On Mon, Dec 03, 2001 at 03:53:34AM +0100, Thomas Uwe Gruettmueller wrote: > However, I don't think your proposal is sufficient: For example, > a manual that contains an invariant (and thus unremovable) > section stating: "Drink alcohol!" or "Satan wants you!", this > will pass your rules, because it is less than 32KB, and propably > less than 5% of the whole text. However, this remark might > render the manual unusable in a primary school, because the > house rules, or maybe some law, might forbid such remarks in > front of children. This type of argument opens a complete new can of worms that, frankly, I haven't the energy to deal with at present. If anyone shares your feelings on this subject I suggest you get together and draft a separate proposal which can be incorporated into whatever document my proposal ends up in. If some folks on this list get their way, that document may be /dev/null, though, so consider yourself warned. :) A far more interesting hypothetical in my opinion would be some person including as invariant text in a package the source code enjoined by the United States Federal 2nd Circuit in the Universal v. Reimerdes case (a.k.a. MPAA v. 2600). That might put Debian in an interesting pickle, or maybe the package would just be shunted to non-us. > I understand that the text of the license document cannot be changed, > because that is the way a license works. Not necessarily. A license *as applied* to a given work, of course, typically requires that it not be modified, which seems reasonable if one accepts the legitimacy of licensing in the first place. It is not fair for someone to alter the license terms on a work and then distribute it to you without informing you of this -- it could expose you to legal liability from the copyright holder, because you didn't know the copy of the license you were given wasn't the same as the one the copyright holder put on it, and you might do things you thought were allowed under the license but which actually were not. On the other hand, it is not necessary that people copyright license documents in and of themselves. The FSF did this with their licenses, as I understand it, because they don't want people to use "forked" versions of their license texts for their own works. > I also understand that free works are accomplished with meta > information, that tell something about the legal status of the > work, e.g. authors, version history, maybe also a table that > indicates which line was written by whom, or a remark like > "Author X disclaims to be responsible for the content of derived > versions"... These things help users to trace back the legal > status of the work, if in doubt. Besides that, they are clearly > seperated from the work itself. Actually, as far as I know, on the GNU FDL has any explicit provisions for rendering such notices non-modifiable. The GNU GPL requires people performing modifications to identify modified versions, but as far as I can recall there is no provision against someone else rewriting these notices, possibly to condense them, or remove irrelevant notices because someone's change was irrelevant or superseded. > What I do not understand is why free works should be accompanied > with any non-free stuff other than the above, and still be > considered free. This is a characteristic of the GNU FDL, and some works licensed under it. I don't think I'm qualified to address this issue; you'll have to ask the FSF. RMS said something to me along the lines of "we have to be sure our message gets through"; beyond that he did not elucidate. > If I have understood it right, you believe that the Emacs manual is > non-free, No, I have made no such statement. Please be careful. > while others do not want it to be categorized like > that, and thus be put on the same level as stuff in [non-free] > that fails any rule of the DFSG, and comes with a special > permission for Debian. > The idea I have in mind about the [non-free light] category is > that non-free-CDs that contain only packages that only fail DFSG > 2, 3 or 6 can still be copied verbatim. So the CD as one big > data chunk cannot become uncopyable. That may be something that the FSF would like to see; I am not certain that the corpus of Debian Developers shares that view. Speaking for myself, I am not sure that I do. -- G. Branden Robinson | The noble soul has reverence for Debian GNU/Linux | itself. branden@debian.org | -- Friedrich Nietzsche http://people.debian.org/~branden/ |
Attachment:
pgpDpKla6VtXQ.pgp
Description: PGP signature