On Tue, Apr 05, 2005 at 08:56:09PM +0200, Josselin Mouette wrote: > Le mardi 05 avril 2005 à 12:50 -0600, Chris Friesen a écrit : > > Josselin Mouette wrote: > > > > > The fact is also that mixing them with a GPLed software gives > > > an result you can't redistribute - although it seems many people > > > disagree with that assertion now. > > > > This is only true if the result is considered a "derivative work" of the > > gpl'd code. > > > > The GPL states "In addition, mere aggregation of another work not based > > on the Program with the Program (or with a work based on the Program) on > > a volume of a storage or distribution medium does not bring the other > > work under the scope of this License." > > > > Since the main cpu does not actually run the binary firmware, the fact > > that it lives in main memory with the code that the cpu *does* run is > > irrelevent. In this case, the Debian stance is that the kernel proper > > and the binary firmware are "merely aggregated" in a volume of storage ( > > ie. system memory). > > It merely depends on the definition of "aggregation". I'd say that two > works that are only aggregated can be easily distinguished and > separated. This is not the case for a binary kernel module, from which > you cannot easily extract the firmware and code parts.
Josselin, please read the thread i linked to in debian-legal, and as nobody really gave reason to oppose it, i believe we have consensus that those firmware blobs constitute mere agregation, provided they are clearly identified and properly licenced, which they are not always. Let's take this to debian-legal only if you want to further discuss it. Friendly, Sven Luther -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]