a reasonable thing for a license to require you to do. So, you'd propose as components of a Debian Free Software Definition that software be licensed in ways that are not: 1) stupid; or 2) unreasonable My problem with this (implicit) proposal of yours is that it's practically tautological. Hardly any Debian Developer is going to accept as DFSG-free a license that he feels is stupid or unreasonably, and hardly any licensor is going to use one that he feels is stupid or unreasonable. That leaves licensors and the Debian Project with precious little in the way of objective grounds upon which to evaluate the terms of a license. I think that, when rejecting a license as non-DFSG-free, we need to be able to say something more about than it's stupid or it's unreasonable. Thus my proposal of adopting the FSF's definition of Free Software, with an as-yet unarticulated fifth freedom that has something to do with privacy. -- G. Branden Robinson| The last Christian died on the Debian GNU/Linux | cross. [EMAIL PROTECTED] | -- Friedrich Nietzsche http://people.debian.org/~branden/ | pgpwblDO0mfrC.pgp Description: PGP signature
Re: Standard non-copyleft free license?
On Mon, Mar 17, 2003 at 12:57:14PM -0500, Don Armstrong wrote: On Mon, 17 Mar 2003, Branden Robinson wrote: License documents that succumb excessively to lawyer's desires to have many sticks with which to beat the licensee should be rejected as non-DFSG-free, because they don't promote freedom. I don't think we really need to worry about whether a license promotes freedom; we should worry whether a license restricts that freedom or not. I disagree. Our Social Contract says that our priorities are our users and Free Software. This means that we expect ourselves to be advocates of and defenders of these priorities. We're not just committed to not making life *worse* for our users and Free Software. These are guiding principles that direct our actions; not merely toggle switches on a veto stamp. We're not merely committed to preventing backsliding. We're committed to forward progress. Licenses that terrorize the licensee and discourage him or her from exercising the rights he or she should be able to expect from a Free Software license are not the sort of thing people should need to worry about coming from Debian main. Certainly. I'm just commenting on the motivation behind the clause. Since the actual action that the clause prevents is (at least in the US) illegal in itself, I don't see a significant problem for Debian. I don't care what's legal or illegal; the law could change tomorrow, and over the past few years in the U.S. this has actually happened a few times. It is not the job of a copyright license to reiterate what is or is not legal in a particular jurisdiction. The job of a copyright license is to *grant permissions*. If you don't want to grant any permissions, you don't need a license at all. -- G. Branden Robinson|Men use thought only to justify Debian GNU/Linux |their wrong doings, and speech only [EMAIL PROTECTED] |to conceal their thoughts. http://people.debian.org/~branden/ |-- Voltaire pgp2nLqn4s8bT.pgp Description: PGP signature
Re: Dissident versus ASP
On Mon, Mar 17, 2003 at 07:30:44PM +1000, Anthony Towns wrote: If your program is not distributed to anyone, then the license cannot require you to distribute it to anyone (no matter how many people use it or for what purpose, etc). Instinctively, this seems a reasonable test to apply to a license. Can anyone think of any currently-accepted-as-DFSG-free licenses that would fail this test? If not, I propose we consider adding this to our battery of tests; maybe we could call it the Towns Test, which is nicely alliterative. :) -- G. Branden Robinson|I have a truly elegant proof of the Debian GNU/Linux |above, but it is too long to fit [EMAIL PROTECTED] |into this .signature file. http://people.debian.org/~branden/ | pgp6d0FKqsPLN.pgp Description: PGP signature
Re: Standard non-copyleft free license?
On Tue, 18 Mar 2003, Branden Robinson wrote: I don't think we really need to worry about whether a license promotes freedom; we should worry whether a license restricts that freedom or not. I disagree. Our Social Contract says that our priorities are our users and Free Software. This means that we expect ourselves to be advocates of and defenders of these priorities. I agree that we should be promoting freedom. However, I don't think that our licenses need to promote freedom, so long as they don't restrict it. That is, I don't think I'll ever see the day where we decide not to package BSD or X licensed software merely because it fails to promote freedom. [If that indeed was the point you were driving at... perhaps I've misunderstood what you were getting at when you used promote.] The job of a copyright license is to *grant permissions*. And often to restrict them, as is the case in the GPL (linking, etc.), and many no warranty clauses. Don Armstrong -- I was thinking seven figures, he said, but I would have taken a hundred grand. I'm not a greedy person. [All for a moldy bottle of tropicana.] -- Sammi Hadzovic [in Andy Newman's 2003/02/14 NYT article.] http://www.nytimes.com/2003/02/14/nyregion/14EYEB.html http://www.donarmstrong.com http://www.anylevel.com http://rzlab.ucr.edu pgpAhe4CI89wn.pgp Description: PGP signature
#153467: libjpeg62: JPEG is patent-encumbered
severity 153467 important thanks Hello debian-legal and Florian, As discussed previously, it seems there is nothing to do about this issue short of moving main in non-US-EU-JP. So I downgrade the severity to important so that we can release sarge someday :). If you have new informations about this issue, please forward them to [EMAIL PROTECTED] Cheers, Bill.
Re: Dissident versus ASP
Scripsit Jeremy Hankins [EMAIL PROTECTED] I'm saying that in the most typical case, these folks will have big servers big pipes. I'm certainly not saying that it's ideal that only folks with big servers and big pipes be able to provide ASP services. What you seem to be saying is that you consider it OK if a (purportedly free) license effectively prevents people *without* big servers and big pipes from using/modifying the software. Well, that's probably not what you mean, but it's the consequence of the this is free enough, because it will not be hard to satisfy for people who do have sexy hardware line of reasoning. -- Henning Makholm*Vi vil ha wienerbrød!*
Re: The Show So Far
Anthony Towns aj@azure.humbug.org.au writes: If you want that formulated as a principle, as though that makes it somehow better, I've already said: ] Sending your tax return, or your latest entries ] in your diary, or whatever, to someone random and sending your changes ] to some program to its author aren't comparable. One's never sensible ] or reasonable, the latter's a good thing that we'd want to encourage ] independent of whether it's required by the license. Which is to say: sending your tax return to someone when you change a program is not a reasonable thing to do. As such, it's not a reasonable thing for a license to require you to do. Ok, so what you are saying then is that only reasonable requirements can be in a DFSG license. The DFSG of course has no such thing in its actual text; but I have no objection to adding it. Of course, now I need to understand why you think the forced-disclosure requirement is reasonable and the tax-return one isn't. I think that *both* are unreasonable, and for the same reason. You think one is reasonable and the other is not; please explain the basis for this determination.
Re: Dissident versus ASP
Anthony Towns aj@azure.humbug.org.au writes: On Mon, Mar 17, 2003 at 10:26:47PM -0800, Thomas Bushnell, BSG wrote: Dissident test + Practical objections == Can't close the ASP loophole You're not making any sense. Because it's logical equivalence, not numerical equality. a^b * a^c = a^(b+c) is true, but a^b * a^c = a^(b+c) isn't true? Or are the rules different for ideas that conform to Peano's axioms, than ideas that prefer modus ponens? = does not mean equal to or greater than; it means implies. I do not agree with the claim that the dissident test is just another way of saying that the only ways your allowed to close the ASP loophole are ones which are practically unreasonable. That's nice. Disprove it. I did. I formulated the dissident test long before I even *knew* of the ASP loophole. I therefore did not formulate it as some strategy for keeping people from closing it.
Re: The Show So Far
Jeremy Hankins [EMAIL PROTECTED] writes: [EMAIL PROTECTED] (Thomas Bushnell, BSG) writes: Jeremy Hankins [EMAIL PROTECTED] writes: But despite the above I do want to point out that the argument about the only thing stopping the possessor can easily (and, IMHO, more justifiably) be used against the GPL and in favor of BSD-style licensing. Simply s/possessor/possessor of source/ to see what I mean. No, and this is an important point. The BSD-license does not restrict modification to the possessor of source. If you have a binary, you can still change it. There is no restriction, and if you are clever enough, you can do it. So the situations are not parallel, and crucially so. How so? If anything, what you're saying here is a further argument against the need for a requirement to pass source along with binaries, since, according to you, we don't really need the source to make changes. Binary only distribution *inhibits* changes, and makes them *harder*, without making them strictly impossible. The GPL says that the costs of including source are trivial--an extra CD, and therefore requires you to share them. The BSD license does not require such sharing, but that does not imply that it views sharing as *wrong*.
Re: Dissident versus ASP
Branden Robinson [EMAIL PROTECTED] writes: On Mon, Mar 17, 2003 at 07:30:44PM +1000, Anthony Towns wrote: If your program is not distributed to anyone, then the license cannot require you to distribute it to anyone (no matter how many people use it or for what purpose, etc). Instinctively, this seems a reasonable test to apply to a license. Can anyone think of any currently-accepted-as-DFSG-free licenses that would fail this test? If not, I propose we consider adding this to our battery of tests; maybe we could call it the Towns Test, which is nicely alliterative. :) I think the principle is a good one. Heretofore, our tests have been thought experiments which demonstrate why a particular license term would be harmful. I would suggest reformulating this in such terms; as it sits, it's really just like an extra DFSG section.
Re: Dissident versus ASP
Jeremy Hankins [EMAIL PROTECTED] writes: Bernhard R. Link [EMAIL PROTECTED] writes: * Jeremy Hankins [EMAIL PROTECTED] [030317 17:31]: Folks who are providing an ASP-style service generally are going to have big web servers and lots of bandwidth anyway; I'm not convinced that distribution of source would be a significant burden for them. Note the generally. You are aware that you imply, that only such persons should have the possibility to do so? I'm saying that in the most typical case, these folks will have big servers big pipes. I'm certainly not saying that it's ideal that only folks with big servers and big pipes be able to provide ASP services. Am I misunderstanding you? I'm simply trying to divide the problem space up a bit. The other two scenarios described situations where ASP-providers didn't have big servers or big pipes. Hardly! Part of the point of free software is that everyone has a software publishing factory on their desk, and it is the noxious copyright regime that blocks them from using it. My server, even if over a tiny pipe, would be useful as an ASP provider for my friends, just as my CD-writer is useful as a software publishing factory for my friends.