Conversation
Notices
-
@fontana,I agree about giving authority to contributors ("empowering" is too corporate-speak IMO). But wouldn't *no* default do that better?
Wednesday, 18-May-11 14:11:28 UTC from web-
just signed the !Fedora Project Contributor Agreement. You should too.
-
@bkuhn Uhm that assumes there is a sub-project.
-
@bkuhn If you take the steps necessary to affirmatively license your contributions default license language is mooted entirely
-
@fontana Indeed the specfiles are the example that defy categorization as contributions to existing sub-projects.
-
@bkuhn what do you mean?
-
@copiesofcopies, I realize your goal there is to ridicule, but in fact defaulting to permissive does seem at least !problematic if not worse
-
#AMD To Support #Coreboot On All Future CPUs | http://bit.ly/iKcgIo
-
@bkuhn I'm not sure what you are getting at. If I choose to affirmatively copyright an original spec file I write as GPL. I can do that.
-
@bkuhn I thought the default was to use the license of the upstream, but I'm not seeing it clearly in the FPCA.
-
@fontana How is that even possible? Isn't the kernel GPL?
-
So the kernel being GPL doesn't mean it's all GPL; the GPL is just the least-permissive license?
-
a module isn't a derived work and only has to be compatible with the GPL, not GPL'd itself AFAIK
-
@fontana ah, if it's only about original (rather than ©-derived) contributions, no problem in this regard indeed
-
@fontana IIUC Fedora would then need a separate license for derived contributions to non-copyleft (and copyleft?) projects, right?
-
@fontana if the FPCA only provides implied permissive licensing to Fedora for non-derived contributions, what about derived ones?
-
@fontana if derived => not covered by FPCA, then separate licensing is required. copyleft base may imply the license, but not otherwise
-
@fontana erhm... but how about a derived contribution that the contributor created herself? e.g. patch for existing file (?)
-
@fontana erhm. now *I* don't understand. what do you mean “derivative work in the GPL sense”?
-
@fontana GPL defers to © as to what's derivative, and its requirements apply to all such derivative works
-
@fontana but it does clarify that works are exempt from the requirements when separate && not derived works under ©
-
@fontana this is not about GPL. my concern is how Fedora gets a license to use patches for non-copyleft works
-
@fontana how can it not match? GPLv2 does not define it at all, it defers to © law
-
@fontana huh? v2 says: "work based on the Program" means either the Program or any derivative work under copyright law
-
@fontana hmm, does “original” in FPLA mean the opposite of “© derivative”? if not, it might address my concern
-
@fontana yeah, but those can't bind you unless you need a license to begin with, and you don't if it's not a derivative AFAIK
-
@fontana anyhow, I get a feeling we're going down a distraction rather than addressing the point (if there is one ;-)
-
@fontana so there seems to be a different problem. AFAICT it has to be either one or the other:
-
@fontana either contributor promises to grant Fedora permissive license on original contributions, even derivatives from copyleft,
-
@fontana or contributor doesn't grant Fedora through FPCA a license to contributions that are not non-derivative works,
-
@fontana or I'm missing something ;-)
-
@fontana ok, so it's the former: GPL-derivative “original” cannot be licensed under X11 (which MIT?), only GPL, IIUC
-
@fontana yep, that was indeed a source of confusion, but see, the initial concern remains after clearing that up
-
@fontana now, to be clear, when I write “source of confusion”, I don't mean “the preferred form to modify the confusion” :-D
-
@clacke GPL requirements cover all derivative and collective works, permitting distribution under GPL only, except for mere aggregation
-
@clacke OTOH, if they're not derived works, then GPL reqs don't (can't) apply to them distributed by themselves
-
@spot, in particular,I'd love http://ur1.ca/435qf to include FAQ entry: "Why does !Fedora need any sort of #CLA / FPCA at all?" cc: @fontana
-
@fontana Just the act of calling it a CLA validates the concept and does a disservice to software freedom...
Bradley M. Kuhn likes this. -
@lxoliva right, what you want is the obfuscated source of confusion then... coming right up...
-
@fontana,the !Fedora / !RedHat #entanglement is already pretty strong. It's clear from various events: Fedora is a RedHat-controlled project
-
@bkuhn Yes, I am aware. Just examining in general what happens if people permissively license contributions to copyleft project.
-
@fontana, yes, there are levels of #entanglement. I agree !Fedora's entanglement is less than #Asterisk's & !Ubuntu's as well, actually.
-
@fontana, I don't believe in keeping consistent positions if one later discovers that his positions were previously wrong & can explain why.
Clacke Moved to Unlimited and Jelle Hermsen like this. -
@bkuhn community-oriented = no non-free firmware?
-
@bkuhn: hobgoblin, etc. ;-)
-
@fontana being 'less shit than anything else' doesnt make your stuff good. Just less bad than other options
-
@clacke point is that, if the work is derivative, it's not “yours” alone
-
@lxoliva The complete and corresponding source and installation info of the confusion in computer-readable, human-understandable form. :-D
-
@clacke 'zactly. does Fedora really expect contributors of patches for copyleft projects to consult with a lawyer every time?
-
@clacke or does Fedora want to induce these contributors to error and legal liabilities?
-
@lxoliva perhaps the combination is a derivative but the work itself is not or at least parts of it is not?
-
@fontana say A posts a patch for GPLed P, patch being derivative of P, and Fedora uses part of patch in X11-licensed Q
-
@fontana author of P sues A and Fedora for copyright infringement. what happens next?
-
@fontana exactly, but Fedora could assume the CA license applies, then blame the contributor when the lawsuit comes
-
@fontana CA places the burden of reporting derivatives and indicating copyleft on the contributor, maybe even shifting liability
-
@fontana which is why I said at first that nobody should sign such an agreement, it's too dangerous and error prone
-
@fontana we've already determined that original per the CA may be © derivative the other day, haven't we?
-
@fontana besides, whether or not it's derivative isn't obvious, and even a lawyer may be deemed wrong by a court
-
@fontana which, in the case at hand, happens to be precisely true: derivative original is subject to GPL conditions!
-
@fontana it sure is convenient for Fedora, but it may create a big problem for the contributor, thus my concern
-
@fontana it would be so much nicer if there was no default, or the default was the same as if the patch had been submitted upstream
-
@fontana no, I haven't read it, I based my concerns on what was posted here
-
@fontana done. it doesn't seem to address at all the problem I described
-
@fontana no, I'm just reading on GPL's requirements for derivative works
-
@fontana got a pointer to this? maybe *this* could ease my concerns
-
@fontana though if one could license *any* *derivative* of GPL work under a permissive license, copyleft wouldn't work at all
-
@fontana oh, I remember having read that before, it's not really relevant to the point at hand AFAICT
-
@fontana right, so it takes one more party to get a lawsuit: Fedora adds derived patch for GPLed P and adds to MIT/X11 Q
-
@lxoliva Catching up with for the help
-
@fontana who'll ultimately be in trouble? B, who trusted Fedora's X11/MIT lic; Fedora, for relying on CA, or A, for signing error-prone CA?
-
@fontana remember: derivative doesn't imply preexisting; original (per CA) doesn't imply non-derivative
-
@fontana you mean e.g. distributing the GPL-derivative patch without a copy of the GPL?
-
@fontana 'cept posting GPL-derivative patches is well-accepted practice, whereas dist'ing GPLed code under X11 then non-Free isn't
-
@bkuhn woah, great trolling :-D I feel welcome to the club ;-)
-
@fontana I don't know about it being customary. if asked, I'd warn against it, unless original portion's clearly not derivative
-
@fontana if so, odds of a problem are low, but FPCA could avoid gratuitously exposing Cs and Fedora to risk, or losing Cs for it
-
@fontana defaulting to permissive when GPL might be required means contributor takes up the risk of forgetting to note GPL reqs
-
@fontana IMHO it would be better for contributor if “unlicensed contributions” for existing copyleft files inherited the copyleft conditions
-
@fontana, they are only incorrect in a "lex mercatoria" sense, though. License violation is still license violation, even if pedantic.
-
@fabsh,due to the so-dubbed "retroactive perfection theory of GPLv2", releasing GPLv4 maybe the only way to get people to adopt !GPLv3. ;)
Clacke Moved to Unlimited and sazius (moved to status.saz.im) like this. -
@bkuhn "its not the newest thing around, so it must be good now" mentality?
-
Clacke Moved to Unlimited likes this.
-
@bkuhn I doubt it. ;)
-
@bkuhn can you start using the #GPLv4 tag when applicable? This can be useful in the future to track discussions on the new license version.
Mike Linksvayer likes this. -
@fontana Nice! Anything published?
-
@fontana Awesome.
-
@fontana patches welcome.
-
@copiesofcopies, license of !SFLC's publications don't really indicate patches are welcome. Would SFLC be willing to license those CC-By-SA?
Mike Linksvayer likes this. -
@fontana, which policy recommendations do you believe are mine in this context? ISTR disagreeing w/ what went into the document ultimately.
-
@fontana I know you wrote it; I assume that's why it's your favorite. But if you do have suggestions for improvement, please send them.
-
@bkuhn yes, I think so. We changed the default site license to CC-BY-SA; should be able to do the same with the docs.
Mike Linksvayer and diablomarcus like this. -
@ruiseabra, it was. I don't know why things turned out that way. It was very unfortunate & I still don't understand the behavior.
-
@bkuhn that is sad and depressing
-
@fontana, I try to consider that anyone presenting a different idea might have useful insight. I learned this by observing Lessig actually.
-
@fontana I suppose one man's "community tradition" is another's "GPL interpretation cartel". Sigh. @FOSSpatents http://ur1.ca/48t7j
-
@bkuhn Lawyercentrism?
-