Talk:GNU Free Documentation License

Has debian-legal finally concluded that FDL with IS is non-free? The last that I heard, it was still a matter of some debate; although things certainly seemed to be headed to that conclusion, I'm surprised that it's already been reached. In particular, is the FSF's documentation for emacs being moved to /nonfree/? -- Toby Bartels 21:17, 7 Sep 2003 (UTC)

They had a survey, which showed (AFAICT) that the majority concluded that GFDL was incompatible with the DSFG *even without Invariant Sections*. In particular, they seem to regard the "anti-DMCA" provisions (cannot use technical measures to prevent further copying and viewing of copies you make or distribute) to be non-free. Martin 23:41, 7 Sep 2003 (UTC)

Using GFDL materials with the GFDL as an invariant section?

I was thinking about including some material from some GNU manuals in several Wikipedia articles. While these materials are released under the GNU Free Documentation License, they use the GFDL itself as an invariant section. Two things:

Isn't the GFDL essentially an invariant section already? You need to include it with any copy of the text. Why is it necessary to include it as an invariant section?
Can I post the material at all, and if so, do I need to include a note such as: This material requires the GNU Free Documentation License as an invariant section, or is that already implied?

Thank you, -- Mattworld 22:04, 14 Oct 2003 (UTC)

Legally a puzzler, but the first question should be: which parts of GNU manuals are encyclopedic? All the bits I can think of are inappropriate for Wikipedia; we should be summarizing the manuals if anything, not dumping them in here verbatim. Stan 22:23, 14 Oct 2003 (UTC)

There was a history section in one of the manuals that would have fit perfectly in Wikipedia; however, I now plan to summarize it, so this isn't that big of an issue. I'm still slightly curious as to the answer, though. -- Mattworld 22:34, 14 Oct 2003 (UTC)

Invariant sections aren't welcome in Wikipedia (though you're welcome to separately distribute a derivitive of a Wikipedia article that does include invariant sections). I'm mighty curious now about the docs... which are they? --Brion 23:37, 14 Oct 2003 (UTC)

In this case though, I think it might not be a problem. You already have to distribute an invariant copy of the GFDL along with any GFDL-licensed material, so including the GFDL itself as an invariant section isn't actually adding any new restrictions--it's just being redundant and restating the same restriction in a different way. --Delirium 01:44, Oct 15, 2003 (UTC)

A GFDL text can be redistributed under a later version of the GFDL. If you put a copy of the license in an invariant section, that that older version will need to be distributed with the text even if you're putting it out under a later version. Can somebody please point out the actual texts under discussion so more than speculation can take place here? --Brion 00:50, 16 Oct 2003 (UTC)

At the GNU site When should a section be invariant? (http://www.gnu.org/licenses/fdl-howto.html) it states that the GFDL is an invariant section [see § 4(H)]. If it is an earlier version of the GFDL that says it can also be released under a later version of the GFDL (i.e. originally ver. 1.1 released under ver. 1.2) then can not the older version be replaced by the newer version as the newer version is already an invariant section in a GFDL work? If it is only released only under ver. 1.1 of the GFDL (does not specifically state that it may be released under a later version) then would that make it incompatible with the GFDL here which is of a "later version" variety? Most of the changes in the later version seems to be just clarifications of the earlier version; is there any reference some one can point to that states how the two versions are incompatible and not just clarifications that have the same legal import? The only two I can find are (1) the five principal author listing (all the authors must be listed in 1.1; this is clearly not problem with Wikipedia as the history pages list all the authors of a particular GFDL work) and (2) 1.2. has the optional warranty disclaimers added so 1.1 is just 1.2 without any disclaimers in that regard; so maybe there is backward compatibility if no warranty disclaimers are used. In the license it is stated that if a use of the license invokes the language "any later version" that a user can decide which version to re-release the material under; as the GFDL is already an invariant section; this indicates that you can release 1.1 under 1.2 even if 1.1 is listed separately as an invariant section because you cannot take away any of the rights granted in the GFDL under § 9 (i.e. the right to use a later version cannot be revoked). BTW if no version number is listed the GFDL (§ 10) states any version can be used; so there is an argument that by making the GFDL an invariant section it then invokes any version of the GFDL even if the version used to release the document does not state that any version can be used. Thus, as a rule, a general statement that the "GNU Free Documentation License" is an invariant section (without reference to the version number) would appear to imply that one can use any version of the license — in such a case it does not appear necessary to provide a copy of the predecessor license as a separate invariant section. — Alex756 05:59, 19 Oct 2003 (UTC)

Some seemingly minor thing - if a material is released under v.1.1, one can name history section as "page history" or "document history," but v.1.2 seems to prohibit that (see at the definition section.) So, some people might prefer ver. 1.1, indeed. Also, isn't there a problem if English wikipedia's article is translated into other wikipedia that uses a previous version of the GFDL? Tomos 22:41, 29 Oct 2003 (UTC)

some questions about compliance

The ver 1.2 of GFDL seems to indicate that Page history section should really named "History." (See the "APPLICABILITY AND DEFINITIONS" part)
I am not sure if it is GFDL compliant when a page is modified and the names of the authors are not listed at the title page. Of course, the title "page" does not exist for Wikipedia. But GFDL specifically mention to such a case, and says that we should use some space between the title and the main text. So it seems that the strict compliance would be to display names of authors after the articles' title. Currently, we have authors' names in page history section.
I am not sure if it is GFDL compliant when a person A uploads an image to a wikipedia, and a person B uses that image in an article. Is the article "modified version" of the image in GFDL's sense? I am really not sure...

Tomos 22:41, 29 Oct 2003 (UTC)

Translations

How do translated works fall under this? --207.103.218.175, 21:37, 12 Feb 2004

History of the GFDL

When was the GFDL "born"? Guaka 22:14, 31 Mar 2004 (UTC)

The article could do with some history. As it currently stands it does not even mention the earlier version 1.1. I have tried to track down some history about the license, but there doesn't seem to be much on line. -- Popsracer 13:22, 13 Apr 2004 (UTC)

Ean Schuessler's invariant section

I've removed Ean Schuessler's contributions, as he was only offering them on the condition that we include an undesirable invariant section. Matthew Woodcraft 20:20, 29 May 2004 (UTC)

Matthew Woodcraft's variation of my invariant section

I still contend that my invariant section was removed illegally. To legally remove my intellectual property you would have to roll the state of the document back to a previous version that was free of the invariant section attached to my input. We are still talking about my invariant section and it is still referenced in the history, therefore you have not "removed" my IP and associated invariant section but have rather masked it behind a confusing web interface that makes access more difficult. This entire site is a composite document and unless it deletes "undesirable" IP with "undesirable" invariant sections Wikipedia is still in violation.

Please take note that my original invariant section was intended entirely to demonstrate the folly of the GNU "Free" Documentation License and not add any useful value in and of itself. Ean Schuessler 19:50, 21 October 2004 (Dallas Texas Time!)

Ean, if you think the intermediate versions should be removed from Wikipedia's database, you should ask the administrators to do that and see what they say. They're unlikely to be reading this talk page. Matthew Woodcraft

Matthew, after further analysis I see that the copyright of Wikipedia disallowed invariant sections at the time of my prank. I apologize for my inability to read the instructions. One might make the argument, however, that you should be required to "click-through" the wikipedia usage agreement before adding content. A side note at best. I'm just making excuses.

Using the information

At the least I'd like to see a link in the article to something that covers this assuming there is such a place but....

The FDL seems like it was designed for software manuals. In that case it makes sense. There is most likely very little data in the manual that is useful outside of the software the manual is for.

But, for an encyclopedia ????? What consistituts reproducing the document? If my 8 year old son is doing a report on Nebraska and he gets info from the Wikipedia is his report now FDLed? If not, why? If he includes a quote from the Wikipedia or just adds the Wikipedia to his bibilography is his report now FDLed?

For the quote, no because it falls under US fair use guidelines. For the bibliography, no because copyright protects expression, not content. Superm401 22:24, 5 Feb 2005 (UTC)

I may sounds silly but I could read nothing in the FDL the specifically gives permission to re-use the info in the document outside of the document without making that info also FDLed.

That permission is unnecessary as no copyright protects against others using information from your work. It only protects the exact way the information is expressed. Of course, it is preferable to cite your sources even when you copy no text from them. However, this is an ethical, not a legal issue. Superm401 22:24, 5 Feb 2005 (UTC)

Like I mentioned, this makes sense for a software manual, even an example that might be useful for another piece of software being lifted out the some FDLed manual I can see why the FDL would frown on that without the other manual becoming FDLed as well but an Encylopedia is a different thing. It would seem to me under the current wording of the FDL that I should never read anything on the Wikiepedia with out full knowledge that any info gained from reading it cannot be reproduced else where without making that new thing FDLed.

I couldn't go researching this history of the middle east on the Wikipedia for an article in the New York Times without making either that article or that entire edition of the New York Times be FDLed. I'm sure some of you will see that as a good thing but an Encyclopedia with those kinds of restrictions is not really useful outside.

As for precident on not sharing info there are plently of private reports with specific licenses that require you not to share the info. The TRST reports from NPDworld.com for example. So the idea that the info on the Wikipedia might not be useable in general because of it's license is definiately a possibility.

I think this needs to be spelled out somewhere. Is the info useable without restriction or not? If I research something here and post the results of my research on my website is my website now FDLed? If not, where is the specific exclusion for that kind of reproduction?

There is a huge difference between copying the text and copying the information. If you copy FDLed text into another text, that other text can only be published under the terms of the FDL. But using the information is no problem, just as with any other encyclopedia or text (unless it contains patented algorithms or stuff like that). Also, if you would copy text from Wikipedia into an article for the NYT or a book, or anything else, I think that only implies that the article that contains the line should be FDLed. Not the entire work. Just like including the Wikipedia no a CD doesn't mean that the entire CD should be FDLed. G-u-a-k-@ 20:18, 13 Sep 2004 (UTC)

mirrors and attribution

Hi - I've asked this question a couple of times in discussions of a specific WP mirror site, wikiverse.org (http://www.wikiverse.org) (see Mirrors and forks) but maybe this is a more appropriate place. Question: for a site using WP text to comply with GFDL, if they have each article on a separate page as we do, do they have to credit Wikipedia on each page? Or is a single mention of Wikipedia on the site's main page adequate? Based on the wording of Wikipedia:Copyrights - "a direct link back to the article satisfies our author credit requirement" - it seems to me that each article is supposed to have some sort of acknowledgement. Yes? —Hob←Talk 03:23, 2004 Aug 15 (UTC)

Invariant Sections Question

I was reading about the GFDL and some of the arguments against (http://home.twcny.rr.com/nerode/neroden/fdl.html) using it. The section about Invariant Sections seemed particularly interesting. My understanding of it is that if a portion of a document licensed under the GFDL is labelled an Invariant Section it cannot be modified or removed from any subsequent versions of the document, i.e. it will be included, unaltered, with the document in question forever. Am I right in this interpretation or am I completely misunderstanding the license (quite possible as I am nowhere near an expert on legal issues)? Basically, I'm wondering if a vandal could post an Invariant Section to Wikipedia and therefore make it be part of the encyclopedia forever? I'm sure that can't be how it works, but thats what it sounds like to me. Where am I going wrong? Some of the discussion above seemed to be about similar questions but I didn't see any consensus... Any help would be greatly appreciated, thanks - biggins 19:36, 6 Sep 2004 (UTC)

Wikipedia does not allow invariant sections. Adding an invariant section would be in violation of our copyright policy so should be removed in the same way as any other violating text. Angela . 00:31, Sep 13, 2004 (UTC)

In any case, rolling back from the version with the invariant section (B) to an earlier version without it (A) ought to allow you to continue to derive from that earlier version without including the invariant section. The really tricky question is whether you could use material that had been added to document B to create its derivative (C) before rollback. I think if you did use that later material you would be bound by the terms added by the invariant section and would be reimporting them into your new document (D) even after rollback. Does that make sense? (No user name; 20 Oct 2004)

Invariant Sections, etc

note: To keep things simple, we don't use Invariant Sections, Front-Cover Texts, or Back-Cover Texts

er, what are invariant sections, front-cover texts and back cover texts? Dunc_Harris|☺ 18:57, 15 Sep 2004 (UTC)

Stuff the GFDL talks about. -- orthogonal 18:59, 15 Sep 2004 (UTC)

They come from book/journal publishing. Sections of a journal that are the same in every issue (editorial board, instructions to authors, that sort of thing) and the text on inside and outside of the front and back cover. Filiocht 07:58, 16 Sep 2004 (UTC)

Fair use and GFDL

From the article: However in some specific cases, commercial re-uses may be fair use and in that case such materials do not need to be licensed to fall within the GFDL if such fair use is covered by all potential subsequent uses.

It is not possible for all potential subsequent uses to be fair use. I am not sure if this is what the above sentence says but English is not my mother tongue. I believe fair use claim for distribution is valid (if it is valid at all) regardless of subsequent uses. Could someone clarify this section, I would do it myself but a native speaker would do it better. at0 22:08, 12 Oct 2004 (UTC)

So quoted material should not be used on wikipedia? How about documentation of material learned at one point from textbooks? I'd really like to help, but I'm not sure what exactly I'm allowed to do.

BIG QUESTION: Moral Rights, Wikipedia and the FDL?

I haven't gone through the FDL properly yet but I wonder: how does it deal with moral rights? Has the situation of a Wikipedia contributor deciding to be difficult and exerting their moral rights to attribution or (God forbid) integrity been anticipated and dealt with? Is this a potential SCO code situation that could cause Wikipedia real difficulties? The Wikipedia enty on moral rights says they can be waived in "some jurisdictions". Are there any where they can't be waived? Is there an explicit waiver clause? Is one needed? (No user name; 20 October 2004)

In the U.S., there is no statutory basis for "moral rights" for text and I am unaware of any case law supporting such a concept. Even in Europe, where there is a statutory guarantee of moral rights, I think it most unlikely that a case could be made, because contributors know exactly how their contribution will be utilized and attributed before they contribute. I think that the most defensible construction is that the Wikipedia text is in its entirety a collaborative work with shared authorship, and that the joint authors have agreed to GFDL licensing.

IANAL and my view is not widely shared in the Wikipedia community. There have been no serious tests of the GFDL in the courts and so many things could change. Because Wikipedia is a worldwide project and because there isn't any case law on copyleft or on massive collaboration, anything is possible. I would expect the courts to rule in such a way as to preserve the project, because Wikipedia is now big enough, important enough, and well enough liked that there would be hell to pay if the courts shut it down over a technicality.

Images are another matter because in most cases the creative effort can be clearly attributed to one person.

uc 13:28, 20 Oct 2004 (UTC)

:o)IAAL (albeit trainee) and an IP one at that. I'm also the same anonymous user that posted the original question. I don't know the answer to it but it's something that should be thought about very carefully. The size and usefulness of Wikipedia may not be enough to outweigh these rights if they do exist and someone tries to exert them. Although it can be difficult for authors to exert moral rights and it might seem an unlikely situation, I expect it's something Wikipedia would want certainty on rather than taking a 'hope and pray' approach. The 'mercilessly edited and redistributed at will' notice on the edit page may be there for that reason? It doesn't seem to mention the right to attribution though.

The size and "importance" and the amount by which it is "well liked" may not be enough to save Wikipedia, especially if what I have heard about the lawsuits (especially by the European Union against Microsoft is correct. Brianjd

A couple things to keep in mind. i) The GFDL was adopted by the project early on and in recognition that it is not perfect, but in consideration that it would have been difficult at best to do any better at that time and with the limited resources then available. ii) There is probably nothing more that can be done now than what is already done. iii) All this pales by comparison to the fact that the project faces legal exposure regarding DMCA violations (of infringing materal posted from elsewhere that hasn't been caught and of supposed "fair use" material where an affirmative defense would have to be mounted) and libel. iv) In light of that, "hope and pray" is probably not the worst approach.

You might consider sharing your thoughts with User:Alex756 who is an attorney and has been involved with the project at times, including some pro bono work. I believe there is also a mailing list for Wikipedia legal issues.

uc 16:22, 20 Oct 2004 (UTC)

The wikilegal-l mailing list is defunct and has been replaced by foundation-l for matters relating generally to the Wikimedia Foundation. Legal issues can be discussed there, but the list is not dedicated specifically to that subject. --Michael Snow 18:12, 20 Oct 2004 (UTC)

I don't think Wikipedia as a project has a particular problem with moral rights (that is, the right of an author to be credited with creating a work, even if he has sold the copyright to someone else). The GFDL requires that the names of authors are preserved, and we keep a full record of exactly who wrote what in the history of each article.

The problem with crediting authors is more an issue for people reusing our content under the GFDL - the problem being that reproducing the history section could take up more space than the actual article in many cases - a real problem for paper copies. The practical solution used by people who reuse our content is to link back to the Wikipedia article, so that readers can look there to see who wrote it. It's an imperfect solution, but one that I think fulfils the spirit of the requirements of the GFDL and (european) copyright law in crediting authors. Neither copyright law nor the GFDL ever anticipated that there could be so many authors to a work that giving credit to all the authors could become a real burden.

Enchanter 17:25, Oct 20, 2004 (UTC)

Moral rights are generally taken far more seriously with respect to works that are highly creative in nature (e.g. art, music, literature). For these works, moral rights reflect a strong social sense that the creativity is associated with the mind of the creator, who is therefore entitled to a say in how the work is presented. Due to the nature of an encyclopedia (do you ever look to see who actually wrote the article in other encyclopedias?), particularly one with policies like NPOV and a highly collaborative model, the amount of creativity associated with any individual author is far smaller. In this context, the claim of moral rights appears less strong. While not guaranteed, I consider it eminently possible that a court dealing with someone trying to assert moral rights against Wikipedia might respond that those rights entitle the person to nothing more than what we already give them (attribution in page history plus the ability to edit the page again). --Michael Snow 18:12, 20 Oct 2004 (UTC)

A messieurs les Anglais !

de:GNU FDL en:GNU FDL nl:GNU FDL sv:GNU FDL

POURRIEZ VOUS MESSIEURS LES ANGLAIS TRADUIRE LES PREMIERS DE FACONS OFFICIEL CE TEXTE EN FRANCAIS ? SVP! MERCI!

-)

There's a French translation of the GFDL at [1] (http://cesarx.free.fr/gfdlf.html). Angela . 15:13, Nov 9, 2004 (UTC)

I removed this bit:

The Debian project no longer uses the GNU FDL for this, and possibly other, reasons.

I don't think Debian rejects the GFDL because it permits adding invariant sections: BSD licenses also allow this, and Debian is happy with those.

Matthew Woodcraft

Overly broad DRM clause

I know that license debates can be heated and I stayed clear of them for now. Nevertheless, the main page had some pecularities under the heading 'Overly broad DRM clause'.

To say or suggest that the FSF has `a hidden agenda to discourage the use of proprietary software' is absurd. This smells like the average conspiracy theory. The FSF is not known for hidden agendas. Instead, they state it very clearly: the FSF discourages the use of proprietary software (http://www.gnu.org/philosophy/categories.html#ProprietarySoftware). So, yes, the FSF has an agenda but certainly not hidden.

Aside from this error, I did not understand the criticism of the quote that is given under this heading. I've dug a bit in the history of this page and I see that the arguments given in the main article are rephrased from the external links. However, the external links were perfectly clear to me, whereas the orginal article left me with confusion.

It seems that three separate points were being made.

The first criticism seems directed at the restrictions on local copies of GDFL licensed documents. People are worried that you cannot save a local copy in a proprietary format even if the original stays available in an open format.

A problem with a broad interpretation of `technical measures' Reading the reference "why you shouldn't use the GNU FDL (http://home.twcny.rr.com/nerode/neroden/fdl.html)", I also see mention of technical measures, where this mention is used as a clarification of the former argument. Therefore, it seems bogus as an additional criticism.

The last concern is about using proprietary software or file formats with build-in DRM features to distribute the content. Indeed, this is a restriction of the GFDL.

I've tried to clarify the text on the first and third point, while dropping the second. Furthermore, I've given the external links a more prominent place, since they are worth a read.

Jan van Male 18:37, 19 Mar 2005 (UTC)

Public domain

What's the difference between GFDL and PD images? There seems to be a strong similarity between the 2. --SuperDude 00:45, 4 Jun 2005 (UTC)

Retrieved from "https://academickids.com:443/encyclopedia/index.php/Talk:GNU_Free_Documentation_License"