Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Mon, 3 Mar 2003 11:51:56 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Mon, 3 Mar 2003 11:51:55 -0500 Received: from h-64-105-34-204.SNVACAID.covad.net ([64.105.34.204]:27296 "EHLO freya.yggdrasil.com") by vger.kernel.org with ESMTP id ; Mon, 3 Mar 2003 11:51:50 -0500 From: "Adam J. Richter" Date: Mon, 3 Mar 2003 09:02:11 -0800 Message-Id: <200303031702.JAA02896@adam.yggdrasil.com> To: esm@logic.net Subject: Re: BitBucket: GPL-ed KitBeeper clone Cc: linux-kernel@vger.kernel.org Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 4699 Lines: 99 [I've dropped all of the direct email addresses from the cc list. I think they'd probably rather just see one copy of this tangential discussion on the linux-kernel list, if at all. -Adam] On Mon, 3 Mar 2003, Edward S. Marshall wrote: >On Sat, Mar 01, 2003 at 06:23:21PM -0800, Adam J. Richter wrote: >> Note that Subversion, in particular, is GPL incompatible and >http://subversion.tigris.org/project_license.html >I don't see anything particularly GPL-incompatible in there; I think the main incompatibility is part 3 (the advertising requirement). Regarding the Apache 1.1 license, which is used by Subversion, http://www.fsf.org/licenses/license-list.html#OriginalBSD says, "This is a permissive non-copyleft free software license with a few requirements that render it incompatible with the GNU GPL. We urge you not to use the Apache licenses for software you write. However, there is no reason to avoid running programs that have been released under this license, such as Apache." >looks pretty much like a BSD-style license to me. FSF has said that they believe that only the "new style" BSD licenses are GPL compatible. They consider the "old style" (with the advertising requirement) to be GPL incompatible. >Something that precludes SVN's use >by GPL'd projects, or precludes integration with GPL'd projects, is >something I'm sure CollabNet and the developers on the mailing list would >love to know about That is not what I mean by "GPL incompatible." http://www.fsf.org/licenses/gpl-faq.html#WhatDoesCompatMean gives the foundation's definition of "compatible with the GPL": | What does it mean to say a license is "compatible with the GPL". | | It means that the other license and the GNU GPL are | compatible; you can combine code released under the other license with | code released under the GNU GPL in one larger program. | | The GPL permits such a combination provided it is released | under the GNU GPL. The other license is compatible with the GPL if it | permits this too. When I say "GPL compatible", my meaning is similar, perhaps identical. By "GPL compatible", I mean that if the contents of a work are comingled with GPL'ed content, even within the same file, that obeying the terms of the GPL with regard to the whole resultant work results in one having at least the permissions of the GPL with regard to the whole resultant work (this, by the way, is what I believe "licensed as a whole" under the GPL means). I mentioned the GPL incompatibility issue to Brian Behlendorf after a he gave at a trade show once, and he said, approximately, "it is the GPL's fault, you know." This discussion took place after the University of California at Berkeley had switched to the new BSD terms for GPL compatibility. At that point, I believed that investing further time in arguing would probably not produce useful results. >Lacking an on-disk format that's actually useful for storing more >information than files and diffs, they invented one. There already were free systems that deal with directory information and change sets on groups of files rather than single files. The Subversion team chose to use a new format that I've heard many people, even people who seem not to care about the GPL incompatibility, says is complex enough to warrant not using svn. Even when one of the Subversion people gave a talk at the Silicon Valley Linux User Group, he acknowledged that many people on some mailing list discussing Subversion design issues, including Larry McVoy, considered their decision to be a big mistake. >[...] svn is architected such that bolting up >to another repository storage system (say, an RDBMS, or even, horrors, a >bitkeeper-compatible SCCS derivative) is really just a matter of writing >the code (with a few caveats, obviously, but that's the basic idea). That is arguably true of most source code control system that try to deal with multi-file commits. For example, Aegis already lets you choose between rcs or sccs. All that said, I do not consider contribution to Subvresion to be something that has a negative sum effect. If you are contributing to Subversion, I'm not saying that that is a bad thing. Adam J. Richter __ ______________ 575 Oroville Road adam@yggdrasil.com \ / Milpitas, California 95035 +1 408 309-6081 | g g d r a s i l United States of America "Free Software For The Rest Of Us." - 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/