Date: Sat, 30 Oct 2004 14:37:48 -0700 From: "David Schwartz" <davids@webmaster.com> To: <tedm@toybox.placo.com>, <chat@freebsd.org> Cc: TM4525@aol.com Subject: RE: GPL vs BSD Licence Message-ID: <MDEHLPKNGKAHNMBLJOLKGENNPGAA.davids@webmaster.com> In-Reply-To: <LOBBIFDAGNMAMLGJJCKNIEJDEPAA.tedm@toybox.placo.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> That is one of the arguments. But, the GPL is concerned with > distribution. > I think this has been raised before with them. I think that the scenario > was, if I make a program that dynamically links into GPL, then I > distribute > both my program and the GPL code that it links into, do I have to put my > program under GPL? I think their answer was yes - they argued that when > the linking takes place and who links it is immaterial, and that the fact > that your program cannot run without their stuff means that when your code > is running, that your program and their stuff become as a single program. This argument is obviously, IMO, absurd. Suppose I make a product that *can* link to a GPL'd library but could, of course, also link to any other library that provided the same interface. The FSF is arguing that as soon as somebody links it to the GPL'd library, it becomes a derived work even though it wasn't one before. Nothing that you can do with a work after it's produced can turn that work into a derivative work of another work. For copyright purposes, it's only the creative content in the work itself that matters. What you do with that work later could create new works that have a different status, but can't change the status of the work itself. The FSF is, of course, free to take any position they wish, but this one is, at least to me, absurd from a legal standpoint. Even if the program had been designed from the beginning to work with an interface that was currently only implemented in a GPL'd library, I would still find the argument absurd (though slightly less so). The program is a derived work from the interface itself, not from any particular implementation of that interface. (And then you would have the very odd case where whether A is a derivative work of B would depend upon whether B was written before or after C!) I can't imagine that anyone who understood what it is that grants you a copyright and what a 'derivative' work really is would find this argument at all persuasive. DS
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?MDEHLPKNGKAHNMBLJOLKGENNPGAA.davids>