Re: DFSG analysis of default LDP license
On Sat, May 17, 2003 at 01:52:57AM -0400, Anthony DeRobertis wrote: On Tue, 2003-05-13 at 02:41, Branden Robinson wrote: Colin Watson helpfully provided this information in a recent mail: 4. The location of the original unmodified document be identified. I feel that this clause might be problematic in a way that clauses 1, 2, and 3 would not be, in that the information in 1, 2, and 3 cannot become false over time. First, it doesn't have to be a network location. So I think we could distribute as original + patches; the location we'd point to would be the .orig file in the pool. Uh, the pool *is* a network location. Or at the very least, it's not something on the installed system. 5) applicable, non-redundant disclaimers of endorsement [3] That's not compelled speech --- I can always remove it (and often have to, if I change the document) if I disapprove of it. You didn't read what I said or follow the link, apparently. I said *disclaimers* of endorsement. BECAUSE THE CONTENT OF THE WORK IS FREELY MODIFIABLE BY ALL THIRD PARTIES, THERE IS NO WARRANTY THAT ANY REPRESENTATIONS MADE WITH IN ARE MADE BY, ON BEHALF OF, OR WITH THE CONSENT OF THE AUTHOR(S) OR COPYRIGHT HOLDER(S). ANY STATEMENTS MADE WITHIN THE WORK ARE NOT NECESSARILY HELD, SHARED, OR ENDORSED BY THE AUTHOR(S) OR COPYRIGHT HOLDER(S). Like that. Surely that must stay in just as warranty disclaimers must, and for similar reasons, no? Endorsement statements themselves must be severable, and indeed serve little purpose if they aren't, and the work is otherwise modifiable. (Sure, I'll endorse ANYTHING!) I recommend dropping this clause. 5. The original author's (or authors') name(s) may not be used to assert or imply endorsement of the resulting document without the original author's (or authors') permission. I don't think it's so critical it be dropped; I think it just restates what the law already does. If modified a work to include hateful propaganda, and didn't make it clear that is not the original author's opinion, I'd be in trouble for defamation. I didn't say it was critical. I'm just identifying a way I think the license can be approved. As I said in my reply to Joey Hess, the BUGs I identified are gray areas, not flagrant violations of the DFSG. But I like clearing up gray areas where possible, because lawyers who write licenses tend to use every square in of gray area as an excuse to write a license clause that takes a square mile. E.g., (36pt) Some Document [ illustration, blank space, whatever ] (16pt) by J. Hacker (9pt) with modifications by others and then putting all those others in the full copyright statement in 8pt type is likely to still be defamation. I'm not really in a position to opine on defamation law, and furthermore I don't find it very relevant to the default LDP license, which doesn't even raise the issue. I think your remark here would be better placed in the droit d'auteur thread. BECAUSE THE CONTENT OF THE WORK IS FREELY MODIFIABLE BY ALL THIRD PARTIES, THERE IS NO WARRANTY THAT ANY REPRESENTATIONS MADE WITH IN ARE MADE BY, ON BEHALF OF, OR WITH THE CONSENT OF THE AUTHOR(S) OR COPYRIGHT HOLDER(S). ANY STATEMENTS MADE WITHIN THE WORK ARE NOT NECESSARILY HELD, SHARED, OR ENDORSED BY THE AUTHOR(S) OR COPYRIGHT HOLDER(S). Not many end-users of a published book will read that. So what? Copyright cartels and shrink-wrap license freaks have been arguing successfully for years in U.S. courts that ignorance of the fine print is no excuse. So, either your point is irrelevant (in the U.S.), or we might get to enjoy the spectacle of the aforementioned special interests reversing themselves someday. It won't much save someone's reputation. The only way this works is if its very close to the places where the authors' names are mentioned. How can you possibly know this? Why is putting this material where the disclaimers of warranty go insufficient? Point me to some case law supporting you assertions about what is and is not a sufficient means to avoid a defamation suit. -- G. Branden Robinson|I have a truly elegant proof of the Debian GNU/Linux |above, but it is too long to fit [EMAIL PROTECTED] |into this .signature file. http://people.debian.org/~branden/ | pgplJMNb1seQe.pgp Description: PGP signature
Re: DFSG analysis of default LDP license
On Tue, 2003-05-13 at 02:41, Branden Robinson wrote: Colin Watson helpfully provided this information in a recent mail: 4. The location of the original unmodified document be identified. I feel that this clause might be problematic in a way that clauses 1, 2, and 3 would not be, in that the information in 1, 2, and 3 cannot become false over time. First, it doesn't have to be a network location. So I think we could distribute as original + patches; the location we'd point to would be the .orig file in the pool. 5) applicable, non-redundant disclaimers of endorsement [3] That's not compelled speech --- I can always remove it (and often have to, if I change the document) if I disapprove of it. I recommend dropping this clause. 5. The original author's (or authors') name(s) may not be used to assert or imply endorsement of the resulting document without the original author's (or authors') permission. I don't think it's so critical it be dropped; I think it just restates what the law already does. If modified a work to include hateful propaganda, and didn't make it clear that is not the original author's opinion, I'd be in trouble for defamation. E.g., (36pt) Some Document [ illustration, blank space, whatever ] (16pt) by J. Hacker (9pt) with modifications by others and then putting all those others in the full copyright statement in 8pt type is likely to still be defamation. BECAUSE THE CONTENT OF THE WORK IS FREELY MODIFIABLE BY ALL THIRD PARTIES, THERE IS NO WARRANTY THAT ANY REPRESENTATIONS MADE WITH IN ARE MADE BY, ON BEHALF OF, OR WITH THE CONSENT OF THE AUTHOR(S) OR COPYRIGHT HOLDER(S). ANY STATEMENTS MADE WITHIN THE WORK ARE NOT NECESSARILY HELD, SHARED, OR ENDORSED BY THE AUTHOR(S) OR COPYRIGHT HOLDER(S). Not many end-users of a published book will read that. It won't much save someone's reputation. The only way this works is if its very close to the places where the authors' names are mentioned. signature.asc Description: This is a digitally signed message part
Re: DFSG analysis of default LDP license
On Wed, May 14, 2003 at 08:20:57PM -0400, Joey Hess wrote: Well, I'm willing to go along with this, but it means adding yet another exception to our no invariant text rule, in addition to the five I already enumerated. I'm having a hard time with the idea of calling a peice of text that we're allowed to keep up to date invariant. Maybe unremovable or something. Okay. I'm more concerned with the concept than the terminology for the purposes of this discussion. We can probably come up with requirements for unremovable text that would be so onerous as to be effectively non-free, but the line is in a different place than the line for invariant texts. Sure. I'd rather see this clause clarified until it can do some good, or stricken entirely so it doesn't serve as bad precedent for more invariant text requirements. Unfortunatly given what Colin said about the LDP being tired of hearing about copyright issues from him, they don't seem very likely to listen or clarify the license. And I would hardly rate this amoung the worst written licenses in Debian. Licenses seem to either be written by a non-lawyer, and thus be vague, self contradicting, and prone to misinterpretaton, and thus suck; or be written by a lawyer, and thus be impossible to understand if you're not one, possibly contain poison pills, and skirt the very edges of the DFSG, and thus suck. :-P I don't disagree with your remarks, but as I said, I felt it was only fair to subject other licenses to same degree of scrutiny as we do the GNU FDL. Just as we should not accept a license from the FSF that we'd reject from anyone else, we shouldn't reserve heightened scrutiny for FSF licenses -- even if the FSF should know better than to publish licenses that abrogate people's freedoms. I think my critiques of the default LDP license are pretty minor and should be addressable without acrimony. If that's not the case then we collectively problem have a problem with this upstream. Are we really sure this is the case? -- G. Branden Robinson| Why do we have to hide from the Debian GNU/Linux | police, Daddy? [EMAIL PROTECTED] | Because we use vi, son. They use http://people.debian.org/~branden/ | emacs. pgp2h52NA9Stz.pgp Description: PGP signature
Re: DFSG analysis of default LDP license
On Tue, May 13, 2003 at 05:35:43PM -0400, Joey Hess wrote: Branden Robinson wrote: [... mail rearranged...] I don't see any flagrant DFSG violations in the above license, but I think some requests for clarification might be a good idea. So you're using BUG to mean this should probably be fixed and not this is non-free. Ok. Correct. I should have been clearer about that. I would have said PROBLEM if I'd seen what I considered an outright DFSG problem as opposed to just something gray that made me a little nervous. [...] 4. The location of the original unmodified document be identified. BUG: Walter Landry has pointed out: [The GNU FDL] requires me to preserve the network location of where Transparent versions can be found for four years. Even if it is no longer correct, and the original author can not be reached. This is probably not uncommon. This does not raise the quality of free documentation.[2] I feel that this clause might be problematic in a way that clauses 1, 2, and 3 would not be, in that the information in 1, 2, and 3 cannot become false over time. (If a document is eventually wholly rewritten, the original author's copyright no longer attaches anyway.) It's annoying, but does not really make it not free, I hope. Remember that we dealt with the FSF snail mail address changing; said address is in the GPL and is in copyright statements that point to the GPL. Many licenses and statements of copyright contain information that will become obsolete, email addresses are another good example. Anyway, this license seems to leave plenty of room for modifying the pointer to the original unmodified document. Unlike the GPL, which may or may not be modifiable (the FSF address is in the preamble, and also in the example use bit at the end, we've heard conflicting things about the preamble modification). And if all other upstream sources go away, I think the license even allows us to change the location pointer to point to packages.debian.org, where after all you can get the .orig.tar.gz. Well, I'm willing to go along with this, but it means adding yet another exception to our no invariant text rule, in addition to the five I already enumerated. I guess it seems vaguely justifiable under the spirit of DFSG 4, given that diff/patch format may not be useful for some works. However, I'd *still* like clarification of the above clause. If the location of the original unmodified document is identified but becomes obsolete, and the modifier is under no obligation to contact all people who recieved modified copies from him and inform them of an updated location[1], then what good does this clause really do? What's the distributing modifier to do when the location of the original unmodified document isn't even under his control? I'd rather see this clause clarified until it can do some good, or stricken entirely so it doesn't serve as bad precedent for more invariant text requirements. [1] such a requirement would be too onerous to be free, I think, especially given that people who distribute Freely-licensed works often have no idea who obtain them -- G. Branden Robinson|Kissing girls is a goodness. It is Debian GNU/Linux |a growing closer. It beats the [EMAIL PROTECTED] |hell out of card games. http://people.debian.org/~branden/ |-- Robert Heinlein pgpBOG7uGdMnT.pgp Description: PGP signature
Re: DFSG analysis of default LDP license
On Tue, May 13, 2003 at 01:41:30AM -0500, Branden Robinson wrote: BECAUSE THE CONTENT OF THE WORK IS FREELY MODIFIABLE BY ALL THIRD PARTIES, THERE IS NO WARRANTY THAT ANY REPRESENTATIONS MADE WITH IN ARE MADE BY, ON BEHALF OF, OR WITH THE CONSENT OF THE AUTHOR(S) OR COPYRIGHT HOLDER(S). ANY STATEMENTS MADE WITHIN THE WORK ARE NOT NECESSARILY HELD, SHARED, OR ENDORSED BY THE AUTHOR(S) OR COPYRIGHT HOLDER(S). Hmm. s/WITH IN/WITHIN/ -- G. Branden Robinson|It may be difficult to to determine Debian GNU/Linux |where religious beliefs end and [EMAIL PROTECTED] |mental illness begins. http://people.debian.org/~branden/ |-- Elaine Cassel pgpVst5WY5rnF.pgp Description: PGP signature
Re: DFSG analysis of default LDP license
Branden Robinson wrote: It's annoying, but does not really make it not free, I hope. Remember that we dealt with the FSF snail mail address changing; said address is in the GPL and is in copyright statements that point to the GPL. Many licenses and statements of copyright contain information that will become obsolete, email addresses are another good example. Anyway, this license seems to leave plenty of room for modifying the pointer to the original unmodified document. Unlike the GPL, which may or may not be modifiable (the FSF address is in the preamble, and also in the example use bit at the end, we've heard conflicting things about the preamble modification). And if all other upstream sources go away, I think the license even allows us to change the location pointer to point to packages.debian.org, where after all you can get the .orig.tar.gz. Well, I'm willing to go along with this, but it means adding yet another exception to our no invariant text rule, in addition to the five I already enumerated. I'm having a hard time with the idea of calling a peice of text that we're allowed to keep up to date invariant. Maybe unremovable or something. We can probably come up with requirements for unremovable text that would be so onerous as to be effectively non-free, but the line is in a different place than the line for invariant texts. I'd rather see this clause clarified until it can do some good, or stricken entirely so it doesn't serve as bad precedent for more invariant text requirements. Unfortunatly given what Colin said about the LDP being tired of hearing about copyright issues from him, they don't seem very likely to listen or clarify the license. And I would hardly rate this amoung the worst written licenses in Debian. Licenses seem to either be written by a non-lawyer, and thus be vague, self contradicting, and prone to misinterpretaton, and thus suck; or be written by a lawyer, and thus be impossible to understand if you're not one, possibly contain poison pills, and skirt the very edges of the DFSG, and thus suck. :-P -- see shy jo pgpehvPr83Xsr.pgp Description: PGP signature
DFSG analysis of default LDP license
Colin Watson helpfully provided this information in a recent mail: The default LDP licence (one which many documents use explicitly and which it's been agreed applies to any documents which don't specify a licence) is a little longer but still quite reasonable, and is also more or less a copyleft. Counting the defaulted documents, it's the most popular licence among the (mini-)HOWTOs, beating GFDL-1.1-no-invariants by a whisker. So, we should probably be fair and subject this license to a DFSG analysis just as we have the GNU FDL. II. LICENSE The following license terms apply to all LDP documents, unless otherwise stated in the document. The LDP documents may be reproduced and distributed in whole or in part, in any medium physical or electronic, provided that this license notice is displayed in the reproduction. Commercial redistribution is permitted and encouraged. Thirty days advance notice via email to the author(s) of redistribution is appreciated, to give the authors time to provide updated documents. I see no problems with the above. The final sentence would be a violation of DFSG 1, failing the desert island test, if it were a requirement rather than a request, but it is not. A. REQUIREMENTS OF MODIFIED WORKS All modified documents, including translations, anthologies, and partial documents, must meet the following requirements: BUG: The requirements should apply to distribution of modified works, not the act of modifying in and of itself, which may not be within the scope of copyright law in some jurisdictions. In the U.S., for instance, one is entitled to arbitrarily modify or deface copies of a copyrighted work in one's possession, such as a Harlequin romance novel or a Dixie Chicks music CD. Furthermore, under the First Sale doctrine (also a U.S. concept), it is not an infringement of copyright to sell or give away such a modified/defaced copy to another party. How the First Sale doctrine interacts with Free Software licenses and other public licenses has not, to my knowledge, been analyzed in depth on debian-legal. It probably merits some thought in a different thread. I recommend rewriting the above as: Modified copies of a document under this license (including translations, anthologies, and partial documents) may be distributed provided each copy meets the following requirements: 1. The modified version must be labeled as such. First Sale analysis notwithstanding, this is not a problem. DFSG 4 recognizes the importance of distinguishing the work of the original author from the work of others, even if the body of DFSG 4, in my opinion, only clumsily gropes around the issue, which I think of as the endorsement principle. 2. The person making the modifications must be identified. First Sale analysis notwithstanding, this is not a problem. This will happen anyway if the modifier is creating a derivative work under U.S. copyright law and wishes to assert his or her copyright. 3. Acknowledgement of the original author must be retained. My analysis of clause 1 applies to this as well. It is interesting that this license requires retention of original-author acknowledgements in modified copies whereas, depending on if and how one uses an Endorsements section, the GNU FDL requires deleting them! 4. The location of the original unmodified document be identified. BUG: Walter Landry has pointed out: [The GNU FDL] requires me to preserve the network location of where Transparent versions can be found for four years. Even if it is no longer correct, and the original author can not be reached. This is probably not uncommon. This does not raise the quality of free documentation.[2] If this argument holds water, I do not see why it wouldn't apply to any license making the same basic requirement, not just the GNU FDL. I feel that this clause might be problematic in a way that clauses 1, 2, and 3 would not be, in that the information in 1, 2, and 3 cannot become false over time. (If a document is eventually wholly rewritten, the original author's copyright no longer attaches anyway.) The Debian Project appears to be moving towards a consensus that compelled speech is incompatible with Free Software. Many of us are only willing to draw fairly narrow exceptions to that rule: 1) applicable, non-redundant copyright notices 2) applicable, non-redundant notices of authorship 3) applicable, non-redundant licensing terms for copyrights, patents, and trademarks 4) applicable, non-redundant statements of warranty (or disclaimers thereof) 5) applicable, non-redundant disclaimers of endorsement [3] [Does anyone on the -legal list have any suggestions for additions to the above list?] I recommend dropping this clause. 5. The original author's (or authors') name(s) may
Re: DFSG analysis of default LDP license
Branden Robinson wrote: 4. The location of the original unmodified document be identified. BUG: Walter Landry has pointed out: [The GNU FDL] requires me to preserve the network location of where Transparent versions can be found for four years. Even if it is no longer correct, and the original author can not be reached. This is probably not uncommon. This does not raise the quality of free documentation.[2] I feel that this clause might be problematic in a way that clauses 1, 2, and 3 would not be, in that the information in 1, 2, and 3 cannot become false over time. (If a document is eventually wholly rewritten, the original author's copyright no longer attaches anyway.) It's annoying, but does not really make it not free, I hope. Remember that we dealt with the FSF snail mail address changing; said address is in the GPL and is in copyright statements that point to the GPL. Many licenses and statements of copyright contain information that will become obsolete, email addresses are another good example. Anyway, this license seems to leave plenty of room for modifying the pointer to the original unmodified document. Unlike the GPL, which may or may not be modifiable (the FSF address is in the preamble, and also in the example use bit at the end, we've heard conflicting things about the preamble modification). And if all other upstream sources go away, I think the license even allows us to change the location pointer to point to packages.debian.org, where after all you can get the .orig.tar.gz. I don't see any flagrant DFSG violations in the above license, but I think some requests for clarification might be a good idea. So you're using BUG to mean this should probably be fixed and not this is non-free. Ok. -- see shy jo pgp0sCGMAuGO9.pgp Description: PGP signature
