Do Debian patches violate the GPL?
The GNU GPL says: 2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions: a) You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change. I guess the interpretation is that we have to alter any file we change to contain a notice of our changes, once we changed a file. However, since we do distribute the modifications as separate files in a .diff.gz patch file, we are not required to add such notices all over the place. BUT: If we distribute scripts (shell, Perl, Python etc.) or PHP files, that are patched before they are packed into a .deb file, they normally don't contain such notices, even though the GPL seems to require them, and we do distribute plain files in these cases. Comments? Regards, Joey -- Every use of Linux is a proper use of Linux. -- Jon Maddog Hall Please always Cc to me when replying to me on the lists.
Re: Do Debian patches violate the GPL?
Martin Schulze wrote: BUT: If we distribute scripts (shell, Perl, Python etc.) or PHP files, that are patched before they are packed into a .deb file, they normally don't contain such notices, even though the GPL seems to require them, and we do distribute plain files in these cases. In general, we distribute the source, a diff, and a modified binary, so, point 3 applies: 3. You may copy and distribute the Program (or a work based on it, under Section 2) in object code or executable form under the terms of Sections 1 and 2 above provided that you also do one of the following: a) Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange; or, [...] The source code for a work means the preferred form of the work for making modifications to it. Question: Isn't a script inside a .deb in executable form? (In such case it would suffice to provide its source, which we already do by providing the original source and the diff).
Unidentified subject!
From ignacio Fri Sep 6 11:56:41 2002 Return-path: [EMAIL PROTECTED] Envelope-to: [EMAIL PROTECTED] Delivery-date: Fri, 06 Sep 2002 11:56:41 +0200 Received: from shannon ([147.156.161.144] helo=localhost ident=ignacio) by shannon with esmtp (Exim 3.35 #1 (Debian)) id 17nFqz-0006XJ-00 for [EMAIL PROTECTED]; Fri, 06 Sep 2002 11:56:41 +0200 X-Sieve: cmu-sieve 2.0 Received: from sello.ci.uv.es [147.156.1.112] by localhost with IMAP (fetchmail-5.9.11) for [EMAIL PROTECTED] (single-drop); Fri, 06 Sep 2002 11:56:41 +0200 (CEST) Received: from idolo.uv.es (idolo.ci.uv.es [147.156.1.38]) by sello.uv.es (8.12.2/8.12.2) with ESMTP id g869swdY016227 for [EMAIL PROTECTED]; Fri, 6 Sep 2002 11:54:58 +0200 Received: from murphy.debian.org (murphy.debian.org [65.125.64.134]) by idolo.uv.es (8.9.3/8.9.3) with ESMTP id LAA08857 for [EMAIL PROTECTED]; Fri, 6 Sep 2002 11:54:58 +0200 (METDST) Received: from localhost (localhost [127.0.0.1]) by murphy.debian.org (Postfix) with QMQP id 45EFD1F6CC; Fri, 6 Sep 2002 04:54:47 -0500 (CDT) Old-Return-Path: [EMAIL PROTECTED] Received: from sello.uv.es (sello.ci.uv.es [147.156.1.112]) by murphy.debian.org (Postfix) with ESMTP id 517DE1F3EA for debian-user-spanish@lists.debian.org; Fri, 6 Sep 2002 04:54:37 -0500 (CDT) Received: from shannon (shannon.irobot.uv.es [147.156.161.144]) by sello.uv.es (8.12.2/8.12.2) with ESMTP id g869sXdY016191; Fri, 6 Sep 2002 11:54:34 +0200 Received: from ignacio by shannon with local (Exim 3.35 #1 (Debian)) id 17nFqY-0006X3-00; Fri, 06 Sep 2002 11:56:14 +0200 Date: Fri, 6 Sep 2002 11:56:14 +0200 From: Ignacio =?iso-8859-1?Q?Garc=EDa_Fern=E1ndez?= [EMAIL PROTECTED] To: Lista de usuarios de debian debian-user-spanish@lists.debian.org, [EMAIL PROTECTED] Subject: Re: Debian registered by a trade as TM in Spain! Message-ID: [EMAIL PROTECTED] Mail-Followup-To: Ignacio =?iso-8859-1?Q?Garc=EDa_Fern=E1ndez?= [EMAIL PROTECTED], Lista de usuarios de debian debian-user-spanish@lists.debian.org, [EMAIL PROTECTED] Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: [EMAIL PROTECTED] User-Agent: Mutt/1.3.28i Organization: Instituto de =?iso-8859-1?Q?Rob=F3tic?= =?iso-8859-1?Q?a?= (Universidad de Valencia) Sender: =?iso-8859-1?Q?Ignacio_Garc=EDa_Fern=E1ndez?= [EMAIL PROTECTED] X-Spam-Status: No, hits=-2.1 required=5.0 tests=IN_REP_TO,PLING,NO_MX_FOR_FROM version=2.20 X-Spam-Level: Resent-Message-ID: [EMAIL PROTECTED] Resent-From: debian-user-spanish@lists.debian.org X-Mailing-List: debian-user-spanish@lists.debian.org archive/latest/48429 X-Loop: debian-user-spanish@lists.debian.org List-Post: mailto:debian-user-spanish@lists.debian.org List-Help: mailto:[EMAIL PROTECTED] List-Subscribe: mailto:[EMAIL PROTECTED] List-Unsubscribe: mailto:[EMAIL PROTECTED] Precedence: list Resent-Sender: [EMAIL PROTECTED] Resent-Date: Fri, 6 Sep 2002 04:54:47 -0500 (CDT) Resent-Bcc: Resent-From: [EMAIL PROTECTED] Resent-Date: Fri, 6 Sep 2002 12:28:05 +0200 Resent-To: debian-legal@lists.debian.org It seems there have been some news. As it has been announced in deb-usr-sp, I'll do a summary here. I also include the full text in spanish. I've annotated the tranlatoin with brackets '[' ']'. In barrapunto it has appered a note that is suposed to be from CEINTEC, the company that was said to have registered Debian. Please, if anyone sees any mistake in the translation or doesn't understand a word, forgive me, and don't doubt to ask :) My English is not very good. The note says: «About the comment published in bp.com on sep-3 2002, CEINTEC wants to point out that: [...] Ceintec did not register the TM Debian. [...] Debian was registered by a person related to CENITEC, by him or herself [not in the name of CEINTEC] on September of 2000, when he/she noticed that some oother names related to Free Soft were being registered by other trades. We [CEINTEC] know that the intention of that person was to protect the mark Debian precisely to avoid others to register it and forbide its use by others (including CEINTEC). We [CEINTEC] know that this person has never oposed to the use of Debian and, of course, that person guarantees that this won't happen in the future (while the GNU philosophy is respected). We [CEINTEC] also know that he/she would give it (now and before) to any Government Institution that guarantees that the whole community can use it freely and that promisses to renew and defend while the distribution [Debian, I understand] exists. [...]» «Respecto al comentario publicado en barrapunto.com el 3/9/02 CEINTEC quiere realizar las siguientes puntualizaciones: Que Ceintec ha defendido y apostado por Linux y el software libre desde 1995. Como se puede comprobar en su Web http://www.ceintec.com Que ha implementado Linux en sus
Re: Debian registered by a trade as TM in Spain!
We [CEINTEC] know that this person has never oposed to the use of Debian and, of course, that person guarantees that this won't happen in the future (while the GNU philosophy is respected). I don't this this guarantee is worth the paper it's printed on. We [CEINTEC] also know that he/she would give it (now and before) to any Government Institution that guarantees that the whole community can use it freely and that promisses to renew and defend while the distribution [Debian, I understand] exists. If the government wanted names to be used freely then the government wouldn't have set up the TM system in the first place, so you can forget that one.
A GNU GPL question (might be slightly OT)
Hello! This may be slightly OT, but I have really looked around for a better place to ask this question, and failed. I'm in a situation where I am trying to get the source code for a program from the company that distributed that program, and this has turned out to be really difficult. Currently, I'm preparing a reply to their lawyer (I have no legal training myself, so this is really difficult) where he talks to me about a three year rule within the GPL. Here's how I understand this issue: (Correct me if I'm wrong.) Jim gives Joe a program licensed under the GPL. Jim does not provide Joe with the source code, but with a written offer to provide that source code upon request. He can do this, according to section 3b in the GPL. However, that only requires Jim to comply with that offer for a period of three years, which is also stated in section 3b. Two years later, Joe re-distributes the program to Jill, and he includes the written offer from Jim. Joe is required to do so, according to section 3c. Two more years pass and Jill decides that she wants that source code. Here's where my question comes: Since FOUR years has passed since Jim originally distributed the program, he feels that he is no longer obliged to comply with his written offer. However, Jill feels that since she got the program only TWO years ago, the offer is still something that Jim is required to comply with. In her timeline, the three-year limit is not passed yet. That is my question. Who is right, Jim or Jill? I think it's a really important question, too. I shall explain why. Consider this: Jim downloads the source code for the gcc compiler. He then modifies that source and creates jimsgcc. But Jim doesn't really want to share the source with anyone. He is evil and he is also smart. So he finds section 3b in the GPL and he contacs Joe, who is also evil and ready to help Jim out when it comes to completing the scheme. Jim distributes the program (jimsgcc) to Joe, with that valid-for-three years offer that section 3b speaks about. Joe then sits on his ass for three years, so that the offer is no longer valid. Joe now starts distributing the jimsgcc (here, working on orders from Jim of course) and if anyone (any third party) requests the source code from Jim according to that written offer, he denies and says that more than three years has passed since the offer was issued and voilá, Jim has managed to short-circuit the GPL. (Yes, three years is a long time for a computer program and jimsgcc not likely to be very attractive by the time it actually is available, but the principle still holds.) Is this a loophole in the GPL? If my question above is answered with Jim, I think it is. If the answer is Jill, it most likely is not. So... What do you all say about this? /Fredrik Persson
Timidity-patches eek
I've been looking at the timidity-patches package, because I've been looking for a decent MIDI patch set for an unrelated project, and that patch set is a good one. The package has no license. Looking into it, timidity-patches turns out to have been put together from patch files taken from the Midia patch set, distributed with the Midia MIDI renderer that ran on SGI workstations. Midia and its patch set can be obtained from: http://www.csee.umbc.edu/pub/midia/ It's released as binary only, and is not supplied with a license, or even a README. The closest I've found is the string: Midia 2.0 (beta) Copyright (c) 1994 ...within the midia binary itself. The patches don't have any license, either, except for doo.pat: Mike's Doo Copyright (C) 1993 Forte. All Rights Reserved. Doesn't this make timidity-patches non-free? In fact, surely it's not distributable at all? Someone please prove me wrong... -- +- David Given --McQ-+ Did you hear about the hard-working but ill sage | [EMAIL PROTECTED]| who got cursed with garlic breath? He was a | ([EMAIL PROTECTED]) | super-calloused fragile mystic hexed with +- www.cowlark.com --+ halitosis. signature.asc Description: This is a digitally signed message part
Re: A GNU GPL question (might be slightly OT)
On 20020906T180308+0200, Fredrik Persson wrote: This may be slightly OT, but I have really looked around for a better place to ask this question, and failed. The FSF may be a better place. They have a mailing address for licensing questions but I forget what it was. That is my question. Who is right, Jim or Jill? I think it's a really important question, too. I shall explain why. I see your point. My initial answer would be that Jim is right, assuming that the offer is properly dated (as it should be). But this does not address the problem. -- Antti-Juhani Kaijanaho, LuK (BSc)* http://www.iki.fi/gaia/ * [EMAIL PROTECTED]
Re: A GNU GPL question (might be slightly OT)
Antti-Juhani Kaijanaho on Fri, Sep 06, 2002 at 07:52:06PM +0300 wrote: The FSF may be a better place. They have a mailing address for licensing questions but I forget what it was. [EMAIL PROTECTED] -- Spencer Hal Visick Please avoid sending me Word or PowerPoint attachments. See http://www.fsf.org/philosophy/no-word-attachments.html
Re: A GNU GPL question (might be slightly OT)
On Fri, 2002-09-06 at 11:03, Fredrik Persson wrote: Is this a loophole in the GPL? If my question above is answered with Jim, I think it is. If the answer is Jill, it most likely is not. So... What do you all say about this? I say that the answer is Jim, but that this is not as serious a problem as you make it sound. The GPL 3c only applies to non-commercial distribution. But I have brought it up here at the FSF, and I believe that future versions of the GPL will take it into account, by saying something like, c) Accompany it with the information you received as to the offer to distribute corresponding source code. (This alternative is allowed only for noncommercial distribution and only if you received the program in object code or executable form with such an offer, in accord with Subsection b above. The offer must still be valid) -- -Dave Turner Free Software Licensing Guru
Re: A GNU GPL question (might be slightly OT)
On Fri, 6 Sep 2002, Fredrik Persson wrote: I'm in a situation where I am trying to get the source code for a program from the company that distributed that program, and this has turned out to be really difficult. Currently, I'm preparing a reply to their lawyer (I have no legal training myself, so this is really difficult) where he talks to me about a three year rule within the GPL. I'm not a lawyer either, but I'll try to answer. stddisclaimer.h. 3b doesn't seem to have a loophole to me, but 3c might be susceptible in a very limited way. Jim gives Joe a program licensed under the GPL. Jim does not provide Joe with the source code, but with a written offer to provide that source code upon request. He can do this, according to section 3b in the GPL. However, that only requires Jim to comply with that offer for a period of three years, which is also stated in section 3b. So far so good. Two years later, Joe re-distributes the program to Jill, and he includes the written offer from Jim. Joe is required to do so, according to section 3c. At this point, as long as the Joe-Jill distribution was noncommercial, there has been no violation of GPL that I can see. At this point, Jill has a copy that she cannot distribute (She can't fulfill 3a or 3b, and 3c only applies to software she got under 3b. Her best bet is to immediately excercise the offer to get source, so she can distribute under 3a or 3b. Two more years pass and Jill decides that she wants that source code. Jill now has a copy of a program for which nobody has a legal obligation to provide source. So does Joe. Heck, so might Jim if he didn't save the source. This cannot be distributed under the GPL. It's possible to argue that Joe can still noncommercially distribute the sourceless binary under 3c, accompanied by the expired written offer from Jim. Nobody who recieved the binary this way could redistribute it (like always, as I read it, for programs distributed under 3c). Fortunately, it's limited to noncommercial transactions and it creates undistributable copies, so there is very little incentive to excercise this loophole. It does mean that Debian shouldn't accept software under 3b or 3c of the GPL. Which we probably wouldn't anyway for logistics reasons. -- Mark Rafn[EMAIL PROTECTED]http://www.dagon.net/
Re: Knuth statement on renaming cm files and Licence violation.
On Wed, 4 Sep, Brian Sniffen wrote: Sadly, I don't own a copy of Computers Typesetting. Can you quote the full copyright page, and give a general indication of the contents of Volume E? Somewhat surprisingly, no-one has done this completely yet. Computers Typesetting, Volume E, Computer Modern Typefaces, is a nicely typeset version of the complete source code to the Computer Modern fonts, typeset in the literate programming style with diagrams of the characters and explanatory notes. I would consider it a canonical distribution of the Computer Modern typefaces. Here is the complete copyright page. The most relevant section (the fourth paragraph) was already posted by Martin Schr??der. --- The quotations on pages 7 and 351 have been excerpted from the Electra file in the Dwiggins Collection of the Boston Public Library. METAFONT is a trademark of Addison-Wesley Publishing Company. TeX is a trademarke of the American Mathematical Society. The programs for Computer Modern are in the public domain, and readers may freely generate and hand-tune their own fonts using the algorithms of this book. However, use of the names is restricted: Any fonts whose names cmr10 or cmbx12 or ... are identical to the standard font names of this book should be fully compatible with the fonts defined here; i.e., fonts with the same names are supposed to have precisely the same character coding schemes and precisely the same font metric files. Library of Congress Cataloging-in-Publication Data Knuth, Donald Ervin, 1938- Computer Modern typefaces. (Computers Typesetting ; E) Includes indexes. 1. Type and type-founding--Data processing. 2. Printing--Specimens. 3. METAFONT (Computer system). 4. Computerized typesetting. I. Title. II. Series: Knuth, Donald Ervin, 1938-. Computers typesetting ; E Z253.K568 1986 686.2'2544 86-1235 ISBN 0-201-13446-2 Incorporates the final corrections made in 1992. Copyright (c) 1986 by Addison-Wesley Publishing Company, Inc. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the publishers. Printed in the United States of America. ISBN 0-201-13446-2 4 5 6 7 8 9 10 11 12 13-HA-9998979695949392 -- (The ... there is really in the original text.) (By the way, volumes B and D are about the TeX and METAFONT programs, respectively. They presumably have similar copyright pages, but I haven't checked.) The statement about the programs for Computer Modern is suprisingly vague for something that was presumably vetted by a lawyer. Let's see it again: The programs for Computer Modern are in the public domain, and readers may freely generate and hand-tune their own fonts using the algorithms of this book. This place Computer Modern in the public domain, and furthermore explicitly grants some of the rights Debian needs (although it leaves out the right to redistribute). However, use of the names is restricted: This is a slightly odd statement, since (AFAIK) names cannot be restricted in the ways that follow. The crucial issue seems to be whether this statement (and what follows) are terms of the grant of permission (above), or merely a request with no force in law. Any fonts whose names cmr10 or cmbx12 or ... are identical to the standard font names of this book should be fully compatible with the fonts defined here; i.e., fonts with the same names are supposed to have precisely the same character coding schemes and precisely the same font metric files. To back up the notion that this is merely a request, I note the words should and supposed. (I also note that there is no mention of specific _filenames_, merely the name of the _font_; and that there is some freedom to modify fonts under the same font name.) I think that, for the fonts to be distributable by Debian under this copyright notice, the statement about public domain has to be taken seriously, since otherwise there is no permission to distribute. Does Debian need legal advice on whether this statement actually places the files in the public domain? Or does it make more sense to approach Knuth directly? If we do approach Knuth, the letter should be carefully worded. Best, Dylan Thurston pgpdiA0RxI0W9.pgp Description: PGP signature
Re: Knuth statement on renaming cm files and Licence violation.
Dylan Thurston [EMAIL PROTECTED] writes: However, use of the names is restricted: This is a slightly odd statement, since (AFAIK) names cannot be restricted in the ways that follow. The crucial issue seems to be whether this statement (and what follows) are terms of the grant of permission (above), or merely a request with no force in law. The names could only be restricted if they are trademarked, which they are not. Computer Modern might be trademarked (I don't know), but cmr10 certainly is not. And, if it were trademarked, trademark restrictions never apply to functional elements.
Re: Knuth statement on renaming cm files and Licence violation.
On Fri, Sep 06, 2002 at 03:35:17PM -0700, Thomas Bushnell, BSG wrote: However, use of the names is restricted: This is a slightly odd statement, since (AFAIK) names cannot be restricted in the ways that follow. The crucial issue seems to be whether this statement (and what follows) are terms of the grant of permission (above), or merely a request with no force in law. The names could only be restricted if they are trademarked, which they are not. Computer Modern might be trademarked (I don't know), It is, as indicated in the text I quoted and you snipped. It's a trademark of Addison-Wesley. but cmr10 certainly is not. Right. And, if it were trademarked, trademark restrictions never apply to functional elements. ... like the file name 'cmr10.tfm'. (Just making everything explicit.) --Dylan pgpqH7RLjqHMA.pgp Description: PGP signature
Re: Knuth statement on renaming cm files and Licence violation.
On 2002-09-06 18:59:45 -0400, Dylan Thurston wrote: On Fri, Sep 06, 2002 at 03:35:17PM -0700, Thomas Bushnell, BSG wrote: The names could only be restricted if they are trademarked, which they are not. Computer Modern might be trademarked (I don't know), It is, as indicated in the text I quoted and you snipped. It's a trademark of Addison-Wesley. Uh? METAFONT is a TM of Addison-Wesley. What makes you think Computer Modern is a TM of Addison-Wesley? Best regards Martin -- http://www.tm.oneiros.de