> David Welton <[EMAIL PROTECTED]> wrote > On Fri, Jan 28, 2000 at 03:00:40AM -0500, Andreas Pour wrote: > > > Errh, I keep hearing this misconception that BSD code can be > > relicensed as GPL code, but can not figure out where it comes from. > > How can you re-license BSD *source code* as GPL code?
IMHO it is a misconception to think that a distribution licence is indissoluably attached to some work (be it source or binary code). What is attached and will remain so in all copies/modifications is the copyright. The licence is just a statement from the copyright holder conditionally lifting the default prohibition of copying of copyrighted material (I would like to note that in the relevant sections of the US code I found no mention whatsoever of the possibility to give a distribution licence, contrary for instance to the possibility to transfer (part of) the copyrights; this would imply that the legal status of licences is vague at least). The copyright owner may at any time decide to redistribute with a different licence; what is not entirely clear to me is if somebody making a derived work has the right to give a it a new licence (or no licence at all, meaning no redistribution is allowed) when this is not explicitly forbidden by the licence attached to the original work. What does seem clear is that (1) GPL explicitly forbids distributing derived works in any other way that licensed under GPL (its "viral clause") (2) the BSD licence does not do so. The relevant phrase in the BSD licence is "Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met:" where the following conditions do _not_ mention that the cited phrase (the actual licence to freely redistribute) must be applied to (modified) copies; it does state that the copyright notice must be retained. This may seem legal nitpicking, but it is the only way I can see that BSD code could legally have been incorporated into proprietary software, which clearly seems to have been done (see the examples given by David Welton). I also believe that the GPL was written precisely to forbid such use of free software. >From this it follows that (unless explicit permission is given by the relevant copyright owners) a work derived both from BSD code and from GPL code cannot be legally redistributed under any other licence than GPL. I'm not actually saying it is legal to redistribute under GPL; since the work is still copyrighted (in part) by the Regents of the University of California, I am not sure whether their permission for such redistribution---which was neither explicitly given nor refused---might be formally required. I'll just say that there are probably plenty of examples where such re-licensing has been done, and it does not seem worse than incorporating code into proprietary software. > > The author of BSD code has told me (and everyone else for that > > matter) that I can redistribute it w/out following the requirements > > of the GPL (all I have to comply with is the far fewer conditions > > placed in the BSD license). Now you want to tell me that, by virtue > > of someone adding a line of GPL code to it, that the whole kit and > > kaboodle has been converted to GPL and I can no longer do what the > > author of the BSD code told me I can do (namely, redistribute it > > without complying with the GPL)? No, by adding GPL-ed code no conversion has taken place to the BSD code, and in fact no distribution licence exists at all for the combined work; but yes, it is true that you cannot legally redistribute the combined work without complying with the GPL (in particular you may not apply the BSD licence to the combined work). If you don't like that, don't redistribute the combined work. > > Well, if you try to do that, I can > > remove that line of code and say forget you, I will distribute the > > code in compliance with BSD but in violation of the GPL. Assuming it was you who added the GPL code you may of course remove it again and the result will be derived only from BSD code, and hence redistributable under the BSD licence. But if the code was combined by somebody else who distributed a copy to you, then he cannot have done so legally without attaching the GPL to the whole; in that case even if you strip away the originally GPL-ed code, what remains can still not be distributed otherwise than under GPL. You may of course go and find the original BSD code in isolation somewhere, to which of course GPL does not apply in any way. > > And what > > is your remedy for this act of mine? Can you stop me from doing it? > > No, obviously not. What does that mean? That the GPL does not > > apply to the "modified work as a whole" -- i.e. to the BSD code. As I've tried to say, no licence applies automatically to combined code. If such a licence is applied to the combined code it is because the author of the combined code decides to redistribute under that licence; she is not however at liberty to choose any licence, because of existing copyrights by others. It is very well conceivable that no licence can legally be applied at all to the combined work, in which case it is simply not distributable. In fact, it is just this which I believe is at the heart of the KDE/Qt licence issue. There is always one exception: if all copyright owners come to an agreement, they may attach whatever licence they like to the combined work; no GPL or other licence can prevent that. [...] > There are some pretty good examples, though, if you don't mind my > saying so, of code where people have forgotten about the original BSD > code and gone with the extended proprietary code - SunOS and > derivatives of early postgres code come to mind. There is even more > temptation if the changes are only GPL'ed - it's still free software. I don't think people can forget about the original BSD code altogether; the original Berkeley copyright is still attached to the derived work. I'm not sure if that has any practical consequences though, given that the restrictions placed by the BSD licence are so minimal. Marc van Leeuwen Universite de Poitiers http://wwwmathlabo.univ-poitiers.fr/~maavl/
