Wikidata talk:Requests for comment/Non-free content

From Wikidata
Jump to navigation Jump to search

Quotations under "fair use"[edit]

This whole subject rely on the assumption that quotations are under "fair use" and can't be "CC0" but I don't see where this come from. Almost all 800 Wikimedia projects use quotation and AFAIK, none consider them to be under "fair use" (and some do quotations while having - more or less explicitly - banned "fair use"), they just consider that for a quotation the original license doesn't apply and place these quotation under CC like any other content, this should apply for data too (and even more for data). So why would quotations suddenly be not-CC ?

Cheers, VIGNERON (talk) 12:52, 10 April 2019 (UTC)[reply]

@VIGNERON: See w:en:Wikipedia:Do not include the full text of lengthy primary sources#Fair use of copyrighted primary sources (a guideline). I only just read this, but it essentially confirms that for each quotation from a copyrighted work, the text is being used under the fair use doctrine, and that the original source material is not now partly CC (the copyright still remains but the quotation does not infringe copyright). I think it helps if you look at it from the lens of "every word someone writes may be copyrighted" rather than "every work someone publishes may be copyrighted". Jc86035 (talk) 15:05, 10 April 2019 (UTC)[reply]
@Jc86035: sorry but I think you're not even wrong. Cheers, VIGNERON (talk) 15:15, 10 April 2019 (UTC)[reply]
@VIGNERON: Is there a different part of US law that exempts quotations from copyright? All the information I can find suggests that the right to quotation of the Berne Convention is enshrined through the fair use doctrine.
Guidelines on the English Wikipedia are meant to have community consensus ("generally accepted standard"). The page has been a guideline since 2005. The first mention of fair use in the guideline was on 31 January 2006, and the section added in that edit has remained since then. Are the guideline's dozens of authors "not even wrong" as well? If you think the guideline is correct but I'm wrong, is there a specific flaw in my reasoning? Jc86035 (talk) 15:23, 10 April 2019 (UTC)[reply]
Also, as noted by MisterSynergy in the RfC, fair use content in CC works is allowed but must be disclosed. As the guideline indicates, Wikipedia does this through the use of in-text attribution. Jc86035 (talk) 15:28, 10 April 2019 (UTC)[reply]
@Jc86035: « the right to quotation of the Berne Convention is enshrined through the fair use doctrine » that does not mean that "right to quotation" = "the fair use", we can use only the right to quotation not the whole the fair use doctrine (and since the first is more common, more specific and stronger than the second, it seems more logical to do so). Plus this assume that quotation are copyrighted so you can use an exception of copyright but you should first prove that quotation are indeed copyrighted (and when, and how, etc.).
The guideline is about "full text of lengthy primary sources" but this is not the subject at hand here and even so, this guideline says "be careful to limit the amount quoted" which is what is already done.
"every word someone writes may be copyrighted" yes, in fact "everything may be copyrighted" but most things are not and especially not words (and everything that is not works, which is often de jure defined as "something that can hold copyright" so if it's not a work it can't hold copyright).
For the last point, again, that's good to know that CC can include fair use content but is there any fair use in the first place? And Wikidata already often does attribution, isn't that enough?
Globally, you suggest solution without even remotely defining the problem.
Cheers, VIGNERON (talk) 16:16, 10 April 2019 (UTC)[reply]
@VIGNERON: Are we talking past each other?
To my understanding, a quotation from a copyrighted work under US law would either be (1) public domain (because the quotation somehow does not meet the threshold of originality or is derived from another work), (2) fair use, or (3) infringing. You are saying that my argument is invalid because the Berne Convention introduces (1a) public domain because the act of quotation means that the original license of the work does not apply. The Berne Convention itself does not state that this renders quotations public domain; it only states that quotation should be "permissible", which is in line with the fair use doctrine. While we may not be infringing on others' copyright by using quotations, we still may not own the copyright of the quotations, which is (in my not-a-lawyer view) why we would need the RfC and possibly the policy. You could infringe someone's copyright in 1,500 characters, right? With LilyPond you could infringe the copyright of an entire song in much fewer than 1,500 characters, and the server could turn it into an audio file for you.
On "most things are not [copyrighted]", this is probably somewhat inaccurate. This paragraph is probably copyrightable, since it's likely that they form a unique combination of words that has not been needed before (and fair use would allow others to quote it even if not for the CC license). How would you propose to define "work"? To me (in quasi-legal terms), if something created by a person is meaningful or significant in some way, then it would form some sort of work; using the same reasoning, the work is quite likely not simple enough to be uncopyrightable unless it is a derivative of another work. Jc86035 (talk) 17:20, 10 April 2019 (UTC)[reply]
@Jc86035: I'm not really saying that your argument is invalid, I'm more saying that you are « putting the cart before the horse ». If we want to make a policy (or even decide if we want to make a policy or not), we need strong and referenced ground, and this is clearly (at least to me) not the case here and without solid premisses, all the rest come tumbling down like a house of cards. Plus, as Wikidata is for the entire world, we should not focus only on US law (especially as US law is quite idiosyncratic, and I'm saying that from France where the copyright is even weirder, for instance "public domain" in inside copyright and still hold some rights forever unlike almost all other countries in the world). Cheers, VIGNERON (talk) 17:50, 10 April 2019 (UTC)[reply]
@VIGNERON: The reason I have focused solely on US law, despite never having been a US resident or citizen, is that it is legally the only set of laws that definitely matters here (w:en:Berne Convention indicates that there is actually no legal certainty on whether other countries' courts have civil jurisdiction over US internet-connected servers). Looking at every country's laws would be a nightmare as it would effectively increase the workload by a factor of 200; if we did this we would have no choice but to follow Commons's example, although it seems Commons prevents fair use uploads but not fair use text. Jc86035 (talk) 17:55, 10 April 2019 (UTC)[reply]
@VIGNERON: For clarity, some of my comments above are not written with the correct implications; see my edit to the lead. Jc86035 (talk) 18:20, 10 April 2019 (UTC)[reply]
@Jc86035: « Looking at every country's laws would be a nightmare » would it? First, I'm not asking to look at every country, then in most countries (if not all, thanks to the Berne Convention) there is some kind of right to quote, so why not simply use the right to quote? (and look further at that) And I'm not talking only about the law here, if we want a policy to be applicable, it should be easily understandable and the right to quote is self-explanatory (while the fair use is not and is unnecessary wide for our needs). For the « legal certainty on whether other countries' courts have civil jurisdiction over US internet-connected servers », I already mentioned the en:LICRA v. Yahoo! case (which is quite complex - as always with lex loci - and quite far from copyright law but seems to indicate that other countries does have some jurisdiction other US servers). Cheers, VIGNERON (talk) 18:43, 10 April 2019 (UTC)[reply]
@VIGNERON: The right to quote for any arbitrary purpose is not necessarily a given; until 2014 UK law had a stricter interpretation only allowing quotation for criticism or review. Other cases, such as w:en:Moberg v. 33T LLC (mentioned in the Berne Convention article), have returned the opposite verdict (in this case, that US copyright law does not apply to foreign works published on the Internet), so I would assume there is no firm legal precedent.
Re. "putting the cart before the horse", I think the visibility of the RfC did generate a wider discussion than would have occurred on the project chat, so at least in that respect it may have been beneficial. Jc86035 (talk) 09:22, 11 April 2019 (UTC)[reply]