Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1764495AbXF1G66 (ORCPT ); Thu, 28 Jun 2007 02:58:58 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1760336AbXF1G6u (ORCPT ); Thu, 28 Jun 2007 02:58:50 -0400 Received: from lsd-gw.ic.unicamp.br ([143.106.7.165]:44501 "EHLO boneca.lsd.ic.unicamp.br" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1760365AbXF1G6t (ORCPT ); Thu, 28 Jun 2007 02:58:49 -0400 To: linux-kernel@vger.kernel.org Subject: Re: how about mutual compatibility between Linux's GPLv2 and GPLv3? References: <20070622013417.GT21478@ftp.linux.org.uk> <20070622041949.GA15625@thunk.org> <20070625132853.GH10008@csclub.uwaterloo.ca> <20070626041013.GG24745@delft.aura.cs.cmu.edu> <20070626162541.GA5464@delft.aura.cs.cmu.edu> <20070628050852.GA16686@delft.aura.cs.cmu.edu> From: Alexandre Oliva Date: Thu, 28 Jun 2007 03:58:35 -0300 In-Reply-To: <20070628050852.GA16686@delft.aura.cs.cmu.edu> (Jan Harkes's message of "Thu\, 28 Jun 2007 01\:08\:52 -0400") Message-ID: User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.1 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 5527 Lines: 117 On Jun 28, 2007, Jan Harkes wrote: >> As for making modifications, I'd like to take this opportunity to >> withdraw, for purposes of interpretation, my earlier agreement that >> TiVo permits modification, even though it doesn't permit modification >> in place. I don't see any distinction in US copyright law between >> modification in place and modification by creating a separate work. > In section 2 where it talks about modifications it at first doesn't seem > to distinguish between source and binary, but it doesn't grant the right > that the resulting modified program will actually work. And yes, you can > still modify the kernel binary on the Tivo harddrive, it just won't boot > with the standard bootloader. > But there is an explicit no warranty clause in the GPLv2. Which I > believe is a good thing. If the license tried to enforce usability, Yup. But will courts see this as a valid excuse for effectively denying users the ability to modify the copies of the GPLed programs they run on tivoized devices, even though the means to accomplish this are based on blocking the ability to run the modified executable, which happens to not require permission from the copyright holder in the US? Even more so when the modified binary provably works just fine on similar computers, and fails to work on the tivoizing device just because mechanisms were introduced to stop the user from modifying the behavior of the software that runs in the device? I.e., it fails because you changed it, rather than because you broke it. Couldn't this be understood by a court as an attempt to exploit a loophole in the letter of the license, rather than as an intentional permission to restrict the ability to run covered programs, an act that the license phrases as "not restricted" and "outside its scope" just because US copyright law doesn't demand permission from the copyright holder to run programs, which results in permission to run being outside the scope of a copyright license? Anyhow, I didn't mean to restart this debate, I just wanted (as stated above) to withdraw my earlier tentative agreement that "modification in place" was something that could be forbidden without infringing GPL. >> When you modify a sculpture, you're modifying it in place, and this >> requires as much copyright permission as creating a modified copy of >> it. Both count as modification. > And when you scribble in the margin of a book you are also modifying it, > so what. Maybe it's fair use, especially for a printed book, of which many copies presumably exist. > I don't think you even need copyright permission, although you > will probably get into trouble if the book was borrowed from a library. > I don't see the relevance in the context of this discussion. Relevance was only to counter the argument that the right to "modify in place", as some put it, was not covered by the GPL, "because that's not how copyright law defines modification." I bought that argument at first, but then I referred back to US copyright law and not only failed to find supporting evidence for this argument, but actually found opposing evidence. So I brought it up to make sure my tentative agreement wouldn't be used in a court as an indication that the argument holds water. >> Here are variations of another scenario I proposed back then: >> >> - Tivoizing device ships with tivoized software, regardless of whether >> the corresponding sources are accessible to the end user or hidden >> inside the box, in a fully-encrypted disk that only that machine can >> read. > I'm assuming no written offer for access to the software is included. Yup, that's the idea. > Source are inaccessible, restricted the recipients rights to copy, > distribute and modify the source code, GPLv2 violation. So, the same standards would apply to the binaries: if they were inaccessible, copyright holder could claim infringement under the same grounds, right? > If it doesn't download the sources then the binary is not accompanied by > the corresponding source code and since we assumed no written offer this > is a violation of section 3. Except, perhaps, for this bit: If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code, even though third parties are not compelled to copy the source along with the object code. So, let's narrow the scenario to: tivoized machine downloads binary from protected site, refrains from downloading sources that it could download, user can still access and copy the binaries, but can't obtain the sources because the machine opted not to get them. Now, the user can't distribute the binaries, because doing so without being able to get the sources to pass them on would be copyright infringement. Would a court see this as a restriction on distribution imposed by the distributor? Or by the copyright holder? -- Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/ FSF Latin America Board Member http://www.fsfla.org/ Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org} Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org} - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/