On Fri, Jun 14, 2002 at 04:50:33PM +1200, Nick Phillips wrote: > On Thu, Jun 13, 2002 at 10:32:00PM -0500, Branden Robinson wrote: > > People might incorporate DFCLed documentation into a C file. Think of > > standards documents, or just damn good manuals and damn poorly commented > > code. > > Hmmm... if you incorporate the whole thing, you may have a problem, but > if you're quoting bits, that's "Fair Use" - maybe "Fair Use" should be > explicitly incorporated for the benefit of those living in countries > that don't have such a concept, but... We don't need Fair Use if we have GPL compatibility. Besides, Fair Use is only a defense you can use after you're in court, unfortunately. > Besides, is it appropriate to use the GPL if you are incorporating the > whole document into the code? What does the GFDL say about this? Nothing. The GNU FDL is not GPL-comaptible and as far as I know there is no intent on the part of the FSF to make it so. > Has anyone asked RMS what his position on incorporating (potentially > large) bits of GFDL-licensed docs into GPLed programs is? That it will > never be necessary? or that it should not be allowed? I don't know that anyone has asked RMS these questions directly. > Or does the GFDL have an exception for this because the only purpose > of the GFDL invariant sections as far as he is concerned is to push > the GNU Manifesto, and if the docs are being incorporated into GPLed > programs, then that fact is judged to be adequate pushing on its own? Neither the GNU FDL 1.1 nor the 1.2 draft have any such exception. > OK, I think if you are really really convinced that you must have GPL- > compatibility (which I am not convinced of myself, but hey...), then you > *might* be able to twist a clause like clause 6 of the GPL to make sure > that the endorsements come back into force if the docs are pulled back out > from the program. I don't think I need to twist clause 6 to do that. The last three paragraphs of clause 2 appear to do the job. > But really, GPL-compatibility is probably a red herring - I think that > GPL-compatibility may well detract from the overall usefulness of the > license, as as others have pointed out, it is likely to render the doc > essentially GPLed. Which a lot of authors will not go for. I'm pretty sure that GPL-compatibility is not something I want to give up. More to the point, I do not expect broad adoption of the DFCL, at least not initially. The OPL and GNU FDL have poisoned the well a bit. People are right to groan when there's Yet Another New License. However, it's not our fault that both openpub.org and the FSF fumbled the ball when it came to a free content license that was DFSG-free no matter how it was applied (no non-free optional clauses). The DFCL *will* need to justify its existence, and will need to be subjected to a reasonable level of scrutinty. > What happens now if you create a document from a program that is licensed > under the GPL? That depends on what you mean by that. Obviously I *can* write a manual for a GPLed program, and not license that manual under the GPL. O'Reilly has been doing this for years. > Is the document forced to be GPLed? But the GPL applies only > to Programs... Actually: This License applies to any program or other work which contains a notice placed by the copyright holder saying it may be distributed under the terms of this General Public License. The GPL applies to "any work", because it applies to any program, or any other thing which is not a program. > someone needs to corner RMS on this (with a very well-defined > list of very specific questions), since we're stuck with the GPL. I agree that it is worth asking RMS why the GPL is good enough for GNU Emacs but not good enough for the GNU Emacs Manual. > I really think is that the GPL is in need of a revision anyway, as it appears > to fail to consider lots of things that are happening today. Like the mess > we are in with the OpenSSL crap - the OS exemption appears to give > proprietary OS vendors a potential advantage over Debian. Mmm, I'm not coming of as the FSF's biggest fan in this discussion, but I disagree here. I think OpenSSL's license problems are entirely OpenSSL's fault. Their lust for advertising clauses is making life hard for all the Linux distributions. Maybe this a cause for celebration among BSD advocates, but I thought BSD license fans weren't supposed to care what people did with the code as long as there was attribution. That some people are following UC's example regarding the advertising clause[1] is just evidence of obstructionism and antagonism. Or maybe just plain old ignorance. "UC did *what*?" > It also does not acknowledge the distinctions (or lack of) between a > Program and any other information which may need to be distributed, > nor does it consider how tightly such things may be integrated with > Programs. Actually, clause 0 of the GPL seems explicitly avoid making a distinction between programs and non-programs. Anything licensed under the GPL is a Program. [1] ftp://ftp.cs.berkeley.edu/pub/4bsd/README.Impt.License.Change -- G. Branden Robinson | Debian GNU/Linux | Extra territorium jus dicenti branden@debian.org | impune non paretur. http://people.debian.org/~branden/ |
Attachment:
pgpD0fMjhPoQs.pgp
Description: PGP signature