Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759390AbXIQWKK (ORCPT ); Mon, 17 Sep 2007 18:10:10 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1756687AbXIQWJ5 (ORCPT ); Mon, 17 Sep 2007 18:09:57 -0400 Received: from smtp2.rz.uni-karlsruhe.de ([129.13.185.218]:52437 "EHLO smtp2.rz.uni-karlsruhe.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756473AbXIQWJ4 (ORCPT ); Mon, 17 Sep 2007 18:09:56 -0400 X-Greylist: delayed 1813 seconds by postgrey-1.27 at vger.kernel.org; Mon, 17 Sep 2007 18:09:55 EDT Date: Mon, 17 Sep 2007 23:39:40 +0200 From: Ingo Schwarze To: Adrian Bunk Cc: misc@openbsd.org, linux-kernel@vger.kernel.org Subject: Re: Wasting our Freedom Message-ID: <20070917213939.GA21548@selene.usta.de> References: <46ED7A8F.1020304@pro-g.com.tr> <20070916195909.GA18232@stusta.de> <20070916203926.GA17863@schlund.de> <20070916211208.GC5502@thunk.org> <20070916231633.GB10339@SDF.LONESTAR.ORG> <20070917093011.GX32213@nudo.bsws.de> <20070917125714.GI18232@stusta.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070917125714.GI18232@stusta.de> User-Agent: Mutt/1.5.16 (2007-06-09) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 4206 Lines: 76 Adrian Bunk wrote on Mon, Sep 17, 2007 at 02:57:14PM +0200: > But stating in your licence that noone has to give back but then > complaining to some people on ethical grounds that they should give > back is simply dishonest. > > Is your intention to allow people to include your code into GPL'ed code > and never give back, or is your intention that this shouldn't happen? > > And whatever your intention is should be stated in your licence. As this is a recurring argument in the present discussion, let's address it, even though it lies somewhat beside the main topic. What i wish and what i try to enforce by legal contracts are two completely different things. In particular, it is _not_ a smart idea to try to enforce all one's wishes by legal means. For example, i wish that as much as possible of the code i write be freely available such that others can use it, too, and i wish that others write useful code and make it free such that i can use it. When i publish code, i wish bugfixes to be fed back to me, and i hope that others might free their derivative works, too. Besides, i might hope that people at large behave in human and rational ways and refrain from doing harm to others. In particular i might wish the fruits of my work not to be abused to harm or oppress people. Quite probably, lots of software developers share similar wishes, whatever licenses they happen to be employing. But this doesn't imply i should be putting any of the above into the license for my code. Once people attach additional conditions to their licences, sooner or later i get stuck when trying to combine different code covered by different licences. However well intentioned, in practice, those additional conditions habitually turn out to be incompatible - even when, regarded seperately, all of them might appear to make some sense. Now doubtless, the two main additional conditions imposed by the GPL - derivative works may only be distributed if they are made as open and as free as the original - are among those making the most sense of all the additional conditions you might imagine, in the sense that nearly any developer of free software will wish that anybody holding the copyright on a derivative work would make that free. Still, when trying to combine code with different licences, even the GPL at times turns out to be a bother. This does not only apply to the case of non-free closed-source commercial code, but also to cases where authors intended to make their code free, but, be it by inexperience or because they failed to restrain themselves, unfortunately added some uncommon condition to the license. Combining such code with ISC or BSD code is hardly ever problem, combining such code with GPL code may well be. Thus, even when wishing derivative works to be free in their turn, i still see a strong theoretical and a strong practical argument to choose the ISC license over the GPL: Theoretically, it's just the categorical imperative: If everybody would be adding her or his favorite condition to her or his license, we would not end up in free software, but in chaos. Practically, i'm quite fed up with GPL license incompatibility issues always popping up at the most inconvenient places, and still more, with all those license compatibility discussions. With the ISC license, there are no incompatibility issues and no incompatibility discussions, it just works. Of course, i lose the option to sue people to open up derivative works, but i keep the hope that some people (especially those engaged in free software themselves) understand and keep up the spirit, and above all, i avoid lots of legalese worries. Ultimately, it's kind of a trade-off. To summarize, there are valid reasons to wish that people would make derivative works free, but to not require it in the license. Just like there are valid reasons to wish that people should not use the code for waging war, but to not require that in the license. - 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/