Wikipedia talk:Non-free content/Archive 65

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 60 ← Archive 63 Archive 64 Archive 65 Archive 66 Archive 67 → Archive 70

Regarding this.

As that commenter said, should we keep large versions of non-free vector images since they can be scaled to any size? Hop on Bananas (talk) 01:17, 25 November 2015 (UTC)

  • All files should satisfy WP:NFCC#3b. Note that the default display size is just an arbitrary number of little importance, so the 'size' of an SVG file needs to be determined in some other way. --Stefan2 (talk) 01:27, 25 November 2015 (UTC)
  • A user removed the {{non-free reduce}} tag (which I originally added) from File:Green Party logo.svg for obvious reasons. It seems that there are several users who agree that large non-free SVGs do not need to be reduced since they can be reduced or expanded to any size, including an acceptable non-free size. Are they right or not? Hop on Bananas (talk) 12:45, 26 November 2015 (UTC)
  • Actually, that was the same user who originally made the comment on the file's talk page. But still, are they right? Hop on Bananas (talk) 12:53, 26 November 2015 (UTC)
    • It has been previously established that if we can legally acquire the logo vector directly from the company that owns the logo (off their website, off a PDF document they have , etc.), that that is an acceptable non-free, and because vectors are resolution-less , it is difficult to apply 3b to them. What we don't allow is the recreation as a vector of a non-free logo (in other words, no user-created non-free SVGs are allowed), and of course, if the only version available are raster images, these have to meet #3b. --MASEM (t) 15:47, 26 November 2015 (UTC)
      • Now, I've always assumed that the ban on user generated SVG versions is because they have too many copyrights attached to them, making them violations of the WP:FREER/NFCC#1 principle ("When using non-free media, the ones with the least restrictive copyright terms should be preferred").Jo-Jo Eumerus (talk, contributions) 15:52, 26 November 2015 (UTC)
        • There is that too, though arguably if the user recreates the non-free logo and then says "my SVG code is CC-BY", that's an "equivalent" free-ness as if the image is directly provided by the company that owns the logo (which covers both the graphical representation and the SVG for the most part). The main reason to avoid user-recreated non-frees is that they may not always be accurate at the small detail level, which can mis-represent the entity. It avoids derivative work issues related to that. --MASEM (t) 16:01, 26 November 2015 (UTC)

While I didn't know why we didn't allow user-created SVG non-free logos before reading those comments, this discussion is about WP:NFCC#3. Hop on Bananas (talk) 22:17, 3 December 2015 (UTC)

  • Is there a way to formulate a "minimal usage" standard for vectorized images? For regular images low size and low resolution are the definition of "minimal usage", but if vectors don't play by the same rule we need a vector-appropriate standard.Jo-Jo Eumerus (talk, contributions) 22:24, 3 December 2015 (UTC)

Talk Page / Meetup Page Usage

For the NYC Wikimedia chapter as follow up to events and as part of the event page clean vector logos from the partner organizations should be able to be used. The benefits are clear: Using the partner logos is professional, gives the pages graphic elements (which are sorely lacking), and if the logos are coming from the Commons, cross-usage should be encouraged. Specifically, the policy should not have to apply to logos of GLAM partner institutions shown on Wikipedia-space meetup pages and talk pages for new editors. This policy needs to be re-considered. BrillLyle (talk) 18:12, 3 December 2015 (UTC)

  • To clarify; BrillLyle is wanting to use File:Queens Library.svg on Wikipedia:Meetup/NYC/AfroCrowd/Press. For an example of intended usage, see this. I've explained that such usage violates WP:NFCC #9 and can not be allowed, and further that if he wants to use the logo in this manner he needs to seek an exemption from the policy here at this talk page. Thus, why it's come here. --Hammersoft (talk) 19:04, 3 December 2015 (UTC)
    • The significant point here is that GLAM partner organizations desire to be identified by their logos, as the Wikipedia namespace is de facto also sometimes an event-organizing platform. While logos should not be used gratuitously in non-article space, I can only think it is appropriate to show the logo of a partner GLAM organization on a particular project (this is very different, for example, from someone showing off their favorite brand on their user page).--Pharos (talk) 19:54, 3 December 2015 (UTC)
  • Even if it weren't for NFCC 9, I don't see how it would pass 8. There's no pressing need for the logo, the name would quite suffice. Might be a "nice to have", but that's textbook decorative use. Seraphimblade Talk to me 20:04, 3 December 2015 (UTC)
    • The more basic question I would ask is: Why do we have this overarching policy in the first place? I would say it is for two reasons, (1) For article pages, to encourage the creation and replacement on non-free images by free images and (2) for non-article pages, to prevent the proliferation and miscellaneous use of non-free images that might be be violating copyright, or that would make third party use problematic. In this case, neither justification really applies. The use of a GLAM logo on a meetup page is not preventing the creation of free images, it's not violating copyright, and it's not endangering third party use in any way. Really, the meetup pages are not part of the encyclopedia itself (such rules are never applied to identical pages on a Wikimedia chapter website), and should be treated differently.--Pharos (talk) 20:14, 3 December 2015 (UTC)
      • We have that policy because this is a free content project, and so we use nonfree content only sparingly, only to enhance educational article content, and only when it's clearly necessary. In an article about a company, the logo is a core part of the identity and branding of many companies, so use of the logo is necessary. As a decoration for a meetup page, it's not, so it's not allowed, since a text list of organizations would suffice to convey the information. What chapters do on their own websites is of course up to them, but all pages on Wikipedia must follow the NFCC rules. Perhaps if you really want to use the logos, you could consider having a chapter host it and linking to that page, that would be perfectly fine. Seraphimblade Talk to me 20:55, 3 December 2015 (UTC)
  • Oppose. This is inconsistent with the Wikimedia Foundation's licensing policy, which indicates that nonfree content may be used only to enhance article content. The proposal doesn't seem to be well thought out at all, since BrillLyle's comment above suggests that partner logos from Commons can't be used on meetup and talk pages, even though it should be obvious that they can be. Moreover, having taken a quick look at the proposed use, I don't see why someone couldn't extract the "Queens Library" typography, which is recognizable without being copyrightable. That would be more consistent with the other logos displayed. The Big Bad Wolfowitz (aka Hullaballoo) (talk) 20:16, 3 December 2015 (UTC)
  • Oppose. No exceptions are allowed by default and the only exceptions allowed by request are non-article space pages "that are used to manage questionable non-free content" (WP:NFEXMP), and the page described in this request surely isn't one. Finnusertop (talk | guestbook | contribs) 20:31, 3 December 2015 (UTC)
  • Perhaps one should encourage these partner organizations to license their logos - I find it unlikely that a logo would need more than attribution and trademarks to serve their purposes. Sort of a NFCC#1 concern this one - instead of accepting non-frees in metaspace, encourage the licensing thereof may be preferable.Jo-Jo Eumerus (talk, contributions) 22:21, 3 December 2015 (UTC)

Non-free TimedText

How should non-free TimedText be handled? I can't find anything about it. Apparently a non-free use rationale can't be added directly to the TimedText page. Does the rationale on the related media file suffice? Or should a rationale be placed on the TimedText Talk page? —teb728 t c 19:14, 20 December 2015 (UTC)

Talk page would be acceptable. It's a fair around the technical limitation. --MASEM (t) 19:23, 20 December 2015 (UTC)

NFCC#1 and recently deceased persons

How does WP:NFCC#1 apply to articles about recently deceased individuals? For example, an individual who meets WP:GNG and is fairly well-known has recently died. A Wikipedia article is created about them and a non-free image of them is used in the infobox. A newly taken freely licensed photo, etc. is no longer possible for obvious reasons, but that does not necessarily mean that a previously taken freely licensed photo of the person, which serves the same encyclopedic purpose, cannot be found or that someone may someday decide to freely license such a photo. Is using the non-free image considered to be acceptable, as sort of a placeholder, until a free image is found? Does it make any difference with respect to non-free use whether the article was created before or after the person died? -- Marchjuly (talk) 04:23, 24 December 2015 (UTC)

It doesn't matter when the article was created. Generally, for a few months after the death of a person, there is a good chance that free media could be found uploaded as a result of the death (by those that had photos of the person from before the death). After some months, about 3 to 6 months, if these photos haven't matetialized, then a bonfire can be justified. --MASEM (t) 14:40, 24 December 2015 (UTC)

NFCC#1 needs revision

Please refer to the ongoing discussion at Wikipedia:Files for discussion/2015 December 27#File:Cape Town Railway & Dock 0-4-2.jpg.

So far, the only thing that has become clear from this discussion is that NFCC#1, as it stands, lends itself to the extremely narrow-minded and, in my opinion, shortsighted and extremely counterproductive interpretation that a copyrighted image (in this instance a dimensional drawing of a locomotive) has to be deleted because someone, somewhere, someday, might get it into his or her head to draw a reproduction and upload it into the public domain on Wikimedia.

If that is the way that NFCC#1 should be interpreted, so be it. If the intention with NFCC#1 should be more constructive with regard to the dissemination of knowledge by Wikipedia, it needs to be revised, or at least reconstructed in a manner that will not allow interpretation so suit individual agendas. - André Kritzinger (talk) 21:48, 29 December 2015 (UTC)

I'm afraid that's the exact intention of #1. If a free file exists or could be created, we don't use a nonfree one, to encourage free content. This is a free content project, so nonfree content should only be used when absolutely no alternative exists. Seraphimblade Talk to me 22:36, 29 December 2015 (UTC)
This is exactly right, and its clear in this case that a free image can be created. --MASEM (t) 23:28, 29 December 2015 (UTC)

Link to Definition of "Free" Does Not Work

The link attempts to go to freedomdefined.org but the link isn't working for me. I've tested a number of other Wikipedia links and other websites from my computer, so perhaps it's not my issue.TheCensorFencer (talk) 01:51, 4 January 2016 (UTC)

It works here. --MASEM (t) 01:53, 4 January 2016 (UTC)
Oops, firewall setting. Sorry about that.TheCensorFencer (talk) 03:01, 4 January 2016 (UTC)

TfD discussion of interest to this board

About that Philippine government copyright template {{Non-free Philippines government}}, it is up for discussion here.Jo-Jo Eumerus (talk, contributions) 11:09, 6 January 2016 (UTC)

Talk:2015 United Nations Climate Change Conference#Eliminate "image_size" parameter?

The discussion is ongoing. I invite all of you to comment. George Ho (talk) 20:52, 6 January 2016 (UTC)

Fair-use status of Timed Text

(Head note, I've WP:IAR aborted all of these MFD's and venue moved them to this discussion board - a clear consensus on the legal and policy implications of these types of files is needed and individual MFD's are not the best place to ensure there is proper involvement. — xaosflux Talk 04:04, 21 December 2015 (UTC))

A batch of timed text files have been considered for deletion, but on fair use violation claims. As copyvio's are speedy candidates, I've aborted the following MFD's in process to centralized the discussion here. (more to come shortly). — xaosflux Talk 03:41, 21 December 2015 (UTC)

@Xaosflux: Can you restore TimedText:Meghan Trainor - Lips Are Movin.ogg.en.srt and add it as an example too? --MaranoFan (talk) 06:19, 21 December 2015 (UTC)
Ping to deleting administrator @JamesBWatson:. — xaosflux Talk 12:53, 21 December 2015 (UTC)
  • File list (current examples)
  1. TimedText:Like I'm Gonna Lose You.ogg.en.srt
  2. TimedText:Dear Future Husband.ogg.en.srt
  3. TimedText:Title.ogg.en.srt
  4. TimedText:Charlie Puth & Meghan Trainor - Marvin Gaye.ogg.en.srt
  5. TimedText:Better When I'm Dancin'.ogg.en.srt
  6. TimedText:WIF & Meghan Trainor.ogg.en.srt
  7. TimedText:Adele - When We Were Young Clip.ogg.en.srt
  8. TimedText:All_About_That_Bass_by_Meghan_Trainor_(sample).ogg.en.srt
  9. TimedText:Lady GaGa-Poker Face.ogg.en.srt
  10. TimedText:Adele - Hello Clip.ogg.en.srt
  11. TimedText:Sample of "Jealous" by Nick Jonas.ogg.en.srt
  12. TimedText:Lady GaGa-Just Dance.ogg.en.srt
  13. TimedText:Meghan Trainor - Lips Are Movin.ogg.en.srt

Initial discussion on MFD

The following discussion is an archived debate of the proposed deletion of the miscellaneous page below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the page's talk page or in a deletion review). No further edits should be made to this page.

The result of the discussion was Speedy Close This discussion is important, but this is primarily a copyright question ("is this fair use") If it is not fair use then this is subject to speedy deletion already. Moving this to a more in depth conversation at Wikipedia talk:Non-free content - will seek additional input there as well. — xaosflux Talk 03:57, 21 December 2015 (UTC)

TimedText:All About That Bass by Meghan Trainor (sample).ogg.en.srt

TimedText:All About That Bass by Meghan Trainor (sample).ogg.en.srt (edit | talk | history | links | watch | logs)

Question has arisen about whether a timed text page like this is acceptable. What has been created is a timed text version of song lyrics which are copyright material. The audio clip it's attached to is copyright as well, but being used with a fair use rationale. Is it acceptable to create a timed text of copyright material? Apparently other texts like this exist, but because timed text is such an unexplored space, I don't see a specific policy spelled out about it. Discussion here could help establish consensus. only (talk) 16:28, 20 December 2015 (UTC)

  • Keep If the song excerpt is fair use, why would the caption not be fair use? Also captioning with TimedText is supported in the WP:ACCESSIBILITY guidelines. Brustopher (talk) 16:40, 20 December 2015 (UTC)
  • Delete !vote contingent on discussion. I also have a concern is over copyright infringement and I would be interested to know if this does fall under same. Pinging Moonriddengirl for her opinion as she has extensive experience with copyright issues. -- WV ● ✉ ✓ 16:42, 20 December 2015 (UTC)
  • Keep per Brustopher and due to existence of several similar files such as this, this, and this. --MaranoFan (talk) 19:45, 20 December 2015 (UTC)
  • Delete: The song’s rationale as I understand it is all about the sound of the song. The timed text contains only the lyrics and does not convey the sound of the song. From the point of view of accessibility, the sound of the song could be conveyed better to non-hearing users with descriptive free text. —teb728 t c 22:37, 20 December 2015 (UTC)
  • But it's barely any text. How (I ask as someone with little knowledge of fair use) is this any less acceptable than quoting a sentence from a book? The average musician's page on Wikiquote probably has far more copyrighted content that your average TimedText. It's not all the lyrics to the song or anything. Also while TimedText on audio may not be too helpful for the deaf, it could be more beneficial to those hard of hearing or people who just can't make out the lyrics of certain songs. Brustopher (talk) 22:44, 20 December 2015 (UTC)
  • It's less acceptable in the sense that media files, unlike text quotations, must have a written rationale on their description pages that prove that they meet all 10 WP:NFCC. Finnusertop (talk | guestbook | contribs) 22:50, 20 December 2015 (UTC)
  • But are TimedText pages media files? They're SubRip text files and in a separate namespace to the File namespace, so aren't they closer to text quotations than the media files which fall under that policy. Also they're used as a component of a media file that already has a written rationale. Brustopher (talk) 22:56, 20 December 2015 (UTC)
  • I agree that the question you pose is what their use (and consequentially the faith of this nomination) hangs on. If they are not 'files', they are textual quotations that must be "properly attributed or cited to its original source or author (as described by the citation guideline), and specifically indicated as direct quotations via quotation marks, <blockquote>, or a similar method" (WP:NFCC). Our non-free content policies only recognize these two types of content - quotations and media - and there is nothing in between. Finnusertop (talk | guestbook | contribs) 23:02, 20 December 2015 (UTC)
This is starting to get into complicated territory. Wouldn't it be best to sort this out in a policy RfC, instead of at MfD? Because if this page gets deleted, it follows that pretty much every single other TimedText page on enwiki for an audio file has to go too. There seems to be 20 different discussion going on at this topic in 20 different places, and it would be helpful if we could find a centralised location to discuss it. Brustopher (talk) 23:57, 20 December 2015 (UTC)
  • If these pages are 'files', then a fair use rationale is needed, per WP:NFCC#10c. If they are not files, then source information is still needed, per WP:CITE. The obvious follow-up question is then where we should place the fair use rationale or source information. The TimedText namespace is a bit special in that you see the wikicode, so it is not ideal to put the information in that namespace. One option is to put the information on the talk page. However, considering that the entire purpose of a TimedText page is to use the page together with a file in the file namespace, and considering that the file needs a file information page with some information anyway, the natural location for this information seems to be the file information page, i.e. the fair use rationale on the file information page should specify why a TimedText page is needed.
The next issue is to figure out when we need a TimedText page for a non-free file in the first place. Since this is a situation which hasn't been discussed a lot in the past, it may be a good idea to start an RfC at WP:VPP or WT:NFC and try to establish some kind of policy or consensus. I don't think that a badly advertised MfD is the best location for writing a policy. --Stefan2 (talk) 01:00, 21 December 2015 (UTC)
The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the page's talk page or in a deletion review). No further edits should be made to this page.


Responses

I am working off the assumption that the .ogg file is being used in compliance with NFCC, that the song sample is being used to support discussion of the work. As long as the clip meets the 30 s/10% limits, I think the TimedText , which by necessity is to provide accessibility for a song sample, where the clip again is within time limits, is not a straight up copyright violation but can fall within our fair use allowances. (Particular these examples which are 4-6 lines of lyrics, at most). I do think that because of the technical limitations of the TimedText namespace (we can't include anything but exactly the timed text) that we need to use the talk page to 1) link to the media file that it is supporting, and 2) perhaps add some standard language via a template to explain that the text is copyrighted, but being used both as fair use for text, and to support accessibility of a non-free file that we also believe is being used under fair use. --MASEM (t) 04:15, 21 December 2015 (UTC)

To add to a comment Stefan made in the previous discussion, this template should identify the url to one of the approved lyric music sites where the lyrics were taken from to prove prior publication, and that we should include, as best we know, the lyric's author and publishing label and other relevant details. --MASEM (t) 04:18, 21 December 2015 (UTC)
Lyrics, as long as, credited to the artist, will fall under fair use and can be used. --QEDK ♠ T ♥ C 07:57, 21 December 2015 (UTC)
If the song is in a language which the uploader understands, then the actual source may be the file on Wikipedia as the uploader may have listened to the sound file and written down the words he heard. --Stefan2 (talk) 10:17, 21 December 2015 (UTC)
Which technically is original research. The fair use of lyrics for TimedText should have an identified published source which could either be the album's liner notes or a recognized legal lyrics site (such as Metro Lyrics, IIRC). The time codes, that's less a problem, but because lyrics are default copyright and often different from the song's copyright owner, we should document that difference a bit more carefully than presume "this is what I heard them say". --MASEM (t) 15:34, 21 December 2015 (UTC)
I think calling a transcription "original research" is too far. Especially as the source content of the description (the sound/video bite) is presented. Should the captioning include more than literal transcription (such as a descriptor like (sad tone) then it may approach OR, but that's not what we are talking about here. — xaosflux Talk 15:42, 21 December 2015 (UTC)
Would the descriptors in (the currently broken and incorrectly formatted) TimedText:GLaDOS_Voice_Change.ogg.en.srt count as OR? If it's not described how the voice changes, a major aspect of what's happening in the sound clip is lost. Brustopher (talk) 15:47, 21 December 2015 (UTC)
That's a great question, and I'm not sure of the answer - I'm still confident that strict transcription is not original research (as it in no way introduces a new new idea), I'm really not sure on this one. I'd tend to personally lean towards it not being OR in general. — xaosflux Talk 16:11, 21 December 2015 (UTC)
I'd have to double check in the game, but it should be noted that there would be subtitles in the game that one could turn on, but off the top of my head I cannot remember if they capture the distinctions. However, that itself is not so much a potential original research part as the actual transcription of the words proper. --MASEM (t) 16:26, 21 December 2015 (UTC)
Not really. Consider the number of infamous song lyrics that are misheard by people [1]. No, in most cases, I don't expect such mistakes to be made, but they can happen. Add in the fact that for most songs the lyrics are copyrighted separation from the actual song, and that to me says that we should source the song's lyrics as we would any other fair use quotation. Which is not a difficult step (again, liner notes work, and if not that, Metro Lyrics), just keeping ourselves out of hot water. --MASEM (t) 16:26, 21 December 2015 (UTC)
Do the (currently incorrectly formatted, titled and non-working) subtitles at TimedText:B.A.P - Warrior.ogg constitute original research? File:B.A.P - Warrior.ogg is in Korean, and the subtitles seem to be a translation of the Korean text. --Stefan2 (talk) 20:50, 26 December 2015 (UTC)
  • As I wrote in the MfD discussion, TimedText pages display wikicode instead of rendering the code, so it is inappropriate to put source information and/or FURs on the TimedText page itself. This information is better placed on either the TimedText talk page or, considering that there always should be a corresponding file, on the corresponding file information page. Since there already should be a FUR on the file information page, the natural location for TimedText source/FURs seems to be the file information page, and the FURs on that page could then be expanded to explain why subtitles are needed.
I don't know in which situations it would be acceptable to add subtitles to non-free files. Apart from this, we need a standardised process for disputing the need for subtitles. --Stefan2 (talk) 10:17, 21 December 2015 (UTC)
  • I mostly agree with respect to subtitle notation on talk pages or the main file page. Beyond that, I'd say that free subtitles to non-free files should be mostly dealt with in accordance to regular encyclopedicity and usefulness criteria. For non-free subtitles, we need to make sure they satisfy fair use law, especially in regards to being excerpts of the whole thing. Beyond that, I'd say that NFCC policy should apply since subtitles are really a part of the media we are representing here.Jo-Jo Eumerus (talk, contributions) 11:10, 21 December 2015 (UTC)
  • Shouldn't this be an RfC if we're trying to figure out a policy for timedtext? Also is this just a conversation about audio captions or video caption too? Also also, would it be worth contacting the WMF about potential changes to the TimedText namespace to make things like this simpler?. Brustopher (talk) 12:05, 21 December 2015 (UTC)
    Getting foundation advice may be prudent. — xaosflux Talk 12:59, 21 December 2015 (UTC)
    @Mdennis (WMF):, talk page invite left. — xaosflux Talk 13:00, 21 December 2015 (UTC)
    Xaosflux, I'll check with legal. :) --Maggie Dennis (WMF) (talk) 13:08, 21 December 2015 (UTC)
  • Probably a simple solution: I think the most simple solution would be to apply all guidelines and procedures that we already have in place for the "File:" namespace and apply them to the "TimedText" namespace. Pages in the "TimedText" will most likely always have some aspect of it that needs to be checked to ensure it is "free vs. non-free". For example, for deletion or free vs. non-free review, TimedText pages could be nominated for WP:FFD or WP:PUF. Also, the speedy deletion criteria starting with "F" could also apply to pages in the "TimedText" namespace. Pages such as WP:NFC and WP:NFCC could be updated to state that TimedText pages also apply to these guidelines. In a nutshell, we already have applicable guidelines in place: we just need to specify that the also apply to the "TimedText:" namespace in addition to the "File:" namespace. Steel1943 (talk) 17:21, 21 December 2015 (UTC)
    • I agree with Steel1943. The only confusing lingo in NFCC I can find is "also applies to the copy in the File: namespace" (NFCC#3) and "image or media description page contains [the rationale etc.]" (NFCC#10). The latter should specify if it's the actual TimeText page, its talkpage, or the file description page of the media it is used with. NFC on the other hand looks okay to me. Finnusertop (talk | guestbook | contribs) 17:36, 21 December 2015 (UTC)
  • WMF Legal thoughts: As always, leading off with the disclaimer that this isn't legal advice about what to do or not do. I would also add that I haven't looked at any specific transcriptions, just the issue in general, so you'll all have to decide what action you want to take, I'm just sharing some thoughts on the issue. From our perspective, an addition like adding timestamped lyrics is best looked at as a separate instance of copying and needs its own fair use justification, but likely has a good one. As you may have seen from the fair use wikilegal posting, there are 4 factors that go into fair use: 1) purpose and character (especially whether it's transformative from the original) 2) type of work 3) how much is used and 4) market impact. In this case, the main purpose of transcribing these short recordings seems to me to be to make them accessible for people who, due to disability or other reasons, can't hear the material on a page, which could be a transformative use. Add to that the fact that transcriptions of short snippets of a work use very little of the original and are very unlikely to have any impact on the market for the work and there is a good overall fair use argument as long as no more is transcribed than necessary to illustrate the article where the snippet is present. There is also a good point of comparison in some parody music cases like Fisher v. Dees, 794 F.2d 432 (9th Cir. 1986) and Campbell v. Acuff-Rose Music, 510 U.S. 569 (1994), which held that copying a short part of the lyrics for a transformative parody was fair use, which would be a similar line of reasoning as copying a short part of only the lyrics to make a page more accessible to the hard of hearing. Jrogers (WMF) (talk) 23:04, 21 December 2015 (UTC)
    • It's good to know that we can treat this as fair use, making the issue more how we document that as fair use on WP due to limitations of the TimedText namespace. It does sound like dropping some type of rationale on the talk page would be an acceptable solution. (I do wonder if we could include noinclude/include markup on TimedText pages across the board as with templates as to transcribe the talk or a doc page like we do for templates without disrupting their feature.) --MASEM (t) 23:27, 21 December 2015 (UTC)
  • Some people above come close to saying that if there is a legitimate non-free use for the media file, there necessarily will be a legitimate non-free use for the timed text. I originally thought that would be so, but what would make a lyric transcription satisfy NFCC would have to be that the encyclopedic purpose of the non-free media is to give the lyrics. And after looking a few timed text files, I suspect that is the purpose for few if any:
As I pointed out in the MfD transcluded above, the cited purpose is to illustrate the sound of the song.
In TimedText:Dear Future Husband.ogg.en.srt the media rationale mentions lyrics, but in fact the lyrics of the song are well described in the article.
In TimedText:Like I'm Gonna Lose You.ogg.en.srt the media rationale also mentions lyrics, but IMO the lyrics of the clip are described by the article.
It seems unlikely to me that the few line of lyrics in a clip would significantly increase reader understanding —teb728 t c 11:14, 22 December 2015 (UTC)
Let's assume that the sound sample that contains lyrics is justified for commentary on the music but not in any detail on the lyrics. The reason to include TimedText at that point is for accessibility then, not so much for non-free discussion. And because they are being included in a text form, we don't require the same "license and rationale" business we require for media files. I still do believe we need to have a template tag that asserts these are being used within fair use tied to the media file (perhaps even spelling out accessibility), and that they should be treated as quotes, and thus must be verifiable to a published sourced and not the editor's own ear. --MASEM (t) 15:15, 22 December 2015 (UTC)
I don’t understand your assumption: If you assume that the use is in accord with the Exemption Doctrine Policy, then I agree. But if you are assuming only fair use to provide accessibility, then I note that the EDP is intentionally more narrow than fair use. As the guideline for textual excerpts says, “Brief quotations of copyrighted text may be used to illustrate a point, establish context, or attribute a point of view or idea.” Although the list of purposes is not exclusive, it surely has to include some encyclopedic purpose. If there is no encyclopedic purpose in showing the lyrics, accessibility could be served optimally by a free textual description of the clip. —teb728 t c 01:16, 23 December 2015 (UTC)
Providing accessibility to both deaf readers (consider if it was used in association with a video file) and for non-native English speakers seems perfectly in line with an encyclopedic purpose. --MASEM (t) 01:36, 23 December 2015 (UTC)
Yes, perhaps for a video or perhaps if the lyrics in a song clip were the point of using the clip, etc. My point is that the use of timed text requires a separate rationale: the legitimacy of using timed text does not follow automatically from the media rationale. —teb728 t c 09:14, 23 December 2015 (UTC)
Except, and here's the key point: NFC does not directly cover text - it is for media files, which TimedText is not. Or more specifically we have no special requirements for rationale/etc. for text, beyond that meets WP:QUOTE - that is, a snippet of the full work and directly cited to the work it came. This type of use is acceptable under CC-BY's considerations of fair use of text works and our own policies. The only issue is that we can't directly attach additional information to the TimedText namespace without screwing up its function, as we would normally have a quote immediately followed by a citation. --MASEM (t) 19:00, 23 December 2015 (UTC)
Some information can be encoded in multiple ways. For example, subtitles can either be uploaded as a page in the TimedText namespace, or they could be directly inserted in a video file. Music could either be included through wikicode (e.g.
c'
) or uploaded as a sound file. In such situations, I don't think that Wikipedia policy should make any difference based on whether the material was included by uploading a file or typing in wikicode. --Stefan2 (talk) 21:08, 26 December 2015 (UTC)
  • I am not a lawyer, and I have no intention of arguing about the legality of the situation. If consensus is that this is acceptable fair use, then I shall not object to another administrator restoring the text that I deleted. However, I shall leave that to others. The editor who uses the pseudonym "JamesBWatson" (talk) 15:28, 22 December 2015 (UTC)
  • Notes: From where this was discussed elsewhere: Commons:Video#Subtitles_and_closed_captioning includes adding captioning of non-words; Wikipedia:Manual_of_Style/Accessibility#Video regarding enwiki accessibility. — xaosflux Talk 19:38, 23 December 2015 (UTC)

Next Steps

So where do we go from here? This started from a batch of MFD's on timed texts that I deferred to this board - has a consensus formed on policy and community standards? — xaosflux Talk 14:46, 6 January 2016 (UTC)

Well, consensus does seem to exist that non-free TimedText is not by default a copyright violation, but can fall under the boundaries of fair use. The next step would be to formulate a practice for maintaining TimedText files and their copyright status.Jo-Jo Eumerus (talk, contributions) 15:00, 6 January 2016 (UTC)
I think there are two things that need to be done:
1) Develop a simple template that can be posted to the talk page of a timed text entry that explains the copyright nature, assures a link to the main file that the time text supports (even if that should be obvious by the page name), a source link where appropriate, and then some type of license statement that the timed text used is only to support the relevant media file.
2) Add language to NFC to explain that for non-free Timed Text, this template should be include or should be found on the talk page of that timed text. --MASEM (t) 15:03, 6 January 2016 (UTC)
We also need a process for disputing timed text pages, and templates for this purpose. Normal templates, such as {{subst:nfurd}}, are inappropriate as they break the syntax of the page and because they display the wikicode instead of the informative text. Templates should probably be substituted so that the informative text appears on comment lines on the timed text page. --Stefan2 (talk) 15:20, 6 January 2016 (UTC)
Hmm, I think this also means that if a non-free file has timed text with it, we should make sure that is linked to the file page, so that if the file is later deleted, the timed text is too at the same time (timed text cannot exist without that file). --MASEM (t) 15:27, 6 January 2016 (UTC)
Unfortunately, there is no notification that a timed text page exists when a user tries to move or delete a file for which timed text is available. Some time ago, I created a database query for finding deleted files with timed text. I found a number of such timed text and tagged most of them for deletion per WP:G8. Then I tried to modify the query to find moved files to move the timed text to the right place, but the query became too slow. I can see if I can fix the query so that I can spot moved/deleted files after the move/deletion and then take action, but it would be nice to have this automatically spotted when someone tries to perform an action on a file. See also phab:T122038. --Stefan2 (talk) 15:41, 6 January 2016 (UTC)
On the other hand, a bot that regularly scans through any TimedText files and finds that the parametrized file page no longer exists (in these templates we'd need to create) should be able to tag such pages for deletion. Though we'd still should make sure that admins involved in deleting files check the file page for the "associated TimedText" template and delete that too. --MASEM (t) 17:26, 6 January 2016 (UTC)
I'm drafting a template for TimedText talk pages here.Jo-Jo Eumerus (talk, contributions) 15:37, 6 January 2016 (UTC)
Addendum: Stefan2, do you perhaps know what the significance of the .srt ending on all TimedText files is? Is it a software requirement?Jo-Jo Eumerus (talk, contributions) 16:07, 6 January 2016 (UTC)
It is the usual file name extension for SubRip files. These pages are essentially files, not ordinary pages. Compare with how we store JavaScript and CSS files in the User and MediaWiki namespaces. --Stefan2 (talk) 16:48, 6 January 2016 (UTC)

I've created a parser function inserted into the abovementioned draft template that is able to detect if the TimedText file is associated with a file. Assuming that the title of a TimedText page is always the same as the associated File page plus 7 characters, that is.Jo-Jo Eumerus (talk, contributions) 17:09, 6 January 2016 (UTC)

The language code in the file name can be either two letters or three letters and may include a two-letter country code if a specific dialect is specified. The number of characters at the end may vary, depending on the length of the language and dialect information. A timed text page should always be associated with a file, so a parser template for showing this isn't necessary. It is useful to have a parser template for finding out if a file is associated with a timed text page, though. That way, a warning can be added to {{rename media}} and deletion tags. --Stefan2 (talk) 17:21, 6 January 2016 (UTC)
Technically, the scope of the parser function was to allow the template to detect if the TimedText it is associated with is orphaned as a way to track these, but if the length of the final string is variable it doesn't work.Jo-Jo Eumerus (talk, contributions) 17:23, 6 January 2016 (UTC)
I think that the length of the final string always is 7, 8, 10 or 11. If you are creating a general template for detecting this for maintenance purposes, then we should also add the template to subtitles of free files as it is also useful to know if a deleted free file still has subtitles on Wikipedia. If possible, also try to detect if the associated file is a redirect or if the file is hosted on Commons as we then need to move the subtitles to the redirect target or to Commons. --Stefan2 (talk) 17:32, 6 January 2016 (UTC)

Now, I've found a way to make the parser function work even when there are more than 7 letters in the suffix, by using Module:String. This way, orphaned TimedTexts can be marked.Jo-Jo Eumerus (talk, contributions) 18:02, 6 January 2016 (UTC)

Tracked in Phabricator
Task T122038
Also check if other projects have better methods for catching orphaned and renamed subtitles. For example, Commons probably has a lot of subtitles. There might be something for us to copy. --Stefan2 (talk) 18:05, 6 January 2016 (UTC)
Noting here phab:T122038 where the request to add a warning for Special:Move and ?action=delete for the existence of TimedText.Jo-Jo Eumerus (talk, contributions) 15:17, 7 January 2016 (UTC)
  • Ok, I've finished up my part of the template here, as an userspace draft. Requesting review.Jo-Jo Eumerus (talk, contributions) 20:40, 6 January 2016 (UTC)
    • The template asks for G8 deletion of subtitles for files which have been moved to Commons. In these cases, we should ensure that the subtitles are moved to Commons before the local copy is deleted. I fixed this by replacing the warning with a different warning if the file was uploaded under the same name on Commons. This check fails if there is a local file information page with a DYK template or similar on Wikipedia (the template doesn't discover that the file is on Commons) or if the file is available under a different name on Commons (the template still asks for G8 deletion). Is there a way to improve this? --Stefan2 (talk) 18:44, 7 January 2016 (UTC)
      I don't know if #2 can be fixed at all, but I'll check. As for #1 - maybe the template can check the content of the file page, but that's something I wouldn't bet on.Jo-Jo Eumerus (talk, contributions) 20:00, 7 January 2016 (UTC)
      • Ok, according to my testing a template cannot distinguish between a file on Commons and a file redirect on Commons, at least not according to the specs I have. However, I can distinguish between "File and Media exists", "File exists and Media doesn't" (G8 cases for the file), "File doesn't exist and Media does" (File on Commons) and finally "File and Media don't exist" (orphaned). I'll implement this distinction.Jo-Jo Eumerus (talk, contributions) 20:59, 7 January 2016 (UTC)
        • I've put up a more general discussion on the talk header proposal here.Jo-Jo Eumerus (talk, contributions) 15:14, 9 January 2016 (UTC)
          • Keeping archive bot away while this is still in development — xaosflux Talk 16:26, 16 January 2016 (UTC)

Draft template in Template:TimedText talk template

Pinging Masem (for proposing the existence of the template, as well as for suggestions regarding the copyright notice), Stefan2 (the template now includes a mechanism to marl orphaned TimedText files as well as TimedText files attached to redirects; I don't think it's possible to detect the existence of a TimedText from the File page) and Steel1943. Right now, it has a basic "general" description text (the Commons template doesn't contain much more) as well as an optional copyright warning.Jo-Jo Eumerus (talk, contributions) 15:07, 7 January 2016 (UTC)

Editnotice

I'm assuming these files are normally created "in the raw" (as opposed to with a wizzard). If so, perhaps we can update the namesapce edit notice to encourage adding licensing tempates to the talk pages? — xaosflux Talk 17:34, 6 January 2016 (UTC)

Notices:
  1. Template:Editnotices/Namespace/TimedText
  2. Template:Editnotices/Namespace/TimedText talk
— xaosflux Talk 17:48, 6 January 2016 (UTC)

Fair-use status of Timed Text

(Head note, I've WP:IAR aborted all of these MFD's and venue moved them to this discussion board - a clear consensus on the legal and policy implications of these types of files is needed and individual MFD's are not the best place to ensure there is proper involvement. — xaosflux Talk 04:04, 21 December 2015 (UTC))

A batch of timed text files have been considered for deletion, but on fair use violation claims. As copyvio's are speedy candidates, I've aborted the following MFD's in process to centralized the discussion here. (more to come shortly). — xaosflux Talk 03:41, 21 December 2015 (UTC)

@Xaosflux: Can you restore TimedText:Meghan Trainor - Lips Are Movin.ogg.en.srt and add it as an example too? --MaranoFan (talk) 06:19, 21 December 2015 (UTC)
Ping to deleting administrator @JamesBWatson:. — xaosflux Talk 12:53, 21 December 2015 (UTC)
  • File list (current examples)
  1. TimedText:Like I'm Gonna Lose You.ogg.en.srt
  2. TimedText:Dear Future Husband.ogg.en.srt
  3. TimedText:Title.ogg.en.srt
  4. TimedText:Charlie Puth & Meghan Trainor - Marvin Gaye.ogg.en.srt
  5. TimedText:Better When I'm Dancin'.ogg.en.srt
  6. TimedText:WIF & Meghan Trainor.ogg.en.srt
  7. TimedText:Adele - When We Were Young Clip.ogg.en.srt
  8. TimedText:All_About_That_Bass_by_Meghan_Trainor_(sample).ogg.en.srt
  9. TimedText:Lady GaGa-Poker Face.ogg.en.srt
  10. TimedText:Adele - Hello Clip.ogg.en.srt
  11. TimedText:Sample of "Jealous" by Nick Jonas.ogg.en.srt
  12. TimedText:Lady GaGa-Just Dance.ogg.en.srt
  13. TimedText:Meghan Trainor - Lips Are Movin.ogg.en.srt

Initial discussion on MFD

The following discussion is an archived debate of the proposed deletion of the miscellaneous page below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the page's talk page or in a deletion review). No further edits should be made to this page.

The result of the discussion was Speedy Close This discussion is important, but this is primarily a copyright question ("is this fair use") If it is not fair use then this is subject to speedy deletion already. Moving this to a more in depth conversation at Wikipedia talk:Non-free content - will seek additional input there as well. — xaosflux Talk 03:57, 21 December 2015 (UTC)

TimedText:All About That Bass by Meghan Trainor (sample).ogg.en.srt

TimedText:All About That Bass by Meghan Trainor (sample).ogg.en.srt (edit | talk | history | links | watch | logs)

Question has arisen about whether a timed text page like this is acceptable. What has been created is a timed text version of song lyrics which are copyright material. The audio clip it's attached to is copyright as well, but being used with a fair use rationale. Is it acceptable to create a timed text of copyright material? Apparently other texts like this exist, but because timed text is such an unexplored space, I don't see a specific policy spelled out about it. Discussion here could help establish consensus. only (talk) 16:28, 20 December 2015 (UTC)

  • Keep If the song excerpt is fair use, why would the caption not be fair use? Also captioning with TimedText is supported in the WP:ACCESSIBILITY guidelines. Brustopher (talk) 16:40, 20 December 2015 (UTC)
  • Delete !vote contingent on discussion. I also have a concern is over copyright infringement and I would be interested to know if this does fall under same. Pinging Moonriddengirl for her opinion as she has extensive experience with copyright issues. -- WV ● ✉ ✓ 16:42, 20 December 2015 (UTC)
  • Keep per Brustopher and due to existence of several similar files such as this, this, and this. --MaranoFan (talk) 19:45, 20 December 2015 (UTC)
  • Delete: The song’s rationale as I understand it is all about the sound of the song. The timed text contains only the lyrics and does not convey the sound of the song. From the point of view of accessibility, the sound of the song could be conveyed better to non-hearing users with descriptive free text. —teb728 t c 22:37, 20 December 2015 (UTC)
  • But it's barely any text. How (I ask as someone with little knowledge of fair use) is this any less acceptable than quoting a sentence from a book? The average musician's page on Wikiquote probably has far more copyrighted content that your average TimedText. It's not all the lyrics to the song or anything. Also while TimedText on audio may not be too helpful for the deaf, it could be more beneficial to those hard of hearing or people who just can't make out the lyrics of certain songs. Brustopher (talk) 22:44, 20 December 2015 (UTC)
  • It's less acceptable in the sense that media files, unlike text quotations, must have a written rationale on their description pages that prove that they meet all 10 WP:NFCC. Finnusertop (talk | guestbook | contribs) 22:50, 20 December 2015 (UTC)
  • But are TimedText pages media files? They're SubRip text files and in a separate namespace to the File namespace, so aren't they closer to text quotations than the media files which fall under that policy. Also they're used as a component of a media file that already has a written rationale. Brustopher (talk) 22:56, 20 December 2015 (UTC)
  • I agree that the question you pose is what their use (and consequentially the faith of this nomination) hangs on. If they are not 'files', they are textual quotations that must be "properly attributed or cited to its original source or author (as described by the citation guideline), and specifically indicated as direct quotations via quotation marks, <blockquote>, or a similar method" (WP:NFCC). Our non-free content policies only recognize these two types of content - quotations and media - and there is nothing in between. Finnusertop (talk | guestbook | contribs) 23:02, 20 December 2015 (UTC)
This is starting to get into complicated territory. Wouldn't it be best to sort this out in a policy RfC, instead of at MfD? Because if this page gets deleted, it follows that pretty much every single other TimedText page on enwiki for an audio file has to go too. There seems to be 20 different discussion going on at this topic in 20 different places, and it would be helpful if we could find a centralised location to discuss it. Brustopher (talk) 23:57, 20 December 2015 (UTC)
  • If these pages are 'files', then a fair use rationale is needed, per WP:NFCC#10c. If they are not files, then source information is still needed, per WP:CITE. The obvious follow-up question is then where we should place the fair use rationale or source information. The TimedText namespace is a bit special in that you see the wikicode, so it is not ideal to put the information in that namespace. One option is to put the information on the talk page. However, considering that the entire purpose of a TimedText page is to use the page together with a file in the file namespace, and considering that the file needs a file information page with some information anyway, the natural location for this information seems to be the file information page, i.e. the fair use rationale on the file information page should specify why a TimedText page is needed.
The next issue is to figure out when we need a TimedText page for a non-free file in the first place. Since this is a situation which hasn't been discussed a lot in the past, it may be a good idea to start an RfC at WP:VPP or WT:NFC and try to establish some kind of policy or consensus. I don't think that a badly advertised MfD is the best location for writing a policy. --Stefan2 (talk) 01:00, 21 December 2015 (UTC)
The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the page's talk page or in a deletion review). No further edits should be made to this page.


Responses

I am working off the assumption that the .ogg file is being used in compliance with NFCC, that the song sample is being used to support discussion of the work. As long as the clip meets the 30 s/10% limits, I think the TimedText , which by necessity is to provide accessibility for a song sample, where the clip again is within time limits, is not a straight up copyright violation but can fall within our fair use allowances. (Particular these examples which are 4-6 lines of lyrics, at most). I do think that because of the technical limitations of the TimedText namespace (we can't include anything but exactly the timed text) that we need to use the talk page to 1) link to the media file that it is supporting, and 2) perhaps add some standard language via a template to explain that the text is copyrighted, but being used both as fair use for text, and to support accessibility of a non-free file that we also believe is being used under fair use. --MASEM (t) 04:15, 21 December 2015 (UTC)

To add to a comment Stefan made in the previous discussion, this template should identify the url to one of the approved lyric music sites where the lyrics were taken from to prove prior publication, and that we should include, as best we know, the lyric's author and publishing label and other relevant details. --MASEM (t) 04:18, 21 December 2015 (UTC)
Lyrics, as long as, credited to the artist, will fall under fair use and can be used. --QEDK ♠ T ♥ C 07:57, 21 December 2015 (UTC)
If the song is in a language which the uploader understands, then the actual source may be the file on Wikipedia as the uploader may have listened to the sound file and written down the words he heard. --Stefan2 (talk) 10:17, 21 December 2015 (UTC)
Which technically is original research. The fair use of lyrics for TimedText should have an identified published source which could either be the album's liner notes or a recognized legal lyrics site (such as Metro Lyrics, IIRC). The time codes, that's less a problem, but because lyrics are default copyright and often different from the song's copyright owner, we should document that difference a bit more carefully than presume "this is what I heard them say". --MASEM (t) 15:34, 21 December 2015 (UTC)
I think calling a transcription "original research" is too far. Especially as the source content of the description (the sound/video bite) is presented. Should the captioning include more than literal transcription (such as a descriptor like (sad tone) then it may approach OR, but that's not what we are talking about here. — xaosflux Talk 15:42, 21 December 2015 (UTC)
Would the descriptors in (the currently broken and incorrectly formatted) TimedText:GLaDOS_Voice_Change.ogg.en.srt count as OR? If it's not described how the voice changes, a major aspect of what's happening in the sound clip is lost. Brustopher (talk) 15:47, 21 December 2015 (UTC)
That's a great question, and I'm not sure of the answer - I'm still confident that strict transcription is not original research (as it in no way introduces a new new idea), I'm really not sure on this one. I'd tend to personally lean towards it not being OR in general. — xaosflux Talk 16:11, 21 December 2015 (UTC)
I'd have to double check in the game, but it should be noted that there would be subtitles in the game that one could turn on, but off the top of my head I cannot remember if they capture the distinctions. However, that itself is not so much a potential original research part as the actual transcription of the words proper. --MASEM (t) 16:26, 21 December 2015 (UTC)
Not really. Consider the number of infamous song lyrics that are misheard by people [2]. No, in most cases, I don't expect such mistakes to be made, but they can happen. Add in the fact that for most songs the lyrics are copyrighted separation from the actual song, and that to me says that we should source the song's lyrics as we would any other fair use quotation. Which is not a difficult step (again, liner notes work, and if not that, Metro Lyrics), just keeping ourselves out of hot water. --MASEM (t) 16:26, 21 December 2015 (UTC)
Do the (currently incorrectly formatted, titled and non-working) subtitles at TimedText:B.A.P - Warrior.ogg constitute original research? File:B.A.P - Warrior.ogg is in Korean, and the subtitles seem to be a translation of the Korean text. --Stefan2 (talk) 20:50, 26 December 2015 (UTC)
  • As I wrote in the MfD discussion, TimedText pages display wikicode instead of rendering the code, so it is inappropriate to put source information and/or FURs on the TimedText page itself. This information is better placed on either the TimedText talk page or, considering that there always should be a corresponding file, on the corresponding file information page. Since there already should be a FUR on the file information page, the natural location for TimedText source/FURs seems to be the file information page, and the FURs on that page could then be expanded to explain why subtitles are needed.
I don't know in which situations it would be acceptable to add subtitles to non-free files. Apart from this, we need a standardised process for disputing the need for subtitles. --Stefan2 (talk) 10:17, 21 December 2015 (UTC)
  • I mostly agree with respect to subtitle notation on talk pages or the main file page. Beyond that, I'd say that free subtitles to non-free files should be mostly dealt with in accordance to regular encyclopedicity and usefulness criteria. For non-free subtitles, we need to make sure they satisfy fair use law, especially in regards to being excerpts of the whole thing. Beyond that, I'd say that NFCC policy should apply since subtitles are really a part of the media we are representing here.Jo-Jo Eumerus (talk, contributions) 11:10, 21 December 2015 (UTC)
  • Shouldn't this be an RfC if we're trying to figure out a policy for timedtext? Also is this just a conversation about audio captions or video caption too? Also also, would it be worth contacting the WMF about potential changes to the TimedText namespace to make things like this simpler?. Brustopher (talk) 12:05, 21 December 2015 (UTC)
    Getting foundation advice may be prudent. — xaosflux Talk 12:59, 21 December 2015 (UTC)
    @Mdennis (WMF):, talk page invite left. — xaosflux Talk 13:00, 21 December 2015 (UTC)
    Xaosflux, I'll check with legal. :) --Maggie Dennis (WMF) (talk) 13:08, 21 December 2015 (UTC)
  • Probably a simple solution: I think the most simple solution would be to apply all guidelines and procedures that we already have in place for the "File:" namespace and apply them to the "TimedText" namespace. Pages in the "TimedText" will most likely always have some aspect of it that needs to be checked to ensure it is "free vs. non-free". For example, for deletion or free vs. non-free review, TimedText pages could be nominated for WP:FFD or WP:PUF. Also, the speedy deletion criteria starting with "F" could also apply to pages in the "TimedText" namespace. Pages such as WP:NFC and WP:NFCC could be updated to state that TimedText pages also apply to these guidelines. In a nutshell, we already have applicable guidelines in place: we just need to specify that the also apply to the "TimedText:" namespace in addition to the "File:" namespace. Steel1943 (talk) 17:21, 21 December 2015 (UTC)
    • I agree with Steel1943. The only confusing lingo in NFCC I can find is "also applies to the copy in the File: namespace" (NFCC#3) and "image or media description page contains [the rationale etc.]" (NFCC#10). The latter should specify if it's the actual TimeText page, its talkpage, or the file description page of the media it is used with. NFC on the other hand looks okay to me. Finnusertop (talk | guestbook | contribs) 17:36, 21 December 2015 (UTC)
  • WMF Legal thoughts: As always, leading off with the disclaimer that this isn't legal advice about what to do or not do. I would also add that I haven't looked at any specific transcriptions, just the issue in general, so you'll all have to decide what action you want to take, I'm just sharing some thoughts on the issue. From our perspective, an addition like adding timestamped lyrics is best looked at as a separate instance of copying and needs its own fair use justification, but likely has a good one. As you may have seen from the fair use wikilegal posting, there are 4 factors that go into fair use: 1) purpose and character (especially whether it's transformative from the original) 2) type of work 3) how much is used and 4) market impact. In this case, the main purpose of transcribing these short recordings seems to me to be to make them accessible for people who, due to disability or other reasons, can't hear the material on a page, which could be a transformative use. Add to that the fact that transcriptions of short snippets of a work use very little of the original and are very unlikely to have any impact on the market for the work and there is a good overall fair use argument as long as no more is transcribed than necessary to illustrate the article where the snippet is present. There is also a good point of comparison in some parody music cases like Fisher v. Dees, 794 F.2d 432 (9th Cir. 1986) and Campbell v. Acuff-Rose Music, 510 U.S. 569 (1994), which held that copying a short part of the lyrics for a transformative parody was fair use, which would be a similar line of reasoning as copying a short part of only the lyrics to make a page more accessible to the hard of hearing. Jrogers (WMF) (talk) 23:04, 21 December 2015 (UTC)
    • It's good to know that we can treat this as fair use, making the issue more how we document that as fair use on WP due to limitations of the TimedText namespace. It does sound like dropping some type of rationale on the talk page would be an acceptable solution. (I do wonder if we could include noinclude/include markup on TimedText pages across the board as with templates as to transcribe the talk or a doc page like we do for templates without disrupting their feature.) --MASEM (t) 23:27, 21 December 2015 (UTC)
  • Some people above come close to saying that if there is a legitimate non-free use for the media file, there necessarily will be a legitimate non-free use for the timed text. I originally thought that would be so, but what would make a lyric transcription satisfy NFCC would have to be that the encyclopedic purpose of the non-free media is to give the lyrics. And after looking a few timed text files, I suspect that is the purpose for few if any:
As I pointed out in the MfD transcluded above, the cited purpose is to illustrate the sound of the song.
In TimedText:Dear Future Husband.ogg.en.srt the media rationale mentions lyrics, but in fact the lyrics of the song are well described in the article.
In TimedText:Like I'm Gonna Lose You.ogg.en.srt the media rationale also mentions lyrics, but IMO the lyrics of the clip are described by the article.
It seems unlikely to me that the few line of lyrics in a clip would significantly increase reader understanding —teb728 t c 11:14, 22 December 2015 (UTC)
Let's assume that the sound sample that contains lyrics is justified for commentary on the music but not in any detail on the lyrics. The reason to include TimedText at that point is for accessibility then, not so much for non-free discussion. And because they are being included in a text form, we don't require the same "license and rationale" business we require for media files. I still do believe we need to have a template tag that asserts these are being used within fair use tied to the media file (perhaps even spelling out accessibility), and that they should be treated as quotes, and thus must be verifiable to a published sourced and not the editor's own ear. --MASEM (t) 15:15, 22 December 2015 (UTC)
I don’t understand your assumption: If you assume that the use is in accord with the Exemption Doctrine Policy, then I agree. But if you are assuming only fair use to provide accessibility, then I note that the EDP is intentionally more narrow than fair use. As the guideline for textual excerpts says, “Brief quotations of copyrighted text may be used to illustrate a point, establish context, or attribute a point of view or idea.” Although the list of purposes is not exclusive, it surely has to include some encyclopedic purpose. If there is no encyclopedic purpose in showing the lyrics, accessibility could be served optimally by a free textual description of the clip. —teb728 t c 01:16, 23 December 2015 (UTC)
Providing accessibility to both deaf readers (consider if it was used in association with a video file) and for non-native English speakers seems perfectly in line with an encyclopedic purpose. --MASEM (t) 01:36, 23 December 2015 (UTC)
Yes, perhaps for a video or perhaps if the lyrics in a song clip were the point of using the clip, etc. My point is that the use of timed text requires a separate rationale: the legitimacy of using timed text does not follow automatically from the media rationale. —teb728 t c 09:14, 23 December 2015 (UTC)
Except, and here's the key point: NFC does not directly cover text - it is for media files, which TimedText is not. Or more specifically we have no special requirements for rationale/etc. for text, beyond that meets WP:QUOTE - that is, a snippet of the full work and directly cited to the work it came. This type of use is acceptable under CC-BY's considerations of fair use of text works and our own policies. The only issue is that we can't directly attach additional information to the TimedText namespace without screwing up its function, as we would normally have a quote immediately followed by a citation. --MASEM (t) 19:00, 23 December 2015 (UTC)
Some information can be encoded in multiple ways. For example, subtitles can either be uploaded as a page in the TimedText namespace, or they could be directly inserted in a video file. Music could either be included through wikicode (e.g.
c'
) or uploaded as a sound file. In such situations, I don't think that Wikipedia policy should make any difference based on whether the material was included by uploading a file or typing in wikicode. --Stefan2 (talk) 21:08, 26 December 2015 (UTC)
  • I am not a lawyer, and I have no intention of arguing about the legality of the situation. If consensus is that this is acceptable fair use, then I shall not object to another administrator restoring the text that I deleted. However, I shall leave that to others. The editor who uses the pseudonym "JamesBWatson" (talk) 15:28, 22 December 2015 (UTC)
  • Notes: From where this was discussed elsewhere: Commons:Video#Subtitles_and_closed_captioning includes adding captioning of non-words; Wikipedia:Manual_of_Style/Accessibility#Video regarding enwiki accessibility. — xaosflux Talk 19:38, 23 December 2015 (UTC)

Next Steps

So where do we go from here? This started from a batch of MFD's on timed texts that I deferred to this board - has a consensus formed on policy and community standards? — xaosflux Talk 14:46, 6 January 2016 (UTC)

Well, consensus does seem to exist that non-free TimedText is not by default a copyright violation, but can fall under the boundaries of fair use. The next step would be to formulate a practice for maintaining TimedText files and their copyright status.Jo-Jo Eumerus (talk, contributions) 15:00, 6 January 2016 (UTC)
I think there are two things that need to be done:
1) Develop a simple template that can be posted to the talk page of a timed text entry that explains the copyright nature, assures a link to the main file that the time text supports (even if that should be obvious by the page name), a source link where appropriate, and then some type of license statement that the timed text used is only to support the relevant media file.
2) Add language to NFC to explain that for non-free Timed Text, this template should be include or should be found on the talk page of that timed text. --MASEM (t) 15:03, 6 January 2016 (UTC)
We also need a process for disputing timed text pages, and templates for this purpose. Normal templates, such as {{subst:nfurd}}, are inappropriate as they break the syntax of the page and because they display the wikicode instead of the informative text. Templates should probably be substituted so that the informative text appears on comment lines on the timed text page. --Stefan2 (talk) 15:20, 6 January 2016 (UTC)
Hmm, I think this also means that if a non-free file has timed text with it, we should make sure that is linked to the file page, so that if the file is later deleted, the timed text is too at the same time (timed text cannot exist without that file). --MASEM (t) 15:27, 6 January 2016 (UTC)
Unfortunately, there is no notification that a timed text page exists when a user tries to move or delete a file for which timed text is available. Some time ago, I created a database query for finding deleted files with timed text. I found a number of such timed text and tagged most of them for deletion per WP:G8. Then I tried to modify the query to find moved files to move the timed text to the right place, but the query became too slow. I can see if I can fix the query so that I can spot moved/deleted files after the move/deletion and then take action, but it would be nice to have this automatically spotted when someone tries to perform an action on a file. See also phab:T122038. --Stefan2 (talk) 15:41, 6 January 2016 (UTC)
On the other hand, a bot that regularly scans through any TimedText files and finds that the parametrized file page no longer exists (in these templates we'd need to create) should be able to tag such pages for deletion. Though we'd still should make sure that admins involved in deleting files check the file page for the "associated TimedText" template and delete that too. --MASEM (t) 17:26, 6 January 2016 (UTC)
I'm drafting a template for TimedText talk pages here.Jo-Jo Eumerus (talk, contributions) 15:37, 6 January 2016 (UTC)
Addendum: Stefan2, do you perhaps know what the significance of the .srt ending on all TimedText files is? Is it a software requirement?Jo-Jo Eumerus (talk, contributions) 16:07, 6 January 2016 (UTC)
It is the usual file name extension for SubRip files. These pages are essentially files, not ordinary pages. Compare with how we store JavaScript and CSS files in the User and MediaWiki namespaces. --Stefan2 (talk) 16:48, 6 January 2016 (UTC)

I've created a parser function inserted into the abovementioned draft template that is able to detect if the TimedText file is associated with a file. Assuming that the title of a TimedText page is always the same as the associated File page plus 7 characters, that is.Jo-Jo Eumerus (talk, contributions) 17:09, 6 January 2016 (UTC)

The language code in the file name can be either two letters or three letters and may include a two-letter country code if a specific dialect is specified. The number of characters at the end may vary, depending on the length of the language and dialect information. A timed text page should always be associated with a file, so a parser template for showing this isn't necessary. It is useful to have a parser template for finding out if a file is associated with a timed text page, though. That way, a warning can be added to {{rename media}} and deletion tags. --Stefan2 (talk) 17:21, 6 January 2016 (UTC)
Technically, the scope of the parser function was to allow the template to detect if the TimedText it is associated with is orphaned as a way to track these, but if the length of the final string is variable it doesn't work.Jo-Jo Eumerus (talk, contributions) 17:23, 6 January 2016 (UTC)
I think that the length of the final string always is 7, 8, 10 or 11. If you are creating a general template for detecting this for maintenance purposes, then we should also add the template to subtitles of free files as it is also useful to know if a deleted free file still has subtitles on Wikipedia. If possible, also try to detect if the associated file is a redirect or if the file is hosted on Commons as we then need to move the subtitles to the redirect target or to Commons. --Stefan2 (talk) 17:32, 6 January 2016 (UTC)

Now, I've found a way to make the parser function work even when there are more than 7 letters in the suffix, by using Module:String. This way, orphaned TimedTexts can be marked.Jo-Jo Eumerus (talk, contributions) 18:02, 6 January 2016 (UTC)

Tracked in Phabricator
Task T122038
Also check if other projects have better methods for catching orphaned and renamed subtitles. For example, Commons probably has a lot of subtitles. There might be something for us to copy. --Stefan2 (talk) 18:05, 6 January 2016 (UTC)
Noting here phab:T122038 where the request to add a warning for Special:Move and ?action=delete for the existence of TimedText.Jo-Jo Eumerus (talk, contributions) 15:17, 7 January 2016 (UTC)
  • Ok, I've finished up my part of the template here, as an userspace draft. Requesting review.Jo-Jo Eumerus (talk, contributions) 20:40, 6 January 2016 (UTC)
    • The template asks for G8 deletion of subtitles for files which have been moved to Commons. In these cases, we should ensure that the subtitles are moved to Commons before the local copy is deleted. I fixed this by replacing the warning with a different warning if the file was uploaded under the same name on Commons. This check fails if there is a local file information page with a DYK template or similar on Wikipedia (the template doesn't discover that the file is on Commons) or if the file is available under a different name on Commons (the template still asks for G8 deletion). Is there a way to improve this? --Stefan2 (talk) 18:44, 7 January 2016 (UTC)
      I don't know if #2 can be fixed at all, but I'll check. As for #1 - maybe the template can check the content of the file page, but that's something I wouldn't bet on.Jo-Jo Eumerus (talk, contributions) 20:00, 7 January 2016 (UTC)
      • Ok, according to my testing a template cannot distinguish between a file on Commons and a file redirect on Commons, at least not according to the specs I have. However, I can distinguish between "File and Media exists", "File exists and Media doesn't" (G8 cases for the file), "File doesn't exist and Media does" (File on Commons) and finally "File and Media don't exist" (orphaned). I'll implement this distinction.Jo-Jo Eumerus (talk, contributions) 20:59, 7 January 2016 (UTC)
        • I've put up a more general discussion on the talk header proposal here.Jo-Jo Eumerus (talk, contributions) 15:14, 9 January 2016 (UTC)
          • Keeping archive bot away while this is still in development — xaosflux Talk 16:26, 16 January 2016 (UTC)

Draft template in Template:TimedText talk template

Pinging Masem (for proposing the existence of the template, as well as for suggestions regarding the copyright notice), Stefan2 (the template now includes a mechanism to marl orphaned TimedText files as well as TimedText files attached to redirects; I don't think it's possible to detect the existence of a TimedText from the File page) and Steel1943. Right now, it has a basic "general" description text (the Commons template doesn't contain much more) as well as an optional copyright warning.Jo-Jo Eumerus (talk, contributions) 15:07, 7 January 2016 (UTC)

Editnotice

I'm assuming these files are normally created "in the raw" (as opposed to with a wizzard). If so, perhaps we can update the namesapce edit notice to encourage adding licensing tempates to the talk pages? — xaosflux Talk 17:34, 6 January 2016 (UTC)

Notices:
  1. Template:Editnotices/Namespace/TimedText
  2. Template:Editnotices/Namespace/TimedText talk
— xaosflux Talk 17:48, 6 January 2016 (UTC)

Clarification regarding general legitimacy of video game screenshots

Apologies if this is a bit longwinded. @Marchjuly: has opened a number of Files for Discussion that I believe follow the same basic rationale. Rather than pointlessly repeat the same points- which I've already had similar discussions about- I'd rather open a more generalised discussion here in the hope of getting more feedback and opinion.

The issue is that of "fair use" screenshots within videogame articles.

Taking "Zool_snes.jpg" as the most typical example, the reasoning used is that the screenshot "is not itself the subject of any sourced commentary within the article so usage fails WP:NFCC#8".

I disagree with this. It's the only screenshot in the article, and in such cases, a well-chosen screenshot is generally the closest one can get to a representation of the subject itself (#1). *Of course* it's been discussed, it's the subject of the article!

The referenced NFCC#8 says "Non-free content is used only if its presence would significantly increase readers' understanding of the article topic, and its omission would be detrimental to that understanding."

Now, I appreciate that "understanding" is open to interpretation in this context (#2), but given that it's impossible to accurately convey the appearance of a visually-oriented arcade game using text alone, I'd say that it's almost a given that a single representative screenshot will almost inherently improve "understanding" of what a game is like!

This applies whether or not the screenshot is the "primary identifier" within the article.

As I commented in a similar discussion regarding a TV programme, rules are always open to interpretation, but if there are (e.g.) two interpretations, one which would leave things broadly as they are, and the other which- if applied consistently- would result in the rejection of the majority of currently-accepted fair-use images, then the former is more likely to be the one implicitly accepted by the community! (And the latter should be discussed as a general policy rather than on an inconsistent case-by-case basis).

I think this is a similar case here; Marchjuly's argument against the Zool image would- if applied consistently- probably result in the deletion of most "fair use" videogame screenshots.

In short, I'm not convinced by Marchjuly's general interpretation of NFCC that seems to underly a lot of these Files for Discussion entries. Is there any previous discussion that can clarify this matter?

(#1) No, the box/packaging isn't the subject; the software is. I dislike the prioritisation of box art over screenshots for this reason, but that's a separate issue at this point.

(#2) Something I already went into in this older FfD, in response to similar arguments from Marchjuly, albeit regarding a TV programme rather than video game.

Ubcule (talk) 23:48, 20 January 2016 (UTC)

I have a bit of bias here since I'm involved with the VG project, but it is generally the case that one non-free screenshot of a video game is acceptable to go along with (what should be sourced) gameplay discussion, to demonstrate what the gameplay is like. This also serves to implicitly show the art style and other factors associated with the game, though it should not be considered a free pass in the way we do for cover art (which itself is used for marketing and promotion, so it is a different situation). The reason here is that unlike the bulk of live-action TV shows, where we have reasonable expectations of what real life looks like, video games are nearly always more abstract between art style, game play, UI elements, and so on, so they do need visual representation alongside their gameplay (at minimum) to help the reader to understand and would make it hard for the reader to understand without. --MASEM (t) 00:12, 21 January 2016 (UTC)
I would also agree.
@Marchjuly: Per policy, the question to focus on is: what does this add to reader understanding?
Showing a "primary means of identification" or being "the subject of sourced commentary" are good reasons for keeping an image; but they are not requirements. (Look up eg Graham vs Dorling Kindersley in the archives of this page [3]).
Of course, if commentary can be found of the visual style, art elements, user interface, gameplay etc compared to the state of the art at the time (eg from contemporary reviews), that is a definite bonus. But for the reasons Masem gives above, especially for early games, which were often very different from each other, perhaps all the more so because of the primary colour limitations of the time, a visual representation of what the game actually looked like really does add hugely to the article. Jheald (talk) 00:30, 21 January 2016 (UTC)
(edit conflict) I was pinged, but not sure if it's appropriate to discuss individual files here and repeat arguments already made in ongoing NFCR/FFD discussions. Those [my] arguments were made based upon my reading of WP:NFC#Meeting the contextual significance criterion and how NFCC#8 has been applied in various NFCR/FFD discussions. Requiring that a non-free image be the subject of sourced commentary when it is not being used as the primary means of identification seems to be the general consensus (according to archived NFCR/FFD discussions I have seen) when it comes to non-free images, regardless of whether it is a former logo, cover-art, a photo, or a screenshot(perhaps among screenshots, video game screen shots are treated differently). The fact that many non-free images are being used in such a way does not necessarily mean that all such usages are automatically NFCC compliant; it could simply mean that their usage has gone unnoticed and their respective non-free use rationales have yet to be evaluated. Anyway, like many things about the Wikipedia, there's quite a bit of room for interpretation so any discussion that helps to clarify the NFCC is most welcome. -- Marchjuly (talk) 00:39, 21 January 2016 (UTC) (Note: This reply was meant in response to Ubcule's original post and ping. I was not aware of Masem's and Jheald's responses while I was editing. -- Marchjuly (talk) 00:50, 21 January 2016 (UTC))
@Masem: @Jheald: You're points are well taken. However, Masem seems to be suggesting that game play discussion should be sourced whereas Jheald seems to be suggesting that is not necessarily always the case. What about articles where there is no discussion, sourced or otherwise, of the visual representation, etc. of the game? Would the usage of a non-free gameplay screenshot satisfy NFCC#8 is such cases? -- Marchjuly (talk) 01:01, 21 January 2016 (UTC)
Per the VG project, the gameplay section should be sourced, and is required for GA/FA. It is nearly always the case that as long as the game is notable, the gameplay has been described to some basic detail in those sources that assert its notability. Even with that, the game is a primary source for the gameplay if that can't be met, but again, it is very much exceptional that a game will not have sources from third-parties to describe gameplay.
In the case where the gameplay is just presented and no one talks about graphics, etc. the case does start to become weaker for the image but I would still argue that it is reasonable to include only to distinguish from infinite possible art styles and the like. There might be a case by case for this where one might not really find a good reason to use a screenshot, but I think in general this is reasonable. (See comment below about software and website screenshots). --MASEM (t) 01:16, 21 January 2016 (UTC)

Can we clarify whether this discussion about the usage of screenshots in general or video game screenshots in particular? The responses of Masem and Jheald seem to be directed more to the latter, while Ubcule has also mentioned a TV programme screenshot in their OP. -- Marchjuly (talk) 00:46, 21 January 2016 (UTC)

I think this specific discussion is limited to video game screenshots, and potentially general software and website screenshots with the same logic that you get one "free" non-free to use to describe the features of the software/website is needed. (but this should still all be sourced). Other screenshots which are primarily otherwise television and film, that's a different beast with different logic to validity. --MASEM (t) 01:10, 21 January 2016 (UTC)
Though as I think about it, there is some similar logic that applies to a television screenshot for the show's page to show the ensemble cast in their roles, as this usually gives some idea of the characterization, show setting and other factors that are also implicit in understanding a long-running show. (eg File:Main_characters_of_Lost.jpg does convey the idea of stranded people and hostility and other character elements that are not directly talked about on the Lost (TV series) page. Similarly File:ST_TNG_Season_One_Cast.jpg gives us character and setting depictions without being necessarily discussed in text on List of Star Trek: The Next Generation cast members) But again here, it is one non-free image that is loosely allowed; any further non-frees must be better justified. It doesn't apply to episode (singular) pages, nor to films, though at the same time this doesn't prevent specific non-frees that are discussed at length to be included. --MASEM (t) 01:28, 21 January 2016 (UTC)
Thanks for the clarification Masem. For reference, I have no problem with an non-free ensemble image being used as the means of identification in an article about the characters of a TV program. I think this is allowed per WP:NFLISTS. I can also see your point regarding "one cover art/one screenshot" in a stand-alone article about a particular video game when there is some commentary (ideally sourced) about the gameplay, etc. within the article. However, I'm not totally following how such a screenshot might be considered acceptable per NFCC#8 when there is no such discussion at all. I guess that might have to do with how I am defining "gameplay". To me, "gameplay" should not simply be a sort of "plot summary" of what happens in the game or what you can do in the game, but it should also include some mention of the technical aspects of the game itself, such as the graphics, etc. I realize this definition may not be the one commonly held when it comes to such articles, but it seems when nothing at all is written about such things that usage tends to be decorative. I am also not sure about how this "One infobox image/one screenshot" principle applies to articles which are not stand-alone articles about a video game or TV program. For example, there are three non-free screenshots (File:Doom ingame 1.png, File:Billdoom.png, and File:Doom gibs.png) being used in Doom (1993 video game). For the sake of argument, let's assume the first one in the "Gameplay" section is acceptable per the "one-and-one" principle. The remaining two screenshots then must be the subject of sourced commentary (which I think they are) to satisfy NFCC#8, right? So, the NFCC usage in "Doom (1983 video game)" seems fine. What about, however, the usage of the same screenshots in the other articles where they are being used? Two of the three are being used in more than one article (FWIW, I'm not trying to open a FFD here about these files. I'm using them as an example to try and gain a better understanding of what is being discussed here) and their usage does not really seem, at least to me, to be NFCC compliant. Each usage is provided with a non-free use rationale, but that doesn't necessarily me they are valid rationales, does it? The "Main characters of Lost" and "ST_TNG_Season_One_Cast.jpg" you referenced above may indeed be considered acceptable for a stand-alone list article or a stand-alone series article. Would their usage, however, be considered acceptable in articles about the TV networks which aired the show or thematic articles about the genre of the show, etc. where the shows themselves may be mentioned, but the screenshot/image itself is not really the subject of any sourced commentary? -- Marchjuly (talk) 07:44, 21 January 2016 (UTC)
For the Doom screenshots in other articles, I would immediately say there is no automatic allowance there - there needs to be sourced commentary about the images to be used anywhere else outside of the game article. Same thing with the character shots - their use outside the show or character list page, such as a genre article, must be supported with sourced commentary. Basically, it comes down to the fact the screenshot of a game only has the reasonable allowance to be used on the article about that game, on the presumption that the gameplay is discussed in a manner to make the screenshot useful, and any other usage will need strong justification. In terms of the first part of your comment, sometimes the graphics or other visual elements are described (a game noted for cel-shading or for using tilt-shift imagery for example), but often just showing the screen can be a way to illustrate the type of graphics used without having to say what they are. It's more that the screenshot with its caption should help orient the reader to understand the nature of the gameplay they are reading about, which is more difficult if there are no visual cues for that (eg omission would harm the reader's understanding of NFCC#8) --MASEM (t) 16:50, 21 January 2016 (UTC)
@Marchjuly: To clarify, this is primarily about "fair use" video game screenshots.
The TV programme link was used because- despite the different subject matter- many of the same principles and arguments applied and I'd already made them there.
In particular, this is an issue because:-
(i) I don't want to have countless near-identical per-file discussions going over the same basic issues; that's obviously unworkable.
(ii) NFCC rules- and their interpretation- should be applied *consistently* to all articles in the field.
(iii) In fact, interpretations of rules should be applied consistently *everywhere*. That is, if a *single* representative example of the appearance of a TV programme (a primarily visual medium) or arcade video game (again, a primarily visual medium) isn't inherently justifiable, then what is the justification for (e.g.) having a "fair use" image of Alistair MacLean when its inclusion arguably adds nothing to understanding his importance as an author- i.e. the reason for his notability- or understanding his literary work? His appearance is irrelevant to that. This strikes me as inconsistent.
So, what I want is some consistency and clarity regarding the use of videogame screenshots because I want to contribute without rehashing the same issues every time.
However, if people are going to argue that their use (in general) isn't justified, then I'd also like to see contradictions like the one above addressed too. Ubcule (talk) 19:16, 21 January 2016 (UTC)
  • Failure: I read Zool#Gameplay (note: I've never played nor even heard of Zool) trying to see what additional information the screenshot might give me that the prose did not. The only thing that perked my interest was the description of the prime character as a gremlin ninja. Of course, that imagery is clearly depicted on the box art. I fail to see anything else the screenshot gives me. There's no discussion anywhere of what we are seeing in the screenshot. This is an abject failure of WP:NFCC #8. So, the only way to include this image would be via a free pass arrangement. This would need to be outlined in WP:NFC. We can't do this per-article. The debates would never end. The line being danced about there being an automatic allowance is too fine of a point. Really, if we accept these images we need to modify WP:NFC to indicate there is an allowance of one non-free screenshot per game article under a 'free pass' arrangement as box art has. Of course, if we do this we're going to have to grant an allowance of one free screenshot per TV episode too. Oh the slippery slopes we tread. To the OP; the idea that we should default to the former case of yours, where we allow lots and lots of non-free content in this case, is false. The default case on Wikipedia is to NOT allow non-free content. We make special exceptions, which are governed by policy and some of which are outlined in guideline. --Hammersoft (talk) 17:11, 21 January 2016 (UTC)
    • An issue is that the gameplay in Zool is far from what is normally expected for a video game article of quality. It can be improved (though I have not played it so I can't do that myself) to describe what actions the player takes in the main game, which would better justify the use of the gameplay image. Now arguably, that should be done now since the file has been flagged, because key again is that these screenshots should be associated with a sourced gameplay section, and this clearly is not. But it is also something that should be fixed, and when fixed to standards we expect, it should be okay, so deleting it presently for purposes of lacking the necessary quality is a problem. And no, this doesn't translate to television episodes, as most television episodes are live actors doing mundane or easily described things that do not have a unique art style or interactivity. Some episodes are unique to that extent, but that's exceptional, not the rule. --MASEM (t) 17:30, 21 January 2016 (UTC)
      • To the last point first; most (all?) television episodic series have art directors. Those directors create a feeling of a universe in which the characters are depicted. This extends to every facet of their environment; clothing, transportation, accommodations, offices, other places of interaction. Thousands of people make their livelihoods developing such environments. If one could make an argument that a screenshot is necessary to depict gameplay, it is equally necessary for episodic series. Any eloquence that would permit prose to describe a scene in an episodic series such as this from Big Bang Theory is equal to the task of describing a game. As to allowing the image pending improvement of an article; with no comment directed at anyone, much less you, such a notion is absurd. Huge numbers of articles languish for years pending improvement. Shall we create another sub (or sub-sub) directory of Category:Wikipedia articles with content issues and hope that someday, somewhere down the road, somebody will improve Zool (a game made for platforms where the most recent was discontinued 13 years ago) to support the image? We do not include non-free content pending the creation of free content that could be created. We should not be in the business of supporting non-free content in the hopes that free content prose will come along to support it. If and when free content prose is created to support it, then a non-free content could be included to support it. Until such time, absolutely not. --Hammersoft (talk) 18:36, 21 January 2016 (UTC)
As I've argued elsewhere, arcade video games (and other ones to a more debatable extent) are primarily a visual product, and generally- as Masem stated- have their own distinct style. It hinges on one's interpretation of the word "understanding" in NFCC, but I think it's clear to say that one can't "understand" the appearance and style of a game from textual description alone.
One can describe it in words, just like one can describe almost any painting in words. And, just as with the painting, I don't think it's arguable that text alone- regardless of how well-written- can convey what a single visual example does.
This- I believe- is what makes (at least) a single, representative example justified in most cases, regardless of whether or not the screenshot itself is the subject of "sourced commentary".
Specific to the "Zool" example, I don't feel the need to seek a "free pass". I think it's clear that one would have a weaker impression of the game even without that virtually-uncommented screenshot, which justifies its use as "its omission would be detrimental to that understanding". The box art shows the character, but nothing else about the game.
Ubcule (talk) 19:37, 21 January 2016 (UTC)
There is a difference between a television series article, and a television episode. Everything you said about the art production of the television episode is really what applies to a TV series overall, and not necessarily to specific episodes. Hence, one non-free image to help express the art style of a TV show on the TV show's article is completely reasonable, but that does not apply to the episode since it is established in the show's article. An individual episode may be of not for its unique art direction, but that's where I would expect to see unique art styles the subject of sourced discussion to explain why the episode is unique and requires a screenshot. Episodes are one of those things that we have refused to allow a free non-free without justification as we do with cover art.
Also do note that with emulation and the like, it's hard to consider a discontinued game having no importance. --MASEM (t) 22:17, 21 January 2016 (UTC)
  • (Came here linked from WT:WPVG) Pardon if I reiterate something from above, but my understanding of NFCC was that for video games contextual significance is basically "gameplay". There is no way to explain gameplay (and sometimes art direction) in prose without an aid of a screenshot for virtually any game. The gameplay is always at least sourced to the game itself, but almost always there is at least basic description in sources (or it wouldn't pass GNG anyway). Even an in-depth gameplay description fails to convey what a single screenshot can. Therefore, such a screenshot gives a lot of context to what an overview description is talking about. In practice, many video game article are terribly underdeveloped and don't explicitly describe the content of the screenshot. Is this a question that WMF's legal should be answering or is it up to us to interpret and apply NFCC? —  HELLKNOWZ  ▎TALK 23:04, 21 January 2016 (UTC)
  • As someone who has written many video game articles, a screenshot is very helpful in illustrating the concept of a game's gameplay. I'm very much so of the same mindset of Masem. The problem isn't so much the images here, but rather, that the video games targeted here are in rather poor shape. Perhaps what we're saying would make a little more sense upon reading some GA's or FA's. Something like Golden Sun, where its got a good caption and good gameplay section, to better demonstrate how its supposed to be implemented. The answer is the improve the content, not delete it. Sergecross73 msg me 14:00, 22 January 2016 (UTC)

Motion to modify WP:NFC

(pinging prior contributors to the above discussion; @Sergecross73:, @Hellknowz:, @Masem:, @Ubcule:, @Marchjuly:, @Jheald:)
Given the apparent overwhelming support for using a screenshot in a game article in addition to the box art, I propose the following be added to Wikipedia:Non-free_content#Images:

6a: Screenshots of gaming products: The use of a single screenshot for visual depiction of the gaming environment in the context of other critical commentary about the gameplay.

--Hammersoft (talk) 21:19, 22 January 2016 (UTC)

Support

  1. Support: Given I'm the only one opposing above, it seems irrational to continue my opposition, and rather assist in crafting the consensus decision wording from above. --Hammersoft (talk) 18:29, 22 January 2016 (UTC)
  2. Support: I believe this is more or less how the WikiProject has been handling it anyways, so its good to get this documented to eliminate future problems like this. Sergecross73 msg me 18:37, 22 January 2016 (UTC)
  3. Support: It logically follows from #6 for software products but I agree that maybe codifying this would be better. I'd only consider crafting the language to be compared to the existing ones, and perhaps pointing to WP:VGIMAGES where we can expand on clarity of what "critical commentary about the gameplay" requires. --MASEM (t) 18:41, 22 January 2016 (UTC)
  4. Support: (edit conflict) This has been the practice and the current state of most video game articles. To avoid future issues, we may as well document that one screenshot is pretty much contextual to the gameplay. (P.S. I'm supporting the general idea and am okay for improving the wording to better align with NFCC points and consensus.) —  HELLKNOWZ  ▎TALK 18:44, 22 January 2016 (UTC)
  5. Support: This seems a reasonable use. The article about the video game presumably discusses the graphics and gameplay. A well-chosen image that illustrates these things would therefore complement the article text and would not be simply decorative. Specifying that it should be one image, at least in most cases, would also be helpful guidance. Seraphimblade Talk to me 23:22, 22 January 2016 (UTC)
  6. Support adding a specific text would make clear what is desired here. Graeme Bartlett (talk) 21:42, 24 January 2016 (UTC)
  7. Support: As someone who struggles when it comes to writing the gameplay section of an article every single time, this proposal is critical. Axem Titanium (talk) 18:06, 25 January 2016 (UTC)
  8. Support This seems to be in line with the standards we use for other topics. HighInBC 18:08, 25 January 2016 (UTC)

Oppose

  • Oppose. In its current form, too arbitrary and artificial, and out of line with the policy.
The question to consider is what the image adds to understanding.
If the image is supposed to show game-play, but the game play in the image cannot be understood by the reader; or if the game play is so obvious that the image adds nothing to comprehension, then the argument that image shows how the game is played is inappropriate. On the other hand, in some cases it may take more than one image to communicate the game play.
A second reason to show images is to show the reader the graphical style and execution of the game. In many cases, that is something that will be apparent to the reader directly from the image, regardless of the accompanying text.
If we are serious about giving an example application of the NFCC #8 criterion, which is what these examples are supposed to do, then the proposed text needs some more crafting. Jheald (talk) 23:15, 22 January 2016 (UTC)
(Note also WP:Voting is evil. It is appropriate to discuss and refine a text, before putting it to a final decision.) Jheald (talk) 23:26, 22 January 2016 (UTC)
  • Oppose for some of the reasons expressed by Jheald above and by myself in other parts of this thread. I think care needs to be taken when creating an "exception" such as this. I am not sure whether a gameplay screenshot or the game's cover art should be used within the infobox, but outside the infobox I think that NFCC#8 requires us to expect a little more contextual significance than only a simple caption or just the image itself. The image and article content should work together so that omitting either would be detrimental to the reader's understanding, but the article content should come first. Simply adding such a screenshot to an article and saying it in and of itself provides all the contextual significance the reader needs may seem fine to those with knowledge of video games, but it seems decorative to me. Whether this content should be sourced may be debated (I think it should to avoid problems with WP:NOR), but no article content at all seems like a gallery-like usage which tends to be considered purely decorative per WP:NFG. Moreover, if the screenshot really does significantly improve the reader's understanding, then it should be easy to add information directly to article to further enhance the image's relevance. Just putting in a short caption means that all the information will be lost if the image is removed or replaced; moreover, if you're going to write a long caption and cite sources, then it seems better to simply add that information to the article in the first place. The person adding the image should is required to provide a valid non-free use rationale which speaks to the specific use of the image per WP:NFCCE, and not some generic nfur copy-and-pasted from another image or another usage of the same image. If they are capable of writing such a rationale, it should be no problem for them to add the relevant content to the article itself. Finally, I disagree that this is the standards we use for other topics. Non-free logos, cover-art, photos, etc. are all expected to be the subject of sourced commentary when they are used outside of the main infobox. Common practices may be different, but there's lots of non-free images being added to articles which is not appropriate per NFCC, just like there are lots of articles created which do not satisfy WP:GNG. The fact such stuff exists does not automatically make it policy/guideline compliant. If you're going to give a single non-free screenshot a pass for video game article, then I think it's going to be hard to refuse other similar requests for other non-free images in other genres of articles. -- Marchjuly (talk) 02:00, 26 January 2016 (UTC)

Discussion

I do not believe that attempting to craft a position where we allow a game screenshot sometimes, and not in others is a tenable solution. It will create an inordinately tangled web of justifications, and unending debate about the use of such images in specific circumstances. We must have either a blanket allowance for one screen shot depicting game play, or no allowance unless there is secondary sourced commentary regarding the interface (as is the current status quo). --Hammersoft (talk) 18:29, 22 January 2016 (UTC)

  • Comment: For reference, I actually think what you [Hammersoft] wrote above begining with your "Failure" post made perfect sense to me. I totally agree with your Oh the slippery slopes we tread. To the OP; the idea that we should default to the former case of yours, where we allow lots and lots of non-free content in this case, is false. The default case on Wikipedia is to NOT allow non-free content. We make special exceptions, which are governed by policy and some of which are outlined in guideline. and We do not include non-free content pending the creation of free content that could be created. We should not be in the business of supporting non-free content in the hopes that free content prose will come along to support it. If and when free content prose is created to support it, then a non-free content could be included to support it. Until such time, absolutely not. statements. So, I guess I'd consider myself to be one who opposes the "free pass" concept, at least for now. I think the problem with making exceptions like this is that most of the posters in favor seem to be editors involved in the video game project. I'm not saying that their arguments should be discounted simply for that reason because I think their arguments do have some merit and have really helped to clarify things for me. The problem is that I feel if a similar proposal was made for granting a "one infobox/one 'free' non-free" exception for sports team logos, company logos, broadcast logos, scouting logos, etc. and only participants from from those respective WikiProjects were invited to discuss (such as like WT:VG#Image deletion discussions you should be aware of by Hahnchen, which seems a little problematic per WP:CANVAS) then you'd probably find that there were more in favor than not of receiving the same exemption. I think it would be better to turn this into a full-fledged RfC just to get more involvement from different parts of the community and develop a broader and stronger consensus. If the arguments in favor of this "one infobox/one screen shot" approach are sound then experienced editors will see it as such and the RfC will pass. It's possible, however, that some editors (outside this particular group of "non-free image jobsworths" and videogame editors) may be able to argue effectively against it. I think it might be a good idea to follow what was done at WT:ETHNIC regarding WT:ETHNIC#The necessity of galleries of personalities in the infoboxes and WT:ETHNIC#Proposal for the deletion of all the galleries of personalities from the infoboxes of articles about ethnic groups. That started out like this as a talk page discussion, but morphed into a full-flegged RfC. There are still some issues left to be worked out with that, but for the most part the consensus appears to be fairly solid since it was challenged here at ANI and determined to be appropriate. -- Marchjuly (talk) 21:01, 22 January 2016 (UTC)
    • (edit conflict) I think some of the reasoning is that it is not an exception or a free pass, but that articles having proper gameplay description are by default giving the sourced context that NFCC asks for (above examples were not good examples of gameplay description). Poor gameplay articles shouldn't have screenshots and I think that's what the wording says ("critical commentary"). I do agree that full RfC is a better idea. —  HELLKNOWZ  ▎TALK 21:17, 22 January 2016 (UTC)
      • Understand. Would the "critical commentary" in the proposed "in the context of other critical commentary about the gameplay" include image captions. The reason I'm asking is because of recent "Keep" arguments made to Wikipedia:Files for discussion/2016 January 19#File:WWF Wrestlemania 1991 Ocean game.jpg seem to suggest that the bare minimum needed is a caption. However, I think if the added context can be included in a caption, then there's no reason why the same contextual content cannot be added to the article and the image located appropriately near said content. Requiring only a caption means that the image can be locted basically anywhere, doesn't it? Moreover, if the image is removed for any reason than the "critical commentary", which might still be quite informative, would also be removed by default. -- Marchjuly (talk) 21:27, 22 January 2016 (UTC)
        • Remember that, per policy (NFCC #8), the point of having images is to improve understanding about the topic. That is the acid test. The focus should not be about having to match a particular stylistic form; nor is the issue about supporting text. The issue is about whether we think the image significantly adds to understanding.
In some cases, eg early 8 bit games, simply the image alone will add to understanding, without any commentary, by showing the (often quite unique) choices that were made, given the restrictions of the platform.
In some cases a caption will be quite enough for the reader to be able to gain the understanding the image can provide.
In other cases it may be enough simply to explain the gameplay in order for the reader to be able to understand what the image shows, for which the game itself is verification.
The great virtue of the law on fair use in the United States, which the NFC policy tries to maintain, is its flexibility to be able to do the right thing in the right circumstances.
This is why the NFC policy focuses on the questions to consider, (similar to the fundamental WP philosophy on rules and guidance, WP:IAR), to retain that flexibility to do the right thing, rather than too many artificial prescriptions and instruction creep. Jheald (talk) 23:03, 22 January 2016 (UTC)
  • Not trying to disagree just to be a pain, but I think it's important to remember that there are two parts to NFCC#8. It could be pretty much argued that adding any relevant image would improve the reader's understanding to a degree (perhaps even a significant one), but the other part of NFCC#8 says that omitting the image would be detrimental to that understanding. For some reason, that part (which I think is just as important) often seems to be discarded. Morever, NFC may be somewhat based upon US "fair use" law, but it clearly says in WP:NFC that "The use of non-free content on Wikipedia is therefore subject to purposely stricter standards than those laid down in U.S. copyright law." -- Marchjuly (talk) 23:24, 22 January 2016 (UTC)
  • I would say (and have said, often) that the two halves are equivalent, just two ways to look at the same question. Understanding you gain from the image being added is understanding you lose if the image is taken away. The important thing (IMO) for the community to focus on is what sort of understanding is considered "significant", as weighed against how much of a copyright-taking the image represents.
(The copyright-taking e.g. from an incidental screenshot being rather less than the taking from the fair-use reproduction of an iconic photograph). Jheald (talk) 23:34, 22 January 2016 (UTC)
The two parts of NFCC#8 are two separate, independent tests. It is always nearly possible to argue for the inclusion of an image to enhance the reader's understanding. But if that enhancement is above and beyond what is encyclopedically necessary for understanding, such that its removal or absence does not harm the basic encyclopedic understanding of a topic, then it fails the second test. We only include NFC to do a necessary job, with a smidgen of leeway in some areas but still far stronger than fair use allowances. --MASEM (t) 23:47, 22 January 2016 (UTC)
NFCC #8 doesn't say "necessary" it says "significant". The question is whether taking the image away will reduce ('detriment') the understanding the reader would have in a way that the community (and, perhaps, hypothetically a court) would regard as significant. "Encyclopedic" (as used in some other policies, though not here) is a kind of circular word. It's basically code for what aspects of what an article might convey are considered to be of significance in terms of what the community thinks an encyclopedia article should convey. Jheald (talk) 00:00, 23 January 2016 (UTC)
Test 2 of NFCC#8 is about omission/removal of an image being determinental to understanding of a topic. This implies "necessary" as relating to having the reader understand the article when they are done reading it. Note that "significant" only applies to the first test, not the second. It is very easy to lawyer and argue that any image "significantly improves" the reader's understand, because of the adage "A picture is worth a 1000 words", but it is much more thought to assess if the removal/omission hampers the article enough to be not useful to the reader. And yes, while "encyclopedic" is sorta circular and vague it still represents a standard that we are not the same as a newspaper, magazine, blog, or other type of media source. We are trying to impart knowledge, and we should only be using non-free when it is necessary to help impart that knowledge (the "educational purpose" that is from the WMF resolution). --MASEM (t) 00:20, 23 January 2016 (UTC)
The circularity is that what knowledge we deem to be "educational" about a topic and worth imparting is pretty much what the community decides is worth imparting. But that said, as a community I think we have evolved a pretty general sense of things we do think are valuable to try to convey.
As to the first point, it is worth flagging that NFCC #8 is all about understanding of the topic, not understanding of the article.
I don't think I agree that the word "significant" doesn't apply to the second half. Any loss of understanding is detrimental; what we're concerned about is whether the detriment is significant.
(Note also that the wording "that understanding" is a pretty clear call-out to the understanding being discussed in the first half of the clause).
But I do agree that there are people who try to claim that any addition is significant, in a kind of abstract way. I think it does throw the question into sharper relief to concretely start with the image, and that to ask: was your understanding of the topic really so significantly affected by its removal? Jheald (talk) 00:48, 23 January 2016 (UTC)
The way NFCC#8 is phrased in the second half is asking a question not about if you'll miss the image if it is removed, but if it is harmful to our purpose as an encyclopedia if it is removed ("detrimental"), which is a far different question, more than just if one is "significantly affected" by its removal. I've seen more than enough FFDs where the editors arguing keep try to justify retention because it is an "important" image or the only image of its type or the like, no other website are tracking these types of images, etc., but that's not the question the second part of NFCC#8. This is why it is very different test than just asking if your understanding is significantly improved by the inclusion. --MASEM (t) 01:01, 23 January 2016 (UTC)
I'm sorry, I don't understand how you equate "detrimental to that understanding" with "harmful to our purpose as an encyclopedia".
"Detriment" simply means loss, damage, or harm [4] -- for example, in the way that something is damaged or harmed if a part of it ceases to exist.
While interesting, it's 1 am here, and this discussion has probably gone on long enough for an issue that is at best tangential to the question at hand. I suggest we hat it and collapse it from 23:47 (or 23:34 if you prefer) onwards. Jheald (talk) 01:17, 23 January 2016 (UTC)

Don't mean to butt in, but all of this discussion of what is "significant" brings me back to what Hammersoft posted above: We should not be in the business of supporting non-free content in the hopes that free content prose will come along to support it. If an image is truly significant to the reader's understanding, then it seems to me that it would be important for such information to be included within the article itself by the person adding the image when the image is added and the reason for doing so specifically added to the nfur. Just seeing the image may be informative, but how is a reader who is not very knowlegable about video games to know what is significant from simply seeing an image. I think the context provided should also be more than what can be expected to be written in a caption. I'm not suggesting it should be an automatic delete of an image when it is lacking such context, but I do think when such images are being discussed at FFD, it should be up to those !voting "keep" to improve the article and add such context per WP:NFCCE. We expect editors adding or readding unsourced content to provided a citation to a reliable source per WP:BURDEN. If someone wants to "keep" an image, then I'm not sure why we shouldn't also expect the burden to be on them to improve the article in a way that eliminates any non-free concerns. Simply adding an image (and possibly a short caption), adding a bolierplate nfur and then hoping somebody else comes along and adds the required content to the article does not seem to be sufficient in my opinion. -- Marchjuly (talk) 01:28, 23 January 2016 (UTC)

Does the reader need to be told why an image significantly adds to their understanding, in order for it to significantly add to their understanding? (For whatever value of Δ(understanding) the community deems sufficient significance to justify inclusion).
I would say: no, not always, it depends on the context. For people, for comic-book characters, for cover images used to promote media, for unbuilt buildings, we have taken the view that simply to know what those things look like is significant and encyclopedic.
I would say that in many cases that also applies to video games -- particularly from the period before quasi-naturalistic rendering was possible, simply to see what the game looked like is something that will add something valuable and worthwhile to topic understanding, even without any prompting or cueing.
Similarly, if an image makes understanding of the gameplay much more concrete, it may take very little explanation of the image, that could even fit into a caption, to unlock that understanding.
There's maybe more of an issue if a modern quasi-naturalistic game has a particular visual character (eg perhaps particularly dark and moody) that the reader might not know was particularly characteristic of the whole game. But even in such a case, the reader nevertheless gets an idea of the quality of execution and art direction. An analogue, perhaps, is images of paintings. We generally do allow an image of a painting that is the subject of an article -- regardless of whether it is naturalistic or abstract, modern or old -- and usually regardless of whether or not there is text discussing the image as an image. One could argue that video games are a similar artificial human-created milieu, and it is quite revealing to the reader even just to see some of the choices that have been made. Jheald (talk) 02:01, 23 January 2016 (UTC)
  • @Marchjuly: (or anyone else for that matter) please feel free to RfC tag this. I didn't think it was really necessary to do so since we've made modifications to the NFC guideline before without having to do an RfC. But, it's fine if someone wants to tag it. --Hammersoft (talk) 21:26, 22 January 2016 (UTC)
    • I've never created an RfC before so not sure exactly how it's done. Maybe it would be better for Ubcule to do since they posed the original question? -- Marchjuly (talk) 03:14, 24 January 2016 (UTC)
      • RfC tag placed on this page, AFAICT, this should result in bot doing the rest. Ubcule (talk) 21:09, 24 January 2016 (UTC)

RfC: Guidelines regarding fair-use images within videogame articles

Discussion on Guidelines regarding fair-use images within videogame articles: Originally intended to cover general issues popping up repeatedly in multiple individual Files for Discussion (specific screenshot for specific videogame) and to achieve clarification on existing policy. However, this has resulted in significant discussion including a "Motion to modify WP:NFC" and further input and opinion on this would be valuable. See Clarification_regarding_general_legitimacy_of_video_game_screenshots. 21:07, 24 January 2016 (UTC)

Image size discussion at Wikipedia talk:Manual of Style/Images

Opinions are needed on the following matter: Wikipedia talk:Manual of Style/Images#Fixing images below the default size. A WP:Permalink for it is here. The discussion concerns whether or not we should keep the following wording: "As a general rule, images should not be set to a larger fixed size than the 220px default (users can adjust this in their preferences). If an exception to the general rule is warranted, forcing an image size to be either larger or smaller than the 220px default is done by placing a parameter in the image coding." The latest aspect of the discussion is the 1.4 Amended proposal (2A) subsection. Flyer22 Reborn (talk) 07:15, 25 January 2016 (UTC)

This discussion has progressed to a WP:RfC: Wikipedia talk:Manual of Style/Images#RfC: Should the guideline maintain the "As a general rule" wording or something similar?. A WP:Permalink is here. Flyer22 Reborn (talk) 21:52, 25 January 2016 (UTC)

Tidying up "di-XXXX" templates for non-free use

Based upon the ongoing discussion at WP:AN#NFCC tagging by bot, and all of my changes after review being undone-request for admin oversight, I am wondering if there is a way to clarify the wording of templates such as Template:Di-fails NFCC so that it's easier to see who is allowed to remove such templates and what are the acceptable reasons for removal. It also might be helpful to add something about Template:Di-replaceable fair use disputed (or a similar template) that can be used by those wishing to contest/challenge the initial tagging of the file. Part of what came out in the AN discussion is that administrators may not check the file's talk page for comments prior to deleting a file (even though the template's wording instructs them to do so), so I think it will be helpful if a way can be found to alleviate such concerns so that templates are not prematurely removed. On the other hand, if such templates can be removed by any editor, then this should be clear as well. Template:Di-disputed fair use rationale is another template which may also need a bit of clean up as well. -- Marchjuly (talk) 01:52, 28 January 2016 (UTC)

Perhaps the way forward here is that if an editor removes such a template, they should be encouraged to notify the person who added it.
Then if there is disagreement, the two can take it to FFD. Jheald (talk) 13:02, 28 January 2016 (UTC)
  • It would be useful to warn admins to check the talk page. {{#ifexist:{{TALKPAGENAME}}|{{imbox|text=Note that there is some text on the talk page:<br>{{ {{TALKPAGENAME}} }}}}}} or some variant thereof could be added to all deletion templates. I don't know if admins regularly check file talk pages when deleting files. Most file talk pages only contain a wikiproject template if the talk page exists at all, so maybe some admins think that it is a waste of time. If my proposed wikicode is used, then the talk page will be automatically transcluded by the deletion tag, so the admins won't have to click on an extra link. This is maybe more useful in the file namespace than in other namespaces.
It would also be useful to have something like {{hang on}} for all DI tags. Maybe we could just redesign and rename {{Di-replaceable fair use disputed}} so that the same template can be used for any DI tag. --Stefan2 (talk) 13:59, 28 January 2016 (UTC)
  • I think it's unrealistic to expect admins to check the talk page -- they haven't in the past, particularly when there's a backlog to resolve; and also given that they may be using semi-automated tools.
Also, there are a number of issues where there are simple formalities to fix, where the user may well be able to fix them, and then removal of the tag may be completely appropriate.
If there is a genuine disagreement about the image, then it should go to FFD, rather than having a countdown to automated deletion hanging over it. I would regard this template as very similar to a WP:PROD, and it seems to me the same rules should apply. Jheald (talk) 15:11, 28 January 2016 (UTC)
  • I think that {{di-fails NFCC}} only should be used for obvious violations of the non-free content criteria but that the admin who processes the tag should be the one who determines if it is an obvious violation or not. If it is not an obvious violation, then it's better to list the file at FFD, but it seems unnecessary to list files at FFD only because the uploader doesn't understand the non-free content criteria. When a file by a new user is tagged as replaceable for violation of WP:NFC#UUI §1, the user sometimes adds a template disputing the violation, and the admin will then only delete the file if it is an obvious violation of the criterion, but listing lots of those files at FFD would fill up FFD with too many files. --Stefan2 (talk) 15:23, 28 January 2016 (UTC)
I think Stefan2's suggested of transcluding the talk page sounds interesting, especially if it helps admins see that the "di" template has been disputed. Also, think that one template for all for all contested NFCC speedy deletions sounds like a good idea. Personally, I'm not sure why {{di-disputed fair use rationale}} and {{di-fails NFCC}} are both needed since they both seem to deal with the same thing, don't they? If a particular usage of a non-free file fails one of the NFCC, then a valid non-free use rationale cannot be written for that usage, right?
As for removing the templates, I realize that sometimes they are added/removed by vandals and other times in good faith. Even so, I don't think they should be simply removed without any good faith attempt at discussing the reason(s) why on either the file's talk page or the page of the user who added the template. Perhaps this could be done by a template, such as is done for {{Deprod}}. If there is still disagreement about whether the problem has been "successfully" fixed, then the discussion should be moved the FFD. Right now, the current wording of "di-Fails NFCC" seems to imply that only an administrator should remove the template (which is why I re-added the templates discussed in the above-mentioned AN thread), whereas "di-disputed fair use rationale" seems to suggest that anyone may remove the template. If it's acceptable for anyone to remove either of the two, even in bad faith like it is for PROD tags, then perhaps they should be similarly worded to reflect that a template should not be re-added once it has been removed (for any reason) and that any outstanding concerns about the file should be discussed at FFD. -- Marchjuly (talk) 00:44, 29 January 2016 (UTC)
{{Di-disputed fair use rationale}} means that someone argues that the fair use rationale is invalid. A valid fair use rationale should specify why the file satisfies all non-free content criteria. If the file fails any criterion, then the fair use rationale is by definition invalid since it doesn't provide a valid rationale for that criterion, so the templates seem to duplicate each other to some extent. The main difference is that {{di-disputed fair use rationale}} requires the user requesting deletion to write in words why the file fails WP:NFCC, while {{di-fails NFCC}} requires the user to specify which criterion the file fails. There is also {{di-missing article links}}, which I tend to use if there is a fair use rationale but the fair use rationale either refers to more than one article or only to articles which do not contain the file. --Stefan2 (talk) 13:53, 29 January 2016 (UTC)
I didn't know about "di-missing article links" and have never seen it used before. At first glance, that seems unnecessary since it is a fairly minor fix to check the "What links here" and fill in |article=. However, I guess even that could be not so simple for the reasons you give. Anyway, the language in that one seems to imply anyone can remove the template if they "fix" the problem which again seems a bit different from "di-fails NFCC". -- Marchjuly (talk) 01:36, 30 January 2016 (UTC)
If the file was uploaded for the purpose of being used in Article A and it has a FUR for Article A but it currently only is used in Article B, then the FUR typically needs to be completely rewritten as the purpose and use of the image doesn't match the new article. This situation is more like {{subst:nfurd}}. {{Di-missing article links}} doesn't mention any notification template. Not finding anything better, I typically use {{di-disputed fair use rationale-notice}} when tagging files with {{di-missing article links}}.
There's also {{di-missing some article links}} which is used in a lot of situations where a file has a fair use rationale for some articles but the file also is in use in other articles for which there is no FUR. It would be necessary to go through all those files and either add FURs or remove files from pages. --Stefan2 (talk) 14:11, 30 January 2016 (UTC)

Jimbo on NFC philosophy

Interesting comment from Jimbo (diff) in a conversation Philosophy of free culture at the WMF on his talk page:

Here's the way I think about this issue - I wonder if you disagree. When fair use competes with free images, we should strongly prefer to avoid fair use. When no free image is possible, then we "should make maximal rather than minimal use of Fair Use". What we don't want to happen is to have free culture contributors disheartened when their perhaps slightly less-good image has to compete with a fair use image of slightly higher quality. And we don't want to say "Well, there's no free picture of X, so we'll grab one and claim fair use" when it would be easy enough to get a free picture of X.

Fair use is a really important principle that we should like to see expanded. Just because someone owns a copyright to something, does not give them absolute and final authority over every possible use - and we should be clear about that. At the same time, a too easy reliance on fair use can harm incentives to great free stuff. So balance is required.

Jheald (talk) 12:58, 28 January 2016 (UTC)

  • Jimbo's opinions on the matter, whether they are pro or con, have as much bearing as the Foundation's at this point (i.e. none). The project is overrun with non-free material. Neither the community here, the Foundation, nor Jimbo are willing to do anything to curb that. Given this circumstance, the status quo is that WP:NFCC is void, and people may use non-free content...to quote Jimbo...maximally. --Hammersoft (talk) 14:12, 28 January 2016 (UTC)
    If we accept the use of non-free content in the Mobile apps (the Search function and some other features at least on Android currently use non-free images for navigation and decorative purposes) we could indeed allow a lot more non-free content here. Or we could remember our principles, reduce non-free material here and complain to WMF developers until the Mobile apps conform to our policies. —Kusma (t·c) 14:20, 28 January 2016 (UTC)
  • The Foundation's previously been approached about the overburden of non-free content on this project. Their response has been silence, except for one instance where their legal counsel chimed in that we had wide liberty to use non-free content under fair use law as an educational resource. Though not intended as a statement of policy about the use of non-free content, the juxtaposition of this response with their silence on other fronts leads one to conclude it is a defacto policy statement about the use of non-free content in general. Few care anymore about limiting non-free content. There were never enough that cared to be a strong voice supporting free content. The few there were have been demonized, and many have been forced off the project, retired from the project, or if still here don't enforce it anymore. --Hammersoft (talk) 14:43, 28 January 2016 (UTC)
  • The statement part "Just because someone owns a copyright to something, does not give them absolute and final authority over every possible use - and we should be clear about that." is actually contrary to US copyright law, which we have to follow. The copyright law gives exclusive authority to a work by the copyright owner. Fair use is a defense, not a right or privilege, to allow limited use of material without having to gain the copyright owner's approval for educational purposes. --MASEM (t) 15:05, 28 January 2016 (UTC)
  • @Masem: Courts tend to construe fair use historically as an extension of the 1st Amendment, so very much a right. The actual statute (see Fair_use#U.S._fair_use_factors) is quite explicit: "the fair use of a copyrighted work... is not an infringement of copyright." So Jimbo is quite right, it is a limitation on copyright -- copyright does not give the holder absolute and final authority over every possible use."
The line "it's only a defense" is sometimes used as a rallying cry by the larger rightsholding organisations, like the MPAA and the RIAA, to try to delegitimise fair use in the minds of the public or politicians. But it doesn't reflect either statute or case-law, and should be resisted. Jheald (talk) 15:47, 28 January 2016 (UTC)
To add: here's the 9th circuit rejecting the "only a defense" argument: [5]; see also "Myth: Fair Use is only a defense, not a right" on page 16 of this from Berkeley. Jheald (talk) 15:54, 28 January 2016 (UTC)
I don't read either of those contradicting the "defense" aspect. The court ruling says that copyright owners should be aware that there are clear fair use defenses that a user can readily claim as to avoid sending frivolous takedown notice (eg don't waste the court's time with obvious cases), and the second frames it from the standpoint of education use which automatically meets one of the four fair use considerations, but still remains a defense. Fair use is always been an exception to the standard copyright law in the US, and you do not have the automatic right to use others' work. I do agree that in practice it is considered a right for some users (educational ones) but if it came to a court case, its on the fair use user to prove they met the defense of fair use, whereas if it was a right, it would be on the copyright owner to prove harm from that right. --MASEM (t) 16:17, 28 January 2016 (UTC)
Fair use does arise procedurally as an affirmative defense which has burden of proof consequences but it is grounded in Constitutional rights expressed in the Copyright Clause and perhaps to some extent in the First Amendment. So, you are both right. I like being able to say that in a discussion. 24.151.10.165 (talk) 18:00, 28 January 2016 (UTC) The Ninth Circuit in Lenz v. Universal Music (2015) (decided on statutory grounds) is on point:

Universal’s sole textual argument is that fair use is not “authorized by the law” because it is an affirmative defense that excuses otherwise infringing conduct. Universal’s interpretation is incorrect as it conflates two different concepts: an affirmative defense that is labeled as such due to the procedural posture of the case, and an affirmative defense that excuses impermissible conduct. Supreme Court precedent squarely supports the conclusion that fair use does not fall into the latter camp: “[A]nyone who ... makes a fair use of the work is not an infringer of the copyright with respect to such use." Sony Corp. v. Universal City Studios, Inc., 464 U.S. 417, 433 (1984).

24.151.10.165 (talk) 19:30, 28 January 2016 (UTC)
  • Actually, Jimbo's statement is consistent with Justice Stevens's language, writing for the majority, in Sony Corp. v. Universal City, 464 U.S. 417, 432-33 (1984):

    [Copyright] protection has never accorded the copyright owner complete control over all possible uses of his work.[13] Rather, the Copyright Act grants the copyright holder "exclusive" rights to use and to authorize the use of his work in five qualified ways, including reproduction of the copyrighted work in copies. § 106.[14] All reproductions of the work, however, are not within the exclusive domain of the copyright owner; some are in the public domain. Any individual may reproduce a copyrighted work for a "fair use"; the copyright owner does not possess the exclusive right to such a use.

    24.151.10.165 (talk) 17:52, 29 January 2016 (UTC)
  • One problem with non-free content is that non-free content makes it prohibited for reusers to use a lot of Wikipedia material outside the United States, although the idea is that reusing Wikipedia material is supposed to be permitted. --Stefan2 (talk) 15:06, 28 January 2016 (UTC)
  • That's a very passé idea at this point. While there are many efforts to republish Wikipedia material, most of them have not a care in the world as to the copyright of the material they are reusing. Several of them strip references to Wikipedia intentionally, which in some cases violates the rights of copyright holders of cc-by material here. Many of the efforts to reuse our material have been rather less than successful, and none are anywhere near as popular as Wikipedia itself...which undermines the idea that reusing it is somehow distributing it any better than Wikipedia can itself. So, the whole notion of Wikipedia as reusable is, in practice, a failed concept. Given that, there's really little motivation to not use non-free content to the "maximal" extent allowable under fair use law in the United States. --Hammersoft (talk) 15:21, 28 January 2016 (UTC)
  • Jimbo's point about using a less good free image, where one exists, rather than a better non-free one still speaks to the desire to have freely distributable content. He doesn't appear to be advocating having fair use swallow the idea of replacing with free images where possible. 24.151.10.165 (talk) 18:00, 28 January 2016 (UTC)
  • I'm going to play the role of Captain Obvious here and say that shifting from "minimizing fair use" to "maximizing fail use" is a seismic shift in approach and one that seems, as Hammersoft points out above, to make the NFCC seem a bit pointless now. Such a change would pretty much make all of WP:NFC#Unacceptable use irrelevant, wouldn't it? It would also seem to make WP:NFCC#3 and WP:NFCC#8 irrelevant as well. Quite a number of files which have be removed or deleted for NFCC reasons, probably would have survived under the goal of maximization. Mroeover, if the common practice in recent years has been to use non-free as much as possible and the WMF now seems to be in favor of such an approach, then maybe the time has come to put the NFCC out to pasture in favor of something more workable and something which more accurately reflects how the community uses non-free content. After all, NFCR was merged into FFD, primarily because the community felt it's time had come and was no longer needed. Maybe the same can now be said of the NFCC? -- Marchjuly (talk) 01:00, 29 January 2016 (UTC)
  • I think Jimbo's statement makes it pretty clear that in the absence of readily obtainable free content we should be using non-free content maximally. I.e., NFCC#1 remains important, but other aspects except those having to do with fair use law are void. Perhaps it's time to place NFCC for merging with Wikipedia:Image use policy. Policy is supposed to reflect good practice. Common practice on the project for years now has been to use NFCC maximally. Jimbo's comments make it clear that practice is not only common, it is now 'good' practice. Policy needs to be amended to reflect that, and allow maximal use of non-free content when free content can not be "easy enough" to obtain. This will allow album covers in discographies (we stopped patrolling for that at least half a year ago), per character images on list and thus vacate WP:NFLISTS (we stopped patrolling for WP:NFLISTS violations a long time ago; more than 50 list articles are on the overuse list now, unpatrolled), codify the enormous overload of currency images, allow per-episode screen shots of TV episodes, allow non-free images that have articles about them to be used liberally in other articles (e.g. Raising the Flag on Iwo Jima image), non-free images in galleries (WP:NFG/WP:NFTABLE goes away), etc. --Hammersoft (talk) 03:04, 29 January 2016 (UTC)
  • Except 1) Jimbo is not the WMF, which set the resolution NFC is fashioned after, 2) Jimbo doesn't set en.wiki policy, either. Further, as I note, his statement really is perhaps an ideal for a free thought arena but doesn't follow with current law, which per the WMF we're required to follow. Unless Jimbo is actively proposing NFC change here, this is just comments and have no impact on the language of NFC here. --MASEM (t) 05:15, 29 January 2016 (UTC)
  • On consistency with US copyright law, see above re: Sony. Agree with respect to policy and language of NFC. 24.151.10.165 (talk) 18:09, 29 January 2016 (UTC)
  • (edit conflict)I would say out of the 10 NFCC, nos. 1, 4 (maybe), 5, 7, 9, and 10 probably could still be applied to some degree, but anything involving context or minimal use would be almost impossible to enforce. Same goes for WP:NFC#UUI nos. 2, 5, 6, 8, 9, 14, 17 since these also seem to be related to NFCC#3 and NFCC#8 more than any of the other NFCC criteria. -- Marchjuly (talk) 05:29, 29 January 2016 (UTC)
  • Agreed. Masem's objections aside, I think we can get wide consensus to move as we suggest. There aren't enough voices left to oppose such a move now. Proof of that is above. Perhaps a stepped process is what we need here. First, RfC to eliminate NFCC#8, then #3. Once those are removed, we then modify NFC#UUI as you suggest, as the underpinnings for those points will be gone by consensus. Then, remove WP:NFG and WP:NFLISTS. With that done, the floodgates will finally be open to allow maximal use. I don't think we can do all of this in one fell swoop. As you note, it's a seismic change. People resist change, even if it's obviously good and right. The vast majority will want this, but may resist changing it even so. So, we have to do it bit by bit to get people acquainted with the idea. --Hammersoft (talk) 14:18, 29 January 2016 (UTC)
  • We can't change NFC as long as the WMF resolution still remains in place, but we also have to further recognize that the extent that NFC applies is still subjective, and most people are satisfied with how NFC works. While there may be more than 500,000 non-free images on WP, that still represents only around 10% of the total article count, and that percentage hasn't significantly changed over the last several years. In other words, people are still respecting a limited use of what would otherwise be fair use images. It could be lower, absolutely, but we can't force it not enforce that with a heavy hand. That doesn't mean NFC is broken and needs to be abolished, just that its subjective application is not as strong as some would like. But it's also not as liberal as others would like as well. --MASEM (t) 14:59, 29 January 2016 (UTC)
  • Yes, we can change NFCC/NFC as we feel the need to do so. If the WMF has an objection to it, they can raise the objection. But, it is highly unlikely they will do so because (a) Jimbo's statement and (b) they've refused to engage us on these points before. I disagree with your statement that most people are satisified with how NFC works. The proof of that being the lack of patrolling now for obvious violations of policy and guideline. Understand, policy is reflective of what happens on the project, not the other way around. What is happening on the project is the usage of non-free content very much in line with what Jimbo states, which obviates the need for NFCC#3 and #8. Your objections to this move ring very true with how things were...five years ago. They are no longer valid. What will you do if we hold an RfC and the consensus is to remove NFCC#3? --Hammersoft (talk) 15:18, 29 January 2016 (UTC)
  • "Obvious violations" is subjective - for example, you brought up the list of denominations for a country's bills and coins which frequently use nonfree images in tables. NFLISTS is not an absolute, and while I believe there are better ways for using NFC, I'm also fully aware that its also a reasonable approach that satisfies the core nature of NFC that generally has consensus. Does this mean NFLISTS is invalid and we should allow images in discographies? Absolutely not, as the money lists are the exception, not the rule. What I do think we have to be aware is that editors are hostile to the idea of non-free images being "patrolled". While clear violations are found they should be dealt with, and questionable images taken to FFD, no question and that doesn't seem to be the problem, it's when we take a police-like attitude towards these (especially since we are talking subjective criteria) that editors become hostile with the idea. And if NFCC#3 was put on the table for removal, we can't since that is hardcoded into the WMF resolution. Even if NFCC#3 disappeared, we would still apply the WMF resolution to keep minimal use of non-free. --MASEM (t) 15:35, 29 January 2016 (UTC)
  • And one last point to stress: Jimbo has no direct say in policy matters or the like. His word is not the word of God. If he wanted to have NFC changed to reflect his opinion, he would have to go through the same channels to do an RFC or the like for that, or would have to have the WMF board to reissue a resolution towards non-free media, but he alone could not do that. Further, he is speaking to an "upper limit" of usage, but does not speak to any lower limit, which we are free to decide for ourselves as long as it does not exceed what the WMF requires. --MASEM (t) 15:46, 29 January 2016 (UTC)
  • Masem, with all respect I think you're missing the point and are still referencing a stance that was valid five years ago, but not now. The time for all the fence sitting that has happened in the past in an attempt to craft a middle road of acceptance has passed. Nuanced answers aren't acceptable anymore. You don't have to quote passages to me from the bible of NFCC. You know full well I was an avid proponent in years past. I know the schpiel. It just simply isn't valid anymore. --Hammersoft (talk) 16:20, 29 January 2016 (UTC)
  • I strongly disagree that the situation is as dire as you make it. Yes, there are fewer people that directly spend time going through NFC to find where some may be violating, and there is probably a need for a bot or two to address the mechanical aspects of NFCC (eg #10). But when you look to all editors overall, there is a healthy respect for NFC, with the loudest complaints coming from newer editors that aren't familiar with the policy in practice (which is true for any policy). I believe it could be stronger (read: less non-free) but there's a balance that editors overall still keep. The projects that deal with works that are frequent uses of non-free (primarily all contemporary publishing like music, books, TV shows, etc.) all have guidelines in place to recommend reducing non-free usage, and the two article quality projects, FA and GA, also respect NFC requirements when evaluating articles. It's just that there is harsh resistance when there is proactive enforcement of NFCC rather than reactive. While proactive enforcement assures fewer non-free, it also sows a lot of ill-will. I saw it before with the whole notability inclusionist/deletionist war, and once people started treating notability in a more reactive manner than proactive, conflicts pretty much died out. For NFCC, this might mean there will exist a good number of images that clearly fail NFC that won't be discovered for some time before they are removed, but that's going to happen anyway, patrolled images or not. I think we are on the middle road, with enough leeway to discuss both additional tighten of non-free usage in some cases, and looser allowances in others, but still all staying on this acceptable and conflict-free middle road that is otherwise not raising any red flags with the WMF. This is a good place to be in terms of NFC as a policy, even if some of us feel we could be better. --MASEM (t) 16:31, 29 January 2016 (UTC)
  • Conflict free???? It's about as conflict free as the Pacific Ocean is dry. --Hammersoft (talk) 16:47, 29 January 2016 (UTC)
  • As per MASEM, let's not fall into WP:Argumentum ad Jimbonem. Jimbo's minimal/maximal language is very loose and open to a number of constructions. A considerably more limited interpretation might suggest loosening NFCC#1 to: "No free equivalent. Non-free content is used only where no free equivalent is available, or could be created, that would serve the same encyclopedic purpose." This change would allow us to make minimal use of non-free images where no free alternative currently exists but would still favor free images created later, serving the Foundation's stated purpose of supporting freely distributable content. A more modest change like this would also allow us to see how it works in practice before just chucking most of the policy. I would consider still disallowing "almost all [non-free] portraits of living notable individuals" to track the Foundations specific language in the authorizing resolution. 24.151.10.165 (talk) 15:57, 29 January 2016 (UTC)
  • In considering the NFCC#1, I think we need to have some type of time frame on "could be created". A good example is a notable building that is in the process of being constructed; once it is built a free image clearly could be created but at the present it can't be, though architect draws do exist. If the building is not to be completed for, say, three or more years, there's fair allowance to use the artist rendering. But if the building is only a few months out from completion, that would be a fair argument to avoid using the non-free and just wait for the building to be completed. It's just a matter of figuring out that reasonable time frame, which I think is going to be on the order of 6 to 12 months. But that's a separate discussion. --MASEM (t) 16:38, 29 January 2016 (UTC)
  • Agree that this is a separate discussion. The major point was that Jimbo's general statement is not a detailed policy proposal. 24.151.10.165 (talk) 18:12, 29 January 2016 (UTC)
  • We're long past the days of "Jimbo speaks and that's the policy." Jimbo is, of course, welcome to provide his input, but he's just one more member of the community. In this case, there's also the WMF resolution to consider; we can't make "maximal" use of nonfree material, because it explicitly specifies the use must be minimal. We'd have to first convince the WMF to change that global policy, and I would not be in favor of attempting to do so. Right now, what we do largely works. We use a few nonfree images where it's crucial and makes sense, but don't splatter them all over list entries, tangentially related articles, articles where free replacements could easily be come by, and so on. As to "nobody cares anymore anyway", just look at the recent flap over Gather, which there's an overwhelmingly supported call to deactivate since it's splattering nonfree images all over its user-created lists. Yes, people do care, and clearly a lot of them as evidenced by participation in the discussion about it. A lot of them may not watch this page, but they do support the principle. That's also certainly evident by the number of inappropriate nonfree images flagged for deletion, and the number of admins who help in deleting them. "Free as in speech" absolutely does matter here, and absolutely should continue to. Seraphimblade Talk to me 11:15, 30 January 2016 (UTC)
    I haven't been able to get anybody interested in the fact that the mobile apps use non-free images in search results. See phab:T124225. —Kusma (t·c) 14:27, 30 January 2016 (UTC)
    I didn't even know about that. (I don't use the mobile interface, I've had "desktop" set on mobile for some time). But just tried it, and you're quite correct. That certainly needs to be fixed; if they can't exclude nonfree images the results need to be text-only. Seraphimblade Talk to me 15:08, 30 January 2016 (UTC)
    Agree. It's hard to protest about something if the feature is unknown. The mobile version has no edit button, so I'm always forced to use the desktop version instead. --Stefan2 (talk) 15:38, 30 January 2016 (UTC)
    In response to the Gather feature, I suggested there should be a magic keyword we can set on a per page basis that lets us specify an image, so that for pages that only use non-free we can use a free icon or similar replacement (eg for video games using the VG project free icon), which should also apply to the mobile version approach too. --MASEM (t) 15:41, 30 January 2016 (UTC)
  • To cite our cliche of a slogan, we are the "free encyclopedia that anyone can edit". Encouraging the creation and dissemination of free content is fundamentally what we are about. Yes, I agree with Hammersoft that in actuality, we're just 'sort of free' (just as we have, with various user right restrictions in place, always been the encyclopedia that 'anyone can edit, but not really anyone '). This proves that we are, and have been, in the middle of the road, and have not passed some critical marker after which it's impossible to maintain that we are both free and inclusive yet practical.
In particular, we should avoid the mistake of thinking that "opening the floodgates" makes it easier for NFC issues and patrolling. Quite the contrary. If a larger share of our images become non-free, we have more and not less images to check against NFC. Remember, we are going to have a NFC policy in one form or another. Even if (at minimum), our policy was to simply comply with what the law says about fair-use, we'd have a surge in cases to patrol and check. And checking against the law is not a walk in the park in comparison to our policy. WMF has lawyers to take care of such things. That's how nuanced it is to "simply" comply with the law instead of our "puritan" NFC policy.
The way I interpret Jimbo's "When no free image is possible, then we 'should make maximal rather than minimal use of Fair Use'" is that every article about a book should have a picture of its cover, every album and film too, every company article should have a logo if one exists. It's an ideal like this that our NFC already provides us tools for, but people are hesitant to go ahead and add those legitimate non-free images, because quite frankly, our policy is so daunting. – Finnusertop (talk ⋅ contribs) 17:44, 30 January 2016 (UTC)

Confusing interwiki links

I find the interwiki links from the pages WP:NFC and WP:NFCC a bit confusing.

WP:NFCC is the exemption doctrine policy of English Wikipedia. Some of the interwiki links go to the EDP's of other projects. For example, there's an interwiki link to zh:維基百科:非自由內容使用準則, the EDP of Chinese Wikipedia. There is also an interwiki link to ja:Wikipedia:フリーでないコンテントの使用基準, a failed proposal to create an EDP for Japanese Wikipedia. However, some projects have pages which are specifically about English Wikipedia's EDP, for example ja:Wikipedia:英語版におけるフリーでないコンテントの使用基準.

WP:NFC is a guideline, and some projects do not have separate pages for both a guideline and a policy. For example, our guideline WP:NFC links to fr:Wikipédia:Exceptions au droit d'auteur, the French EDP. Also, our guideline is mapped to a Japanese page about English Wikipedia's NFC guideline, ja:Wikipedia:英語版におけるフリーでないコンテント, which thanks to Wikidata now is incorrectly mapped to the French EDP.

Shouldn't the pages WP:NFC and WP:NFCC link to pages which discuss non-free content on English Wikipedia instead of linking to pages which discuss non-free content on the local project? For example, it would seem more natural if WP:NFCC links to ja:Wikipedia:Usage criteria for non-free content at English Wikipedia instead of ja:Wikipedia:Usage criteria for non-free content. However, we wouldn't want ja:Wikipedia:Usage criteria for non-free content at English Wikipedia (a Japanese page describing English Wikipedia's non-free content criteria) to be mapped to it:Wikipedia:Non-free files (an Italian page describing Italian Wikipedia's non-free content criteria).

How can we clean up this mess? --Stefan2 (talk) 18:37, 15 February 2016 (UTC)

Why not fair use photos of living people for leads?

This discussion is currently ongoing at WP:BLPN#Why not fair use photos of living people for leads? I know NFCC#1 has been discussed many times before on this talk page, but I'm just adding the link in case anyone more familiar with the rationale and origin of NFCC#1 would like to comment. -- Marchjuly (talk) 08:10, 16 February 2016 (UTC)

Looks like User:Iridescent has already dealt with it pretty well. Worth keeping an eye on, but unless the topic gets any more comment, probably best to let it just fall into the archive (IMO). Jheald (talk) 08:33, 16 February 2016 (UTC)

Non-free images on temporarily transcluded pages

An interesting problem has come up at Redirects for discussion; since it's likely this will come up again, I thought I'd bring it here. Please see this discussion for background. A non-free file is in use on an article (a logo representing the article's subject), and that article has been selected as a "selected article" for the front page of a related portal. As such, the image is being displayed on the portal page as well, but no non-free content criteria is available for the portal. Presumably the portal displayed a different article yesterday and will display a different one tomorrow, so the non-free image is only visible on the portal today. Is this use violating the non-free content criteria? Ivanvector 🍁 (talk) 15:18, 18 February 2016 (UTC)

@Ivanvector: Yes, I think this is a violation. An easy fix is to add <noinclude>...</noinclude> tags to the article so that the image is not displayed. Most portals, though, only display an edited copy of part of each article, and do not try to display the whole thing. -- John of Reading (talk) 15:23, 18 February 2016 (UTC)
  • Per WP:NFCC#9, you may not use non-free files in the portal namespace. It would be nice if the infobox template somehow could detect non-free content automatically and thus automatically hide the image whenever someone copies an article to a sandbox or transcludes it on a portal page. --Stefan2 (talk) 15:27, 18 February 2016 (UTC)
Ah I see, well that's pretty clear then. Is it possible we could make non-free images automatically wrap with <noinclude> tags? Or is that a WP:VPT question? Ivanvector 🍁 (talk) 15:30, 18 February 2016 (UTC)
It is easy for a template to display one image in the article namespace and another image if the same wikicode is placed in a different namespace by using {{main other}}. However, I don't think that it is possible for a template to detect if a file is unfree or not. We'd only want a namespace switch to hide images which are unfree. Someone could maybe answer the question at WP:VPT. --Stefan2 (talk) 15:42, 18 February 2016 (UTC)

It's Fair Use week!

An annual celebration of the important doctrines of fair use and fair dealing.

It is designed to highlight and promote the opportunities presented by fair use and fair dealing, celebrate successful stories, and explain these doctrines.

http://www.fairuseweek.org/

See the introduction, and then the daily round-ups on the site, for links to blogs and other info-pieces being posted this week, mostly from university libraries across the United States, as to what is and what is not fair use, and why it's important.

There's some good stuff there! Jheald (talk) 20:37, 24 February 2016 (UTC)

Nice find, Jheald. They even cite our logo policy as exemplar fair use! (on their brochure) – Finnusertop (talk ⋅ contribs) 21:08, 24 February 2016 (UTC)
Thx. The credit is due to Slowking4 however, who dropped a passing mention to it on the Wikipedia Weekly facebook group. Jheald (talk) 22:19, 24 February 2016 (UTC)

Reusing non-free images in the same article

Hey all, I have a question I'd like y'all input on. I'm currently doing a major reworking and cleanup of an old FA of mine, Myst. In addition to archiving old refs and changing the date formatting I've been pulling in a lot of new references and such. The current non-free situation is that there are four images—one illustrating the gameplay, and then three showing the game in its various incarnations as its graphics have been reworked. I was thinking that at least one of these images could definitely go as it has a weaker rationale (File:Myst-library and ship.jpg), but I think I've got enough current and new references to support the three comparison images once I'm done. With that said, I'd really want one of these images up in the gameplay to serve as an illustration of what is being talked about there, but I'd also like to keep the comparison images clustered together. To my knowledge I've never seen non-free images utilized multiple times in a single article—is there any precedent on this? Der Wohltemperierte Fuchs(talk) 16:10, 25 February 2016 (UTC)

The only immediate one that comes to mind is the symbol Prince used and on the Love Symbol Album, but only used twice, establishing the title in the infobox and then once in the lede; however, this is a vastly different situation. If you are going to use the image twice (and I'm not saying this is necessarily allowed), I think we'd need separate rationale for both uses on the same page (noting the section it is used in). --MASEM (t) 16:52, 25 February 2016 (UTC)
I think using the same non-free image twice in the same article is possible but you need a lot of commentary to go along with the images.Jo-Jo Eumerus (talk, contributions) 18:51, 25 February 2016 (UTC)
  • WP:NFCC#10c requires a FUR for each use, so if the file is used twice in the same article, then you need two FURs for that article. Also, if the same file is used twice in the same article, then one of the uses would almost always violate WP:NFCC#3a as you could easily refer to the other section where the image is used. --Stefan2 (talk) 15:08, 2 March 2016 (UTC)

When it can't be helped

A few hours before starting the RFC on whether tiny images violate NFCC, Alsee blanked the part of the guideline that more obviously related to his question.

So until yesterday, the guideline said: "Non-free content may also appear in search results, certain categories (such as Category:Publicity photographs with missing fair-use rationale), or special pages (such as Special:ListFiles). Fair use rationales are not required for such pages."

Alsee's edit summary says that the "second half is redundant to the exemptions above, and the first half is a significant expansion that should not be made with no visible discussion here". I conclude from this, that although it was posted in full compliance with the process stated in the relevant section of the policy on writing and updating policies, and although he has no objection to the substance, that he is making a bureaucratic objection.

Does anyone object to the actual substance? For example, is there anyone who believes that FURs should be written for every file that might appear in a special page? WhatamIdoing (talk) 23:01, 16 March 2016 (UTC)

NOGALLERY is routinely applied to categories that can contain non-free files, except in maintenance categories which are much easier to work on when in gallery form (c.f WP:NFEXMP). As for the search results and special pages I don't think it has been discussed extensively, although for Special:NewFiles the point about NFEXMP would likely apply.Jo-Jo Eumerus (talk, contributions) 23:09, 16 March 2016 (UTC)
Yes, the NOGALLERY approach is why User:Finnusertop specified "certain categories", rather than all of them.
Searching the File: namespace would be unreasonably difficult if you couldn't see any of the images that were returned in the results. WhatamIdoing (talk) 23:20, 16 March 2016 (UTC)
I was objecting to the substance. Decorative images on navigational links is one of the main things prohibited by the policy, deliberately minimizing use of non-free images. I was rather surprised to see such a direct reversal of policy made with no discussion. I don't think it's necessary or desirable to expand the use of non-free images like this. Alsee (talk) 18:58, 17 March 2016 (UTC) (Meaning, images in search results. Alsee (talk) 19:49, 17 March 2016 (UTC))
P.S. Special:NewFiles clearly falls under the intent of non-article administrative/management exemptions, I certainly support adding it to that list if it isn't explicitly covered. Alsee (talk) 19:15, 17 March 2016 (UTC)
The 'special' namespace is something we can't control. If you want to change something in that namespace, then it has to go through mw:. It would seem to me that Special:Search, Special:ListFiles and Special:NewFiles would be useless in the file namespace if they do not show images, so those pages would probably have been added to the category of exemptions if it were possible to categorise special pages. --Stefan2 (talk) 22:11, 17 March 2016 (UTC)
Alsee, just how – exactly how, as in, by what practical method – is an editor supposed to find an image to use, if you're going to take away all the images in Special:Search? Just click on dozens or hundreds of pages until you find the right one? Imagine doing this exact search to find a specific non-free Wikipedia logo (e.g., to put on a page in the Wikipedia: namespace to welcome new editors), and being given only a list of file names, instead of thumbnails that make it possible to figure out which one you want to use. Does that sound even remotely functional to you? WhatamIdoing (talk) 01:11, 18 March 2016 (UTC)

Related discussion

Alsee appears to be involved in a related discussion about unexpected sexual images appearing in search results at mw:Talk:Wikimedia Discovery#Images in search results. WhatamIdoing (talk) 16:41, 17 March 2016 (UTC)

Brief summary, I question whether search results should be including decorative images at all. I firmly support noncensored policy and I firmly reject singling out objectionable images for special treatment. The indiscriminate use of images in search results is going to create unneeded headaches when people searching for other topics unexpectedly get "porn" or images of Muhammad or whatever else, in a non-relevant non-educational context. A main point upholding noncensored against opponents of the policy is that anyone seeing the objectionable content was explicitly looking for that topic and they should dang-well expect to find relevant content there. Alsee (talk) 19:10, 17 March 2016 (UTC)
You are using your own editorial judgement in all of your language, plus slippery slope rhetoric to boot. You do not get to decide what search results are "non-relevant non-educational context." You simply don't. I do not either, neither does the community, neither does the Discovery team. That's for the reader to decide. That's the point of why we're not censored. Your explanation and language show a fundamental lack of understanding of one of our root policies. Keegan (talk) 23:39, 17 March 2016 (UTC)
Keegan there is almost no resemblance between what I wrote and what you read. I am a stanch opponent of anyone seeking to apply "editorial judgment" or "deciding" on any sort of selective image content in article search results. Alsee (talk) 05:27, 18 March 2016 (UTC)
I'm pretty sure everyone read what you wrote the way that I do. "I'm against censorship, just think of the children." Keegan (talk) 16:31, 18 March 2016 (UTC)
Images in Special:Search aren't "decorative". If you still truly believe that displaying images in search results is just about making things pretty, then I invite you to go to c:Commons:Village pump and ask whether they would like to have all those bandwidth-eating "decorative" image thumbnails removed from the search tool there. WhatamIdoing (talk) 01:15, 18 March 2016 (UTC)
WhatamIdoing your argument is that it would be dumb to remove images from image-search run on file pages? Ok. I agree. It would be dumb to remove images from image-search run on file pages. Alsee (talk) 05:27, 18 March 2016 (UTC)

CLARIFICATION: Ok, now I see why there's confusion. We're talking about two different things. This started with the Discovery's teams work, where the basic article search is slapping images on all possible results. It even starts spitting out images while you type, and the first two letters you type can throw up a closeup of naked genitals. Alsee (talk) 06:43, 18 March 2016 (UTC) 

Exemptions from non-free content policy are made for the use of non-free content on media file search results, special: pages for encyclopedia management, and certain administrative non-article space pages as necessary to creating or managing the encyclopedia, specifically for those that are used to manage questionable non-free content. Those pages that are exempt are listed in Category:Wikipedia non-free content criteria exemptions.[1]

References

  1. ^ Due to software limitations, TimedText pages for non-free video files will automatically include the video file, and as such, pages in the TimedText namespace are presumed to be exempted from NFCC#9.

How's that? Alsee (talk) 07:22, 18 March 2016 (UTC)

Most pictures of genitals are free content, as are most images of the prophet Mohammed and just about every other "offensive" topic that you bring up. This means that your argument has nothing to do with protecting free content, instead it's simply "THINK OF THE CHILDREN!!!!" This attitude of censorship has been roundly rejected, repeatedly, by the English Wikipedia and Wikimedia in general. AFAIK, there is a single wiki which censors images of Mohammed, and that is all. Your position is contrary to that of the community no matter how you spin it, leave any engineering team out of it. Keegan (talk) 16:24, 18 March 2016 (UTC)
Wikimedia Foundation guidelines are to follow the Principle of least astonishment in relation to potentially offensive imagery. It is not a non-free content issue. From Alsee's arguments at mw:Talk:Wikimedia_Discovery#Images_in_search_results, they believe that to exclude potentially offensive material by omitting MediaWiki:Bad image list images from mobile search would constitute censorship. And that to avoid censorship, we should remove all images from mobile search. Which reads to me as cutting off the nose to spite the face. Again. - hahnchen 17:00, 18 March 2016 (UTC)
Got it, thanks for breaking the two issues apart, as they should be. Keegan (talk) 18:13, 18 March 2016 (UTC)
Yep. I didn't realize how confused things became when it got posted here as a "related issue". Anywho, crazy idea, this discussion could focus on the proposed edit to the free-image policy. :) I believe the edit I proposed matches actual practice by both the Community AND the Dev team. My proposed edit contains basically two differences from WhatamIdoing's edit. (1) "Search results" became media file search results - note that non-free images are already excluded from Discovery Team's new article search images. (2) "Special pages" was changed to special pages for encyclopedia maintenance. This means something like Gather doesn't get an automatic pass on non-free images. The Reading team had already moved to exclude non-free images from Gather, when Gather wound up being removed. Alsee (talk) 18:36, 18 March 2016 (UTC)
  1. I don't see any value in excluding non-free images from any search results. People actually do use those images to figure out which article they want to read. The difference between Cars (painting) and Cars (film) would be much more obvious in mobile search if mobile search included a tiny thumbnail from the first.
  2. It sounds like Gather isn't installed here and doesn't display non-free images. In that case, banning it would be pure pointless WP:CREEP, with the additional downside of risking disputes over whether this or that Special: page is "really" for encyclopedia maintenance and the technical problem that Special: pages do not technically exist (which means that all the maintenance scripts would be incapable of coping with it). If we ever found a use that this community truly thought was inappropriate, no matter what namespace it was in, then we can and should deal with it then. But I see little point (and not just a little room for concern about WP:POINTiness) in pre-emptively banning something that might happen and, if it happened, might also be something we didn't want. WhatamIdoing (talk) 02:47, 19 March 2016 (UTC)
What? Banning?? This has become extremely confused and I'm baffled by the narrative you're writing for me. I'll consolidate in the discussion below. Alsee (talk) 07:23, 20 March 2016 (UTC)
An alternate way to look at this: we should not prevent non-free where there is automatic gathering of the non-free image alongside non-free without any discrimination (such as search results); no user is putting the image there directly and would be akin to the "necessary" function of the encyclopedia. But this use would be clearly separate from where there is human-directed navigational elements (eg images on a discography page or a category without a NOGALLERY tag, since this is purposely being created as such. I don't know best to word that at the moment, but that would distinguish uses here. --MASEM (t) 03:33, 19 March 2016 (UTC)
  • I think Alsee's wording is fine. It accurately captures the current situation, without either allowing or disallowing any hypothetical future use of non-free media. Let's just deal with that when the situation arises. —Ruud 10:03, 19 March 2016 (UTC)
    • The first change that Alsee proposes is meant to ban non-free images in the search results unless you are specifically searching for a file, which is neither a completely accurate description of the current situation or what editors above express as a desirable, principle-compliant situation. (It also won't solve the problem that prompted all of this, because the lead image for Pearl necklace (sexuality) is GFDL and CC-BY-2.5 – aka "free content".) WhatamIdoing (talk) 19:22, 19 March 2016 (UTC)
      • @WhatamIdoing: No, Alsee's phrasing would explicitly allow search to show images when searching the File namespace. It doesn't say anything about searching other namespaces. As the search engine doesn't display any images, other then when searching the File namespace, I'm not really seeing why it should? I really cannot read the comments by anyone on this talk page (other than Stefan2) as a willingness to make the text of this policy broad and unambiguous enough to cover currently existing uses of non-free media. I think I'm seeing a lot less disagreement than you are? (The selection of "inappropriate images", whether free or not, is a separate issue. I think it's best to discuss that issue on the MediaWiki talk page, so as to not further conflate these. Both are complicated issues, but for very different reasons.) —Ruud 01:12, 20 March 2016 (UTC)
        • Yes, Alsee is is narrowing this to explicitly allow users to see images when searching the File: namespace. It implicitly disallows showing images in any and all other searching, i.e., the images that are shown in Special:Search on the Mobile desktop site. To understand Alsee's purpose for this change, you need to go to https://en.m.wikipedia.org/ (click that exact link), type "pearl" into the search bar, wait two seconds, and see all the images that appear on the screen. Those images – the ones right there in the search results – are what Alsee is trying to ban with this "[only] when searching the File: namespace" language. WhatamIdoing (talk) 21:54, 20 March 2016 (UTC)
          • WhatamIdoing somehow you latched onto an incorrect assumption, and I'm not sure how I can make it any clearer than I already did. Please stop making wildly faulty statements about what you think my intent is. I am utterly baffled how you could possibly have made the bizarre statement you just did. You just suggested that some proposed edit here implicitly disallows showing images in any and all other searching. We all know that is not remotely true. This is page is policy on non-free images, and it only applies to non-free images. It has exactly zero connection to our typical free images, regardless of whether they are images of puppies or porn. I am baffled that you've apparently pegged me as such a supervillian that you so totally lost sight of the scope of non-free image policy. 05:35, 21 March 2016 (UTC)
      • WhatamIdoing, what's with banning banning?? Instead of your narrative for what you think I'm doing, how about I give my narrative of what I've been doing?
        1. I've been working with the Reading Team, for which they have thanked me almost a dozen times. I noticed the Project manager say non-free images were being excluded from Hovercards and that he thought it would require a policy change to include them. I said "I think I can help", and I explained why I thought non-free images in Hovercards were perfectly fine under existing policy. That's the opposite of "banning", and isn't even proposing a policy change. The project manager said they would love to see that idea explored, so I opened the RFC above. I presented the case that non-free images were fine in Hovercards, seeking clear community backing for my reasoning. So far one person didn't address the question, one person believes Hovercards are not acceptable under existing policy, and there is otherwise unanimous agreement that non-free images are fine in hovercards. I am eager to tell the project manager that this is confirmed to be a non-issue, and I'll likely get another thanks. They are considering changing the PageImage function so that features like as Hovercards can choose whether non-free images should be included or not.
        2. I had posted a completely unrelated comment to the Discovery team questioning whether it was a good idea to add images to generic article results. I think quite a few people would consider it a a surprising violation of the WMF's "principal of least astonishment" that naked genitals are thrown at them while they are tying the first two letters of Vulcan. Trying to type Anaesthetic throws up an image of anal sex, typing Zoo instantly shows a dog's penis in a woman's vagina. I don't think it is "banning" to want a little thought and discussion put into it before boldly adding bestiality-porn onto every search starting with Zoo.
        3. While working on #1, I stumbled across the rather sweeping exemption you added to the policy here. WhatamIdoing you probably consider me a pain in the butt from some previous encounters, but I do respect you as a valued long standing member of the community. (I wish more WMF staff had your experience.) I also understand it must sometimes be awkward trying to navigate the gray area between WhatamIdoing and WhatamIdoing_(WMF). But this doesn't look good. I don't want to make assumptions, but it looks like you learned of the WMF's plans before they became generally known, it looks like you made an overbroad sweeping rewrite of policy to facilitate those plans before they were generally known, and it looks like you disagree with the consensus behind the policy. I agree that pages like Special:NewFiles obviously fall under the existing encyclopedia-maintenance exemption, but I oppose a pre-emptive blanket policy exemption for anything that might ever get dumped into Special. If something needs to go into Special and it's not currently covered, then if appropriate we can add a new exemption for it. I personally have unresolved mixed thoughts about non-free images in general article-search results, but my overriding position is to back consensus. It looks like a core intent of policy is that non-free images should not be used to decorate navigational links. At this time the WMF isn't including non-free images in article-search results, so I think the simplest Policy language is to accurately reflect accepted current practice. If the WMF wants to add non-free images to article-search results then I welcome that discussion. I might even decide to support it, after more carefully weighing the competing concerns. Alsee (talk) 07:23, 20 March 2016 (UTC)
          1. I agree that Hovercards' use of images is fine. I also agree that your Hovercards discussion and your reversion are not the same discussion, although there are overlapping issues.
          2. I realize that you were surprised by the search results that you found. I realize that you are surprised that WP:NOTCENSORED means what it says, even to the extent of applying to search results (i.e., a context in which we can be reasonably certain that we are providing information that the reader doesn't want and isn't specifically asking for). As a way of addressing these concerns, you may be interested in reading the advice at Wikipedia:Manual of Style/Images#Offensive images and especially the WP:LEADIMAGE section (which I also helped write ;-). It might be good to update that page directly to mention the effects on mobile users; if you agree, then you should WP:PGBOLDly add that, so that more editors will become aware of the effects of their choices.
          3. In the end, it's impossible for a feature to require a change in a local guideline (not policy!). The official written description is in WP:CONEXCEPT for years (since before my time, although the current language is largely my creation), but fundamentally, this is a mere common-sense application of the facts of life: The English Wikipedia is not the whole wiki-world, and is it impossible for any software to "obey" every community's guidelines and policies on every (sometimes self-contradictory) point. The devs should create a good product (whatever that means), and the local community should decide whether they're willing to accept it.
            I don't follow what Reading or Discovery are doing (or even care about it, except that I hope that desktop search will continue to improve). I do, however, follow half a dozen discussions pages and noticeboards, and I frequently fix confusion about matters that involve different policies and guidelines. I have done this for many years. You can see me making such changes here, for example, where the only "discussion" is after the fact in edit summaries; here's another and another and another – and actually, glancing through my contributions, I don't see any evidence of me discussing in advance even a single change to a policy or guideline this calendar year. When you know what the practical consensus is (=what editors actually do), and when you know how that particular page gets used, and when you are making that page more accurately reflect reality (and if you're any good at writing policies and guidelines, which requires a significantly different mindset from writing an article), then discussions aren't usually necessary for moderate changes.
        • When I compare what happened to you (unwanted NSFW search results) and your proposed solution (keep getting all the NSFW search results, but ban useful results like album art cover), I cannot support this change. It's bad policy writing, it doesn't solve your real problem, and it makes things worse for people who are trying to figure out which Pearl Jam-related page they want to read. If you want to get rid of the NSFW search results, then get rid of NSFW search results. There are ways to do this that won't make search results less functional for other users. WhatamIdoing (talk) 21:54, 20 March 2016 (UTC)
          • WhatamIdoing please strike your comment. That "proposed solution" is all yours, please stop inserting your imagination into my mouth. I don't see how this discussion can proceed productively unless you consider the possibility that there has been a miscommunication here, and consider that maybe I'm not saying what you thought I was saying. Alsee (talk) 06:12, 21 March 2016 (UTC)
            • Well, let's talk about that. You write above – and, as I have said before, WP:Policy writing is hard, and maybe you didn't write what you meant – that you wish to change this guideline so that it prohibits the display of non-free images in search results, except when the user has specially chosen to search the File: namespace by ticking the box. This change (if the devs implemented it, of course; see WP:CONEXCEPT) would prohibit the inclusion of non-free images such as album art covers, famous modern paintings, and corporate logos in search results. It would have no effect at all most of the NSFW images that currently appear in those same search results. That is the actual meaning of your proposed sentence, "Exemptions from non-free content policy are made for the use of non-free content on media file search results". That sentence means "I ban the inclusion of album artwork (which is non-free), and I accept the inclusion of NSFW images (almost all of which are free)." Is that what you wish to accomplish? If it's not, then maybe we should talk about what you want to accomplish, and see if we can find some wording that will do this. WhatamIdoing (talk) 17:00, 21 March 2016 (UTC)
Thank you. Let's see if we can mend some fences here. I apologize if my consecutive mention of this policy elsewhere created confusion. I believe you and I agree that this policy has zero interaction with NSFW. You were upset when you thought I was making a crazy argument, then I was upset that you were putting a crazy argument in my mouth even after my efforts to deny it. I think it would be helpful if we agree that NSFW is irrelevant here, and it disappears from this discussion.
Regarding search results. My intent was to acknowledge media files are obviously covered, without addressing the hypothetical & less clear case of article search results. (i.e. neither an explicit ban nor an explicit exemption on article search results.) Could you / would you be willing to suggest language like that? Either that, or get consensus for a blanket search exemption? I'll even commit to staying out of it.
Conexcept: Firstly, we all need to try to work together as partners. The WMF mandated that each community establish an Exemption Doctrine Policy, with an explicit directive that EDPs minimize non-free content use. It also states It may not be circumvented, eroded, or ignored by Wikimedia Foundation officers or staff nor local policies of any Wikimedia project. For the WMF to disregard those EDPs would be, in the most generous possible terms, "morally dubious". If you are trying to claim Conexcept as some sort of "trump card" claiming dominance in a conflict-model.... I hesitate to feed into the conflict-model.... but the Community is in fact holding the "trump card". A simple consensus on a very reasonable RFC has the power to remove all EnWiki non-free images from search results. All it would take is consensus to adopt the policy from any of the various wikis that don't include non-free images. Alsee (talk) 00:57, 22 March 2016 (UTC)
Looking at the comments above, do you really think that the case of article search results is "hypothetical & less clear"? It's pretty obvious to me that most community members are satisfied with seeing tiny thumbnail images of album artwork in search results, as a means of identifying which article is about the album versus the band versus something unrelated (e.g., see what User:Hahnchen has written here and elsewhere).
One of the hidden problems in writing policies and guidelines is that when you say "X is okay", many editors interpret this as meaning "not-X is not-okay". Your intentions don't matter, and neither do talk-page comments that say things like "We will figure out Y and Z some other time" – "X is okay" is routinely misunderstood as "not-X is not-okay". Consequently, my recommendation is to settle it now, if at all possible. WhatamIdoing (talk) 15:29, 22 March 2016 (UTC)
Looking above I don't see many people in this section, but I'll tell ya what. I'll drop it if Ruud thinks there's zero chance that a narrow-use consensus might prevail regarding non-free images in article search. To clarify: The Discovery team *has* built article search that returns images. You can see it at https://www.wikipedia.org/ Just type letters in the box and you start getting images. It currently does not return non-free images, but I believe there is strong interest in changing that. I know of no plans to implement it on en.wikipedia.org search box, but if it is deemed an "upgrade" I don't see why it wouldn't eventually come here too. What do you think? Is it worthwhile, or a waste of time, to get broader input on non-free images in article search results? Alsee (talk) 00:16, 23 March 2016 (UTC)
This page does not apply to www.wikipedia.org, which is not part of the English Wikipedia. The community here gets to have a say in what happens here, in the search boxes on the desktop and mobile sites for the English Wikipedia only, not in the search boxes for any other site.
My recommendation is that you restore as much of the original text as you feel is reasonable for this stage. After all, nobody disagrees with the facts as originally stated, e.g., that non-free files actually do appear in certain categories and that we actually don't require FURs for those uses. (As a point of good policy-writing, I discourage you from trying to declare that use to be "necessary", as that language risks a needless debate over whether it's truly necessary or merely convenient. I deliberately chose the purely factual "Non-free content may also appear" language to avoid that pitfall.) WhatamIdoing (talk) 18:08, 28 March 2016 (UTC)
How about edit?[6] Things like Category:Publicity photographs with missing fair-use rationale are already explicitly exempted in Category:Wikipedia non-free content criteria exemptions. Things like Special:ListFiles are utterly non-controversial because of the management-purpose, and something like Special:Gather got strong objections because of the content-purpose. Alsee (talk) 14:50, 29 March 2016 (UTC)
The grammar's a bit off. Let me think for a while about how to fix it (anyone else who sees an easy solution should feel free to jump in). WhatamIdoing (talk) 04:02, 31 March 2016 (UTC)

Exceptions for living individuals

From time to time in discussions about the use of non-free images of living individuals, it is noted that we permit the use of non-free images of living people in certain circumstances. These are limited to individuals who are either (a) a known recluse (as substantiated by reliable sources), (b) on the run from the law or (c) incarcerated for life. I've noted these exceptions myself. Yet, we've never codified them either in policy or guideline. I can't even find the sustaining consensus decisions that support those exceptions. We just note them, as if they exist...yet, they don't.

I've taken a look at possible articles where these exceptions would exist.

  1. Recluse; We do not have an image for Bill Watterson, Darby Conley, Naoko Takeuchi, and Tsugumi Ohba. In fact, I can't find a noted recluse BLP where we use a non-free image.
  2. On the run from the law; I examined every article in Category:Fugitives wanted by the United States (66 total), and found just 4 articles with non-free images on them (1, 2, 3 and 4 (though he is probably dead). The vast, vast majority do not have non-free images. Quite a few (~five times as many) actually had free images. The exception is, at best, infrequently used and seems undermined by the presence of free images for a number of these fugitives.
  3. Incarcerated for life; I took at look at Category:Prisoners sentenced to life imprisonment by Pennsylvania and found five of 17 (though not all 17 are alive) using non-free images (and 1 using a free image), and Category:Prisoners sentenced to life imprisonment by New York, where I found 9 of 28 (again not all 28 are alive). This usage of non-free content seems to be more common.

Thoughts? --Hammersoft (talk) 16:42, 28 March 2016 (UTC)

Overall, I think we should have some statement of how "if a free alternative can be obtained" NFCC#1 to be clear in broad cases, which includes BLP photos too, we should be more explicit: for example ,while we all recognize that images (should) have their copyright expire in time, we aren't going to ask editors to wait out 75 years to include an otherwise-acceptable copyright image, but we can ask editors to wait 3-6 months for some building to be constructed or the like instead of using design plans. There's a number of other such cases.
But to BLP: I think we should definitely codify these. All cases should start that if a free image is known to exist in the first place, the non-free in inappropriate (taking care of the "on the run" cases Hammersoft lists). But as long as one can document reclusiveness or the like, non-free should be acceptable: the lack of it on those articles may simply be due to the case that people don't recognize this as an allowed exception. --MASEM (t) 16:52, 28 March 2016 (UTC)

"low resolution" guidance on non-free images with standing granted permission

BlueRasberry and I have been discussing the O'Reilly Perl Camel trademark image, as I have a new, highly detailed SVG that I'd be willing to contribute, if appropriate. The discussion can be found here. In a nutshell, the question boils down to: Does the low resolution guidance for non-free images apply to this image, given the extensive standing permission O'Reilly already granted in its Perl Camel FAQ. If I understand correctly that the purpose of this guidance is to reduce image fidelity, to make the image less commercially valuable from a piracy perspective. Given that O'Reilly widely distributes this image, and grants extensive reuse of this image for non-commercial purposes, is this purpose (and the guidance based upon it) valid in this context?

Thanks --Linux_dr (talk) 22:31, 5 April 2016 (UTC)

Unfortunately no. Non-commercial licenses are still treated as non-free on Wikipedia (as they limit what reusers of our content can do with them, which may include commercial applications). And as SVGs are infinite fidelity, we would not allow its use. The only exception we'd make for non-free SVGs is when it is the logo of an entity as provided by that entity directly. As we appear to be talking about Perl the language and not any entity, in addition to the fact that the image is still belonging to O'Reilly (co-opted under the non-commercial license by the Perl community to represent them), that would not enable us to use that image here as an SVG. A low resolution PNG would be acceptable, however. --MASEM (t) 00:06, 6 April 2016 (UTC)
Masem, thanks for the quick response. While I feel compelled to clarify a few minor points, it sounds like a definitive decision has been made. To clarify, this is Wikipedias's, and not O'Reilly's policies preventing this image from being used; Correct?
This does almost feel like a common enough case where an automated mechanism to store high quality images tagged with a non-commercial reuse licenses such, they would be automatically down-converted in a "commercial reuse" mode. (Just a suggestion to think about.)
As you said the SVG is from (me) an independent member of the Perl community at large, and not from O'Reilly. The SVG was reconstructed from a 300dpi scan, it is at best approximately original fidelity, rather than "infinite fidelity" (though I will choose to take that as a compliment, none-the-less). Thank you for listening to my ramblings. Can you think of a non-commercial image repository where an image like this would be welcome.
Thanks again --Linux_dr (talk) 5:05, 6 April 2016 (UTC)
m:NonFreeWiki/interested_people is a possibility. How about a gist? –Be..anyone (talk) 06:12, 8 April 2016 (UTC)

Hovercards

The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion. A summary of the conclusions reached follows.
The result of this discussion is yes, it is acceptable to include non-free images in Hovercards. A lot of participants have pointed to similar non-free image functionality in the longstanding WP:NAVPOPS tool, which has never been objected to before. The consensus here is that the use of a non-free image in a Hovercard or a Navpop is considered the same use as the image in the article itself, since Hovercards (and Navpops) are automatically generated representations of those articles. As such, the same fair use rationale for using a non-free image in an article would extend to the quick summary in a Hovercard/Navpop. Therefore, there has been no breach of the non-free content guideline as written, and no change to the guideline is necessary. Supporters also note the important role that non-free images play in visual identification of subjects. The consensus is that including non-free images in Hovercards would be beneficial to readers and improve the encyclopedia by providing users with that visual identification. I will note, however, that nothing in this consensus expands the scope of non-free content beyond Hovercards or other similar popups. Respectfully, Mz7 (talk) 23:56, 14 April 2016 (UTC)
Screenshot of Hovercards in use

Hovercards is a feature currently in betatesting. You can test Hovercards by enabling it at Special:Preferences#mw-prefsection-betafeatures. When a user hovers over an article link a popup displays the lead text, and when available, the lead image of that article. The WMF has already developed the functionality required to exclude non-free images.

There is currently a proposal at Village Pump: Enable Hovercards on whether we want the feature enabled by default. If you have an opinion on that question, please comment there. The discussion here should focus strictly on interpretation of non-free image policy.

Independent of whether a decision is made to enable Hovercards at this time, is it acceptable to include non-free images in Hovercards? The image would only appear when a user specifically requested a preview of the article containing the image. The image is displayed in a pop up mini-view of the article itself, along with some of the text from the article. It is clearly not being used for a "navigational" purpose. The image is arguably being viewed "in" a stubified version of the article itself, and it is arguably still serving the educational purpose of informing the reader about the topic and the related text that they are viewing in the mini-preview. See the image at right, the Hovercard (bottom center of image) is showing a preview of the intermediate value theorem. The image in this case is a free image, but it well illustrates how the presence of image in the preview is important to understanding the text.

Would Hovercard article-previews qualify as an acceptable place for non-free images to appear? Alsee (talk) 16:51, 16 March 2016 (UTC)

  • Support I believe this is within the spirit and intent of the policy, and that it is logical and useful for the article's lead image to appear in a mini-view of the article itself. Alsee (talk) 16:51, 16 March 2016 (UTC)
  • Comment @Alsee: While I would support Hovercards displaying non-free images if they come from the lead section or infobox, I think this point is moot now. As far as I know Hovercards also uses PageImages, which will no longer be displaying any non-free images soon. —Ruud 17:01, 16 March 2016 (UTC)
    Per phab:T124225. --Izno (talk) 17:02, 16 March 2016 (UTC)
    Ruud/Izno I spoke with the project manager.[7] He was very interested in exploring this possibility. PageImages could return a flag indicating whether the image was non-free, then the images could be selectively included or excluded where appropriate. Alsee (talk) 17:09, 16 March 2016 (UTC)
    Yeah, it's probably good to have this discussion in any case, but it should probably also be wider than just about this one extension. I'm in general not opposed to using a non-free image as the "hero image" of a particular page, but it should be done within reason. Picking the non-free image from the lead or infobox is probably going to be more acceptable than picking it from somewhere lower down on the page. Displaying it in search results or on hovercards is probably going to be less of a probem than using them in the Related Pages feature (the first two are only visible if the user "asks" for them, in the latter case we are slapping non-free images on pages that otherwise would have none.) Etc. —Ruud 17:19, 16 March 2016 (UTC)
  • Comment While this is in the spirit and should be okay for NFC, this is related to the issue of the Gather extension in that it is pulling the first image (free or not from a page). A lot of this could be solved if we had a magic word or similar ability to specify which image should be used alongside the hovercard, Gather entry, or whatever other potential extension could come into play, including the possibility of no image. --MASEM (t) 18:50, 16 March 2016 (UTC)
    MASEM see above. If non-free is acceptable here they may include non-free here and exclude non-free in anything like Gather. And yes, they're working on a keyword for image selection. The current plan is using the keyword with no image will force no image (if none of the images are appropriate), and a non-free image with no keyword will default to no image. (To avoid blindly defaulting to a random-bad next image in the article.) And of course keyword with image will use that image. Alsee (talk) 19:04, 16 March 2016 (UTC)
    Then that's fine. Here, as it is only a temporary thing on the screen, it's difficult to call it as a fixed medium that we arguably would have any problem with fair use aspects so from the nature of what NFC should serve, its not an issue (we've had the Popup gadget for so long that that would have raised concerns before), but it's the Gather side that poses a larger problem since that can end up in a fixed medium. But having the image keyword option would solve all concerns. --MASEM (t) 19:10, 16 March 2016 (UTC)
  • Of course it's acceptable to display tiny thumbnails in mini-articles. WP:NAVPOPS has done exactly that for years, and AFAICT nobody has ever complained about it. WhatamIdoing (talk) 22:54, 16 March 2016 (UTC)
    This might not make a difference, but the images displayed by Hovercards are substantially larger than those on Navigation popups. Having non-free images displayed so prominently outside the target article makes me uneasy. --Yair rand (talk) 23:13, 16 March 2016 (UTC)
    They also show different images. Hovering over The Hershey Company gives me a thumbnail-wide (actual thumb-size) non-free logo in NAVPOPS, but Hovercards gives me a 2.5-inch-wide free image. WhatamIdoing (talk) 23:17, 16 March 2016 (UTC)
    • The difference in image-size between Hovercards and Navigation popups seems like nitpicking to me. A bigger difference is that Navigation popups are a power-user script, that is enabled by individual users and where we as a community have less control over. Hovercards, on the other hand, may well soon be visible to all readers. So that is a good opportunity to reflect on where we stand on the "free" vs. "encyclopedia" issues in this "free encyclopedia" thing. That said, I think displaying a non-free image from the lead or infobox of the article together with an extract of the lead in a hovercard is completely in the spirit of this policy. —Ruud 01:25, 17 March 2016 (UTC)
  • Allow non-free images in Hovercards and search results - The English Wikipedia serves up 3 billion mobile views per month, I'm not aware of any complaint until phab:T124225. Mobile search and hovercards had been serving readers with non-free content for over a year, and now instead of updating policy to match reality, we've convinced the WMF to spend time and resources degrading user experience with T124255. What a pitch, "Donate to Wikipedia, we'll spend your money making it worse!". Lead images are identifying images. They are the best image, providing clarity at a glance as to what the subject is, our non-free rationales defend the use of these non-free images because they're the only reasonable identifying visual. A reader sees the image and knows what the article subject is without having to open the link or read through the lead. If you search for Parallel Lines on Google, you'll see a non-free image thumbnail of Parallel Lines, because its the best identifying artwork. That thumbnail comes from Wikipedia. Hovercards now show no image, mobile search shows no image, the WMF have made an engineering change which has degraded our search capabaility against off-site search, when instead we could have just edited Wikipedia:Non-free content. Let's edit it now. - hahnchen 23:03, 16 March 2016 (UTC)
    • That's a good point, hahnchen. Also, if the image is "identifying", then it's partly "navigational". One of the best things about images in NAVPOPS is that the image can help you decide, with half a glance, whether that's the page you want to read next. WhatamIdoing (talk) 23:14, 16 March 2016 (UTC)
  • This. I'm actually annoyed that NFC paranoia is removing a great feature. We already go above and beyond US law regarding fair use, here's another cool thing gone. Why are y'all doing these negative things because you don't like something? Keegan (talk) 23:47, 16 March 2016 (UTC)
  • While I note I'm not opposed to this (see my comments above), talking about images as navigational aids is a slippery slope argument. In fair use on an encyclopedic work, you readily can defend it in a court of law so it's not a legal issue, but it is an issue related to the free content mission. I can see this argument "navigational non-frees should be allowed" meaning that "Oh, this discography would be enhanced by adding the cover art to help the reader navigating", which would likely clear fair use defense but utter fail free content mission as well as going against established consensus on the use of such images in lists. What I think is important is to have the ability to set an alternative image for pages that are exclusively using non-free images. For example, an album page may be indicated with a generic free record icon so that it still can be differentiated from other topics of the same name. --MASEM (t) 01:14, 17 March 2016 (UTC)
  • @Masem: do forgive me and note that my comment was out of annoyance, as prefaced :) I think Tigress223 puts it well. Keegan (talk) 01:35, 17 March 2016 (UTC)
  • @Keegan and Masem: Furthermore, we have had some difficult time in Community–WMF relations lately concerning the two aspects at the heart of this issue: our freedom to chose what tools developed by WMF we adopt (cf. Media Viewer, Gather, Related pages), and in our responsibility in making sure the WMF follows its own stated mission (cf. Knowledge Engine). Having conservative local policies – particularly in the area of non-free content – has proven to be a successful strategy in helping the Community maintain its independence from the Foundation. Gather was disabled and Related pages modified by the WMF because the Community cited their noncompliance with the local non-free policy of the English Wikipedia. From the point of view of the Community, it makes a great deal of sense to maintain conservative and strict policies, for perhaps what can be seen as the "wrong" reasons. – Finnusertop (talk ⋅ contribs) 01:53, 17 March 2016 (UTC)
  • Thank you for providing your perspective, I appreciate it. Keegan (talk) 02:33, 17 March 2016 (UTC)
  • If we disabling or degrading useful features for the sake of expressing "independence" from the foundation, we are cutting our noses to spite the face. - hahnchen 16:10, 17 March 2016 (UTC)
  • Oppose Just leave NAVPOPS alone, (Redacted)--Stemoc 23:33, 16 March 2016 (UTC)
  • @Stemoc: just for the record, if I didn't happen to work for the WMF, you'd be blocked right now for using the word rape in relation to WMF tech staff. If you're looking for the line, it's far behind you. This website still has policies on civility and no personal attacks, and I would appreciate administration no matter whom the person is targeting. Keegan (talk) 23:40, 16 March 2016 (UTC)
    • And the fact that we don't have admins that do so, is the reason i stopped being an admin. —TheDJ (talk • contribs) 12:06, 17 March 2016 (UTC)
  • Sure. Next time I will follow the written rules instead of the unwritten ones. Keegan (talk) 19:12, 17 March 2016 (UTC)
  • Support - I agree with what Hahnchen said about the importance of lead images. I mean if I looked at a hovercard for a movie, and saw the picture of a few rivers and streams (where the film was filmed perhaps) instead of the poster, I would slightly doubt myself whether I am at the right page or not. And even if I realize that it is the page I'm looking for, I'd rather associate a movie with its poster instead of a picture of the Niagara Falls.Tigress223t@lk 00:45, 17 March 2016 (UTC)
  • Comment in my opinion, if we don't show it in popups and search results, then we should seriously consider if it still belongs in articles to begin with. I'm not in favor of throwing out all out non-free content, but if it is causing discussions like this, wasting time like this, then I'd rather just toss the whole lot out. We should seriously ask ourselves how many people still read articles where we are writing them (and thus showing NfC images). This is changing, and as such limiting ourselves to this very narrow definition of an article where we are allowed to use NfC material is more and more stupid. (As are resolution limitations btw, recently I thought an image was broken, but it just wasn't high resolution enough to show properly on a retina screen). —TheDJ (talk • contribs) 12:19, 17 March 2016 (UTC)
  • Comment 2 this won't affect navpopups btw. As in, no one is gonna change how navpopups behaves. There are 2 modes available all images on, or all images off, but there is no one who is gonna rewrite that thing to take into account NFCC. —TheDJ (talk • contribs) 12:19, 17 March 2016 (UTC)
  • Oppose Incompatible with WP:NFCC#10c, which requires a fair use rationale for each article in which the file is used. If the file appears in all articles which link to the file's article, then lots of new FURs need to be written. Additionally, it's a violation of WP:NFC#UUI §6: we are supposed to link to articles instead of including non-free images in other articles. --Stefan2 (talk) 22:05, 17 March 2016 (UTC)
  • I don't understand your reasons. Hovercards, and navpopups, are not new articles and thus NFCC#10c does not apply. Same goes for your second reason: this is not including non-free images in other articles, it is showing non-free images from the articles which they already exist, to illustrate the article. Keegan (talk) 22:31, 17 March 2016 (UTC)
    • Even if it is incompatible with current NFCC, the scope of this discussion is whether we should update NFCC to allow for non-free images in Hovercards and mobile search. - hahnchen 22:36, 17 March 2016 (UTC)
  • The point of this discussion seems convoluted. Is this RfC saying that we have to modify NFCC to allow non-free images into Hovercards, even though they've existed in Navpops for over a decade and have nothing to do with NFCC since neither Hovercards nor Navpops are new uses of NFC but are summation of existing content? For what? So we can go to the WMF engineers and say, "Okay, now you have our permission?" By removing functionality as a negotiating tactic, the project is actively being hurt. Do you hear that? You are hurting other editors. That is the definition of WP:POINT. Please explain yourselves, and how you have the right to do this on behalf of the community. Keegan (talk) 23:26, 17 March 2016 (UTC)
    • Maybe this discussion is too convoluted and we should start again, if so we should be clear that we are seeking a definite exemption (WP:NFEXMP) for PageImages and the features that use PageImages (Hovercards, mobile search). - hahnchen 00:08, 18 March 2016 (UTC)
      • Yes. The question at the start assumes that this violates any policy to begin with that needs exception; even worse, it runs with this assumption as truth. It's not neutral, it's not how RfCs are supposed to be formatted, and frankly it's a waste of time coatrack for power/control/ownership of the English Wikipedia against WMF engineering. Make it go away, and we shouldn't tolerate this Byzantine coatracking any further. Keegan (talk) 02:08, 18 March 2016 (UTC)
        • Keegan you just made a batch of bad assumptions on why I opened this discussion. I opened this discussion because the PROJECT MANAGER had the impression that Hovercards violated policy. I told the project manager why I thought it didn't violate policy, and the project manager said they'd love to see that idea explored. So I offered to help. This discussion was opened because I wanted to clearly establish whether my reasoning was generally accepted. The only "policy change" I see would be maybe adding Hovercards to the list of examples of accepted uses. Based on my comments, the project manager is considering a software change to ADD non-free images to Hovercards. The PageImage function could flag whether an image was free, then other features like Hovercards can easily decide whether to include or exclude non-free images as appropriate. Alsee (talk) 21:38, 18 March 2016 (UTC)
          • @Alsee: yup, I did, because assume good faith is not a suicide pact. To quote Jimbo, "We need to take due process seriously, but we also need to remember: this is not a democracy, this is not an experiment in anarchy." For the past eighteen months, in the self-proclaimed interest of protecting the project, you have process and policy wonked hundreds and hundreds of wasted staff and volunteer hours. For every little victory you claim in your Quixotic crusade, you are causing immeasurable harm with your misunderstandings of said policy and process to staff and community alike. Eventually the community is going to tire of you opening a new RfC every week to pursue some meaningless thing, and it will let you know so very vocally. Keegan (talk) 21:47, 20 March 2016 (UTC)
            • Keegan I'm glad we cleared up my intentions on this RFC. I believe the project manager sees my efforts as and positive and helpful. I hope you'll consider the possibility that this isn't some one-off anomaly. Regarding my "weekly RFCs" that you think may get me sanctioned, could you helpfully clarify with some recent examples? Thanx. Alsee (talk) 07:44, 21 March 2016 (UTC)
  • Comment Seems to me that Hovercard/Popups panes are just another way to display the article that actually contains the NFC. As such, I don't think that runs afoul of NFCC or FURs, since the content remains part of the article for which it is justified, and is just being presented in a different way. Otherwise, it would be like needing a separate FUR for the web-based article and the mobile article. I think I generally err on the skeptical side of NFC but I don't see this implementation being a problem. CrowCaw 23:35, 17 March 2016 (UTC)
  • To Keegan's point just above me, and why I'm waffling my bolded comment between Comment and Support: As I basically said in my post above, Hovercard/Popups are just alternate methods to display a page that has had its FURs all sorted. As such, I don't think any change to NFCC is required or warranted, as we're not changing where the NFC is located but just providing another method of viewing that location. So setting my comment back to Comment, as I think the premise of the RFC is making assumptions that are not necessarily valid. CrowCaw 23:50, 17 March 2016 (UTC)
  • Comment: I don't object to the general idea as such, if and when the non-free image in question serves to "visually identify" the content of the article, in the way much of our non-free items do (e.g. cover art at the top of an infobox etc.) I don't really hold with the NFC-fundamentalist argument that each dynamic popup would constitute a new use that would require rationales for that new context. The way I see it, the NFCC are worded for use on wiki pages, and use in dynamic popup elements is somehow outside their scope. Or, under a different angle, I would still see the popup more as a temporary ad-hoc instantiation of its source article (for which the image has its justification) than as an element of the host page over which it pops up (where it doesn't). However, what I do have a problem with is the popup use of non-free items that don't have this "visual identification" function. Say, a non-free image appears somewhere down in an article, where its non-free justification is based on its specific function of illustrating some text next to which it stands, in conjunction with a specific explanatory caption. Nevertheless, it might still be the "first" image of the article, and thus would be selected for the popup. The popup would then make it appear in a context which is not covered by its NFC rationale, and where indeed it might not make any sense to the reader at all. This needs to be avoided. Fut.Perf. ☼ 20:10, 19 March 2016 (UTC)
  • Acceptable: It's simply re-presentation of the article (or part of it). Opposition to this borders wikilawyering. The intent of the copyright policy is not to prevent reuse of WP content (especially internally), it is to prevent violation of actual copyright laws, which this does not.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  06:25, 21 March 2016 (UTC)
  • Support – hovercards show an actual partial preview of an article, and the number of non-free images shown at a single time would be limited. sst✈ 07:19, 21 March 2016 (UTC)
  • I have been a long-time advocate for fewer usages of non-free images, and I support this proposal. The long-standing policy has been that non-free images are limited to the article namespace, and hovercards and popups are still showing content from the article namespace, just in a different format. —howcheng {chat} 17:40, 23 March 2016 (UTC)
  • Support the retention of image in popups, no matter the status of the image. The popup is the top part of the article, and if someone is hovering over the link there is an expectation of the article as someone has a specific interest, therefore we are showing it to them. We are not building a separate summary page of unrelated articles, or auto-popping an article. It is direct relevance, and it is mini, and it is still fair use. — billinghurst sDrewth 06:14, 23 March 2016 (UTC)


Main page and rationale of "some of the article is displayed"

Why is the rationale of "some of the article is displayed, and so we should allow non-free images" not also applied to the main page? --Izno (talk) 11:18, 21 March 2016 (UTC)

Or portals, for that matter. If Izno is arguing that the fact that some of the article is displayed is not relevant, I agree. If there is a distinction, it's that images on the main page and in portals are chosen by editors as content, but hovercards are requested by the user so that "content" is created on the go and at the user end only. – Finnusertop (talk ⋅ contribs) 11:26, 21 March 2016 (UTC)
Just employing a little Socratic method to make sure I-and-anyone-else understands the distinction. Also, yes, the main page is a special case of a portal. --Izno (talk) 12:37, 21 March 2016 (UTC)
It's why, below, I think the distinction is between a result that is generated by a search engine and thus lacks human interaction to put the image there, and the actual point of featuring an article on a portal/main page which is a choice made by a human, including how the article's trimmed down to show that section. --MASEM (t) 14:23, 21 March 2016 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

{{Non-free historic image}}, WP:NFC#UUI #5 and WP:NFCC #8

Since, as WP:NFC#UUI #5 says, using "[a]n image whose subject happens to be a war, to illustrate an article on the war" is unacceptable, there must be something else that's making non-free, non-historic images that depict historical events suitable for inclusion in Wikipedia articles. What is that something else?

Let me turn the question around with a clarifying example: this non-free picture of the February 2009 Kabul raids depicts a historically significant event, but it is merely "[a]n image whose subject happens to be a war, to illustrate an article on the war." There appear to me to be large numbers of analogous images tagged {{Non-free historic image}} that are also images whose subject happens to be X and which are merely being used to illustrate an article on X (and where, additionally, the uploader appears not to have considered the difference between "historic" (the language of the template) and "historical", which are not simply synonyms. I also note that the failure to observe a distinction between these two words mars a good deal of the discussion of related issues in the archives). I don't see that this image is "unique" (also the language of the template) or "historic", so I don't believe the image even merits the tag let alone passes all the requirements of WP:NFCC. The image doesn't convey any information not readily available in the text (it's just a bunch of soldiers on a roof) or "significantly increase readers' understanding of the article topic". So what I really mean is I don't see any "something else" in this and a lot of other cases.

If this were a unique case the solution might be simply to dispute the FUR and nominate the image for deletion, but there are lots. And lots and lots. Candidly, I think that in a lot of cases the uploader's only rationale for using this tag is a) the image is from the past, and b) he or she would like to include it in an article. What can be done to better combat the abuse of the tag and the simple bypassing of WP:NFC#UUI #5 and WP:NFCC #8 in cases like this? -- Rrburke (talk) 16:30, 26 February 2016 (UTC)

I think this is important to recognize the two prong tests of NFCC#8: that inclusion helps the reader (which nearly always is the case for inclusion of any image), while omission may harm the reader's understanding. It's generally why we really would like sourced commentary on historical images, though this is not the requirement, and particularly with war pictures, it might not be the picture itself but the events within the picture of interest. For an hypothetical example, if there were a photograph depicting the Allied storming of Normandy beach from WWII even if the picture itself is not of discussion, it's reasonable to recognize the importance of that military action and illustrating it with a non-free clearly meets NFCC#8. In the specific image above, it's armed men with guns on a rooftop. It might be an old photo, but I completely agree it is not historic and fails NFCC because it's depicting a random shot from the event but nothing that really helps the reader's understanding, and certainly doesn't harm it if it is removed with the article in its current state. The template should not be considered a free pass from NFCC#8 (and possibly NFCC#1) requirements. --MASEM (t) 16:42, 26 February 2016 (UTC)
Hi and thanks for your reply. Obviously I agree with you about this specific image.
But doesn't NFCC#8 actually set a higher bar than that, in requiring that the inclusion of the file "significantly increase readers' understanding of the article topic"? First, I'd point out that "significantly" is a substantial burden to meet in and of itself. Second, the inclusion of the NFC must "increase readers' understanding": so it a) can't merely complement the article text, because the policy specifies increasing the reader's understanding and b) it must increase their understanding: in other words, it can't merely improve the article in some other way that doesn't touch on their understanding, such as merely enhancing the reader's experience, for example. Finally, it must "significantly increase readers' understanding of the article topic" -- that is, the understanding of the main topic of the article, not of some ancillary passage or section. -- Rrburke (talk) 17:34, 26 February 2016 (UTC)
The thing I've found with the first part of NFCC#8 is that unless we're talking a totally unrelated picture, it is always possible to reasonably argue that inclusion significantly helps understanding, based on the concept "a picture speaks a thousand words"; it is very difficult to counter any arguments to this specific part of the test. Taking a common example, a random screenshot from a television episode in an article about that episode can nearly always irrefutably be described as helping the reader understand the article because it shows what the episode looks like. But that's where the balance of NFCC#8 as an episode just dealing with "talking head" scenes doesn't require the reader to see a shot of "talking heads" to understand the episode. I could see the issue in the image you use. An editor wanting to keep it could argue it shows the military presence in the event, which is a fair point for meeting the first part of NFCC#8. But unless there's further discussion about armed men on rooftops, it's not necessary to see that image to understand what is described in the text. --18:43, 26 February 2016 (UTC)
I have to agree with Rrburke. The question, per policy, is not "to understand what is described in the text", but to better understand the topic. And I have never been convinced by your great distinction between the two halves of NFCC #8 -- to my mind (and most others I think), they are simply two (valuable, complementary) perspectives for looking at the same thing: what do you understand with the image, that you did not understand without it; what would you not understand without the image, that you do understand with it.
The point of NFCC #8 is to get people to ask: what does this image actually contribute to reader understanding? And is that contribution significant in the context of the article as a whole? Ultimately those are questions for the community to consider, but as Rrburke indicates, the bar is not a low one. Jheald (talk) 16:37, 27 February 2016 (UTC)
I agree that the first clause implies the second. However, in the interests of parallelism, shouldn't the second clause read "and its omission would be significantly detrimental to that understanding"? The issue is moot, I suppose, because the first clause already specifies "significantly" and the second merely restates the requirement in different terms, so the stronger clause governs, but wouldn't it clarify the point if they both took the same adverb? -- Rrburke (talk) 15:43, 28 February 2016 (UTC)
But surely there must be some difference between just "increase" and "significantly increase", otherwise why include "significantly"? How, in practice, are uploaders expected to meet the burden of that distinction, and how is the distinction enforced? And to return to the example, to know that there were soldiers on roofs, which we probably could have guessed, doesn't aid the reader's understanding at all, let alone "significantly". -- Rrburke (talk) 15:57, 27 February 2016 (UTC)

@Rrburke: in this context, don't forget that when WP:NFCC#1 requires that "no free equivalent is available, or could be created", it's not only about and equivalent image, but also equivalent text. That is, if one can use free text to convey the same information as the non-free image, the image fails WP:NFCC#1. And this is the case with most of those not-so-historic images. damiens.rf 03:44, 22 April 2016 (UTC)

@Damiens.rf: Good point; I hadn't thought of that. It seems to me then that a large proportion of the files tagged {{Non-free historic image}} have no business carrying this tag. Category:Wikipedia non-free historic files is probably in need a large-scale cleanout, but that would undoubtedly raise hackles. There's also the practical issue: there are over 12k files in this category. Incidentally, how do you understand the difference between historic and historical? I've been through the archives of this page and seen them used interchangeably, which seems to have resulted in a large expansion of Category:Wikipedia non-free historic files well beyond what ought reasonably to be included. Having been through the archives, I know I'm beating a dead horse, but historic image to me means this and this, not that and that. -- Rrburke (talk) 10:50, 27 April 2016 (UTC)
The general point you make about some images not being historic is likely true, but the latter two examples of biography portraits of deceased individuals you cite are likely mis-tagged and are allowed if they meet the criteria of a different provision, WP:NFCI §10. See also {{Non-free biog-pic}}. Note that this license tag was only created in 2011, so there are many older biography portraits that used other tags such as the historic image one before that date. 24.151.10.165 (talk) 15:43, 27 April 2016 (UTC)
Wow! This template is terribly mistaken! The "biopic for deceased individual" is an internal thing in Wikipedia. It has nothing to do with fair use and licensing. We accept fair use images as illustrations of biographies when the individual is deceased, because a free equivalent can not be produced. But the death doesn't make us believe that the image "qualifies as fair use under United States copyright law". I'm going to nominate it for deletion. --damiens.rf 23:24, 2 May 2016 (UTC)
Nominated for deletion here: Wikipedia:Templates_for_discussion/Log/2016_May_2#Template:Non-free_biog-pic.damiens.rf 23:38, 2 May 2016 (UTC)

Two questions relating to music articles

(1) What is the criteria for displaying cover art in articles for singles or albums? The guide from WikiProject Albums states that cover art should be no more than 100,000 pixels in size, so effectively for a square single or album cover that's 300 x 300 pixels. But I don't know if this is just a guide or an actual fair use legal requirement (the cover art obviously being copyright), and I can't find anywhere on Wikipedia that states explicitly that 100,000 pixels is the maximum to comply with copyright law. I ask because some new editors have recently being uploading cover art for some records by Colombian artists, but they have uploaded it from a site where the cover art is 500 x 500 pixels, so 250,000 pixels in total, and I don't know if this means the cover art will have to be deleted or reduced in size.

(2) Another experienced editor and I are debating the legal status of the pictures uploaded by the same new editors to illustrate the biographical articles for two Colombian artists. One is the picture for Anasol (File:Anasol live.jpg), which has been taken from a fan website: obviously the fan doesn't own the photo, but given that the fan website probably obtained the photo illegally off the internet presumably means that the use of the photo on Wikipedia is equally invalid? The other is for Soraya (musician) (File:Soraya_musician.jpg) – here the issue is that the artist has been dead for ten years and that therefore the photo's use has been justified on the grounds that no free image can reasonably be obtained, but the picture used is still an official promotional photo from when the artist was alive, and doesn't that mean copyright is still in place? If you look at the non-free rationale used for each photo you'll see that the justification used is sketchy in the extreme.

We'd be grateful for any clarification on the above matters. If any or all of the images are deemed unusable on Wikipedia, I'd also like to know how to go about nominating them for deletion, as I've never had to do it before: obviously I can delete them from the articles in question, but that doesn't delete the files from Wikipedia's library. Thanks. Richard3120 (talk) 20:33, 17 May 2016 (UTC)

On 1, the reason we stick to about 300 px is that all default ways to show images, including User set thumbnails, never go larger than 300px. Considering all various aspects ratios, the 0.1 MP guideline is a good handle to respect the max default display for images and for most media types, assures low resolution. Low resolution is a means to assure we have a strong fair use defense.
On 2, a random image from the Web is going.to presumed to be copyright and non free, and if the performer is still alive, not allowed to be used, since a free photo can be taken. For a deceased artist, a copyright photo is presume to be acceptible for nonfree, as long as it is not a press corp/Gettys type image. Most promo photos made for artists are fine. --MASEM (t) 21:45, 17 May 2016 (UTC)
Thanks for replying. So does that mean only the Anasol photo is in breach of guidelines? Richard3120 (talk) 21:49, 17 May 2016 (UTC)
(edit conflict) Per WP:NFCC#3b, non-free files should not be unreasonably large. 100,000 pixels is often but not always a good guideline. You will have to consider if there is any need for a larger image or if it would work equally fine with the smaller image. If an image is too large, then just tag it with {{non-free reduce}}. A bot will then make the file smaller within a day. I would say that File:Anasol live.jpg is too large but that File:Soraya musician.jpg is acceptable size. Accordingly, I have added the template to the first file. File size issues are described in detail at WP:IMAGERES.
According to WP:NFCI §10, it is permitted to use non-free images like File:Anasol live.jpg and File:Soraya musician.jpg provided that the depicted person is dead and provided that the image doesn't fail WP:NFC#UUI §7. I can't find any evidence that Anasol is dead, and the article is written in the present tense, suggesting that she is alive. Accordingly, I have tagged the file for deletion. The other picture shows a person who seems to be dead, so it would seem to be acceptable under the guideline. Note that I have never seen any evidence that WP:NFCI §10 is compatible with copyright law (and I haven't seen any evidence to the contrary either).
According to WP:NFCC#4, the file must have been used somewhere with permission from the copyright holder before it was uploaded to Wikipedia. File:Anasol live.jpg appears to be sourced to a website where it is used without permission. However, I assume that it has been used with permission at some point in the past as the other website must have found the picture somewhere. Since I have already tagged the file for deletion for another reason, I don't see any need to check if the file also fails the policy for this reason. File:Soraya musician.jpg appears to be sourced to a website where it is used with permission, so WP:NFCC#4 is obviously not violated. --Stefan2 (talk) 21:58, 17 May 2016 (UTC)
NFCI §10 is very much only required under WMF for all Wikimedia projects, and has no bearing on copyright law outside of how an image is defined as free or non-free. We want to encourage free use over non-free, but when a person is dead, free images are very unlikely to be found, hence why §10 exists. --MASEM (t) 22:13, 17 May 2016 (UTC)
WP:NFCI §10 is about using material, not about avoiding to use material, so your statement above seems to be wrong. --Stefan2 (talk) 22:45, 17 May 2016 (UTC)
Its about a specific case where we know its near impossible to ask for NFCC#1 to be met (because with a deceased person, there's no way a new free image can be made, period), so it outlines where we specifically allow such non-free images, as to distinguish from, say, WP:NFC#UUI #1. Arguably NFCI#10 is not really needed, as it should follow, but its been added to be clear that such images of deceased persons are nearly always considered as allowed non-free so that the issue is fully clear. --MASEM (t) 23:08, 17 May 2016 (UTC)
WP:NFCC#1 is unrelated to this issue as it does not have anything to do with copyright law but is something which is imposed by the Wikimedia Foundation. --Stefan2 (talk) 23:13, 17 May 2016 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Non-free_content/Archive_65&oldid=1137428263"