On Wed, 2004-06-16 at 21:59, Joe Wreschnig wrote: > Step by step, tell me where you start to disagree: > > If I write a program that contains the entire ls source code as one > large C string, and then prints it out, that is a derivative work of the > ls source. > > If I write a program that contains the entire ls source code in a > similar manner and uploads it to a non-display device, it is still a > derivative of ls; the only thing that has changed is the device I am > printing to. > > If I write a program that contains a nontrivially long, copyrightable C > string and uploads it to a non-display device, it is a derivative of > that copyrightable C string. This is a generalization of the last > statement to any copyrightable material. I suspect this is where you > will disagree with me, but I am curious why. > > Firmware can be represented as a copyrightable C string, and the > firmware in question is. This is a fact. > > That firmware is not released under the GPL. This is a fact.
I forgot to note something important here -- the firmware is not only not released under the GPL, it is released under a GPL-incompatible license (in most cases, no modification is allowed). The remainder of the argument does not hold if we have the firmware (in source or binary form) under some GPL-compatible license terms like the 2 clause BSD license. I still believe such things infringe upon the GPL's terms, but *this* argument does not show that. > But the rest of the program source (i.e. everything but the C string) is > licensed under the GPL. Although not all of Linux is, a lot is; this is > also a fact. > > So the program is a whole is licensed partly under the GPL, and partly > under GPL-incompatible terms. This follows from the previous two. > > Therefore, the program as a whole has an inconsistent license, and > Debian cannot legally distribute it. This follows from the previous > conclusion and copyright law. -- Joe Wreschnig <[EMAIL PROTECTED]>
signature.asc
Description: This is a digitally signed message part
