Thomas Dalton writes:
If the contributions in question are extensive, that involves rewriting whole articles - ie. we lose enormous amounts of content.
I think this problem is overstated. Not all edits are additions of content.
Like I've said before, it only takes one.
I think running a project in fear of the one case where someone decides to impose massive litigation costs on us is probably not optimal decision-making strategy. There is so much we already do that poses greater risks of litigation that the migration would that I can't take seriously the proposition that a migration to new GFDL significant new problems.
So we can be reasonably confident about winning any such cases. Are you equally confident about being able to win full legal costs?
That's not how legal risk assessment works.
An American court would enforce a foreign law that doesn't exist in the US? Maybe. I have very little experience of the American legal system, however my experience of the US in general would suggest that they wouldn't give a damn about foreign laws...
Yes, an American court might enforce a foreign judgment that would have led to a different result if brought initially in an American court. That's first-year civil procedure (i.e., an early lesson in the basic training of every American law student).
--Mike
On 24/11/2007, Mike Godwin mnemonic@gmail.com wrote:
Thomas Dalton writes:
If the contributions in question are extensive, that involves rewriting whole articles - ie. we lose enormous amounts of content.
I think this problem is overstated. Not all edits are additions of content.
No one said all edits were additions of content. It's the ones that are (and there are a lot of them - all content was added at some point, after all) that will cause a problem.
Like I've said before, it only takes one.
I think running a project in fear of the one case where someone decides to impose massive litigation costs on us is probably not optimal decision-making strategy. There is so much we already do that poses greater risks of litigation that the migration would that I can't take seriously the proposition that a migration to new GFDL significant new problems.
The number of people doing something, and the likelihood of someone doing it are different things. I think there is a significant (while still probably quite low) chance of at least one person suing, and as I've said, that's all it takes to cause a very big and expensive problem.
I think there is a key part of the equation that we're missing - how big would the differences between the current GFDL and the new one be? If they're all very minor, there's a good chance no-one will bother doing anything, if they're quite major, it's a much bigger issue. Until we actually have a provisional new license in front of us, we really can't be sure. My guess (and that's all it is at this stage), is that the changes required to unify GFDL and CC-by-SA would be fairly large, and it would end up being dependant on people's views of the "spirit" of the license.
So we can be reasonably confident about winning any such cases. Are you equally confident about being able to win full legal costs?
That's not how legal risk assessment works.
Would you care to enlighten me? Risk assessment involves assessing potential damage and the chance of that happening. Is legal risk assessment somehow different? I can't see how it would be. Legal costs are just as damaging as any other similar sized cost.
Yes, an American court might enforce a foreign judgment that would have led to a different result if brought initially in an American court. That's first-year civil procedure (i.e., an early lesson in the basic training of every American law student).
I'm not saying it wouldn't be legal, I'm asking if it would actually happen.
On 25/11/2007, Thomas Dalton thomas.dalton@gmail.com wrote:
I think there is a key part of the equation that we're missing - how big would the differences between the current GFDL and the new one be? If they're all very minor, there's a good chance no-one will bother doing anything, if they're quite major, it's a much bigger issue. Until we actually have a provisional new license in front of us, we really can't be sure. My guess (and that's all it is at this stage), is that the changes required to unify GFDL and CC-by-SA would be fairly large, and it would end up being dependant on people's views of the "spirit" of the license.
The licence is a defence.
Plaintiff: "Defendant is using my copyright content unlicensed." Defendant: "Plaintiff licensed it under GFDL 1.2 Or Later, and I'm using it per GFDL 4.7, which is covered by Plaintiff's release." Plaintiff: "But 4.7's not in the same spirit as 1.2!"
I can't see this flying.
They'd need to bring an action against the FSF and have it stick before they could put those reusing under an "or later" provision in serious danger for doing so in good faith. I would think.
(not a lawyer, much less a US lawyer, but have been around this stuff way too long; correction welcomed from those who are)
- d.
The licence is a defence.
Plaintiff: "Defendant is using my copyright content unlicensed." Defendant: "Plaintiff licensed it under GFDL 1.2 Or Later, and I'm using it per GFDL 4.7, which is covered by Plaintiff's release." Plaintiff: "But 4.7's not in the same spirit as 1.2!"
I can't see this flying.
They'd need to bring an action against the FSF and have it stick before they could put those reusing under an "or later" provision in serious danger for doing so in good faith. I would think.
What action could they bring against FSF? The only claims I can see being relevant are claims of copyright infringement or contract violation, neither of which would (directly) involve FSF, since FSF doesn't own, or use, any of the content, and they are not a party to any of the contracts. The only way I can see the "same spirit" clause being enforced (and if it isn't, FSF could do whatever they like with anything released under GFDL, which would be very bad) is if it can be used as a counter argument for the defence you describe.
On Nov 25, 2007 2:33 PM, David Gerard dgerard@gmail.com wrote:
On 25/11/2007, Thomas Dalton thomas.dalton@gmail.com wrote:
I think there is a key part of the equation that we're missing - how big would the differences between the current GFDL and the new one be? If they're all very minor, there's a good chance no-one will bother doing anything, if they're quite major, it's a much bigger issue. Until we actually have a provisional new license in front of us, we really can't be sure. My guess (and that's all it is at this stage), is that the changes required to unify GFDL and CC-by-SA would be fairly large, and it would end up being dependant on people's views of the "spirit" of the license.
The licence is a defence.
Plaintiff: "Defendant is using my copyright content unlicensed." Defendant: "Plaintiff licensed it under GFDL 1.2 Or Later, and I'm using it per GFDL 4.7, which is covered by Plaintiff's release." Plaintiff: "But 4.7's not in the same spirit as 1.2!"
I can't see this flying.
But...it's a strawman. A better response would be "No I didn't license it under GFDL 1.2 Or Later; furthermore, even if I did, I retract it."
But...it's a strawman. A better response would be "No I didn't license it under GFDL 1.2 Or Later; furthermore, even if I did, I retract it."
But that would be simply untrue, on both counts. It clearly was licensed under GFDL 1.2 or later (unless "or later" clauses are found to be completely invalid, regardless of spirit, but that's a separate issue), since that's what it says right above the "Submit" button, and there are no legal means (to my knowledge) of retracting the GFDL.
On Nov 25, 2007 3:45 PM, Thomas Dalton thomas.dalton@gmail.com wrote:
But...it's a strawman. A better response would be "No I didn't license it under GFDL 1.2 Or Later; furthermore, even if I did, I retract it."
But that would be simply untrue, on both counts. It clearly was licensed under GFDL 1.2 or later (unless "or later" clauses are found to be completely invalid, regardless of spirit, but that's a separate issue), since that's what it says right above the "Submit" button, and there are no legal means (to my knowledge) of retracting the GFDL.
That's not what it says right above the "Submit" button. Above the "Save Page" button, on the English Wikipedia, it currently says: "Content that violates any copyright will be deleted. Encyclopedic content must be verifiable. You agree to license your contributions under the GFDL*." Then, at the bottom of the page it says "^ GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation; with no Invariant Sections, with no Front-Cover Texts, and with no Back-Cover Texts." And it hasn't always said this.
I certainly would not stipulate that clicking on such a button constitutes agreeing to a license. And actually, I don't recall ever having clicked on a button with such a text on that page, though it might be possible for you to gather records showing I had done so.
As for "or later" clauses being found completely invalid, regardless of spirit, that's not necessary. The GFDL license itself defines what "or later" means - "If the Document specifies that a particular numbered version of this License "or any later version" applies to it, you have the option of following the terms and conditions either of that specified version or of any later version that has been published (not as a draft) by the Free Software Foundation. If the Document does not specify a version number of this License, you may choose any version ever published (not as a draft) by the Free Software Foundation." The term "has been published" is sufficiently ambiguous that one could argue it only includes versions published at the time of the agreement, that would certainly make sense. Furthermore, the license implies that the "or later version" language should be specified in the Document itself, not agreed to by some clickthrough license.
As for retracting the GFDL, first of all, rights under the GFDL are automatically revoked if "Any other attempt to copy, modify, sublicense or distribute the Document" is made. Surely it could be shown that the WMF at one point made a copy or distribution outside the terms of the GFDL. The article-only dumps don't comply with the GFDL, do they?
Secondly, there is a significant question as to whether the GFDL is a contract or a waiver. If it is a waiver, it can be withdrawn at any time. If it is a contract, then there must be consideration, and any unconscionable parts would be void. Entering irrevocably into a contract which can be changed by a third party at any time with no possibility to opt-out and which does not require future contracts to be in the same spirit of the original contract is surely unconscionable.
In the end, I seriously doubt a court would find that by clicking on a button that says "Save Page" you have irrevocably agreed to do anything the FSF ever decides to say you've done. Maybe they'll deny punitive or even actual damages for past actions as long as the WMF is acting in good faith (*), but I don't see how they could allow the WMF to do anything it wants in the future simply because you clicked on some button ten years ago when the GFDL was completely different.
(*) Although, in a real world situation most likely a DMCA takedown notice would come first, which would make it much harder to claim that you weren't aware of the license revocation.
That's not what it says right above the "Submit" button. Above the "Save Page" button, on the English Wikipedia, it currently says: "Content that violates any copyright will be deleted. Encyclopedic content must be verifiable. You agree to license your contributions under the GFDL*." Then, at the bottom of the page it says "^ GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation; with no Invariant Sections, with no Front-Cover Texts, and with no Back-Cover Texts." And it hasn't always said this.
So you have to read the footnote, big deal. I don't know what it used to say, but I'm fairly sure it always said you were releasing it under the GFDL.
I certainly would not stipulate that clicking on such a button constitutes agreeing to a license. And actually, I don't recall ever having clicked on a button with such a text on that page, though it might be possible for you to gather records showing I had done so.
If clicking the button doesn't constitute agreeing the the license then the whole of Wikipedia is a copyright violation, and changing licenses isn't going to make a damned bit of difference. As for not having clicked the button - you must have never contributed to Wikipedia, then, because there's no other way of getting content on there (without direct database access, anyway).
As for "or later" clauses being found completely invalid, regardless of spirit, that's not necessary. The GFDL license itself defines what "or later" means - "If the Document specifies that a particular numbered version of this License "or any later version" applies to it, you have the option of following the terms and conditions either of that specified version or of any later version that has been published (not as a draft) by the Free Software Foundation. If the Document does not specify a version number of this License, you may choose any version ever published (not as a draft) by the Free Software Foundation." The term "has been published" is sufficiently ambiguous that one could argue it only includes versions published at the time of the agreement, that would certainly make sense. Furthermore, the license implies that the "or later version" language should be specified in the Document itself, not agreed to by some clickthrough license.
That's the first time I've ever heard that interpretation, and it's clearly not the intended one, since it would be a completely redundant clause.
As for retracting the GFDL, first of all, rights under the GFDL are automatically revoked if "Any other attempt to copy, modify, sublicense or distribute the Document" is made. Surely it could be shown that the WMF at one point made a copy or distribution outside the terms of the GFDL. The article-only dumps don't comply with the GFDL, do they?
You're misunderstanding that clause. The license to the person violating the license is revoked, not the license as a whole (so maybe the WMF could no longer use it, but a fork could carry on doing so), and you can't choose to revoke it, it either happens automatically, or it doesn't happen at all. And as far as I know, the article-only dumps do comply with the GFDL, is there a specific part you think is being violated?
Secondly, there is a significant question as to whether the GFDL is a contract or a waiver. If it is a waiver, it can be withdrawn at any time. If it is a contract, then there must be consideration, and any unconscionable parts would be void. Entering irrevocably into a contract which can be changed by a third party at any time with no possibility to opt-out and which does not require future contracts to be in the same spirit of the original contract is surely unconscionable.
It *does* require future contracts to be in the same spirit, what are you talking about?
On Nov 25, 2007 6:24 PM, Thomas Dalton thomas.dalton@gmail.com wrote:
That's not what it says right above the "Submit" button. Above the "Save Page" button, on the English Wikipedia, it currently says: "Content that violates any copyright will be deleted. Encyclopedic content must be verifiable. You agree to license your contributions under the GFDL*." Then, at the bottom of the page it says "^ GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation; with no Invariant Sections, with no Front-Cover Texts, and with no Back-Cover Texts." And it hasn't always said this.
So you have to read the footnote, big deal. I don't know what it used to say, but I'm fairly sure it always said you were releasing it under the GFDL.
It didn't always say Version 1.2, and it didn't always say "or any later version". Anyway, I expect your arguments to be accurate. If it doesn't say something "right above the submit button", but instead says it in a footnote, don't say it says it "right above the submit button". If it's no big deal, there's no need to exaggerate.
I certainly would not stipulate that clicking on such a button constitutes agreeing to a license. And actually, I don't recall ever having clicked on a button with such a text on that page, though it might be possible for you to gather records showing I had done so.
If clicking the button doesn't constitute agreeing the the license then the whole of Wikipedia is a copyright violation, and changing licenses isn't going to make a damned bit of difference.
That's probably true to some extent, although I'd say there is an implicit license given to the WMF when hitting submit. I'd say that implicit license is only valid if the WMF doesn't significantly alter what it's doing, though, and that it can be revoked at any time.
As for not having clicked the button - you must have never contributed to Wikipedia, then, because there's no other way of getting content on there (without direct database access, anyway).
Well, that's not true, as I can send a POST request to the server in many ways, but that's not what I was getting at. I was getting at the fact that I don't know for sure if the text read that 1.2 or any later version nonsense last time I hit submit. But for the purposes of argument, let's assume it did, and that the WMF can show that it did.
As for "or later" clauses being found completely invalid, regardless of spirit, that's not necessary. The GFDL license itself defines what "or later" means - "If the Document specifies that a particular numbered version of this License "or any later version" applies to it, you have the option of following the terms and conditions either of that specified version or of any later version that has been published (not as a draft) by the Free Software Foundation. If the Document does not specify a version number of this License, you may choose any version ever published (not as a draft) by the Free Software Foundation." The term "has been published" is sufficiently ambiguous that one could argue it only includes versions published at the time of the agreement, that would certainly make sense. Furthermore, the license implies that the "or later version" language should be specified in the Document itself, not agreed to by some clickthrough license.
That's the first time I've ever heard that interpretation, and it's clearly not the intended one, since it would be a completely redundant clause.
I've never heard that particular clause interpreted before, but "has been published" is what it says, and it's hard for me to see how it can be interpreted as "will be published in the future". It's only a redundant in certain instances. If I release something under GFDL 1.1 or any later version, it's not redundant.
As for retracting the GFDL, first of all, rights under the GFDL are automatically revoked if "Any other attempt to copy, modify, sublicense or distribute the Document" is made. Surely it could be shown that the WMF at one point made a copy or distribution outside the terms of the GFDL. The article-only dumps don't comply with the GFDL, do they?
You're misunderstanding that clause. The license to the person violating the license is revoked, not the license as a whole (so maybe the WMF could no longer use it, but a fork could carry on doing so),
In this particular hypothetical situation the defendant is the WMF, isn't it? If not, then, well, the whole part about "clicking on submit" is going to be a whole lot tougher, since presumably only the WMF and myself are parties to that agreement.
and you can't choose to revoke it, it either happens automatically, or it doesn't happen at all.
Right, this is an additional argument, though.
And as far as I know, the article-only dumps do comply with the GFDL, is there a specific part you think is being violated?
The current version article only dumps. Do they contain the GFDL? Do they have a compliant section entitled history? Do they even list all the authors, anywhere?
Secondly, there is a significant question as to whether the GFDL is a contract or a waiver. If it is a waiver, it can be withdrawn at any time. If it is a contract, then there must be consideration, and any unconscionable parts would be void. Entering irrevocably into a contract which can be changed by a third party at any time with no possibility to opt-out and which does not require future contracts to be in the same spirit of the original contract is surely unconscionable.
It *does* require future contracts to be in the same spirit, what are you talking about?
Sorry, I thought you were arguing that the future licenses don't even have to be in the same spirit.
By the way, once you've shown all of this, and that the WMF has received a license under the GFDL 4.7 (which didn't even exist at the time of the agreement), that this can't be revoked, and that it hasn't automatically been terminated, now the burden is on you to show that you're actually in compliance with the license.
Anthony wrote:
And as far as I know, the article-only dumps do comply with the GFDL, is there a specific part you think is being violated?
The current version article only dumps. Do they contain the GFDL? Do they have a compliant section entitled history? Do they even list all the authors, anywhere?
I expect them to be. You can provide a GPL program in a binary only form, *if you make available the source code by the same way*. So if i can download history dumps and article dumps, and choose the dump without the history, it's not WMF's fault. The problem would arise because history dumps fail very often for large wikis. About containing the GFDL, the en: dump will have the GFDL at [[Wikipedia:Text_of_the_GNU_Free_Documentation_License]], which is also interesting because the GFDL text itself is not under the GFDL.
On Nov 27, 2007 5:46 PM, Platonides Platonides@gmail.com wrote:
Anthony wrote:
And as far as I know, the article-only dumps do comply with the GFDL, is there a specific part you think is being violated?
The current version article only dumps. Do they contain the GFDL? Do they have a compliant section entitled history? Do they even list all the authors, anywhere?
I expect them to be. You can provide a GPL program in a binary only form, *if you make available the source code by the same way*. So if i can download history dumps and article dumps, and choose the dump without the history, it's not WMF's fault.
The GPL and GFDL are different licenses, and the requirement to provide a transparent copy (which would be analogous to the GPL requirement to provide source) is not the same as the requirement to list authors in the section entitled history. I don't think it's GPL compliant to provide a GPL program in binary form without at least listing the authors somewhere. I'm not sure if the GPL requires you to include a copy of the license or not.
The problem would arise because history dumps fail very often for large wikis.
That'd be another problem.
About containing the GFDL, the en: dump will have the GFDL at [[Wikipedia:Text_of_the_GNU_Free_Documentation_License]], which is also interesting because the GFDL text itself is not under the GFDL.
I wasn't sure if it was in the article only dump, which is why I asked, but I guess it is.
wikimedia-l@lists.wikimedia.org