Talk:Licensing update

From Meta, a Wikimedia project coordination wiki

Archive Archives
1 |

Although this is a current discussion, some of it has been archived! (Note the archive link(s) to to the right.) Also, some discussion has been filed under Questions and Answers/Oppositionalarguments. (With this note, I'm simply trying to avoid duplication and obfuscation, which is why I'm putting this at the top.) --Elvey 03:22, 25 April 2009 (UTC) Reply

The current version contains a link to CC BY-SA 3.0 unported. I assumed the license to be migrated into would be CC BY-SA 3.0 US, which was mentioned in the Foundation's resolution foundation:Resolution:License_update.

Sorry if I missed any discussion on this, but unported license is more of a prototype to work on than a finished license, so I thought it might be better to choose US or let each local wikis chose the easiest jurisdiction-specific license for them to deal with.

Tomos 15:35, 1 March 2009 (UTC) Reply

NO! There should be one global license. Yet ANOTHER reason this is a terrible idea! -24.98.65.137 11:02, 28 March 2009 (UTC) Reply

Mr. 24.98.65.137, when you find an all-in-one, global, free license that would serve the best interest of the Wikimedia project, please tell us. The way I see it, the GFDL isn't providing enough for us, so the CC-by-sa license is needed as well (just my opinion). --Michaeldsuarez 18:49, 29 March 2009 (UTC) Reply

It might be stating the obvious, but many CC licenses include the provision that unless agreed mutually in writing, the license is the whole of the agreement. No interpretation or understanding about the license exit.

See, for example, 8.d. of CC-BY-SA 3.0 unported [1] or 8.e. of CC-BY-SA 3.0 US [2]

So there is a chance, I suppose, that provision on how to properly attribute might turn out to be wrong as an interpretation of the CC license.

I do see that having some guideline would be helpful for downstream users, but we (including the Foundation) certainly do not want to give false impression that certain attritbution practice complies with the license where such certaintly does not exist. Tomos 16:27, 1 March 2009 (UTC) Reply

I don't think anyone is suggesting that we have certainty in our interpretation. Considering that wikis are a very unique class of intellectual property, however, it seems pertinent that we offer at least some basic guidance on how we think these licenses actually apply to our content. This is not a new practice for Wikipedia. We have plenty of guidance on how to interpret the GFDL, fair use, and other legal concepts for which we are not considered an actual authority. Kaldari 19:12, 1 March 2009 (UTC) Reply

Clearly, further analysis is necessary before leaping! -24.98.65.137 11:01, 28 March 2009 (UTC) Reply

when we vote this?

I think that double licensing leaves out many contents licensed under GFDL that might be added to Wikisource, a double licensing is good idea for Wikipedia, but a projects like Wikisource not. Other example, Wikinews, is currently not GFDL, so i have a question,

Is double Licensed good Idea for all projects? I think that might do exceptions for imported contents in some projects like Wikisource: if the content to import is GFDL, could clarify that is not licensed doubly. After all, in Wikisource all the content is imported, for example a something in the public domain can not be licensed under GFDL or CC.

For other side, in Wikinews in spanish we import from a Newspaper with http://creativecommons.org/licenses/by/2.1/es/ , question: this licence remain compatible? If not, then we lose an important resource. Regards Shooke 19:13, 1 March 2009 (UTC) Reply

The only case in which having GFDL content on WikiSource would be a problem is if that content was imported into WikiSource after November 1, 2008. Personally, I'm not aware of any content on WikiSource that has been imported from a GFDL source that was not another Wikimedia project. Do you have any examples? Indeed, this should actually make it easier to import more work into WikiSource since Creative Commons works will now be allowed. Public domain sources will still be treated the same - they will not be licensed at all and will appear with a template notifying the viewer that the work is public domain. I don't have enough knowledge of WikiNews to answer your other question. Also, the vote will be held some time before April according to the WMF. Kaldari 19:25, 1 March 2009 (UTC) Reply

A survey regarding the attribution issue has been started. Please register your opinion there. Kaldari 00:22, 3 March 2009 (UTC) Reply

It looks like it's not available anymore, is it already over? --Elitre 15:48, 8 March 2009 (UTC) Reply

Yep, it was just a small "feeler"-survey. Not the official vote or anything. Cbrown1023 talk 16:04, 8 March 2009 (UTC) Reply

Yes, I found some answers on foundation-l. And can't believe my eyes. --Elitre 16:10, 8 March 2009 (UTC) Reply

I'm adding Erik Moeller's message about results and next steps. --Elitre 17:44, 8 March 2009 (UTC) Reply

«If you make modifications or additions to the page or work you re-use, you must license them under the Creative Commons Attribution-Share-Alike License 3.0 or later»: that's not true, one can choose the GFDL if the content is available under the GFDL, can't he?
Attribution: what does «where possible» mean? I can add a link on a book or a DVD for offline use, but readers would not be able to reach the page or history, while it wouldn't normally be too onerous to list all authors (the real problem is to include a full copy of the GFDL, especially on gadgets such this: see last page). --Nemo 10:17, 15 March 2009 (UTC) Reply

Should we note in our footer something along the lines of (Edits produced prior to %IMPLEMNETATIONDATE% are released solely under the GNU Free Documentation License) for at least some period? xaosflux ^Talk 03:55, 17 March 2009 (UTC) Reply

Why? If the transition occurs, all past and future edits (with the exception of certain previously published and imported content) will be GFDL + CC-BY-SA. Dragons flight 04:06, 17 March 2009 (UTC) Reply

Thanks, forgot this was also a relicensing effort, I did read over this enwp: recap, and looks like there may be some timelines in play? (e.g. Must be complete by 01AUG2009, may not cover content made between 3NOV2008 and current). We we require a means to not relicense the last few months worth of contributions? (Will this cause document forks?)

The November date affects previously published GFDL-only content that is imported. Normal edits, i.e. original content created by users, will all be relicensed regardless of date. Imported GFDL-only text added between Nov 3rd and the implementation date will probably need to be deleted. Hopefully this is only a tiny precentage of all content. Dragons flight 14:45, 17 March 2009 (UTC) Reply

Thanks for the quick replies DF! xaosflux ^Talk 01:34, 18 March 2009 (UTC) Reply

So you're going to DELETE it all? Why is this needed again? JUST ANOTHER EXAMPLE of how terrible of an idea this is! -24.98.65.137 10:59, 28 March 2009 (UTC) Reply

Don't be silly. Please see wikipedia:Special:Log/import. The English Wikipedia imports very little and not very recently. --Michaeldsuarez 18:44, 29 March 2009 (UTC) Reply

Implement a bugtraq:? Is there a bug open to include a specific means to mark future contributions licensing status, as we will be multi licensing by default, but only single licensing certain types of contributions? xaosflux ^Talk 10:53, 17 March 2009 (UTC) Reply

Not to my knowledge. Our goal is that all text content will by CC-BY-SA. Most text will also be GFDL. The current proposal places the responsibility on the reuser to figure out whether a document contains CC-BY-SA-only text or not. Since it only affects imported previously published content, one should be able to identify such imports from the history if they are annotated the way they ought to be. Individual communities can augment this if they wish, for example by adding a CC-BY-SA-only Category tag to articles subject to that restriction. Dragons flight 14:55, 17 March 2009 (UTC) Reply

I've created a chart comparing GFDL and CC-BY-SA: Licensing update/License comparison. Dragons flight 06:23, 17 March 2009 (UTC) Reply

I welcome this change. Well done! – Kaihsu 00:10, 20 March 2009 (UTC) Reply

It is known that Wikipedia follows strictly statist and capitalist rules. It means that Wikipedia won't import articles with incompatible license. But if a project with such license allows Wikipedia to do it (the project promise that it won't take the matter to court), will Wikipedia import or not? Caesarion 06:16, 20 March 2009 (UTC) Reply

License compatibility needs to be ensured both for us and for the downstream users of Wikimedia content. If you have a specific proposal in mind, I'm sure we'd look at it, but our requirements are pretty firm for importing external content. Dragons flight 08:19, 20 March 2009 (UTC) Reply

If the project can "promise that it won't take the matter to court", then they must have the legal ownership of the text and can put it under an applicable license as well as whatever it is now. But I agree with DF, we should always err on the side of caution and legality – the text on Wikipedia is intended for other use, so "we won't sue Wikipedia" is not good enough. Cbrown1023 talk 01:07, 21 March 2009 (UTC) Reply

Wikipedia isn't "statist and capitalist". It is simply obeying the laws, as they exist. If it doesn't obey the laws, it can be shut down, and many of us won't be able to use the results (and thus would work against the whole point of Wikipedia). Dwheeler 14:57, 14 April 2009 (UTC) Reply

In one place it says "we will strive to indicate clearly to you on the article or the description page for the file" but in other places it says this can be done in the page history. It needs clarifying which one is actually the case. In reality, it needs to be the latter as in-article attribution gets removed as "spam" on en.wp now. Angela 06:58, 24 March 2009 (UTC) Reply

This has been addressed in the text. Dragons flight 19:41, 27 March 2009 (UTC) Reply

Having thought about the issue a bit, it seems that, in the absence of a perfect solution (the hope fo which is always justification to do nothing, however awkward the status quo), it seems that dual licensing is about the best that can be done in this situation. I cannot see any harm to Wikiprojects, and there is the possibility of Wikiimprovement. I do wonder whether the server storage capacity planning will be able to keep up though...

There are some glitches in the proposed policy formulations and so there should be a serious improvement, bug detection pass, before adoption by the Board. Or perhaps I missed the opportunity to tune up the phrasing? Ww eng 08:43, 25 March 2009 (UTC) Reply

As noted, the WMF reserves the right to tune things for clarity and elaboration, but the essence of the proposal is essentially fixed. So if you have suggestions for clarifications and better wording feel free to offer them (though they may or may not be acted upon until after the vote). Changes in the structure of the proposal are probably out of the question unless this proposal ends up being rejected by either the community or the Board of Trustees. Dragons flight 07:51, 27 March 2009 (UTC) Reply

Has anyone on the Wikimedia Foundation Board ever indicated that they might not or do not support the transition to GFDL 1.3? 69.228.88.127 00:47, 28 March 2009 (UTC) Reply

I am unaware of any public comments to that effect, though there have been statements that the Board is unlikely to adopt this should it fail to achieve majority support in the editing community. Dragons flight 03:38, 28 March 2009 (UTC) Reply

Let's have a poll then. I think a day or two should be sufficient to show near unanimity, as long as we advertise on the largest Village Pump(s). People with concerns have had ample time to share them already, and if there are any unresolved concerns, I haven't seen them. 69.228.88.127 08:30, 28 March 2009 (UTC) Reply

A poll will start soon, but a day or two is way to short :) Huib talk 18:50, 29 March 2009 (UTC) Reply

I'm closing this. Per the timeline, the official community vote is intended to start in less than a week. There is no reason to clutter things with an unofficial poll right now. If people want to continue to express opinions, such as the one in the section below, I am okay with that, but no voting please. -- Dragons flight 17:00, 28 March 2009 (UTC) Reply

This is a TERRIBLE idea. It only confuses the hell out of licensing. I, for one, would sue if you misappropriated my content by relicensing it under something other than the GFDL. I'd spend every dollar I had fighting your changes to licensing of my contributions (from many IP addresses all over Georgia Tech). I would cease making edits and tell every person on campus to do the same. If the community really understood what havoc you're causing with this change, they'd be in an uproar! -24.98.65.137 10:53, 28 March 2009 (UTC) Reply

Could you describe the havoc you are issuing legal threats about? ▫ JohnnyMrNinja (talk / en) 18:26, 28 March 2009 (UTC) Reply

All talk, but no proof or examples? Where's your explanation? --Michaeldsuarez 22:33, 28 March 2009 (UTC) Reply

You might want to save all those dollars you had, because according to U.S. copyright law you will need to wait, "35 years after the execution of the grant or, if the grant covers the right of publication, no earlier than 40 years after the execution of the grant or 35 years after publication under the grant (whichever comes first)," before you can revoke the grant that you made when you submitted your work. 69.228.190.230 03:49, 29 March 2009 (UTC) Reply

But we're not revoking the GFDL; the content will still be licensed under the GFDL. Anyway, you should read more about the GFDL (specifically, Section 11); the FSF is allowing its clients to switch to or duel-license with a Creative Commons license. --Michaeldsuarez 18:26, 29 March 2009 (UTC).Reply

If relicensing is a problem, would the proposed Libre Puro approach work for new resources? These new resources could be relicensed under an existing free license.

If so, what would it take to make it official? K 08:05, 13 May 2009 (UTC) Reply

Here's what Wikimedia Foundation General Counsel Mike Godwin says:

GFDL 1.2 expressly says this: "Each version of the License is given a distinguishing version number. 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." Seems clear to me that there's no need to amend 1.2 to allow use of 1.3. (That's certainly what the FSF intended.)

HowDoIUseUnifiedLogin? 21:19, 30 March 2009 (UTC) Reply

I have this wonderful Firefox add-on that helps me translate stuff called Foxlingo. Check this out for example (and bear in mind that I don't speak Spanish).
Centralnotice in English: ==qqq== {{cn translation status|qqq|ready}} * Please participate in a vote to determine the future copyright terms of Wikimedia projects (vote ends April 23, 2009). * Vote now!
Centralnotice in Spanish: ==qqq== {{traducción estado cn | qqq | listo}} * Por favor, participar en una votación para determinar el futuro del derecho de autor de los proyectos Wikimedia (votación termina 23 de abril de 2009). * Votar ahora!
Let me know if there's anything wrong with the translation. Whip it! ^{Now whip it good!} 03:49, 31 March 2009 (UTC) Reply

Thanks, but we'd rather not use internet translators. :-) Usually the result is not free (some are copyrighted based on the program used to derive it) and not as good as it could be with a real person. Cbrown1023 talk 20:45, 31 March 2009 (UTC) Reply

Second part of second point claims the relicensing is about avoiding FUD while the text itself is FUD. It should be rephrased to say why the new licensing scheme is better instead of making unverified claims about the old one. There should also be something substantial about the legal part of relicensing content, everything I've seen so far is basically "we changed the GFDL so we do whatever we want". Jeblad 10:40, 31 March 2009 (UTC) Reply

This page is essentially brief; the Licensing update/Questions and Answers goes into significantly more detail.--Eloquence 16:16, 10 April 2009 (UTC) Reply

Second point (Terms on edit screen) says that all new content is dual licensed, while point four says it may be licensed CC-by-sa only. Perhaps the text could be modified to say that the text uploaded on this specific edit screen is dual licensed but content uploaded through other means may be CC-by-sa only.

Yes, I think we could try to insert a very simple clarification regarding CC-BY-SA-only content there.--Eloquence 16:43, 10 April 2009 (UTC) Reply

Third point (attribution) says very small contributons may be filtered out, I like that as it avoid the opposite to identify the main authors, yet it may be necessary to say something that such filtering should make some safeguards as to satisfy any local laws about identification of main authors. That is, filtering out contributions less that 100 chars in a 500 chars article is not the same as filtering out contributions of 100 chars in a 50K article

Fundamentally, when people create author lists, they must apply their own judgment as to who to (not) include. I think that at least the standardized licensing terms aren't the place for this, though it's a question that's well suited to the development of community guidelines. That said, I don't anticipate that this will be much of a problem as most people will probably choose to attribute by URL instead.--Eloquence 16:43, 10 April 2009 (UTC) Reply

Third point (attribution of rich media) says media created through substantive collaborations can be treated as text. What is the real meaning behind "created", is it the creation of the original (musicians recorded) or the creation of the edited copy (someone postediting the recording)? The first one seems acceptable to me, while the second really are about attribution of a derivative work. The later may also be used to bypass attribution of the original creator.

It's intended to be the latter for contributions made through Wikimedia. However, given the restriction to substantive collaboration, the attribution-by-URL wouldn't be possible for mere attempts to inflate the author count without also substantially and usefully deriving a new work. The intent here is to develop language which can work well with future massive collaboration scenarios around rich media.--Eloquence 16:43, 10 April 2009 (UTC) Reply

Third point (attribution of externally contributed content) should say something about what is externally contributed as almost everything will somehow be related to the community. One possible interpretation is that "externally contributed content" is something not produced for Wikimedia, which is a very difficult interpretation as it is impossible to verify, an other one is that copyright/ownership of such content is held by a non-person (ie a museum).

A very good point; I agree that we should try to come up with a good definition of what "external" means here. I would suggest a definition that essentially refers to content that the uploader "cannot with reasonable effort obtain under terms equivalent" to those required of all Wikimedia contributors. I think we can resolve this question collectively after the licensing update as it relates to the general question of what our policy on externally contributed content is. The LU doesn't answer that question; it merely defines basic parameters of the general Wikimedia approach to licensing.--Eloquence 16:43, 10 April 2009 (UTC) Reply

False. Under your proposed terms of use external content cannot be introduced into wikipedia.Geni 11:27, 12 April 2009 (UTC) Reply

Erik, this point needs to be clarified. While there seems to be a general agreement to te CC-BY-SA licensing of our content, I think that introducing two categories of contributors ("normal contributors" and "third parties") is unacceptable. What do I have to do to become a "third party". Decision of this cannot be postponed after the vote.  « Saper  // @talk  »  11:18, 14 April 2009 (UTC) Reply

Third point (share alike) makes it possible to make content that may be impossible to reimport given one of the possible interpretations of point two.

It will definitely be possible to re-add CC-BY-SA only content (see above), and there shouldn't be any obligations for continued dual-licensing by third parties. I am open to adding this as a suggestion, though.--Eloquence 16:43, 10 April 2009 (UTC) Reply

An that seems to be all from me! Jeblad 11:03, 31 March 2009 (UTC) Reply

If the licensing is changed i would advocate to an additional change. In fact to the unkown authorship, it is currently not clear how IP-contributions are to be handled. A new license should be aware of this problem. If the foundation irrecovable would be representative for all IP-contributions all licensing questions could be handled without respect to unknown authors. --Mijobe 14:30, 9 April 2009 (UTC) Reply

In practice, where IP authors are listed, we've generally credited them as "x anonymous authors". Crediting any authors except by means of a hyperlink is not mandatory, but it can't hurt to just express the aforementioned terms in a couple of places explicitly. I don't consider that an essential change and am comfortable making it after the vote if we go ahead with the implementation (since we're pretty much ready to go, I want to avoid additional burden on translators at this stage).--Eloquence 16:14, 10 April 2009 (UTC) Reply

Realistically saying "this edit/content was written by Badmouse923" and "this edit/content was written by 76.101.42.5" seem (superficially at least) equal as to attribution. If a selected random string of text can be deemed to legally meet a need to "identify" an authorthen, if that author chooses to edit under a random string of 4 binary numbers, that too "identifies" an author, at least in principle. That is how they have chosen to be identified, or their preferred attribution, for that edit. FT2 ^{(Talk | email)} 23:06, 13 April 2009 (UTC) Reply

As an IP contributor with well over 25 edits at multiple IPs (not this one, it's a Blackberry), I am quite apalled that you are not giving me the opportunity to vote. The GFDL requires you to ask the whole community, not just those who have given you their contact information. 25 edits from an IP should let that IP in to vote, period. If someone logs in and nullifies that vote with a "shared IP, and I disagree" button, then I suppose you are left without a vote for that IP. Until then, you will count my vote or I will never forget. I am not anonymous, I am the IPs I come to your server from every day. If I could afford to sue, I would. You cannot have my content under your terribly written, faulty CC license! -206.53.144.51 11:43, 19 April 2009 (UTC) Reply

But this is the issue. Even looking at your contributions to all Wikimedia projects, we see only 14 edits. On one project 'you' seem to be a small-time vandal, yet on another 'you' are a constructive contributor. I expect that these two 'people' are not the same person: you do not vandalise on one wiki and edit constructively on another. But if we allow 'you' to vote from an IP, which person do we see on the voting page? The one whose thoughts and opinions we value, or the one we'd much rather didn't edit Wikipedia at all? You claim to have other contributions on other IPs, but how can we confirm this? It is simply not possible.

If you wish to sue, we cannot stop you. But you were very clearly presented with a statement that read "you irrevocably agree to release your contributions under... the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation...". Version 1.3 of the GNU Free Documentation License, as published by the Free Software Foundation, states that "The operator of [a wiki site licensed under this license] may republish [the wiki content] contained in the site under CC-BY-SA on the same site at any time before August 1, 2009..." Now I'm not a lawyer, but that position of yours doesn't look very legally strong to me. Happy‐melon 17:17, 19 April 2009 (UTC) Reply

As of today, the suggested text for the footer says:

All text is available under the Creative Commons Attribution/Share-Alike License. Text may also be available under the GNU Free Documentation License. See ((link to copyright policy)) for further details. See the page history for a list of authors. Media files are available under different licenses; click the file for more information.

I believe "file" is misleading here. The notice does not primarily refer to displaying entire media object in the File: namespace, but to embedded media aggregated with a normal page. In the latter case, "click on the file for more information" makes no sense.

I propose:

... Media objects are available under different licenses; click the image, video, etc. for more information. --G.Hagedorn 08:12, 10 April 2009 (UTC) Reply

How about "click the embedded media file for more information"?--Eloquence 16:08, 10 April 2009 (UTC) Reply

I think "file" does not make sense here from a consumers point of view. The image is simple a part of the page on the screen - the fact that in html this is realized by relating to multiple files is a very technical perspective. Perhaps media "object" or "item"? --G.Hagedorn 20:11, 12 April 2009 (UTC) Reply

---

Related: is the statement true at all? It seems true for images only, but not for sound or movies. --G.Hagedorn 20:11, 12 April 2009 (UTC) Reply

Dear colleagues,

I suppose that we should explicitly specify versions of the licenses in the Terms of Use texts.

It should be "1.2 or later" for GFDL and "3.0" for CC-BY-SSA.

Dr Bug (Vladimir V. Medeyko) 13:22, 10 April 2009 (UTC) Reply

Yes, you're correct. Since we're about ready to start the vote and this isn't an essential change, we'll defer it to later, but all references to the licenses will be versioned.--Eloquence 16:05, 10 April 2009 (UTC) Reply

Sorry because I don't know exactly where to put these comments in, but I think there is error in vote server's authentication. I just tried to vote for the 2nd time for fun, but the server accepted, it didn't show me an error as I expected. Can anyone else confirm this error and file it to technical group? Vinhtantran 01:34, 13 April 2009 (UTC) Reply

It cancels your prior vote and replaces it. Dragons flight 02:33, 13 April 2009 (UTC) Reply

At least it should inform users about the replacement. I remembered I had translated those kind of messages in Translatewiki.net. Vinhtantran 07:50, 13 April 2009 (UTC) Reply

I get the following message trying to vote Set $wgShowExceptionDetails = true; at the bottom of LocalSettings.php to show detailed debugging information.

Retrieved from "https://wikimedia.spi-inc.org/index.php/Special:SecurePoll"

I have to post here anonymously, but I'm registered at the English Wikipedia as user oaktree_b, and should be eligible to vote. (I don't want to create an account here just for this problem). Please advise, is that a browser problem, or is the website at fault? --204.153.84.10 14:06, 22 April 2009 (UTC) Reply

Someone should reply, I want to vote but can't because of the error above. Please advise.--204.153.84.10 16:31, 30 April 2009 (UTC) Reply

It is a problem with the vote software. It was a reported quite a while ago, but as far as I know has not been diagnosed. You are only the third individual who has reported this error (versus 15000+ ballots successfully cast). When it occurs the error appears to persistent and fatal, and we don't have any technical solution. If you want to email me your preference, I'd be willing to manually add it to the final count. Dragons flight 18:41, 30 April 2009 (UTC) Reply

The aforementioned header on the English Wikipedia isn't there. 85.196.111.46 03:06, 13 April 2009 (UTC) Reply

And yes, logging in doesn't work either. 85.196.111.46 03:08, 13 April 2009 (UTC) Reply

I was able to find the header by enabling JavaScript, but apparently I need to "check my spelling", because my user doesn't exist. 85.196.111.46 03:15, 13 April 2009 (UTC) Reply

Well, what do you know, suddenly it started working. 3rd time's the charm! 85.196.111.46 03:18, 13 April 2009 (UTC) Reply

Here, look up a few section for answers. OhanaUnited ^{Talk page} 04:12, 13 April 2009 (UTC) Reply

I added a link to the actual vote page (diff). This was curiously missing. By the way, it would be nice if the SecurePoll page would link back to Licensing_update, so people coming there directly can find more info easily. -- Duesentrieb 11:35, 13 April 2009 (UTC) Reply

Thanks for fixing that link. Huib talk 11:38, 13 April 2009 (UTC) Reply

Hm, got reverted on that [3]. Why do I need to come to the page via my "home wiki", and what does that even mean?... Maybe links for the largest wikis could be offered? or at least there could be an explanation how to find the vote page. As it is, it's really confusing. I come to the page, read it, figure out i want to vote, but it doesn't tell me how. That's pretty bad. -- Duesentrieb 12:08, 13 April 2009 (UTC) Reply

A link definitely needs to be here to the vote, and from the sitenotice and vote page to here. The link to the meta page will work if you have sufficient edits here (fairly likely due to the low number required); otherwise the home wiki link needs to be used it seems (although in these days of Single User Login, I'm not quite sure why). A small table of links might be useful, if that's the only way to do it. Mike Peel 12:13, 13 April 2009 (UTC) Reply

(ec) Your right to vote (e.g. having 25 contributions, etc.) is checked against the local wiki when you visit Special:SecurePoll on your home wiki. Having a link here would only work for Meta Wikipedians. I suppose we could have a list of local links, but that would be a very long list. Dragons flight 12:18, 13 April 2009 (UTC) Reply

Ok then. Let#s just have that explanation in a prominent spot, with the instructions "go to Special:SecurePoll/vote/1 on your home wiki". But that should really bet there. Telling people they can vote, but not where and how, is really confusing. -- Duesentrieb 12:31, 13 April 2009 (UTC) Reply

I find the global notice for the licensing change is likely to cause a little confusion on Wikinews, where the license is CC-BY (yes, the plain CC-BY with no suffix).

The actual vote page specifies that it is all currently GFDL projects, but it does not anywhere specify that there are exceptions, and what they are. I feel that either a Wikinews-specific global message would be appropriate, or something on the page it takes you to.

It is indeed in the interest of Wikinewsies to see the change go through, the compatibility it would give with WP material would be most useful. --Brian McNeil / ^talk 16:44, 13 April 2009 (UTC) Reply

It does affect Wikinewsies insofar as they have uploaded GFDL (without "1.2 only") multimedia directly to Wikinews or Commons, but sure, feel free to reword the notice on Wikinews to be more clear for its particular context. :-) --Eloquence 18:40, 13 April 2009 (UTC) Reply

The instructions on how to make an attribution when reusing a Wikimedia project text say: "Attribution of text: To re-distribute an article page in any form, provide credit to the authors either by including a) a hyperlink (where possible) or URL to the article or articles you are re-using ... or c) a list of all authors"

URLs are often long and obscure, especially when the article is written in a non-Latin alphabet. For example, this is the URL of the history page of the article about the Wikimedia Foundation in the Hebrew Wikipedia: http://he.wikipedia.org/w/index.php?title=%D7%A7%D7%A8%D7%9F_%D7%95%D7%99%D7%A7%D7%99%D7%9E%D7%93%D7%99%D7%94&action=history. It is very easy to get this URL wrong when citing the article, and it doesn't say much to the average reader.

The list of authors includes at least 10 major contributors, and if exclude only the really minor edits, we'll get to 15 or maybe even more. These contributors identify themselves in all kind of nicks, using different set of characters, and some of them are known only by an IP address.

I think we should allow people to give an attribution by stating the name of the article, the specific project name, and the date and time of the cited version, e.g. "Taken from: קרן ויקימדיה, The Hebrew language Wikipedia, 5 April 2009 01:09". In case non-Latin characters are unavailable to the redistributer, I would even allow: "Taken from the article about the Wikimedia Foundation on the Hebrew language Wikipedia, 5 April 2009 01:09", or alternatively using a known transliteration, e.g. "qrn vyqmdyh (transliterated), The Hebrew language Wikipedia, 5 April 2009 01:09". Dror_K 17:34, 13 April 2009 (UTC) Reply

http://he.wikipedia.org/wiki/קרן_ויקימדיה is a functioning URL, however, and also http://he.wikipedia.org/w/index.php?title=קרן_ויקימדיה&action=history. --Nemo 18:46, 13 April 2009 (UTC) Reply

These are not legal URLs though and not every medium will support these characters; furthermore, unwieldy URLs are inappropriate in various settings, such as twitter or Braille or spoken text. I agree with Dror that, in order to maximize reuse, we should interpret the attribution requirement of CC-BY-SA as liberally as possible: clear identification of the project, article and version are sufficient. AxelBoldt 22:47, 13 April 2009 (UTC) Reply

I tend to think that you are getting too liberal. The attribution clauses requires a reasonable effort to identify the authors. This could be done directly (e.g. listing authors) or indirectly (e.g. tell someone where to find the authors), but if you reduce the reference to just "project, article and version" with no additional explanation, then that strikes me as largely inadequate under CC-BY-SA, especially if the space available would allow for a more direct reference (such as a url). Project + article + version identifies a document, which might lead one to the authors if you know how, but it requires extra effort and particular knowledge. Under most circumstances I would expect "reasonable" attribution to be more direct than that. One could add an explanation of how to find authors given the article and version number, but I don't think this is what you had in mind. You could debate that project + article + version is "reasonable" under some tightly constrained circumstances, but I would consider it unreasonable under many others. Hence I am plainly against the idea that project + article + version is an adequate general purpose form of attribution given the expectations of CC-BY-SA.

You and the original poster should keep in mind that the attribution advice and terms of service being offered by Wikipedia are meant to express recommended behavior that should always be okay. The content is still covered by CC-BY-SA, and reusers can still choose to interpret and utilize that license in other ways besides those that we recommend, but in that case they will bear greater responsibility for ensuring the license is used appropriately under their particular circumstances. Dragons flight 23:34, 13 April 2009 (UTC) Reply

This could be made explicit in the language: author list or URL are recommended attribution methods in case they are reasonable for the intended use case. AxelBoldt 00:24, 14 April 2009 (UTC) Reply

The common practice when citing academic articles is writing one or two major authors and adding "et al." (i.e. "and others") or giving an exact reference to a book/an article and writing "various authors". Do you really suggest that the cc-by-sa license goes beyond that to demand a full list of all authors? Dror_K 12:13, 14 April 2009 (UTC) Reply

We're not talking about citation. You can cite a Wikipedia article the same way you cite any other website or article. Only when reusing a Wikipedia article in a way that would be illegal with an ordinary article/website you have to obey to the stricter rules mentioned here. --217.225.245.54 12:55, 14 April 2009 (UTC) Reply

As I see it, this need to attribute all authors who have made significant contributions is, in a project like Wikipedia, likely to be more onerous (for some pages) than distributing a copy of the GFDL. I've just taken a look at the contributors to the main Linux page - there are over 200 authors who have made 5 or more edits. I'm voting nay. Robhogg 23:29, 14 April 2009 (UTC) Reply

I voted for the change. However, the thought occurs to me, if the media is dual licensed, doesn't that mean that we keep the old problems with the GDFL and attribution while potentially adding new problems. My vote is still case for the change, but this nagging thought does worry me a bit. 98.119.207.176 21:14, 13 April 2009 (UTC) Reply

Any given reuse will generally only need to follow one or the other, and hence they can choose whichever creates fewer problems within that application. Dragons flight 22:11, 13 April 2009 (UTC) Reply

I personally have the feeling, that dual-licensing of Wikipedia diminishes it's potential. If I have found usable content, released under CC-BY-SA or GFDL only, we can't add it to this project. It has to be dual-licensed! I'd prefer a total relicensing to CC, as the GFDL v1.3 suggested. I didn't find the answer to this question anywhere, but I hope someone can explain, why we would want to stick with the GFDL (other than for historic reasons).--Yamavu 13:41, 14 April 2009 (UTC) Reply

No, we can add CC-BY-SA-only text, but not GFDL-only text. See also here. --Nemo 14:30, 14 April 2009 (UTC) Reply

Nemo, to me, that is exactly the problem. Who cares what you can import or not? Or to be more specific, who cares if someone is not able to import CC_BY_SA stuff. If the content (I don't like this word) is ABSOLUTELY essential, then go ahead and use it as fair use. That does not hinder our project. If it is not absolutely critical, don't use it! Recreate it. Its not like someone has a patent on ideas. Why complicate matters by adding stuff like "content which includes imported CC-BY-SA-only information cannot be used under the GFDL." What next? Kushal one 13:51, 15 April 2009 (UTC) Reply

Sorry, but we have to respect the law. Fair use has to be fair, and fair use content is not free content, while Wikipedia is free... --Nemo 21:10, 15 April 2009 (UTC) Reply

It is fair if there is compelling reason to use it. (Am I missing something?) Kushal one 05:13, 21 April 2009 (UTC) Reply

The comparisons seem self-explanatory, with the exception of the "Open file formats" where it says simply "doesn't exist under CC-BY-SA". But this seems a non-trivial section.

It guarantees that anyone republishing the content beyond a fairly tiny level, must also ensure the text is available for easy public access by anyone, and remain so for a significant period after publication ceases. That's a very non-trivial requirement.

I'd value an explanation why this is as it is (it seems a strange removal of what is a useful, possibly significant, and negligible effort to comply with, requirement), and what factors go into this change. Thanks.

FT2 ^{(Talk | email)} 23:11, 13 April 2009 (UTC) Reply

I agree it is a non-trivial difference. I can't give a definitive explanation, because I was not privy to the considerations that went into formulating CC-BY-SA. However, reading the license suggests a likely guess for what is going on.

CC-BY-SA is meant to be super general. In theory it can apply not only to text, but also art, music, interpretive dance, and absolutely any other creative effort where copyright can apply. In some cases the author's means of communication may even be quasi-ephemeral (e.g. spoken word presentations, live concerts). In order to accomodate that generality, CC-BY-SA avoids making any assumptions about the nature of the content or how it will be initially communicated. If one is going to work from that starting position, then it is impossible to add requirements that presume the existance of any files at all. Hence, my guess is that the authors of CC-BY-SA chose to forgo the requirements for machine readable documents in open formats because those requirements weren't generalizable to the range of works that they wanted to cover.

This is perhaps the one major area where the fact that the GFDL was tailored to written works has actually been an obvious advantage. GFDL requires machine readable copies. The reduced constraints of CC-BY-SA allow licensed written works to be distributed simply as fixed print with no files at all. Dragons flight 00:18, 14 April 2009 (UTC) Reply

If that's so, I'd have expected some kind of requirement that IF the source material existed in [durable or] electronic form, and more than X copies were distributed, then there is an obligation to make reasonably available the source material for a period, as it was required under GFDL. Obviously a bit late to add it in now, but thats the one obvious point that to my mind should have somehow been in there, because it guarantees that if you widely circulate a work or adaptation or something that is in a suitable form for this, then the source material is certain to be available for a while too. Disappointed that such a strong and valuable point for preserving the record, that is anticipated in other licences under discussion in this context, is not carried over into CC-BY-SA. Too late now? FT2 ^{(Talk | email)} 01:45, 14 April 2009 (UTC) Reply

Yeah, not much that can be done now. It is already moot for content added to Wikipedia, since anything added here will already be electronic. It might have been nice to have such clauses to cover downstream reuse though. Incidentally, there is also something of a philosophical divide between FSF and CC that may be relevant here. FSF is generally focused on spreading maximal freedom through the application of copyleft. CC is generally more concerned with providing authors with flexibility to decide how their works are used (hence the large number of CC licenses and the options for things like NC and ND). Since CC tends to be more author-centric, rather than free-centric, perhaps that influenced their decision not to have burdensome format and access requirements.

Incidentally, there has been talk of creating a CC-BY-SA-strong license that would provide a stronger copyleft and might address this issue. (Though the primary motivation is actually related to issues with how SA media interacts with text.) I'm not sure if that will ever come to pass, and even if it did I'm not sure we would necessarily use it. Dragons flight 03:59, 14 April 2009 (UTC) Reply

I'm curious how it would be possible to distribute text in a non-open format.  — Mike.lifeguard  | ^@en.wb 01:56, 14 April 2009 (UTC) Reply

For the purposes of GFDL, an open format is one whose specification is publicly known and where editors capable of handling that format are widely available. Proprietary and obscure document formats would be considered non-open. Dragons flight 03:59, 14 April 2009 (UTC) Reply

Right, so how does that relevantly apply to text?  — Mike.lifeguard  | ^@en.wb 02:39, 15 April 2009 (UTC) Reply

For example, if I give you a printed book under the GFDL, then I am also required by the GFDL to make available a machine readable electronic copy of that book. That copy must be in an open format that can be widely read. I could not, for example, use a proprietary format and charge you 50,000ドル for the software needed to read it. A software file protected in that way would not meet the machine readable distribution requirements. Dragons flight 03:19, 15 April 2009 (UTC) Reply

"Dragons flight"; speaking on your example, if I give you a printed book under the CC, then I can, for example, use a proprietary format and charge you 50,000ドル for the software needed to read it? --Mahmudmasri 20:20, 17 April 2009 (UTC) Reply

CC has no machine-readable file requirement at all, so you would be under no obligation to give me any file (for 50ドルk or any other price). Once I recieve the printed book, that is the end of it. Dragons flight 20:32, 17 April 2009 (UTC) Reply

(削除) (English) Hey, Not link in the page and I visit Special:SecurePoll/vote/1 and reject me, i have more 20000 edits in Wikipedia in spanish and work in several Wikimedia projects like Wikinews Commons, etc . Other users can have same problems and not become to vote. Consider delay this, Shooke 23:50, 13 April 2009 (UTC) (削除ここまで)Reply

(削除) (Spanish) Hey, no hay un link en la página y yo visité Special:SecurePoll/vote/1 y me rechazó, tengo más de 20000 ediciones en Wikipedia en español y trabajo en varios proyectos de Wikimedia como Wikinews. A Otros usuarios tal vez le ocurra lo mismo y no vuelvan más a votar. Consideren en retrazarla hasta que funcione bien el sistema de votación. Shooke 23:50, 13 April 2009 (UTC) (削除ここまで)Reply

Visiting SecurePoll at Meta only works for established editors at Meta. You need to visit SecurePoll at ES, i.e. http://es.wikipedia.org/wiki/Special:SecurePoll/vote/1 The correct link should also appear on the sitenotice shown at ES, etc. Dragons flight 00:22, 14 April 2009 (UTC) Reply

Thanks i can vote!! Shooke 16:37, 14 April 2009 (UTC) Reply

Why invite everybody, only to tell most of them that they are rejected? Since there must be a list somewhere with the few who actually are invited, why not contact them more them directly instead of pestering everybody and then insulting the vast majority of them when they try to accept the invitation? At the very least the banner splashed to everybody could have had a warning that only a small group was eligible and which group that was. Aaron Walkhouse 11:22, 15 April 2009 (UTC) Reply

Did you try to vote on Meta or English Wikipedia? I believe you should be able to vote if you use the English Wikipedia since your account has +100 edits and is created in 2006. Your meta account has only one edit and that is not enough. Huib talk 12:45, 15 April 2009 (UTC) Reply

I think the point isn't that this person can't vote but that the banner is extremely misleading.

I am curious about a couple of the technical details of the voting process. I did my thing and voted and got my PGP vote receipt, but is there anything I can do with this receipt? Can I do some kind of verification? Or can someone else verify my vote? Poking with GnuPG reveals that it was encrypted with public key 35A64240B98FD85E. This doesn't seem to be very useful. The public key isn't on any of the key servers and I can't find it anywhere else. If I did find it, I don't see what I could do with it anyway. What's the point of the receipt? I can't find any information about it. --Mosquitopsu 15:55, 14 April 2009 (UTC) Reply

Talk:Licensing_update/Bugs#Vote_receipt. Dragons flight 16:01, 14 April 2009 (UTC) Reply

Thanks! That's what I was looking for. --Mosquitopsu 16:38, 14 April 2009 (UTC) Reply

So... this means there is nothing we can do with the signature :-( --WikiSven 18:29, 18 April 2009 (UTC) Reply

I believe that it is not the right way for us to allow contributors to allow to license their contributions under CC only.

"to inform re-users that content which includes imported CC-BY-SA-only information cannot be used under the GFDL". seems to imply that.

What about requiring contributors to license under both licenses from now on? Could someone enlighten me about it? Kushal one 23:25, 14 April 2009 (UTC) Reply

Do you see that, according to the proposed terms of use, when you save your edit, you are explicitly asked to license your contribution under both licenses? The first chunk says that. If you are just adding your work to Wikipedia, it must be dual-licensed, as I understand from this. But then if absolutely everything has to be dual-licensed, we cannot bring in other people's works even when they are licensed under CC-BY-SA or GFDL. The part you are concerned addresses this question, and says that you can add works of others if they are under CC-BY-SA.

Although I have no authority on this, or source to back up my point, I think this reading makes good sense. Hope it helps. Tomos 10:59, 15 April 2009 (UTC) Reply

Well actually, reading Nemo's answer in the section #Dual licensed, doesn't that mean we keep the old problems while adding new? might help.. Tomos 11:09, 15 April 2009 (UTC) Reply

for me "content which includes imported CC-BY-SA-only information cannot be used under the GFDL" is the bone of contention. We should stand firm on this. We must MAKE contributors dual-license everything and disallow "imported CC-BY-SA-only information". Maybe someone will come tomorrow and say "well, I have a great license, XYZ. GFDL is incompatible with my XYZ license. So, if I add a sentence from my "scholarly work" that is under XYZ, I can make the whole article, including work of hundreds of volunteers not available under what they chose to make it available as ... Even if my "scholarly work" gets disputed and removed, the article from that point is only available under XYZ. For the win! Kushal one 16:05, 15 April 2009 (UTC) (If the above is not how it is, I apologize. Please enlighten me.)Reply

That's not possibile, if XYZ is not CC-BY-SA. All contributors will have to release all ther contribution under both CC-BY-SA and GFDL, but if e.g. Encarta is released under CC-BY-SA-only we will able to import its content, and the articles edited in this way will be available under CC-BY-SA-only (except for the old revisions). --Nemo 22:06, 15 April 2009 (UTC) Reply

Thanks for your reply. I agree. Still, it seems we are effectively transitioning from GFDL to CC. (Agreed?) I used the vague XYZ because it is hard to argue against the well-intentioned people who created CC. Kushal one 22:20, 16 April 2009 (UTC) Reply

I've made well over 25 anonymous edits, many significant, yet I cannot vote. Pretty lame requirement.

Could you suggest a better alternative? Kushal one 16:05, 15 April 2009 (UTC) Reply

Hello, there is no way to check if you did the 25 edit's because ip numbers can change from owner. Only registerd users can vote. Sorry, Huib talk 17:21, 15 April 2009 (UTC) Reply

I believe making anonymous edits amount to or should amount to relasing one's work to public domain - and I don't see why IP users should have a say here.--194.182.87.97 17:07, 16 April 2009 (UTC) Reply

I am unhappy with this:

provide, reasonable to the medium or means You are utilizing: (i) the name of the Original Author (or pseudonym, if applicable) if supplied, and/or if the Original Author and/or Licensor designate another party or parties (e.g., a sponsor institute, publishing entity, journal) for attribution ("Attribution Parties") in Licensor's copyright notice, terms of service or by other reasonable means, the name of such party or parties;

because, wth is the "and/or" for? If the licence is satisfied by "or", what is the "and" there for in the first place? And the "or" will mean that this parses as "provide some party mentioned in Licensor's copyright notice". This means that the original author can legally drop out of the attribution after a single iteration. Say "licensor" Wikipedia states "from Wikipedia", and this is imported by "Crappypedia" with the credit "from Wikiepia" (no authors) and the addition "from Crappypedia", then anyone can take the text from Crappypedia and simply say "from Crappypedia". This is not what I bargain for when I expect attribution.

The second objection would be "reasonable to the medium you're utilizing". This means you can get rid of any attribution by going through some low bandwith channel. The upshot of this is, in my understanding, that the "Attribution" advertised in the licence isn't worth shit.

I would welcome this as a good idea in principle, but the "reasonable" and "and/or" clause renders this licence worthless in my book. --92.104.139.142 17:26, 15 April 2009 (UTC) Reply

I don't believe you're interpreting the clause correctly. You are the Original Author of any contributions you make. You are also the Licensor. You have complete control over both of the and/or clauses. And Wikimedia is not asking you to "designate another party or parties" for credit. In your example, "Crappypedia" is not the Licensor, they are a Licensee; while they can insert the additional requirement that any re-reusers also attribute "from Crappypedia", they cannot legally remove the attribution requirements that have been put in place upstream. You as the ultimate Licensor will impose the requirements for attribution. Wikipedia, as the primary Licensee, will provide guidelines for how the authors of a page can be best attributed, by a link to the original Wikipedia page, a link to a stable copy of the page which itself links to the original list of authors, or a complete list of authors. These requirements cannot be removed by downstream parties. So Crappypedia could require that any reusers of its content must also include a tagline "from Crappypedia", but they must include the original links to the Wikipedia authors. Reusers of our content do not suddenly get to rewrite the ground rules when re-relicensing it for re-reusers; all downstream parties are bound by the requirements put upon them by the ultimate copyright holder.

With regards your second point, there is no medium where it's possible to reuse the content without it being possible to provide attribution. Even if you were sending a page by semaphore you would have enough bandwidth for it to be "reasonable" to send over the URL of the original article, and 'unreasonable' not to. There are media in which there is no "reasonable" way to provide attribution. It's not possible in those media for our content to be reused in the first place. Happy‐melon 11:34, 16 April 2009 (UTC) Reply

Without them I get an "No account data received. In order to vote in this election you must visit Special:SecurePoll at your home wiki. It will verify that your account has the proper qualifications to vote and then transfer you here with that account information attached. You can not vote at this page directly." error which doesn't mention cookies. -- Jeandré, 2009年04月15日t18:44z

Cookies are required to log into Wikipedia at all. It shouldn't be surprising that the voting site also uses them. It would be a very unusual scenario to have cookies enabled for Wikimedia but not for other sites, and not one I'm inclined to worry about. Dragons flight 19:06, 15 April 2009 (UTC) Reply

I missed the grey text in a box at the top of pages for several days, only finding the vote link after reading the Signpost. I suggest removing the box at least. -- Jeandré, 2009年04月15日t18:44z

Why not just send a bot message to all members? People tend to notice new message banners at least. Noian 00:35, 16 April 2009 (UTC) Reply

Making the notice the same color as the message template would produce a similar result. I think, however, people are nervous about the %.01 that flip-the-hell-out because of "spam" or "annoying banners". ▫ JohnnyMrNinja (talk / en) 04:03, 16 April 2009 (UTC) Reply

The french wikipedia doesn't have (and never had) the "or any later version" mention, so the GFDL 1.2 still applies. So, no update to GFDL 1.3, and the update to CC By-Sa can't be possible. So, why asking contributors of the french wikipedia to vote for that update? Because it's only applying to the english wikipedia, doesn't it?

--82.66.149.70 10:29, 16 April 2009 (UTC) Reply

Hello,

The French Wikipedia doesn't have a note about the version it uses or that it moves to a newer version. But when you read the license you will see this part:

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.

Since the French Wikipedia only gives a note that the information is released under GFDL is it possible to uses any version ever published.

The license update is for all uses globally, so it will affect all Wikipedia's and other Wikimedia projects, therefor we are asking the complete global community to vote.

Best regards,

Huib talk 17:20, 16 April 2009 (UTC) Reply

Er we are? "Veuillez participer au vote pour déterminer les futures clauses de droits d'auteur sur les projets Wikimedia (le vote finit le 3 mai 2009). Votez maintenant !" which according to Babelfish says "Please take part in the vote to determine the future clauses of rights d' author on the Wikimedia projects (the vote finishes on May 3, 2009). Vote now!" Clicking the link directs you to Fr:Spécial:SecurePoll/vote/1 Nil Einne 19:29, 16 April 2009 (UTC) Reply

Yes you are. You can't vote from the French Wikipedia because (like me) you're not a contributor there; you don't have the 25 edits needed to be elegible to vote. If you go to the same page on a wiki where you do have 25 edits, you will be able to vote. Happy‐melon 19:36, 16 April 2009 (UTC) Reply

Sorry for any confusion resulting from my comment. I thought Abigor was saying that we should (but weren't) inform all language wikiprojects of the proposed change so was simply trying to point out were were informing all languages, but it appears I read the comment by Abigor wrong at the time Nil Einne 14:01, 18 April 2009 (UTC) Reply

Actually, the text about the license on wikipedia FR was mentioning the GFDL version, but I just realized someone changed it 10 days ago... You can see the difference here: [4]. That doesn't seem really fair to me... --82.225.170.198 21:42, 16 April 2009 (UTC) Reply

(削除) That needs to be reverted ASAP (削除ここまで), they're screwing with the whole WMF's licensing policy there. Does someone from the LiCom want to either undo it themselves as an office action, or get a steward to do so? Happy‐melon 09:21, 19 April 2009 (UTC) Reply

Scratch that, it's not their copyright notice, just their copy of the GFDL. Happy‐melon 09:22, 19 April 2009 (UTC) Reply

Actually, it is part of the copyright notice, because the footer text is saying "All the texts are available under the terms of GNU free documentation license" with a link to this page, and the removed text on the page was saying "Here is the text of the GNU FDL 1.2. All texts available on wikipedia are available under the terms of this license." (quick translation). *This* license means the GFDL 1.2. So yeah legally there is clearly a problem here, because this page will be considered by the law as a part of the copyright notice. Because the footer text is not the license, it just says "this text is licensed under the terms of the license on *this* page", so this page really matters, legally speaking. I think this situation could lead easily to some legal troubles. --82.225.170.198 11:19, 19 April 2009 (UTC) Reply

The way I see it, it's not clear cut. It says (more or less) that the content is licensed under the GFDL which is a relatively complete licensing statement. It then provided a link to the GFDL 1.2 license. The question is would a court consider the link an intrisic part of the licensing statement, or just intended as a copy of the GFDL 1.2 license, which the content is licensed under (but not exclusive) and the statement itself the accurate licensing statement (which only mentions GFDL and therefore implies any version)? Also from what I can tell Fr:Wikipédia:Droit d'auteur which is also linked also only mentions GFDL since the beginning [5]. In any case, I think the legal implications here are best dealt with by the our lawyer or the foundation not us Nil Einne 07:11, 23 April 2009 (UTC) Reply

Hello,

You should be able to vote when you use the link on the English Wikipedia or Commons. Please let us know if that doesn't work. Huib talk 20:34, 16 April 2009 (UTC) Reply

I think it's important that everyone is informed of exactly what this decision involves before voting. To that end, I have created a blog post on WikiLog outlining the details. --Heebiejeebieclub 18:58, 16 April 2009 (UTC) Reply

I'm not clear on some aspects, perhaps someone can walk me through this. Does the latest version of GFDL automatically license all media previously licenced under previous GFDL "or later" versions multi-licensed under cc-by-sa-3.0? If so, then it would seem to me that there is no vote here needed; the change to multi-licensing has already happened. If on the other hand the latest version of GFDL does not automatically multi-license all earlier "or later" GFDL works, by what authority can votes by anyone other than the original copyright holder change the license under which their work is released? What am I missing here? Puzzled, -- Infrogmation 01:47, 19 April 2009 (UTC) Reply

GFDL 1.3 grants the operator of a wiki licensed under the GFDL the option of relicensing that wiki's content under CC-BY-SA. Hence the Wikimedia Foundation must decide whether or not to migrate the license. Wikimedia has said that their decision will be contingent on community approval. Hence the vote is to advise the WMF about whether we want them to exercise the migration clause. Dragons flight 02:26, 19 April 2009 (UTC) Reply

Um. Say I took a photograph. I and I alone am the copyright holder. I scan it and post digital copies of it on the web, with notice of my copyright. One of those copies is to Wikimedia Commons, where I have agreed to license use of the image under GFDL. It seems to me the image remains MY content, not Wikimedia's content. Wikimedia and anyone else has been granted permission by me to reuse it with attribution under GFDL. How would it be legal for Wikimedia or any other reuser to relicence my content without my permission? And issue of legality aside, how could that be considered ethical? -- Infrogmation 20:38, 19 April 2009 (UTC) Reply

I'll address the legal "how". The photograph you sent to Commons probably says it may be used under the GFDL version 1.2 or any later version. (If you didn't include the "or later version" clause then it can't be relicensed and the discussion is moot.) In particular, it would then be usable under GFDL 1.3 [6]. GFDL 1.3 section 11 says that if the work was published to "massive multiauthor collaboration site" (read: wiki) then the operator of that site may choose to relicense the GFDL 1.3 content as CC-BY-SA 3.0 (subject to some caveats, not relevant to your question).

By asserting "or later versions", you gave the FSF permission to change the license. The FSF changed the license in a way that allows the WMF to further relicense your work. The argument that this is ethical relies on the observation that GFDL and CC-BY-SA are quite similar, and so a similar set of rights are protected in either case. Dragons flight 21:54, 19 April 2009 (UTC) Reply

In the example of an image I copyrighted and shared under GFDL, I still don't see how WMF could change the licensing without my permission. Okay, if I licensed under GFDL "or later version", I can see that GNU Free Software Foundation could hypothetically make later versions of their licenses changed however they wish. I can see how the new GFDL allows Wikimedia to change their wikis to multi-licensed. However my photo is NOT a "Massive Multiauthor Collaboration". It is a specific work by one single author. It is my work. In uploading a copy to Wikimedia Commons I have authorized my work to be reused under such licenses as I specified, but I have NOT transfered copyright to Wikimedia nor granted them authority to change the license. I can see that the proposed change could change the text of the Commons description page where my image is displayed to multilicensed, as part of the MMC, but not my copyrighted image itself. -- Infrogmation 15:29, 22 April 2009 (UTC) Reply

MMC is defined, in a less than intuitive way, as any work published to a MMC Site where it could be edited by anybody. The fact of whether something actually has been edited collaboratively is irrelevant. (This is important for dealing with single author stubs and other text works, but also applies to images.) By publishing to a wiki where it could be editable by anyone, you made your image part of an MMC and eligible for relicensing. Or so the legal reasoning goes anyway. Dragons flight 15:53, 22 April 2009 (UTC) Reply

Thank you for your replies. So if I support the proposal, I would be voting to change the licensing of some other contributors to something other than what they explicitly stated, contrary to their wishes and without their permission. I'm still not certain this is strictly legal, but I am certain that it is unethical, an act of bad faith against good faith users, and a very poor precedent. Whatever merits the "licensing update" has, I can't support that. -- Infrogmation 13:20, 23 April 2009 (UTC) Reply

I double voted by mistake from enwiki and simplewikiquote (didn't read the "Don't test" thing), could someone strike my simplewikiquote vote? Thanks. MBisanz ^talk 10:02, 20 April 2009 (UTC) Reply

Consider it done. Huib talk 10:38, 20 April 2009 (UTC) Reply

To my knowledge, the jurisdictional variant URL wasn't intentionally chosen in the hyperlinks in the Board resolution. The language of the proposed migration, which has been vetted by the Board, has consistently referenced the "unported" version. -- From Erik on the Foundation list.

Anthere 19:58, 20 April 2009 (UTC) Reply

If this vote does pass, and the Board of Trustees do pass this, then all content on Wikipedia will be under a license that states that if you'd like to republish this work, then you must give credit to Wikipedia? Or will you be able to use the GNU Free Documentation License with any content? 69.68.196.73 02:43, 22 April 2009 (UTC) Reply

You will still be able to use the GFDL with all local content if you find the license terms easier to satisfy. Local contributors will be required to dual-license their contributions under both the GFDL and CC-BY-SA to facilitate that. Happy‐melon 09:24, 22 April 2009 (UTC) Reply

What is the difference between a local contributor and any other contributor? 69.68.196.73 20:47, 22 April 2009 (UTC) Reply

We can import content from other projects that use compatible licenses. So for instance the reason no one at Wikimedia is remotely concerned about the growth of Citizendium is because when they finally get around to writing a nice article, we can in many cases (will be all cases if this resolution passes) copy it to our own wikis, with suitable attribution, of course, because Citizendium's license is compatible with our own (GFDL for articles they pinched from us, CC-BY-SA otherwise). Now, if we move to dual-licensing, we will require anyone who edits directly to Wikimedia to dual-license their contributions, but we can't require that from outside sources. To avoid severely restricting our choice of sources, we will accept any imported content that is licensed under at least CC-BY-SA. So some of Citizendium's articles are GFDL 1.2+, which exactly matches our own. Other articles are 'only' CC-BY-SA. If we import one of those articles to Wikimedia, obviously that content can only be relicensed under CC-BY-SA. So the layman description is "all content can be relicensed under CC-BY-SA. Most content is also available under GFDL". Happy‐melon 21:08, 22 April 2009 (UTC) Reply

I just learned that we'd be cutting ourselves off from every other GFDL source in the world. I don't think there are that many CC-BY-SA sources that we'd want to incorporate into Wikipedia to make it worth losing every other GFDL source. I want to change my vote to be against it -- how can I do that? From the IRC chat group:

according to gnu.org/licenses/fdl-1.3-faq.html... If GFDL content was published somewhere other than a public wiki, it can only be used under CC-BY-SA 3.0 if it was added to a wiki before Nov 1 2008. So nothing added after that date can be relicensed as CC-by-SA and will need to be removed before we switch or as soon as possible afterwards

My response was:

So with the license transition, we won't be able to use anything added to pretty much every other wiki, unless they also switch to this cc-by-sa thing? I mean, pretty much every other wiki is using GFDL licensed content.

Honestly, why would we want to cut ourselves off of all GFDL sources? There just isn't enough CC-BY-SA stuff out there that we'd want to incorporate to make it worth it. Banaticus 18:35, 24 April 2009 (UTC) Reply

I think you'd be surprised how many sites will follow our lead, whichever way we choose to go. Yes, there are other wikis out there, but we are without a doubt 'the daddy'. It won't be universal, naturally; but it probably will be widespread. Also consider that the new license will open up a whole new set of wikis and sources that weren't previously available; for instance, we'd be in license-compliance with Citizendium, so we really wouldn't need to 'worry' about their 'competition' (not that we really do at the moment!). Happy‐melon 18:54, 24 April 2009 (UTC) Reply

But what about, say, Ballotpedia? What if other sites don't follow Wikipedia's lead? I mean, it apparently took a team of lawyers and the GFDL people had to specifically be on board for Wikipedia to make the change, how is anyone else honestly expected to be able to do that? But even if they're able to come in under the same terms that Wikipedia just had, they have to go back and review all stuff that was added since November for copyright violations? Again, it comes down to lawyers, funding, people -- how are other places supposed to take care of all that? Other than other wiki sites who are under the meta umbrella who would possibly have it all handled for them, I really don't think many other sites are going to go through all the time and trouble of switching. Banaticus 20:15, 24 April 2009 (UTC) Reply

Simply vote again (preferably from the same starting wiki). In the event of multiple votes from a single account, only the most recent one is counted. Dragons flight 20:48, 24 April 2009 (UTC) Reply

Thanks, I found the link at en.wikipedia by going to w:Special:SecurePoll/vote/1 Banaticus 04:38, 25 April 2009 (UTC) Reply

I'd like to see some substantiation of the assertion that "pretty much every other wiki is using GFDL licensed content". I've personally seen far too many CC-licensed ones to take that statement at face value without further verification. – McDutchie 20:15, 27 April 2009 (UTC) Reply

The proposed termes of use include the following: "If you import text under the CC-BY-SA license, you must abide by the terms of the license; specifically, you must, in a reasonable fashion, credit the author(s). Where such credit is commonly given through page histories (such as wiki-to-wiki copying), it is sufficient to give attribution in the edit summary, which is recorded in the page history, when importing the content." No, it is not sufficient to give attribution in the edit summary. Authors of text under the CC-BY-SA licence published outside WMF-projects might have dicided on specific ways how they want to be given attribution. Furthermore, re-users applying the "expectations for re-users" will almost certainly fail to meet the CC-BY-SA requirements in this case.

Let me give an example: Netizen A writes an interesting article and publishes it on his private website. Since he wants to share his knowledge, he publishes the text under the CC-BY-SA licence, but requires attribution-by-name. Netizen B imports this text into a WMF wiki - let's say, the English Wikipedia -, crediting User A in the edit summary. Netizen C copies the article from the Wikipedia with a link to the article, thus using attribution-by-URL. By doing so, netizen C does not credit netizen A in the way that netizen A had required (to be precise, netizen A is not given credit at all!).

And if netizen A decides to re-use his own text in another context, everybody will think he has only copied from the Wikipedia... Adrian Suter 14:38, 28 April 2009 (UTC) Reply

Same applies to imported images, i.e. tons of flickr-content. We cannot assume all flickr-contributors agree to the TOS. -- smial 12:07, 29 April 2009 (UTC) Reply

There are some users of wikipedia too, that don't agree the proposed terms of service. The TOS may not change the licence in that way, or much content has to be deleted. --Stefanbs 15:25, 21 May 2009 (UTC) Reply

It is possible that, at some future point, GFDL 1.2 media may be disallowed. However, this will only happen if CC-BY-SA is modified to make it more explicitly a "strong copyleft" license for embedded media, requiring the surrounding content to be licensed under CC-BY-SA. Currently both licenses are somewhat ambiguous in this regard.

This means that in this occasion it could be a storm of deletions of all file licensed with GFDL, does this? I think it's legally impossible for other to change a license if decided by the author of the file. --Roberto Segnali all'Indiano 07:34, 2 May 2009 (UTC) Reply

There aren't that many GFDL-1.2-only media currently on WMF; none on de.wiki, 12,000 on Commons. Disallowing GFDL-1.2 would be conducted as a 'phasing out' rather than an outright cutoff to prevent precisely the deletion spree you mention. Happy‐melon 13:38, 6 May 2009 (UTC) Reply

On an important note: all changes will be conducted over a period. Things won't necessarily change dramatically overnight. Things will gradually phase out – and in. —Anonymous Dissident ^Talk 08:14, 11 May 2009 (UTC) Reply

Should we put a banner of some kind on top of the lemma which tells that the voting has stopped and where to find the announced results on March 10th? I for one am not sure where to look at that day (though it most probably will be in the media anyway... But I'd like to read the regular announcement too). Konsumkind 00:13, 6 May 2009 (UTC) Reply

Do you mean May 10? And we have placed the message on Meta's home page. OhanaUnited ^{Talk page} 01:29, 6 May 2009 (UTC) Reply

Yepp, meant May 10th (sorry). You're right with the home page, though some people set direct links to this page. Some sort of message that it is obsolete now might be a benefit. Konsumkind 09:50, 6 May 2009 (UTC) Reply

A notice to tell people when the results are available is currently being translated. It will be used when the results are posted and will contain a link to them. :-) Cbrown1023 talk 01:37, 6 May 2009 (UTC) Reply

Ah ok, thx :-) Konsumkind 09:50, 6 May 2009 (UTC) Reply

It's may 10, any word on the results? 189.105.47.35 10:47, 10 May 2009 (UTC) Reply

Seems it becomes May 15. 百家姓之四 01:48, 12 May 2009 (UTC) Reply

+1 Syrcro 06:27, 11 May 2009 (UTC) Reply

+1 -- smial 07:11, 11 May 2009 (UTC) Reply

Hello, You are placing here messages like +1, but that will not make us work more fast. Please remenber that we are just humans just like you ;-) We are doing what we can, please have a little bit patiënce. Huib talk 07:22, 11 May 2009 (UTC) Reply

I wonder exactly what has to be done? Tallying is done by software I guess, vote eiligibility seems to be done by software, what is exactly what has to be done manually? (The question is based on curiousity and not on the wish to hurry you by any means.) --grin ✎ 07:47, 11 May 2009 (UTC) Reply

Double votes have to be manually extracted. More than that, there have been or would have been problems with sockpuppetry. —Anonymous Dissident ^Talk 08:11, 11 May 2009 (UTC) Reply

Thanks for the info. Anything on how many votes approximately were cast? --grin ✎ 11:53, 11 May 2009 (UTC) Reply

This is not a forum, doing messages like "+1" are really immature even in the forum environment, let alone here. Please don't rush us. Rushing only increases our pressure and further reduces productivity. To answer your question, a total of 18,707 ballots (see graph on the right). That's the grand total, including socks and double-voting. OhanaUnited ^{Talk page} 15:59, 11 May 2009 (UTC) Reply

I have to say, that's a bit mean :D Raising our hopes like that... Happy‐melon 21:45, 15 May 2009 (UTC) Reply

I thought of adding another +1 and decided that it's silly, but this "mid-may" thing starts to look pretty bad.

And why are votes changed manually anyway? Votes? Changed? Manually? Where else in the world votes are changed manually? --79.176.241.146 06:47, 16 May 2009 (UTC) Reply

I think you understand us wrong, we removed double votes manually. We didn't change votes :-) Huib talk 07:03, 16 May 2009 (UTC) Reply

Is there a way to get the results with and also without your manual handling? I agree in 79.176.241.146's doubts. -- Sozi 07:54, 16 May 2009 (UTC) Reply

Struck votes, which are mostly duplicate votes where the same individual voted multiple times, are never opened. As a security matter, vote content is encrypted and no one has access to the actual vote content during the period that duplicate votes are handled. There are also multiple people available to review this process, and a written (though not public) reason is required for each removal. It would be very difficult to remove enough votes inappropriately to influence the result without being noticed. (Incidentally, a better system could probably do >90% of the duplicate vote reduction automatically, but that's not the system we have.) Dragons flight 16:10, 16 May 2009 (UTC) Reply

That's certainly not the level of transparency I expect from an organization that promotes open knowledge. And I think that I'm not alone. The overall picture of the whole voting process is, well, not the best. Very disappointing... --Berntie 11:44, 18 May 2009 (UTC) Reply

I think it's entirely misguided to even begin to suggest that anything under the table or out of form is occurring in the manual checking of the votes here. It's for the quality and the accuracy of the vote that this time is taken. As it is, this process has been completed; all that remains is for the formalities to be executed with the Board before things go public. Patience. —Anonymous Dissident ^Talk 12:08, 18 May 2009 (UTC) Reply

Just to make it clear: I'm not impatient, I don't care if we have to wait a few weeks for the result. What I do care about is that nobody can follow what's going on during these weeks. Nothing is made public. --Berntie 11:31, 19 May 2009 (UTC) Reply

Pray tell, what exactly did you expect? Happy‐melon 12:20, 18 May 2009 (UTC) Reply

The current voting process is brand new. It has some serious limitations. It also has some design flaws that could be improved for better transparency. I think we certainly do want to improve this going forward, so if you'd like to offer constructive feedback about what you'd like to see in the future, I'm sure people will take it into consideration. Dragons flight 17:58, 18 May 2009 (UTC) Reply

What I'd like to see in the future? It would be a good start if you could just provide the result before the "cleaning". That's not a technical issue and I don't see why that should be impossible or harmful. --Berntie 11:31, 19 May 2009 (UTC) Reply

I know I can see why it would be harmful to release the result before cleaning – because it's extremely subject to change. You could have an entirely different result by the end. —Anonymous Dissident ^Talk 12:14, 19 May 2009 (UTC) Reply

That's exactly the point. I want to know what was changed during the non-public cleaning. What's harmful about telling people: "The raw voting result is: ... After removal of multiple the result looks like: ..."? As I said before, that doesn't have to be right now. But when the final result is announced, I can't see any reason for not giving both results. (Well, I do see a possible reason, but a rather unpleasant one: If the result before the cleaning does not reflect the desired outcome, and the result after cleaning does.) The community should be allowed some controlling of what the foundation did with the votes. --Berntie 14:36, 19 May 2009 (UTC) Reply

The number of removed votes and the reasons for removal is already public. There is no "raw voting result", because the votes were removed prior to the vote being counted (when they were removed, we had no idea whether someone was voting +, -, or =). Cbrown1023 talk 15:07, 19 May 2009 (UTC) Reply

Having now seen the final tally, I suppose it doesn't make a difference since if you were to place all the rejected votes in any category, you would not change the outcome (in addition to the fact that it was done blinded). --Falcorian 06:45, 21 May 2009 (UTC) Reply

Exactly. The whole point is that no one knows how those duplicated votes were cast, so any bias in the reviewing team is not able to affect the outcome. Correct me if I'm wrong; my understanding of the process is that the vote was held on external servers (run by the (削除) FSF (削除ここまで) SPI) so that the Foundation was not in control of the data: once the vote was over, the (削除) FSF (削除ここまで) SPI released the data of who had voted, and from where, but not how each vote was cast. So the team from Wikimedia who went through and removed duplicate votes only saw that a vote had been cast, not how it had been cast. Once they compiled a list of which ballots to count, the (削除) FSF (削除ここまで) SPI reunited each ticket with the actual vote, compiled the results, and returned them to licom, which has now sent them to the Board. I assume that licom does not, therefore, know anything more than aggregate results? That is, you don't have a list of who voted how, only separate lists of who voted, and what votes were cast? Happy‐melon 15:35, 19 May 2009 (UTC) Reply

The external server was run by the non-profit Software in the Public Interest (FSF arguably has a stake in the outcome). Otherwise you are basically correct. Neither WMF nor Licom knows who voted in what way. As you say, there is a list of who voted and a separate tally of the votes cast, but they were prepared in such a way that no one on our end has access to the details of who voted in what way. Dragons flight 19:22, 19 May 2009 (UTC) Reply

Doh <facepalm/> Glad I got the basics right though. Happy‐melon 20:39, 19 May 2009 (UTC) Reply

The result of the vote has been shared amongst members of the Licensing Update Committee and with the WMF Board of Trustees. In accordance with our prior agreement, the Board will be given a couple days to prepare a reaction before the tally is announced to the public. Dragons flight 18:54, 16 May 2009 (UTC) Reply

The results of the poll have been published at the results page. Dragons flight 06:04, 21 May 2009 (UTC) Reply

On the Norwegian wiktionary, the link to 'see the result' is towards a local page which does not exist, rather than towards the meta page, where the result actually is announced. V85 08:25, 21 May 2009 (UTC) Reply

Where is the raw numeric data available at for File:WMF licensing update vote ballots.JPG? (added and created by OhanaUnited). Would it be possible to have a breakdown of votes per day (Like "first day, 420 +, 69 =, 24 -", etc)? Thanks. -- Jokes Free4Me 09:37, 21 May 2009 (UTC) Reply

I rounded the figures to nearest 100. It would be impractical to get the exact votes per day. Besides, the graph is intended as an assistance tool, so I don't get why someone is complaining about me using wrong image file type for the graph. The graph is not a big deal! OhanaUnited ^{Talk page} 18:36, 21 May 2009 (UTC) Reply

I'd have opposed. Not that my vote would have mattered, but perhaps I could have made an impact by sharing this. Eh, who am I kidding? Anyway, this stinks. --Leavestock 07:05, 23 May 2009 (UTC) Reply

Hello,

We tryed very hard to make sure that all users did know about the vote (multible emails to mailinglists, globalsitenotice, messages in local villagepumps) I am very sorry to hear that you didn't know about the vote. Huib talk 08:14, 23 May 2009 (UTC) Reply

Thanks. It's no biggie. --Leavestock 20:29, 23 May 2009 (UTC) Reply

Why doesn't it include the word "irrevocably"? This is an important thing that must be in there, because that is the case. ViperSnake151 18:59, 24 May 2009 (UTC) Reply

Hello,

I am not sure if I understand what you mean.

Huib talk 16:50, 26 May 2009 (UTC) Reply

He means that the proposed text for the message under the edit screen says "By submitting an edit, you agree to release your contribution under the terms of the Creative Commons Attribution/Share-Alike License and the GNU Free Documentation License.", while the current version usually says "By submitting an edit, you agree to irrevocably release your contributions..." I think he might have a point; even if it's not legally necessary, it's language a layman can understand, which is helpful for dealing with complaints along the lines of "i revoke my permission, remove all my edits". I agree the word should be retained. Happy‐melon 17:07, 26 May 2009 (UTC) Reply

So, presumably, all edits done to, say Wikipedia, were done so under GFDL. Now that it's being changed, does a user have the right to have his work removed from the project because he doesn't consent to a license that wasn't an option when the user contributed? - ALLST✰R ▼^echo _{wuz here @} 07:45, 29 May 2009 (UTC) Reply

Retrieved from "https://meta.wikimedia.org/w/index.php?title=Talk:Licensing_update&oldid=1497756"