Wikipedia talk:Non-free content/Archive 54

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Archive 50

←

Archive 52

→

Opinions on image use at The Epoch Times

I am soliciting opinions of other editors on a discussion about fair use images being used in The Epoch Times article. The discussion is at Talk:The Epoch Times#The Malaysia edition image. Note that I am an involved editor. You input would be appreciated. -- Whpq (talk) 13:12, 28 September 2011 (UTC)

WP:Arguments to avoid in image deletion discussions has been nominated for deletion

The nomination, along with a slew of other pages, can be found at Wikipedia:Miscellany for deletion/Wikipedia:List of policies and guidelines to cite in deletion debates.

I believe that this would be a pity. Jheald (talk) 11:30, 4 October 2011 (UTC)

Input

I want to know if ~~File:DS M&S Dream Canoe.jpg~~ File:DS Mario and Sonic Dream Canoe event.png fail nfcc criteria. Its inclusion to Mario & Sonic at the Olympic Games was requested sometime ago at FAC, and fulfilled. The image was later removed. « ₣M₣ » 06:17, 4 October 2011 (UTC)

As it stands at the moment, the image is not used in any article, so fails NFCC #7. Apparently this has been the case all the time since April [1], which I find quite surprising, since I thought we had a bot to seek out such unused items of NFC and mark them for deletion.

As for whether that edit in April, removing it from the article, is correct, that is also something we could look at. (The last edit where it did appear is this one.) Personally, I think it probably can add something valuable to an article like this to show the look & feel of the game as actually played -- for someone like me, that knows nothing at all about the game, I feel that to show at least something of this does add to my understanding of the topic. The use-rationale for the image identifies various aspects of the Dream Canoe part of the game which it says (with a source) are unique -- though I am not clear whether that is supposed to mean "unique within the context of the Olympics game" or "unique within the context of DS games". The latter proposition could certainly be quite strong grounds for inclusion, the former one rather less so (why go out of our way to show what is not characteristic of the bulk of the game?). On the other hand, as the edit rationale for the removal notes, the case for the image is significantly weakened when this point is not something considered worth mentioning in the article itself, nor even the caption for the image in the article. (For the same reason, the caption for the "trampolining" image currently used probably also needs to be improved to reflect the quite sensible information for its inclusion presented in its rationale, which currently the reader of the article is likely to be oblivious of).

Bottom line: if the gameplay shown is particularly characteristic to this game, and/or the onscreen look is particularly distinctive, then it may be possible to make quite a good case to include this image. But if the reader is left unlikely to consider that either is particularly noteworthy, then they are unlikely to pick up much from the image that they might consider a "significant addition to their understanding" about the game. Jheald (talk) 07:35, 4 October 2011 (UTC)

Re-uploaded as File:DS Mario and Sonic Dream Canoe event.png. I had the image recovered to try and improve the rationale. « ₣M₣ » 04:08, 5 October 2011 (UTC)

NFCC exception for product logos in a table?

Direct me to the right place to discuss this, if this isn't it. At Intel Core is a table of the many variations of the product, and the corresponding many subtle variations of the cramped little logos Intel used to differentiate them. They were all deleted, citing WP:NFTABLE. I'm of the opinion that the images are quite useful and encyclopedic, and should be allowed, because the images convey information that the text cannot. (I would not be here if the images did not convey unique identifying information). As both an amateur encyclopedist and a long-time system builder and user of PCs with nearly every processor vintage produced, IMHO this table loses a very great deal of its value without the images. So - how to proceed? Are free versions of these logos even possible? I tend to think not. --Lexein (talk) 04:42, 7 October 2011 (UTC)

WP is a free content encyclopedia with a mandate by the Foundation to minimize non-free use. These logos all have the general style and shape, so iterating all the variations, while within fair use, is a failure of minimizing non-free. You are likely right that the logos are too complex, passing the threshold of originality and thus all non-free. You likely can use one logo as a representative sample of the general logo/chip branding, but WP cannot support iterating through all the logos. --MASEM (t) 04:58, 7 October 2011 (UTC)

Gah. Ok, then citing the logo, in a ref, should be fine, so that the information is still available, but not on WP's servers. --Lexein (talk) 06:31, 7 October 2011 (UTC)

Article passing GA with serious non-free issues

The Byrds - GA review at Talk:The Byrds/GA1. 25 non-free usages including 18 music samples. I've dropped a note to the GA reviewer but would it be better to (a) cull the overuse straight away, or (b) just return it to WP:GAR? Black Kite (t) (c) 17:40, 2 October 2011 (UTC)

I was the reviewer and wasn't really aware of WP:NFCC#3a as it is not an explicit part of the WP:WIAGA criteria. I would suggest contacting the main editor, User:Kohoutek1138 and culling the overuse. You could send it to GAR, but as it is not an explicit failure of the GAN guidelines, it will possibly just languish there. Jezhotwells (talk) 20:51, 2 October 2011 (UTC)

What we're talking about is a use that is legal and acceptable in the real world, but not per a controversial point in a WP guideline. Why should GA criteria be modified to weigh in on one side of the controversy? North8000 (talk) 20:55, 2 October 2011 (UTC)

Minimal use is required by the Foundation. We're not asking GA reviewers to be as thorough as FA is handled (25 non-frees won't fly at FA for sure), but they should be aiming the editors to consider non-free reduction. --MASEM (t) 21:06, 2 October 2011 (UTC)

Policy, not guideline. The problem is that WP:WIAGA mentions images but not music samples. If a GA candidate had 25 non-free images it would fail, the only difference here is that 18 of them are not images. Black Kite (t) (c) 22:18, 2 October 2011 (UTC)

In this particular case, I think the issue is more where to put these music samples than if they belong on Wikipedia. I think given the context of the article, most of these music samples could likely be moved to the respective articles of the songs being described. A short sample of each song, on the article describing the song and its history, certainly does fit withing not just the letter but the spirit of what fair-use non-free content is all about. On the article about the music group performing the song seems to be less necessary, where links within the article and a nav box (on the article anyway) should be sufficient for somebody wanting more details about the song. In short, we don't need to delete this material but rather re-arrange where it is found on Wikipedia as I think acceptable compromises can be made. One or two sound samples demonstrating musical styles might be appropriate on such an article... especially in sections which describe those styles. --Robert Horning (talk) 22:08, 2 October 2011 (UTC)

OK, looking at sample usage

"Mr. Tambourine Man" and "Eight Miles High" - used in their own article but could be argued that they are signature tunes for the band
"Old John Robertson", "You Ain't Going Nowhere", "Hickory Wind" - used in album article
"Tulsa County" is used in the article for the guitar accessory it depicts being used
The other 12 files are not used anywhere else. This suggests they would have to have an excellent rationale to remain in the article. The previous four I believe can be safely removed. Thoughts?

Black Kite (t) (c) 22:33, 2 October 2011 (UTC)

I think that I agree with y'all on the particular case in question. But, on a general note, even what the foundation statement is controversial and considered by some to be detrimental to Wikipedia, and one should not rule out that it should be changed. But, more to the point, the policy oversteps what even the foundation said and is even more controversial/questionable, and such should be recognized. North8000 (talk) 22:25, 2 October 2011 (UTC)

Nevertheless, the Foundation statement and the policy are what we have to work with; it's not something we can ignore because we disagree with it. This, as you say, is a simple case; there are too many non-free usages, as can be seen if one looks at Featured Articles on musical bands. It's just a question of which ones should stay and which (the majority, I think) should go. Black Kite (t) (c) 22:33, 2 October 2011 (UTC)

I think it's worth pointing out at this point, in reference to some comments made by North and Jezhotwells, that the adherence to the NFCC is a part of the GA criteria. The criteria require that "images are tagged with their copyright status, and valid fair use rationales are provided for non-free content" (and, of course, "[o]ther media, such as video and sound clips, are also covered by this criterion."). It is the NFCC which determine whether or not a rationale is "valid", so, yes, articles which use content contrary to the NFCC most certainly are not eligible for GA status. Surely, this is obvious. J Milburn (talk) 23:11, 2 October 2011 (UTC)

It doesn't make much sense to me that a "use" of audio files would be treated the same as a use of images within an article. Including an image in an article means displaying a copy of it when the page loads. So from that, there is some reason not to repeat its use when there is a main article use that can be linked to--each use is effectively a copy. But an audio file doesn't autoplay in an article when you load it--the article does not contain a copy in any meaningful sense. So while we would have a problem with indiscriminate linking (say, my favorite sound clips on my user page), I don't see a need or benefit to excluding links from articles to which they are integral (like the band's article) just because they are used in another article (on the song itself). postdlf (talk) 23:56, 2 October 2011 (UTC)

Wikipedia's servers are hosting the non-free clip, when the goal is to be free content. Just because they aren't auto"shown" like images, they still cause conflict with the Foundation's mission. Thus their uses should also be minimized. --MASEM (t) 00:04, 3 October 2011 (UTC)

I don't see how that's a response to my comment. postdlf (talk) 03:48, 3 October 2011 (UTC)

Your statement seems to suggest the policy is based on the image being autodisplayed (automatically served) to the user while an audio file is not. That's not the basis of the policy; it's about the hosting of the images. --MASEM (t) 04:05, 3 October 2011 (UTC)

No, it's not, because I was talking about multiple articles using the same file. A sound file that is already linked in one article does not become two hosted files on the Wikimedia servers when a link to that file is added in a second article. postdlf (talk) 04:25, 3 October 2011 (UTC)

From whatever "legal" standpoint there is, yes, you're right - once on the server, multiple uses probably do not strain anything with fair use. But that said, we still strive towards minimal use on en.wiki, in both number of unique files, and the number of times each file is used. Now, that is only based on our policy here, not the Foundation and thus subject to consensus, but given that we require justification for each use, we do not reuse a file that has at least one appropriate use indiscriminately across WP. --MASEM (t) 05:12, 3 October 2011 (UTC)

Each usage of a file has to be judged on its own merits- each usage has to meet the NFCC. I'm unconvinced about arguments as to whether the file appears elsewhere- either way, that should not influence our decision to use it in a particular article. Some files may conceivably belong in several articles, other files may belong in none; their appearance/non-appearance in other articles is irrelevant, as far as I'm concerned. J Milburn (talk) 08:47, 3 October 2011 (UTC)

Responding to Masem, The general objective of getting free content is often at odds with article quality and also at odds with the concept of allowing people to earn a living from the things that they produce. And policies try to implement that tradeoff....IMHO this one has been tilted too far one way by being written by the insiders/"enforcers" on the topic who listen mostly just to each other. Sort of like having the police write the laws instead of the legislators. (Failure to allow a more limited permission (I.E. just for Wikipedia/Wikimedia, or just for non-commercial use) is the biggest problem.) Too bad this has come up an an article which is not a good example, i.e. where it looks like overuse. North8000 (talk) 15:09, 3 October 2011 (UTC)

NFC has nothing (directly) to do with respecting copyright and assuring the creators are credited for their work: it is that we want to have as much material that can be redistributed without any gotchas across the world and minimize the amount of material that cannot be freely redistributed. And note that we don't disallow images that are "for Wikipedia only" or "non-commercial use", it is just that that means their redistribution rights are just as limited as an unlicensed, fair use reproduction; ergo they fall under non-free content. What is the soft point is what the balance is between being helpful to the reader as an encyclopedia and minimizing non-free content. That's a point defined by consensus, and I would argue one that really hasn't changed much since .. oh about 2009 where the removing of images for discographies and equivalent was made. --MASEM (t) 15:18, 3 October 2011 (UTC)

"Redistribution" right is not the main issue; the details of the policy wording making it very difficult to use them in Wikipedia IS the issue. I don't knwo if that is new or old. North8000 (talk) 15:24, 3 October 2011 (UTC)

Actually, redistribution is at the very heart of the matter. It's part of our m:Mission. --Hammersoft (talk) 15:32, 3 October 2011 (UTC)

That's not what I meant. I meant that the biggest barrier isn't "their redistribution rights are just as limited as an unlicensed, fair use reproduction" it's policy wording details. Actually the biggest barrier is "quick test" wording which is not a policy statement but which gets interpreted and quoted as if it was. North8000 (talk) 16:55, 3 October 2011 (UTC)

Are you saying that there are people that enforce NFC that are disallowing/XFDing images that have "for WP use only" or similar non-commercial licenses? Our policy does not disallow these images, only that they are non-free content and subject to the same evaluation for use as any other non-free. --MASEM (t) 17:01, 3 October 2011 (UTC)

No, I'm saying that it is extremely difficult and painful to use non-free images in Wikipedia. Policy wording is tougher than what the foundation says, and the non-policy wording that gets quoted as policy is tougher than policy. North8000 (talk) 17:37, 3 October 2011 (UTC)

Non-free images are supposed to be exceptional (per Foundation), which is why there are a lot of bits and pieces that we need to show. But, unless you've got an egregious case, or that you're taking the article to FAC where image use will be put under a microscope, we are generally very lenient towards inclusion. What exact "non-policy" text is being used to dismiss non-free use? --MASEM (t) 17:43, 3 October 2011 (UTC)

It's: "Could the subject be adequately conveyed by properly sourced text without using the non-free content at all?" If the answer....is yes, the non-free content probably does not meet this criterion" This sets a standard of having to show that it is impossible to even just adequately convey it via text. Taken literally, no image use could meet this criteria. Sincerely, North8000 (talk) 18:33, 3 October 2011 (UTC)

You said the non-policy wording is even tougher, but what you quoted is from policy. ? --Hammersoft (talk) 18:40, 3 October 2011 (UTC)
- I think there probably are cases where sometimes a gloss of policy is stated that is actually tougher than policy itself requires (eg claiming that content can only be used if it is "essential" to understand the topic; or can be used only if it has been discussed by an identifiable external reliable source). But I don't think the example you give is one of them. For one thing, as Hammersoft notes, it is actually policy. For another, the word "adequately" should probably be understood in conjunction with NFCC #8 -- i.e. to mean: without any significant increase in understanding of the topic being lost that the reader could gain from the image. Jheald (talk) 00:57, 4 October 2011 (UTC)

Well, maybe there's the ambiguity. It is not written as an operative statement and at least by real world norms such would not be considered a statement of policy. North8000 (talk) 01:06, 4 October 2011 (UTC)

I have informed User:Kohoutek1138 of this discussion as no-one else has bothered to do so. Jezhotwells (talk) 01:25, 4 October 2011 (UTC)

(Thanks for the heads-up regarding this discussion, Jezhotwells!)

Hi! I'm the editor who was responsible for dragging The Byrds article up to GA standard and as a result, I was also responsible for adding the non-free media samples that we are discussing. Let me first of all say that I was not really aware of WP:NFCC#3a, I was just operating under the guidelines listed at Wikipedia:Manual of Style/Music samples - guidelines that I have followed very strictly, I might add. Let me first of all make it quite clear that I have not simply added audio samples to this article for "decoration", in most cases they have only been added when they genuinely illustrate an aspect of The Byrds' music that is next to impossible to adequately describe in words (to paraphrase the Non-free media information and use rationale for each sound sample).

However, if we accept that there is a problem with the audio samples being too numerous, I agree with Jezhotwells that reclassifying the article as one in need of Good Article reassessment is probably not the best course of action, given that having too many audio samples is not an express violation of GAN guidelines. I am much more apt to agree with Black Kite's suggestion that some of the audio samples should be culled from the article, and as Robert Horning suggests, re-located to their own respective articles.

Having said that, I strongly disagree with the proposed list of culls that Black Kite has put forth, because some of the samples are (in my view) indispensible in providing the reader with a clear understanding of certain aspects of the Byrds' sound. I'm thinking specifically about the "Mr. Tambourine Man", "Eight Miles High" and "Tulsa County" samples. The article would be immeasurably poorer without those because they illustrate distinctive aspects of the Byrds' sound and musical development...aspects that are discussed in detail in the article text.

To me, the fact that these audio samples are used elsewhere on Wikipedia seems slightly irrelevant to this discussion, given that they all have multiple non-free use rationales. Surely the aim of using audio samples in a given article should be to improve a reader's understanding of that article, not to showcase sound samples that aren’t found anywhere else on Wikipedia.

Therefore, if consensus is to cull some of the files, I would like to propose the following list of removals or relocations...

"I'll Feel a Whole Lot Better" - to be moved to the song's own article.
"So You Want To Be A Rock 'n' Roll Star" - to be moved to the song's own article.
"Everybody's Been Burned" - to be moved to the Younger Than Yesterday album article.
"Goin' Back" - to be moved to the song's own article.
"Hickory Wind" - to be removed altogether, since it is already being used in an album and song article.
"Ballad of Easy Rider" - to be moved to the song's own article.
"Chestnut Mare" - to be moved to the song's own article.
"Full Circle" - to be moved to the song's own article.

The above 8 audio samples can, I feel, stand to be lost without too drastically affecting the quality of the article. That would leave us with 11 non-free audio samples. If, however, further culling is required, I would suggest that the following 3 could also be removed...

"The Only Girl I Adore" - to be moved to the Preflyte album article or perhaps deleted altogether.
"You Ain't Goin' Nowhere" - to be moved to the song's own article.
"Tiffany Queen" - to be moved to the Farther Along album article.

However, I feel that the loss of the above 3 audio samples would diminish the quality of the article overall, but their loss would bring the total down to 8 non-free audio samples (which is the same number that are currently being used in The Beatles article, which is a FA rated article). Let's call these culls that I've proposed "Phase 1" and "Phase2"...perhaps I could get some feedback on whether both phases of the cull need to be implemented in order to satisfy the criteria at WP:NFCC#3a, or just the first stage.

Whatever is decided, I would very much like to be heavily involved in the process of making whatever changes are necessary in order to satisfy the people here. I have a detailed and comprehensive knowledge of the article subject and therefore, I think I can better identify exactly which are the most valuable audio samples to retain, in order that the article still provides the best possible learning/research experiece for the reader.--Kohoutek1138 (talk) 11:42, 4 October 2011 (UTC)

"I can better identify exactly which are the most valuable audio samples to retain" sounds suspiciously like original research—why can't these be identified via secondary sources? Moreover, why, when I buy a book about the Byrds or the Beatles, does it not come with a CD of music samples? Uniplex (talk) 12:01, 4 October 2011 (UTC)

What a strange and singularly unhelpful response to my post. My comments sound nothing like original research...they sounds like what they are, my opinion regarding my own knowledge of The Byrds, in relation to this particular discussion. My statement has nothing to do with the article per se, so therefore can't be deemed original research, surely? I'm just saying that my familiarity with the subject matter (as illustrated by the fact that I wrote vast swathes of The Byrds article) means that I have a very good idea of what the defining characteristics of the band's sound are and hopefully that will be an asset in the context of this discussion. In addition, these defining characteristics of which I speak are identified by secondary sources...all throughout the article.

As for your comment regarding why you don't get a CD of music samples with books about the Byrds or Beatles, I can only assume from this that you are in favour of removing all non-free media samples from all music related articles on Wikipedia. If that is the case, I would suggest that you start another discussion somewhere else because such wide ranging changes to music related articles are surely above and beyond the scope of the current discussion. Like I say, a strange and singularly unhelpful response. --Kohoutek1138 (talk) 12:44, 4 October 2011 (UTC)

Seems you forgot to WP:AGF (or maybe I forgot to WP:AAGF). Both my previous comments are intended to help you: to minimise the amount of non-free content per policy—The Beatles article is not the yard-stick—and to choose which if any should remain (which ideally should be evident from the sourced text of the article). Uniplex (talk) 13:02, 4 October 2011 (UTC)

Oh no, I assumed good faith...I just couldn't discern any helpful input. But I agree that choosing which audio samples should remain should be based on the sourced text of the article. That's really what I was getting at when I said "some of the samples are (in my view) indispensible in providing the reader with a clear understanding of certain aspects of the Byrds' sound" and that some of the audio files "illustrate distinctive aspects of the Byrds' sound and musical development...aspects that are discussed in detail in the article text."

The list of samples I'm proposing we retain are the ones that illustrate important aspects of The Byrds' music, as detailed in the article text and supported by reliable, 3rd party inline references...like "Mr. Tambourine Man", "Eight Miles High" and "Tulsa County", for example. It is precisely this criteria that I have used in order to draw up the list in my above post regarding which samples I feel should or could be dispensed with. I think that basically we're both on the same page here, we were just perhaps talking at cross-purposes. --Kohoutek1138 (talk) 13:35, 4 October 2011 (UTC)

My opinion, Kohoutek has done an excellent job reevaluating the audio files; I would see no problem with the cull (phase 1 and 2) to reduce 19 audio files down to 8 with the sourcing he's backing up with. I would caution that FAC may find more to cull, but at the level of a GA nom, the original issue has been dealt with, and while I'm sure there's a few of us on NFC that would like to see fewer, I don't think we have much we can push on at the present time; its still a high number of uses, but nothing unheard of and accepted elsewhere on similar types of articles (with the cautionary WP:OTHERSTUFFEXISTS warning, of course). --MASEM (t) 13:42, 4 October 2011 (UTC)

Uniplex is right. Which samples to be retained should be clear from secondary sources. If it's not, the sample shouldn't be there. --Hammersoft (talk) 13:39, 4 October 2011 (UTC)

I agree Hammersoft and as I explained in my last post, that is exactly the criteria I have used in drawing up my shortlist of proposed audio file removals. The core 8 files that would be left after phases 1 and 2 of the proposed cull, are all explicitly mentioned in the article text as notable representations of the band's evolving sound at various points in their development and thay are all supported by secondary sources. --Kohoutek1138 (talk) 14:00, 4 October 2011 (UTC)

I would go with the 11 removals from this article. But, if they are unsupportable in this article, there's not much case for them to be supportable in other articles. Moving them around to other articles doesn't reduce our burden of non-free content. It just masks it. Non-free content has to be strongly justified wherever it goes. --Hammersoft (talk) 14:14, 4 October 2011 (UTC)

I agree on the general advice that simply moving a piece of non-free to a different article doesn't make it immediately appropriate, but I would say that in this case, moving a song sample from a band's page to a song's page may make it more relavent and thus meet NFC better. I would not expect the band page to go into critical discussion of every song they released, instead aiming for the highlights and their top hits, but conversely, a non-top hit single may still be critically commented on and justify the use of the sample. This is not to be taken as 100% assurance here, there are likely a few cases among the 11 non-free samples that would still remain inappropriate for use on the song's article page. --MASEM (t) 14:22, 4 October 2011 (UTC)

Like Masem, I quite agree that simply moving an audio sample to another article won't automatically make it appropriate (the suggested relocations that I listed above were just that - suggestions.) Obviously, each sound sample would need to be considered on a case by case basis in relation to its destination artcle. With this as a given, do we have enough of a consensus here for me to go ahead and begin phase 1 and 2 of the culling and relocating (where appropriate) of audio samples from The Byrds article? --Kohoutek1138 (talk) 10:34, 6 October 2011 (UTC)

Fine with me. --Hammersoft (talk) 15:56, 6 October 2011 (UTC)

OK, well it looks as if there are no other objections, so I'm gonna begin the cull of audio samples now. --Kohoutek1138 (talk) 13:16, 7 October 2011 (UTC)

Help deleting image:File:RUTH WOMAN.JPEG

Could someone XfD File:RUTH WOMAN.JPEG for me. I don't know how. Also File:MyKe.a.g.Hawke.jpg seems a bit sketchy, but it might be OK. These are images of the hosts of a reality show. - Peregrine Fisher (talk) 16:45, 8 October 2011 (UTC)

I just removed File:RUTH WOMAN.JPEG from its article; it will then get deleted as an orphan. No need to waste process on it. File:MyKe.a.g.Hawke.jpg is on Commons, so that's not our bother here. If you question its authorship and licensing, then go to Commons and deal with it there. postdlf (talk) 16:56, 8 October 2011 (UTC)

Wikipedia:Graphic Lab/Illustration workshop and WP:NFCC compliance

Regulars here may be interested in This WP:AN/I discussion and this WP:GL/I discussion. --Hammersoft (talk) 22:16, 12 October 2011 (UTC)

NFCC#9

I generate a report daily about NFCC9 violations, its been over 50 violations consistently since 2011 07 23 (and I was ArbFucked™ in early July and can no longer take care of these issues myself) and its now approaching 300 violations. Can someone please take an axe to tools:~betacommand/nfcc/NFCC9.html ?? ΔT _{^{The only constant}} 15:30, 13 October 2011 (UTC)

I see some of them are audio files; is the consensus that linking to one in Template:Listen on a page constitutes a "use", in that it allows the file to be played directly from that page? postdlf (talk) 16:08, 13 October 2011 (UTC)
- Yes; it doesn't matter how the non-free is wrapped up; if it is otherwise directly viewable or playable on the page, that must meet NFCC. (colon-linking, on the other hand, is fine wherever, though that's not considered a use if its an issue with orphaning). --MASEM (t) 16:11, 13 October 2011 (UTC)
  - That seems counterintuitive when compared to image usage in the article, but I don't really have a workable alternative standard to suggest, so meh. postdlf (talk) 16:14, 13 October 2011 (UTC)
    - Again, we're talking non-frees being used outside of article space - a no-no regardless how its presented. We need the colon-link ability to obviously discuss problematic images or files and other encyclopedia-building tasks, but having them directly present on a non-article page is never appropriate. --MASEM (t) 16:18, 13 October 2011 (UTC)

There are some false positives in the report, because of mislabelled images (such as File:Bluecanvas logo.jpg, which I just corrected as PD-textlogo, not non-free) so anyone working from it should look out for that. postdlf (talk) 16:14, 13 October 2011 (UTC)

DASHBot Task #5 works on the issue of removing non-free images from places outside of main article namespace. But, it does not do very many of them, averaging about 7 per day in August for example. The creation per day rate of these violations is more than that. I removed 20 myself today, but Δ is right; the list keeps getting longer and longer. --Hammersoft (talk) 16:37, 13 October 2011 (UTC)
- Does it remove them outright, comment the links out, or add a colon? I went through about 20 or so just now too, and most of them were userspace article drafts, so removing outright is not the most constructive solution. Comment tags is probably the best bot-derived solution, as the use of the colon won't work if, for example, if the image is used in a template that doesn't have the full filename (Example.jpg instead of File:Example.jpg) as a parameter. postdlf (talk) 16:52, 13 October 2011 (UTC)
  - I think it always uses the colon masking technique. Personally, I don't touch userspace drafts if its obvious the editor is continuing to work on it. I really think there should be a special exclusion to userspace drafts IFF the draft is being edited, and it's not been hanging around indefinitely. I think it is very hostile for an editor trying to create a draft to have other people come in and strip images out of it. But, I previously suggested such an exception and the suggestion was shot down. --Hammersoft (talk) 16:57, 13 October 2011 (UTC)
    - I agree. Can you show a link to that past discussion? The same thing should probably go for articles for creation pages too. postdlf (talk) 17:03, 13 October 2011 (UTC)

Wikipedia_talk:Non-free_content/Archive_44#Non-free_images_on_sandbox.2Fuserspace_developing_articles. --Hammersoft (talk) 17:20, 13 October 2011 (UTC)

Thanks. Y'know, now that I've thought about it more, there really isn't a big benefit to allowing display in userspace drafts. You can still collect all the files you plan to use with colon-linking, and if you want to see what the layout looks like, just remove the colon and preview without saving. Speakng as someone who writes quite a bit in userspace drafts, that's not a real burden. And if we were to allow display in userspace, then we'd probably just slide into ridiculous and interminable arguments over whether the draft's use complied with NFCC at various stages of completion ("there's no commentary." "no, but there will be"). So the current rule is the probably for the best, and is at least the easiest to deal with. postdlf (talk) 17:50, 13 October 2011 (UTC)

From an experienced user's standpoint, I 100000% agree with you. From the perspective of a new editor here, not so much. There's such a massive learning curve here. It is incredibly easy for a new editor to bump into walls. There's signs all over the place telling them what a screw up they are for violating this, that or the other policy/guideline. We tell users to create userspace drafts, then beat the crap out of them for not using references properly, not citing sources, using non-free images improperly. It's like, "Welcome! We're glad you're here!" and then ... "don't mind the piano that's about to drop in your head! Have fun!". If we had a bot that went through userspace and looked for pages that were tagged with a userspace draft tag, and how long it had been tagged with that, it could make a choice on when to delete. Letting a bot manage that so such pages didn't get more than (arbirtray number) days old, it'd be a LOT less hostile towards newer editors here. --Hammersoft (talk) 18:14, 13 October 2011 (UTC)

Depends on the usage. If there's a userspace draft where the non-free images are going to be valid when the draft goes live, then I'd agree with you. However, most of the ones I see is where a user has copied an existing article into their userspace so they can edit it, but the images are irrelevant to what that article is eventually going to be. In that case, there's no problem removing them. Black Kite (t) (c) 17:44, 17 October 2011 (UTC)

What constitutes contextual significance when it comes to alternate cover art?

When I brought an issue concerning perceived vandalism to WP:ANI regarding an IP who kept removing the alternate cover I added to Who You Are (Jessie J album), Black Kite decided to remove it, citing that it did not meet the minimum use (which he states is zero non-free images) and contextual significance requirements, as well as stating that there is a free alternative to the image. He elaborated that articles on albums should only have one version of the album artwork unless an alternate version has some sort of historical significance (citing Electric Ladyland's other album cover due to censorship).

As far as I have been aware, this is not the general practice among those who work on these articles, and the template {{Extra album cover}} was created to locate the image in the article's infobox.

Surely, there are plenty of album covers (alternate or main) that have no contextual significance other than being the cover for that particular album. Does that mean we should remove them from the articles, replacing them with a description of the cover because they have no contextual significance and are above the minimum usage threshold of 0?

What is the proper course of action here? Should {{Extra album cover}} be nixed because it encourages people to violate this policy? Should all album art be removed unless it is free or unless the cover was subject to some form of critical commentary? Or are all album art variations suitable usage of non-free content?—Ryūlóng (竜龙) 18:22, 27 September 2011 (UTC)

As I said on my talk page {{Extra album cover}} is occasionally valid, if there is sourced comment about two different covers (my example would be Electric Ladyland, although that doesn't use this template). Otherwise, WP:NFCC#3a applies - the image is being used only for identification, so we don't need two - especially, as in this case, where they're practically identical. Black Kite (t) (c) 18:30, 27 September 2011 (UTC)
- Black Kite's got it. Alternate album cover needs to be a discussion of the art/cover itself to be allowable for inclusion. See above section why we allow one cover to be used even if the cover is not discussed. --MASEM (t) 18:38, 27 September 2011 (UTC)
  - But what is a sourced comment? Does it have to be "Oh, there was another cover because of [controversy]"? Or does "The album came in a standard and (a) deluxe edition(s)" suffice, particularly where the difference between images is not merely color or posing, as was the case with the Who You Are article? What is and what is not allowed?—Ryūlóng (竜龙) 18:46, 27 September 2011 (UTC)
    - The comment needs to either explain why there was an alternate cover, or if there was commentary from critics or the like on the cover itself. A comment "the deluxe edition has this cover" with no other comment is not sufficient. In other words, the commentary about the cover need to come from secondary sources or from the primary source itself; you cannot use original research to assert an alternative cover is appropriate to include. --MASEM (t) 19:02, 27 September 2011 (UTC)
      - So either the alternate artwork has to be discussed by someone or the artist has to say "I gave this an alternate cover because [bar]" in order for alternate artwork to satisfy contextual significance? I know that on some articles I have worked on, the music news sites report that there's two (or rarely more) versions of the release and then they include the two (or more) variations of the artwork in their news article. Does that satisfy the contextual significance or is something else required?—Ryūlóng (竜龙) 19:15, 27 September 2011 (UTC)
        No, it absolutely doesn't - the "main" image is fine for the infobox, but you're going to need to have actual critical commentary (i.e. a significant and reliably-sourced discussion on why the alternate is important) to have it shown. What certainly isn't going to pass NFCC is the fact that different covers are used in different territories, unless (again) those covers are referenced with critical commentary in the article (not just "there was a different cover in Outer Mongolia"). Black Kite (t) (c) 19:19, 27 September 2011 (UTC)
        Fine. That covers international editions. But what if there is no single main image? What if there's a standard release and a deluxe release within the same territory that have equal prominence within their markets? Randomly selecting an article from the news source I usually go with, this EP/single has 3 different versions, each with its own version of the cover, and none more prevalent in the market than the others. The same can be said for this release.—Ryūlóng (竜龙) 19:27, 27 September 2011 (UTC)
        In that case, if none of them are more useful for identification that another, we use none of them. Minimal use = zero. Black Kite (t) (c) 19:29, 27 September 2011 (UTC)
        Now that doesn't seem right. If all can be used to identify the album, but none more than the other, why should the minimal use be zero?—Ryūlóng (竜龙) 19:31, 27 September 2011 (UTC)
        
        Or as in the case of the Video Games project, we have a set of guides based on country of first English publication (or otherwise of the developer/publisher's country) to select the most representative cover for the work.
        
        Or, if the publisher (the one holding the copyrights on the cover art) has actually published a single-image montage of all the cover art, that single image can be used instead. You as a WP editor cannot create the montage as that ends up being X pieces of non-free media against the 1 provided by the publisher. --MASEM (t) 19:33, 27 September 2011 (UTC)
        That also doesn't make sense. If something is released only in one market, and it has several pieces of cover art, what is supposed to be done? For the second link I posted, all I can find is this. So does that mean that tiny set of images to the right should be cropped out and utilized on this page (for which they have chosen the CD-only yellow-colored release)?—Ryūlóng (竜龙) 19:37, 27 September 2011 (UTC)
        Well, no, the montage solution is not always ideal, though I've seen it used successfully in one case (I just cannot recall which). When you start getting into alternative cover art, you need to ask "how much different is this from the original" and "how much of this is easily abstracted by text". The Tone album covers are different - - but they should the same two artists in only slightly different poses, no background, and different color filters. There is no effective difference between them, so you simply pick one and say that's what it is. This why the VG project developed rather objective guidelines to avoid endless wars about which one to use. This would be the case where i'd think the albums project could have the same. For album art that varies on the type of release (normal, limited, +dvd, etc. etc.) my gut tells me to stick to the one that is the CD album cover - the most common denominator for the work, but that's just me. --MASEM (t) 19:42, 27 September 2011 (UTC)
        
        (edit conflict) An editorial judgement has to be made. For intance, there are at least two covers for this album, but, ultimately, that isn't that important. No one gave any real commentary on the covers, or mentioned them in any significant way. As such, I made a judgement call, and used the first one released. The article's now featured, and, I hope, would not have passed FAC with multiple non-free cover images. A collage doesn't really solve anything, as the same amount of non-free content is still being used. J Milburn (talk) 19:45, 27 September 2011 (UTC)
        What about the "Oh My Gah!" (this release) situation? While two of the covers are similar, they still feature the people in different poses and costumes, and the third (the CD-only release) is radically different from the others.—Ryūlóng (竜龙) 19:49, 27 September 2011 (UTC)
        Make an editorial decision. Again, I point out the VG project has made an objective set of guidelines to prevent too much fighting over which cover to use, I'm surprised there's not a similar approach for albums. My only advice is that because you are getting a "free" (as in, will likely not be contested by NFCC) use of an album image, use the one that conveys the most information that you talk about in the article. The cases you have don't have this, but I can image if one case of two "equal" album covers is one an abstract piece of art, and the other based on the band posing, the latter is going to be the better choice. --MASEM (t) 19:59, 27 September 2011 (UTC)
        Again, what if both are equally relevant to the discussion of the article? I know that one single that has an article I am working on is going to have two variations of the cover art. They images have not been released to the public yet, but based on precedent I know that one will feature the performer, and the other will feature a character from the program that the song is used as the theme for. For example, these two images are the covers for a single for which the English Wikipedia does not have an article. The first shows the performers and the second shows the characters. Which is more useful in an encyclopedic sense?—Ryūlóng (竜龙) 20:12, 27 September 2011 (UTC)
        You'll have to make a choice barring any additional commentary on either image. My suggestion: as the artists are alive (I assume) and likely already have free media, the character version may be better since now you can also show the connection to the program, which otherwise would still require non free media to show. --MASEM (t) 21:12, 27 September 2011 (UTC)

Section break

Guidance which has been proposed in the past and found wide support, and which I believe also reflects prevailing standards at WP:FFD is:

...ensure that if you add additional non-free images, that the use complies with the non-free content criteria. Essentially, an alternate cover that is significantly different from the original and is widely distributed and/or replaces the original passes the criteria for identification. Also, an alternate cover that is the subject of specific (sourced) critical commentary passes the criteria for inclusion."

The reason we show album covers is essentially twofold: (i) for identification by the reader, so that they see and are reminded of the album cover they know, which assures them they are at the right article, and by helping to unlock what they already know about the subject adds to what they take away from the article as a whole; the second (ii) is to show how the album was presented and marketed, as this is an important part of the story of the whole product.
When you have an album that has a very different cover e.g. in UK/Europe and in the United States, or very different covers at two different stages of its release history, the consensus has been that the above objectives are best served by showing both covers -- because otherwise only half the readership will see the cover that they "know"; and they will only get half the perception of how the album was released and marketed. This position has been upheld in a slew of decisions at FFD when people tried to push it, conforming to NFCC #8 as judged in context of the extreme slightness of the copyright taking, being that these images were specifically created for the identifiability and promotion of the albums in question; and since then appears to be the standard that has stably persisted in articles. Jheald (talk) 22:02, 27 September 2011 (UTC)

Which this example fails, for both reasons. However two points (1) in all circumstances the second cover still needs to pass WP:NFCC, which of course overrides this guideline, and (2) a cover doesn't get a free pass just because it was widely distributed, especially in a non-English speaking country - this is en.wiki after all. Black Kite (t) (c) 22:17, 27 September 2011 (UTC)
- This is what I don't get. Black Kite is saying that policy doesn't allow it. However, Jheald has brought forward the more common practice on the project which apparently conforms with the policy. Which is correct?—Ryūlóng (竜龙) 23:27, 27 September 2011 (UTC)
  - You're missing the point. There is no hard and fast rule except WP:NFCC. If a second image passes all ten criteria of this, then it can stay, otherwise it goes. JHeald is pointing out situations where a second image may pass WP:NFCC, not where it will. The vast majority of alternate covers will fail, usually (as in this case) because they're so little different from the first one. Black Kite (t) (c) 06:06, 28 September 2011 (UTC)
    - But does File:Jessie J - Who You Are (Japan Edition).jpg fail WP:NFCC when File:Jessie J - Who You Are.jpg is also used on Who You Are (Jessie J album)? Because you have been the only person to suggest that and I still do not agree with your assessment.—Ryūlóng (竜龙) 06:20, 28 September 2011 (UTC)
      - Do you know how many different cover variants there are? Is the Japan release the only alternate, does it use the same cover art as other non UK-US releases, or was there a different cover photo used in every country? postdlf (talk) 07:02, 28 September 2011 (UTC)
        It is the only alternate cover that I have found, because it is the only non-English speaking nation in which the album was released in this format.—Ryūlóng (竜龙) 18:42, 28 September 2011 (UTC)
        
        There are several reason not to have alternate covers many involve the WP:PAGESIZE and making the articles too long. Too many albums, singles, etc, have very many album covers and several of them aren't even English release covers, their mainly international. I've tried this conversation before and trust me there's no winning so please for the love of God give it up and let it go already, a majority rules. I stick by the standards it's a violation of the Non-free content and that's that. JamesAlan1986 *^talk 08:48, 28 September 2011 (UTC)
        But if there's just alternate versions within the same nation (neither one the main version), that are drastically different images but they have equal encyclopedic value in inclusion, why should we limit it to one 300x300 (or whatever the minimum size is) image when both convey relevant information?—Ryūlóng (竜龙) 18:42, 28 September 2011 (UTC)
        We need to start with the premise: most cover art does not contain relevant information - it is never referred to in the article, but it is only the nature that consensus agrees that one piece of cover art should be included on notable published works. I don't necessary agree with this view, but this is what consensus recently (within the last year, if I recall correctly), so I will stand by that consensus. This means that for any alternative art to be used, its relevancy must be seen through sourced discussion of the cover art. It doesn't matter if there's equally "official" cover art for the same product in the same reason (much less differences in international regions and/or special editions); if there's no discussion of the cover art, you only get one "free" cover image. Remember, there are plenty of places on the internet where readers can find all the variations on cover art. Furthermore, to an earlier argument, I disagree that we are including cover art to have readers assure they are on the right page. We are not a printed work, where such images would be useful when you're flipping through pages. We use our leads effectively to assure via text that the user is at the right page, and if they aren't, they likely can link to the page they were trying to look for via navboxes and the rest. --MASEM (t) 19:31, 28 September 2011 (UTC)

Masem, I'm not 100% sure which previous discussion you were specifically referring to above, but if it was the RfC on Cover Art from January 2011, then the closer's summary was
In general, artwork used in infoboxes qualifies under non-free content criteria #8 in that the article and image work together to justify its use. Therefore detailed dicussion of the artwork itself is not necessarily required in the body of the article,
addressing the main point of contention in the RfC. The closer did not determine that "one piece of cover art should be included on notable published works" (your emphasis). That may have been a line that you sought to press, once it became clear that more fundamental change to the status quo was generating no traction, but it gained little interest or comment in the discussion, participants apparently seeing it as of little relevance to the main question the RfC had been convened to discuss.

Furthermore, it is worth noting that the RfC did not endorse the cover art images on the basis of any kind of exceptional dispensationalism, but rather on the basis that WP:NFCI#1 directly conformed with the requirements of WP:NFCC#8: the cover art was upheld because it was judged to be part of the significant understanding of the topic such an article should convey -- for the kind of reasons I've set out above, in my comment above of 22:02 on 27 September. And to repeat what I wrote there, the judgement has been, at least for music albums, that when an album has very different covers for its release in different parts of the world, or at two different stages of its release history, then the look of both has been considered part of the knowledge that it is appropriate for the article should seek to convey.

The wording on this I cited above was developed in an open RFC that went on for several months. It reflected the balance arrived at after a run of contentious IfDs, and represents a line that has held -- on the one hand, those seeking to preserve images have accepted that those that do not meet these criteria should not be kept; on the other hand, those who initiated those original runs of IfDs have since tended to nominate only those that do not pass these criteria. And it is perhaps worth noting that when the wording was presented here (August 2009), you yourself Masem called it "a completely reasonable and high bar for image inclusion, and a good faith effort to commit to NFC policy".

Of course you're welcome to change your view, and to try to change the consensus position. But the way to do that is through an open RfC specifically on this question, and then see what the community thinks. You can't just write your personal opinion as a footnote into the guideline text, as you did on 27 September (diff -- my apologies that it's only now that I've spotted this). Accordingly I've replaced this text with the previously understood position, and suggest that if you now think this is not correct the appropriate way to move forward would be a formal RfC. Jheald (talk) 09:01, 7 October 2011 (UTC)

Would this also be the same RFC where it was summed up that there should be "Stronger discouragement of alternate cover art just because it exists"? :) If you can explain how an alternative cover "significantly increases the reader's understanding of the article", as opposed to a piece of text mentioning that there was an alternative cover, then I'd be happy not to remove them whenever I see them through a failure of NFCC#3a and NFCC#8. As far as I can see, the only thing that a second cover increases the understanding of is what that cover looks like. Black Kite (t) (c) 09:48, 7 October 2011 (UTC)

Masem doesn't get to close his own RfC. As I wrote above, that may have been the new proposal Masem chose to press, once it became clear that his original expectation of a radical change to the status quo was generating no traction, but it appeared to gain little interest or comment in the discussion.

As to your second point: indeed, what a second cover increases is the understanding of is what that cover looks like. And the RfC established that such understanding can sometimes be exactly the understanding to be considered significant, in the context of NFCC #8. The question is in general when should that understanding be considered significant, and this was the question considered in the IfDs and subsequent lengthy discussions in 2009, reaching the view that for album covers (assuming of course the differences are not so slight they can be conveyed in words) it should be when the alternate cover can be considered to be widely distributed in relation to the other cover, and/or when it totally replaced the other cover. Jheald (talk) 10:52, 7 October 2011 (UTC)

That would suggest that this discussion apparently agreed to modify our policy, yet there is no mention of it at the guideline WP:NFCI, let alone WP:NFCC. How can editors be expected to be able to police non-free issues if people are making up their own guidelines and not publicising them? Black Kite (t) (c) 11:05, 7 October 2011 (UTC)

Edit: Oh I see, you've changed NFC itself. I reverted that, not out of a purely political standpoint, but because your wording is horribly open to being gamed (not to mention that I don't believe it does represent either practice or consensus). What does "widely distributed" mean? One large country? Many smaller countries? Should we be looking at sales figures? That's just too vague. I think it's also open to abuse. Many major album releases have different covers for Japan, and quite a few also have different ones between the US and Europe, not to mention other markets. Three or more covers for a single album is clearly excessive. Black Kite (t) (c) 11:14, 7 October 2011 (UTC)

(I've removed the footnote entirely, pro tem, because reverting it back to Masem's personal addition is misleading).

Remember to start with that we're explicitly considering covers that are "significantly different from the original" -- which I think that we are in agreement that Ryūlóng's were not. Going back to the 2009 discussions at WT:ALBUMS (which you yourself participated in, and more than once -- they weren't held in a cupboard), what is a major or minor territory had to be considered in the context of the album. A release cover only used in a minor territory (Yugoslavia was the example somebody gave) should not normally be a "widely distributed cover" in the context of a worldwide release. On the other hand, if that was the singer's home country, where the album was originally released before it was taken up more globally, then that should be considered a wide and significant release; equally, if that was the one territory where the band really had particularly significant sales.

Further up you write that "this is en.wiki after all". But we have an entire wiki-project devoted to broadening WP's coverage from just the English-speaking nations' view of the world. And it should also be remembered that en is the most important second language or swing language for most of the world -- we're not just here for native speakers. Indeed just last week, the Signpost was noting that en.wiki is often the preferred wiki in the area of popular culture even for German speakers, because de.wiki's coverage tends to be so stunted. Furthermore, even en.wiki's home populations can be expected to have a worldwide interest in the subject -- note for example the long-term popularity in domestic record shops of Japanese "import" releases of mainstream Western artists. So I don't think we can dismiss a cover just because it was only "big in Japan".

I do agree that the case for alternative covers gets harder to make after two or perhaps three covers for a particular album. Beyond that, I suspect the principal understanding being conveyed to readers is just that "this album has had a lot of covers", rather than what those covers were. But I think, even in this day and age of multiple versioning, it is rare that an album would have so many significantly different covers that can truly be called "widely distributed". Yes, at the end of the day the language is ultimately only sketching out the kind of issues that need to be considered; it is not always going to act as a clear bright line. Particular circumstances will sometimes apply. But I do think the language is enough to give a good steer as to what consensus expects. Jheald (talk) 12:25, 7 October 2011 (UTC)

Except that the Music Wikiproject seems to be the only project that feels it needs to have alternative art. I don't see the issue at Film (which certainly would benefit from multiple advert posters around the world), books, video games, whatnot. Yes, works are released around the world with various different covers, to a point where its the norm, not the exception. Without any sourced discussion of any of the cover art (or perhaps even the grouping of cover art, explaining why there are different covers in different regions of simultaneous releases), it should be sufficient - for the reasoning that we allow cover art - that one representative image is enough to show the style of marketing and branding implicit with the article. Cover art is not there to assure the reader has a one-to-one connection between the published work and the cover art so that they know what to look for when they want to buy it in the store. Without any other text to support it, the only purpose cover art supports is to provide some type of context of the branding while reading the rest of the article; whatever connection the reader gets out of that is out of our hands, and we cannot force that to be "to explicitly show what the reader to look for to buy". Ergo, if you have multiple covers: you chose one by some objective means with consensus overriding, and seek out sourced discussion if you want more, otherwise it violates NFC and rubs against the Resolution. --MASEM (t) 12:46, 7 October 2011 (UTC)

If the marketing and branding is very different in different territories (or in different decades), how is it that one image is supposed to be represent that?

Regarding the different media, it is simply the nature of the beast. The rule is for albums to have a particular defining image, which for the most part typically is used worldwide, remains unchanged for decades, and becomes strongly associated with the product. A few albums sometimes have two very different covers used worldwide, or are re-released with a very different cover and presentation. But it's rare. One can conjecture why this should be; it may be because music and sound are otherwise comparatively un-concrete and intangible that a strong visual image becomes so strong a part of an album's identity. But that's the way it is.

I don't think what we're trying to do here is "to explicitly show what the reader to look for to buy". Rather, I think an important part of what we're trying to do is to unlock knowledge and associations the reader may already have about the album, that helps build their understanding of the topic from our article. That, to my mind, is why NFCI#1 and its forerunners have always highlighted the value of these images "for identification", as well as showing how the labels wanted their product to be identified.

Finally, nobody is setting out to violate NFC here. NFCC #8 provides that an image must add significantly to reader understanding of the topic (implicitly, as judged against the degree of the copyright taking). We're just trying to formulate ground-rules here as to when generally the community believes the alternate cover will be significant in the context of the topic -- appropriate ground rules, as you yourself recognised in 2009. Jheald (talk) 13:18, 7 October 2011 (UTC)

First, the marketing edge is what came out of the RFC, and why we allow cover art in the first place without any specific discussion of it; we can imagine that every article with cover art implicitly has text "this is a cover of the published work that the publisher used to brand their product". The statement "the marketing effort is vastly different in different countries" as to support additional covers is, effectively, original research, and thus without a source, you're making the presumption that this is the case; that's not good enough to support a second cover.

Cover or identifying image are not used to help connect the reader to something they already have seen or possess, either. There's a reason we stripped cover and identifying art from discographies and episodes, where the images clearly would be used to support this identification. And if that's the only reason to have the images, we might as well publish every single variation of the cover (translated text, different format/aspect ratios, etc.), which of course we can't do. A well written lede is the assurance that the reader has landed at the right page, not the cover art image. --MASEM (t) 13:26, 7 October 2011 (UTC)

The fact that the cover presentation and image is radically different is directly verifiable from the cover image itself, so hardly OR.

As for the lede, yes that can speak to the reader's intellect. But like Proust's madeleines, the image can unlock a lot more than that (see involuntary memory); and we have long recognised that is part of the value of having it.

And pretending that if you have any then you have to have all is just silly. We only accept NFC that adds something significant to the reader, over and above what they get from the other content on the page; and over and above the kinds of small differences that could be explained in words. A very different key image with different presentation satisfies that. Mundane differences, such as translated text, different format/aspect ratios, etc., clearly do not. Jheald (talk) 13:53, 7 October 2011 (UTC)

But why is it radically different? Visually, yes, I can tell that would be the case between alt covers, but without any explanation (likely necessary to have sources for this), you're presenting images without supporting information. Second, I do recognize that visual images can help people understand articles, but we have to consider that NFC and the Resolution are all about balance between maintaining the free content mission and using free (which may include "zero") alternatives, and the aid to the reader. Sourced discussion of a alternate cover: strongly agree this helps. Providing a radically different alt cover from a different region without additional comment, very much unlikely. Again,I point to the fact that nearly every other project that deals with commercially published works stick to one identification image, with the music project being the outsider here. --MASEM (t) 14:04, 7 October 2011 (UTC)

Of course it would be a nice thing to have discussion about how covers came to be chosen or created, for any of the covers presented. But that is a bonus: it is not principally to support such discussion that we present such covers. As the RfC clearly confirmed, the key value in showing people covers is in the covers themselves, so people then know what they look like; while further text would be nice, it is simply what the cover or covers in themselves looked like that is held to be "significant" understanding about the topic imparted by the article.

Does this actually impact our "free content mission"? Hardly; though it seems to be a slogan that some people trying to stampede the community sometimes get behind. As was hashed out at length in 2004 and enshrined here, our free content is technically no less free for having appropriate non-free images aggregated with it. In practical terms what matters (and what WP:NFC is here to protect) is our redistributivity, our reputation, and of course our legal position. These images impact none of those things, because they are accepted worldwide -- for example, even on fr.wiki, operating under the stringent restrictions of French law and European law. Ultimately the balance at WP:NFC reflects the community's view of what best supports our m:mission, because the more rounded and comprehensive our articles are, and the more WP is the first choice go-to site for information, then the more likely it is that people will contribute further content here for us to freely disseminate effectively and globally.

Finally, on the question of covers for albums versus other media, I've already addressed this above. Jheald (talk) 14:34, 14 October 2011 (UTC)

I doubt the legal question is even one we have to worry about; displaying cover art alongside discussion of the published work is (as I recall) actual case law supporting fair use within the US. So legally, not an issue. The thing to remember is what basically the Foundation has said of non-free media, per point 3 of the resolution: Such EDPs must be minimal. Their use, with limited exception, should be to illustrate historically significant events, to include identifying protected works such as logos, or to complement (within narrow limits) articles about copyrighted contemporary works. An EDP may not allow material where we can reasonably expect someone to upload a freely licensed file for the same purpose, such as is the case for almost all portraits of living notable individuals. Any content used under an EDP must be replaced with a freely licensed work whenever one is available which will serve the same educational purpose.. Clearly, the Foundation themselves suggest themselves that cover art can be acceptable, and the last RFC has shown that one cover certainly is on en.wiki. But the question that is on the table and yet to be answered is is about the minimization of non-free uses, and as soon as you start talking about a second cover image, that should be raising red flags as to why a second cover is needed. Such a case is certainly acceptable if there's discussion about the second cover that, upon seeing it, will make it easier for the reader to understand than by text alone, and thus fits the framework of both the Resolution and NFCC. But adding that second cover without any comment at all begs the questions, "is it replacable by 'nothing' (which does qualify as a "freely licensed work")?", and "is the reader's understanding of the article harmed by its omission?" Reasons being given along the lines of "so people then know what they look like" are very touchy-feely, particularly after one cover has been displayed without question, and lead to a whole host of slippery-slope needs for inclusion. We must assume our readers are aware that cover art may vary region to region, edition to edition, and know that what cover is shown is a representative one. --MASEM (t) 14:45, 14 October 2011 (UTC)

Remember, the test set is not "reader understanding of the article" -- a mis-statement that (for whatever reason) you seem to make again and again; rather it is "reader understanding of the topic", of which the community has determined that knowledge of the cover art is a significant and relevant part. In the (comparatively rare) cases where an album has more than one cover strongly associated with it, half that knowledge and understanding of the cover art is lost if we show only one cover, that is entirely un-representative of the other.

As for a "slippery slope", en-wiki has been going for almost ten years now. We've included album covers, and alternate covers for most of that time. There's no evidence that it has led to a slippery slope. Sure, occasionally someone tries to add so much NFC to an article that it lights up like a Christmas tree. But when it's out of balance with the topic it swiftly gets removed. The position on alternate covers has been fairly stable for years. All the evidence is that practice along the lines of the guidance that WP:ALBUMS came to that I quoted above is entirely unproblematic and pretty much uncontroversial. Jheald (talk) 15:42, 14 October 2011 (UTC)

"topic" or "article", it's the same thing for purposes of NFCC. I still am at a lost to understand, that if neither cover is discussed, how the reader loses out from the purposes of an encyclopedic work, to be not seeing the second cover. If we were a cataloging service or a storefront, sure, I can completely see that, but as an encyclopedia, I challenge that assertion; the fact that several other projects recognize that alternative covers aren't needed normally shows that it goes against the grain. As for the stability, I agree that NFC is stable, but both consensus can change (the reason I started that RFC in the first place because there was a sign of that in the air from FAC reviews), and that we should always be looking at ways to help the Foundation minimize non-free use, even if that means a use accepted in the past is no longer allowed (eg discographies/episode lists). Alternative cover art presented without comment seems to be of similar nature and thus of questionable allowance with the NFC. (And yes, I know I'm aware that probably 2-3yrs ago, I accepted that cover art that differed significantly could be used, but my feelings have changed as well). --MASEM (t) 17:34, 14 October 2011 (UTC)

Honestly it seems pointless to discuss this to me but I'm gonna give another input. I personally don't agree that there shouldn't be alternate covers as for example How Do I Live has two English covers, the US and the UK and we are suppose to post English differences. So why not post the 2 different English ones so it shows reference to both? But that's my opinion but as I posted above on 08:48, 28 September 2011 (UTC) it's a worthless case to try and talk about as a consensus will never be made on this but that's me. This is my final opinion on the matter I do not intend to make another comment on this. Hope you all get something worked out. JamesAlan1986 *^talk 08:53, 8 October 2011 (UTC)

So is that the conclusion—the music project needs to step in line? Either way, we need a statement for the guide if we don't want this to be a perennial time-sapper. Uniplex (talk) 04:37, 13 October 2011 (UTC)

I don't think that's the real conclusion. It's just what Masem wants to happen and was never given a consensus. The reason that the music project is not in line with the other projects concerning media is that it is more common to find alternate identifying images. However, there are definitely many other subjects that fall under this issue. Catherine (video game) has two covers, but only one has been chosen for reasons I don't know why. The same goes for the various Pokémon games; it seems that there was at one point a decision to only use the first of the versions in the pairs (Red, Gold, Ruby, FireRed, Diamond, HeartGold, Black) when both could be used as neither is more prevalent than the other. The truth is if there's more than one version of the cover artwork, and both are equally valid as items to identify the subject and neither is more prevalent than the other, then there is no reason both images cannot be used. This qualifies for the description of the policy on WP:NFC and WP:NFCC "for visual identification only in the context of critical commentary of that item" and "Multiple items of non-free content are not used if one item can convey equivalent significant information" (one item cannot convey the fact that two versions of the item exist, in my opinion).—Ryulong (竜龙) 05:59, 13 October 2011 (UTC)

And if there's absolutely zero discussion of any cover image from the sources, then "Multiple items of non-free content are not used if one item can convey equivalent significant information" means you only use one image, period. At the VG project, we have set rules to decide which of several possible covers to use based on objective criteria so that's there's no infighting on that and minimize NFCC. Take the Pokemon example: there is zero benefit to the educational understanding to show both the Pokemon Red and Green covers; either adequately conveys how the cover appears and its not difficult - either through imagination or clicking through to a vendor site - to find the other cover. That's another thing that we have to remember - we should not and cannot be the last repository for information a topic; we purposely summarize a topic, but provide links and references to where more detailed informaton can be found. That should include alternative covers if there is no sourced discussion about them. We are not here to make it easy for a reader to shop for a published work; we are not here to help a reader know they're on the right page by just a visual image alone. Remember, non-free media is supposed to be exceptional per the Foundation. Accepting one non-free cover art for notable works seems appropriate, but you have to justify anything beyond that, otherwise it's no longer exceptional. --MASEM (t) 06:09, 13 October 2011 (UTC)

That is not what the policy pages state. You may want the policy to state that you only get one "free use of cover art", but there is nothing on WP:NFC that says that any alternate cover art must be the subject of critical commentary. The only thing that has to be the subject of critical commentary is the album/song/video game/film itself. And there is most certainly nothing on WP:NFCC that defines "minimal use" as any particular number. You may interpret the policies this way, but there is enough leeway in the wording that your interpretation is not the writ rule.—Ryulong (竜龙) 06:27, 13 October 2011 (UTC)

WP:NFCC does contain a requirement for minimal usage. If we were to permit any and all covers for a particular album, it is possible there could be dozens of covers for releases all around the world. Obviously this is not acceptable. --Hammersoft (talk) 11:31, 13 October 2011 (UTC)

Minimal usage can still be 2 or 3, particularly if there are 2 or 3 equally valid covers. There's no reason Pokémon Red and Blue cannot include the cover for the Blue version or the Japanese Green version. If two versions of something exist, then they can both qualify for WP:NFCC#8.—Ryulong (竜龙) 19:11, 13 October 2011 (UTC)

Further, WP:NFCC#8 requires each image to have significant benefit to the reader in comprehending the article and would be detrimental to that understanding without it. I've already pointed out that I can accept one cover image that is presented without any additional discussion as reasonably fitting this as long as the published work is notable (Even the Foundation calls this as a possible use), but you immediately hit a wall when you start trying to have this work with a second, alternative cover presented without discussion; you meet the first part of NFCC#8 that it aids in the reader's comprehension but fail the second part as the reader's comprehension is no long harmed by omitting the alternative cover if you have one already in place. --MASEM (t) 12:25, 13 October 2011 (UTC)

Why do you have an immediate "this is not allowed" reaction to alternate artwork? If one edition qualifies, why can't the second when it is equally as relevant in the discussion of the subject? Again, there is nothing written in policy that says an alternate version requires sourced commentary about it. The only thing that is required is that the object it represents be the subject of sourced commentary.—Ryulong (竜龙) 19:11, 13 October 2011 (UTC)

Not speeking for Masem, but for my part increasing amounts of non-free content usage demands increasing justification. So there's a red version and a blue version of the cover. Does carrying either of the covers as a second cover bring more encyclopedic understanding of the subject to the reader? Absolutely not. It's a pointless second image. --Hammersoft (talk) 19:17, 13 October 2011 (UTC)

One cover is just as justified as the other. It's rare that there are different editions of the same game released simultaneously. So this rarity should be reflected in the rare occasion that we allow alternate cover artwork. Showing just one half (or other fraction) of what is known to exist is a disservice to our readers. The mere fact that we have sources that say "there are two versions of this" should be justification enough to show the other artwork.—Ryulong (竜龙) 19:24, 13 October 2011 (UTC)

And with that we're skittering down the slippery slope. If one cover is as justified as the other, then we can include as many as we like with the same justification. That fails WP:NFCC. Sorry. The point is, one cover fine...past that, you need stronger justification. --Hammersoft (talk) 19:25, 13 October 2011 (UTC)

What slippery slope is there when there are only two versions of the cover? When there's only a Red version and a Green version? Or if there's only a standard release and a deluxe release? I know I'm not about to go make an article on an album I know of that has 6 cover variants and include all 6 variants in the article. But two versions, when we state that there are two versions, should not fail WP:NFCC, and the general practice is that they don't until some twat IP removed my upload of the alternate release cover of Who You Are (Jessie J album).—Ryulong (竜龙) 19:30, 13 October 2011 (UTC)

So your proposed rule is it's ok to break NFCC requirements so long as there's only one alternate cover? If not, then we're on the slippery slope as I noted. --Hammersoft (talk) 19:47, 13 October 2011 (UTC)

Such a rule already exists. But I am not proposing anything. I am merely saying that if there's one version, and another, both are valid under what is written out in the policies we have.—Ryulong (竜龙) 22:07, 13 October 2011 (UTC)

You may be willing to stop at two, but it is such a slippery slope that it will be gamed. For example, The Simpsons Game was released with five (at least four) different covers, near simultaneously. At some point, our article showed all 5 covers, despite the fact that the general theme (one of the major Simspons family member in the fore-front, with minor cast behind them, and all in a yellow tint) was all the same. Given that we have plenty of media on the Simpsons as a notable franchise, there is absolutely no need to show all 5 to understand the branding that went into the game.

What I'm seeing happening is putting too much emphasis on this idea that the reader needs to see the cover to fully appreciate the work. I disagree but fully respect the consensus that a cover helps to identify branding, an implicit part of any published work. But when the arguments for why we need cover (main and alternative) are going "to make sure the reader landed on the right page" or "to help the reader shop for this item" we are failing the educational purpose that the exceptional use of non-free media is to serve. The Foundation does identify that "for identification of published works" as a reason to allow for NFC, but that point needs to be taken to heart. This is where I do see the partial logic that if alt. cover is so radically different to make impossible to recognize the work represented by one cover when presented with the other cover, but at the same time, the only cases I've every see this happening is typically when we're talking a limited or special edition and not the one that is for mass marketing - and ergo not appropriate to include unless the branding of that cover is discussed in detail.

But I'm going to stress this point again: NFCC#8 requires two conditions for each image use: it helps to understand the reader's comprehension of the article; and that the reader's comprehension is harmed without that image. Assuming one cover image (without sourced discussion) is already present, how does the lack of a second cover image that may exist (again, without sourced discussion) harm the the reader's understanding of the article? This is a question that has yet to be answered and seems to be obviously answered that there is no harm. --MASEM (t) 00:27, 14 October 2011 (UTC)

I agree with Masem. For WP, secondary sources are key: the scope of the topic of an artistic work is determined by how secondary sources discuss that artistic work. Typically, this comprises discussion of how the artistic work was created, and the effects that the artistic work had on others. Rarely, does the marketing (including cover artwork) of an artistic work constitute a significant part of the discussion; even more rarely, would the inability to directly view such marketing material significantly impact the discussion of the marketing, let alone the overall topic. A carte blanch allowance of even one non-free image per article is highly dubious in this context. Uniplex (talk) 05:24, 14 October 2011 (UTC)

You are of course welcome to your personal opinion. But what the RfC established was that the community does see the cover art as a significant part of the understanding an album article may convey. Where two very quite covers have been used, both widely circulated, with differences that are more substantial than small tweaks which can simply be conveyed from text, then that is understanding that we fail to convey if we only show the one cover. Jheald (talk) 14:49, 14 October 2011 (UTC)

It's not a ‘personal opinion’; it's an observation: I observe that secondary sources, taking a scholarly approach in discussing an artistic work, do not in general, devote significant coverage to the marketing/cover-art. The 'understanding' that we at WP seek to convey to our readers, is that of a summary of what is conveyed by secondary sources. Editors' views of how different, or how significant, alternative covers may be, are not important in this context. Uniplex (talk) 06:47, 20 October 2011 (UTC)

Cover art is allowed because of WP:NFC#Images. And regarding NFCC #8, it harms the reader by forcing Wikipedia to only present one version of the item when it is clearly evident from reliable sources of all types that there are two variations of the item. Why should we implicitly provide false information by only showing one half of the visual representation of the released media merely because no one states anything about the packaging?—Ryulong (竜龙) 06:34, 14 October 2011 (UTC)

You have it wrong. NFC, specifically. WP:NFCI, is a guideline, providing the common cases where images will likely fall under the policy of WP:NFCC, but as it states there, NFCC still has to be met for each and every case. As for the harm, if it is clearly evident that there are multiple covers for a work, then you can say this, even possibly as a caption under the image ("European release cover") and thus implicit that there are multiple images without having to show multiple images - eg, per NFCC#1, a string of text talking about multiple covers is a free equivalent serving the same educational purpose as actually showing multiple images, and thus we use that over the non-free situation. If the packaging is that important for the reader's comprehension, then you can find sources that talk about it and ergo can show multiple covers, otherwise it becomes original research to assert that multiple cover art is needed to show this. --MASEM (t) 12:34, 14 October 2011 (UTC)

If the packaging is that important for the reader's comprehension, then you can find sources that talk about it -- this is the position you called the RfC to discuss, and which the RfC roundly rejected, because knowledge of the cover art in itself was seen as a significant part of the understanding an album article conveys. If the differences between the two covers are so slight (as they sometimes are) that such understanding can be conveyed fully in text, then text would fulfil that function. But if the second is a completely fresh design, then the understanding of what it looks like is best conveyed by the cover itself, and something entirely in line with all of the NFCC criteria. Jheald (talk) 14:58, 14 October 2011 (UTC)

Wheeeeeeeeeeeeee! Off to add covers back into discographies! This will be sooo cool! --Hammersoft (talk) 15:15, 14 October 2011 (UTC)

Should we take that non-sequitur as a sign that you've run out of actually, like, relevant contributions to make? :-P Jheald (talk) 15:47, 14 October 2011 (UTC)

It might interest you to know that your arguments for retaining alternate album covers can equally be used to support album covers in discographies. --Hammersoft (talk) 18:14, 14 October 2011 (UTC)

Not really. We have a clear community consensus position. There's a difference in contextual significance, in the language of NFCC #8. Per the explicitly sought and clearly expressed view of the community, cover art is considered to add significantly and valuably to the topic of the article when the topic of the article is intended to be a rounded and comprehensive treatment of an album itself. But individual cover art is not considered to make a sufficiently valuable contribution when the topic of the article is simply a list of albums, with no critical commentary of each one (as NFCI #1 specifically disclaims). I may or may not agree with that, but vox populi, vox dei. That is the well-known longstanding line that has been drawn, and I see no chance of that position changing any time soon. Jheald (talk) 18:53, 14 October 2011 (UTC)

Rather the opposite. But, be that as it may, the same arguments could be applied to a broad swath of article types. I cited TV series below as an example. There's also books, games, comics, tv/radio station logos, you name it. Just because something is associated with a topic doesn't mean we have to have a non-free image of it. Such a "line" permits carte blanche use of non-free images. That isn't what the Foundation wanted, even if consensus supposedly says otherwise. --Hammersoft (talk) 19:03, 14 October 2011 (UTC)

Section break 2

And here we have another permutation of this, of how badly things start to go off the rails. See Undercover Boss. It's not enough that we have a title screen in the box from the series. We have to have three of them. Oh but that's ok! The rationales say for purpose of use, 'infobox'. Oh perfect! So, so long as we say we're going to use it in an infobox, it's ok to include as much non-free content as we like. Afterall, we don't want to confuse our poor readers from (insert country) who might get lost because they don't see their particular country's version of the titlescreen in the infobox. Our readers are flippin' idiots, and won't be able to understand they've reached the "Underover Boss" article if they don't see their own permutation of the title screen. The show's already got version for Norway and Germany. We'd better include those title screens too. Oh, but let's not stop there, the series is now producing versions for Denmark, Spain, Turkey, Sweden, Belgium, The Netherlands and Israel. God help those poor readers if they come to the article and don't see their title screen. So, we'll have an infobox with 12...count 'em...12 title screens. Don't forget, loyal followers, all the title screens are in itself signifiicant parts of understanding the show! --Hammersoft (talk) 15:45, 14 October 2011 (UTC)

Per my comment further up the thread [2], it's a mistake to make the same rule for all media -- to assume that because album covers and alternate album covers have a particular significance, the same necessarily applies to other media. The fact is that albums tend to have a particular defining image, which for the most part typically is used worldwide, remains unchanged for decades, and becomes strongly associated with the product. Evidently if a TV franchise has a different title caption in each and every national market, the same can't be said. Though it may sometimes be of interest to present how the branding has been adapted to different locales. Jheald (talk) 15:59, 14 October 2011 (UTC)

CERTAIN album covers are iconic, and there's secondary sources to readily support that. Beyond that, it's complete original research to making any assertion like that, and apply it carte blanche across an entire genre of media. I can just as well say that the title screen in any given country becomes strongly associated with the product. This is the slippery slope your ideas are putting us on. --Hammersoft (talk) 16:05, 14 October 2011 (UTC)

Stop refering to a slippery slope. Undercover Boss clearly shouldn't have those three title screens. However, there should be no reason to prevent me from making an article on the Anna Tsuchiya single/EP "Unchained Gir" and using both equally prevalent versions of its cover art in its infobox. Now, I don't expect to be allowed to do the same for the single "Double-Action CLIMAX form" and its six variant cover art versions, even though they too are all valid and equally prevalent in the market.

It is a rare occasion that there are different cover art editions exist. Why can't the exception be made for when there's a drastic difference between editions (released simultaneously)? I don't expect it to be done for situations such as Born This Way where its special edition cover is just a zoom on its standard cover or the slight differences between The E.N.D.'s two variants. But if we have rare occasions such as what happened with Who You Are (Jessie J album) where the versions are no where near identical and what commonly happens in the Japanese market (the third largest music market in the world), variant artwork should not be forbidden to use just because the cover art is not the focus of critical commentary. As much as Masem and Hammersoft keep pressing for this, there is no policy in place that states that critical commentary must exist for alternate cover artwork which does not even hold for a bulk of the non-free images that exist on their own.—Ryulong (竜龙) 19:41, 14 October 2011 (UTC)

I will continue to use the slippery slope analogy as I see it necessary to highlight the problems with the proposals being made here. There's an apparent effort to describe albums as some sort of special class. They're not. If the arguments to include alternate covers on albums are valid, then the same argument applies to things like Undercover Boss. I don't see any reason to make an exception for 'rare' cases where there are multiple covers. If it applies to rare exceptions it could apply across a huge swath of articles with similar sorts of constraints, as previously noted. No, I don't see a need for exception. --Hammersoft (talk) 20:08, 14 October 2011 (UTC)

Well, I've fixed Undercover Boss which should definitely not have had any of those images (they're fine on the individual articles though). Albums are indeed a special case because there are rarely alternate covers. When there are, why shouldn't we be allowed to include them?—Ryulong (竜龙) 20:33, 14 October 2011 (UTC)

Why should we? That's the question, not why shouldn't we. --Hammersoft (talk) 20:41, 14 October 2011 (UTC)

While it is true that the words "critical commentary must exist for alternate cover artwork" do not appear in any policy or even the Foundation, you must understand that this is based on how NFCC#8 is treated. I hate to keep pointing back to the RFC on cover art, but it important to understand that the reason we allow one cover art image without discussion of the actual artwork is that the cover is a means of representing the published work, which is the subject of sourced discussion (presuming that we've passed WP:N). The (or "a") official cover presents the branding that the publisher has decided to use for that work, and thus implicitly connected. You do not need two or more pieces of cover art to represent the same work, which is what the argument is trying to be used here. The alternative, when there is actually sourced discussion on the second cover is that that cover is no longer being the representation of the published work but an image on its own merit to be understood in the context of the larger topic. --MASEM (t) 20:45, 14 October 2011 (UTC)

Sorry, but you are pulling this out of thin air. The RfC was asked to rule on why we allow cover art, and the resounding answer that came back was that it was considered knowing what it looks like adds significantly to user understanding. You keep italicising the word one cover, but that was not a proposition that got a plurality of support, or even the support of a tenth of the people that responded. There is no backing for the change in practice you are claiming is mandated -- not in policy, not in guidelines, not in past discussions, not in the RfC, not in logic that was supported by that RfC for why we show covers. (Answ: so people know what they look like). So stop flogging a dead horse. Jheald (talk) 21:19, 14 October 2011 (UTC)

Yes. I cannot see anything that says "only one" anywhere other than in your comments Masem. The only thing that exists in writing is "minimum" which has no exact value.—Ryulong (竜龙) 21:31, 14 October 2011 (UTC)

Minimum does have an exact value: zero. Of course, no one involved in this discussion is saying that we should never allow non-free media on WP. What we are saying is that a second cover image provided without comment fails several parts of NFCC, particularly NFCC#3a and #8, that the first image is generally taken as acceptable. In other words, there's what is a point of diminishing returns on the education value of that second cover image when one is already present - the work has been identified, there's no need to identify it a second time with a second image. Arguing that our policy or guideline doesn't spell this out completely misses the point of NFCC; it is a philosophy that we all need to learn to appreciate due to the free content mission and the Foundation's goals. Yes, there is consensus needed, but everything to date has pointed towards the consensus that one cover image is enough: the fact that all other projects involved with copyrighted works agree to this, the fact that we do allow one representative image on lists like discographies or the like, and the fact that we disallow the use of historical logos/callsigns/whatnot without discussion. The only point that sticks out is the music projects.

Again to make it clear: if we are only using cover art to identify a work, "one" is clearly doing a better job than "zero", but "two" does not gain any significant benefit over "one", and thus why "one cover image" is the default standard and what every article on published works should consider as the limit, barring sourced discussion of alternate covers. --MASEM (t) 00:32, 15 October 2011 (UTC)

10c violations

In analogy to the above thread, there is this. This page is a list of all 10c violations (ie all images lacking a valid rationale for some or all of their uses). Currently I go through this list manually and remove violations from the articles where they are used on. However lately I do not have much time for this anymore. Furthermore the list seems to grow faster than it can cut it back and the number of editors performing this task seems to be extremely limited. I would like to explore the possibility of perhaps having a bot for this task. I would be happy to monitor the bots talk page and respond to messages left there. As I feel some people might be opposed to such a bot, I would like to discuss the possibilities and reach a consensus on how this bot (if there were a consensus to run one) should perform this task so that we can come up with a clear-cut proposal that can be presented at WP:BOTREQ and is backed up by consensus. Thanks. Toshio Yamaguchi (talk) 09:37, 20 October 2011 (UTC)

While #10c should be completely enforceable by a bot (it's 100% objective - either the name's there or not), when this was even done by hand, a large number of editors complained because of the rationale having an article name but due to page moves, typos, or other things, the name didn't match the image use, and thus it should have been "obvious" to the editor removing the #10c that it was just as easy to fix the rationale. Mind you, this was also largely directed at Delta would, before, was under certain editing restrictions, but Hammersoft's also received a fair number of complaints. I'm not saying we shouldn't have a bot for this, but be aware you are going to have a great deal of resistance due to this. --MASEM (t) 12:44, 20 October 2011 (UTC)

One plus to this; I've seen people complain like a stuck pig when a given action is done, but if a bot does the same exact thing, not a peep. Food for thought. --Hammersoft (talk) 13:08, 20 October 2011 (UTC)

One more thought from me: We have XLinkBot which also seems to receive a large number of complaints and that bots actions are not even backed up by policy (note that WP:ELNO is a guideline rather than a policy). On the contrary the actions of this bot would be 100% backed up by WP:NFCC, which is official policy. If the complaints of people who don't like this are the reason not to do it, then we could also simply trash WP:NFCC and let people include as much NFC as they want without the need to provide a rationale. Either we should do it right or we should trash NFCC. I cannot see the logic behind a policy that is unenforceable on a large scale without unproportionally high hurdles for those enforcing this policy. And again WP:NFCC is in line with the resolution approved directly by the Foundation, so those who are unwilling to stick to the rules under which a community is supposed to function should be excluded from the community. I think it is unreasonable to expect a community to bend its rules just to fit the needs of some who do not like, do not understand or for whatever additional reasons are unable to follow those rules. Toshio Yamaguchi (talk) 14:04, 20 October 2011 (UTC)

It is not the case that people are dismissing #10c. The history is long and difficult to work through but it is a combination of : such changes (removal from the article) having little to no notification; the lack of any bot/script functions to consider the case of page moves w/o redirects that many images are broken upon completion; the conflict between attitudes of those that aggressively enforce #10c and those that its those #10c enforcers that need to correct problems and not the uploaders or editors using the work, and a general apathy towards what non-free content is on WP (with the common mistake that it just like fair use and we're being overly aggressive on that). Again, a bot should be and can be able to do it, but likely its functionality is going to be needed to be transformed to work within the expectations of those that don't like automatic removal of images.

Also be aware: it is en.wiki that enforces #10c. There's nothing directly in the Foundation Resolution that requires what we have of #10c. --MASEM (t) 14:19, 20 October 2011 (UTC)

The exact wording of "Resolution:Licensing policy#Resolution 4." is "Media used under EDPs are subject to deletion if they lack an applicable rationale." I understand this to be a reference to the non-free use rationale required by 10c or am I somehow misreading this? Toshio Yamaguchi (talk) 14:34, 20 October 2011 (UTC)

And Resolution 4 talks about deletion. Does "Deletion" here mean Removal from a page (such as an article) or does it mean "Deletion of the File from the File namespace". I admit I am confused. Toshio Yamaguchi (talk) 14:38, 20 October 2011 (UTC)

Regarding "...the conflict between attitudes of those that aggressively enforce #10c and those that its those #10c enforcers that need to correct problems and not the uploaders or editors using the work". WP:NFCC#Enforcement does not say or even imply that the 10c enforcers have to correct the problems. I would say this part of the policy is quite unambigious unless I am missing some obvious implication of this part of the policy. Toshio Yamaguchi (talk) 14:46, 20 October 2011 (UTC)

Let me re-preface: I fully support the idea of a bot for #10c enforcement and completely agree that it's policy to do so, I am just trying to prevent another breakdown between those of us on this side of maintaining NFC, and those editors that dismiss NFC or are fearful of automatic NFC enforced (despite what our policies otherwise say).

To be clear: only within en.wiki do we require that the article title be present in the rationale (the #10c part); there has to be a rationale per the Foundation, but we have the extra step. When #10c is failed, we remove the image from the article(s) it is failing on. That may leave valid uses elsewhere, or it may leave the image orphaned, at which point the orphaned image bot will pick it up and go from there.

And the last point, we generally point to WP:BURDEN - those that want to retain information need to justify its retention - when people argue that #10c enforcers should be the ones to correct images. Yes, if its blatantly obvious, its good faith to do so, but we had people arguing that "oh, you should spend the five to ten minutes to shift through all the history to see it was on the right page two page moves ago, and thus should fix it yourself!". --MASEM (t) 15:13, 20 October 2011 (UTC)

@Toshio Yamaguchi - a) XLinkBot does not get a 'large number of complaints', it is significantly less than 1 a day, and there are sometimes weeks that the bot does not get a complaint (I was away for >2 weeks, and found 5 new threads, 4 of which were concerns about, IMHO correct, reversions by the bot), and b) the edits are backed up by WP:NOT#REPOSITORY, WP:NOT#LINKFARM and sometimes by WP:COPYRIGHT (all three policies), WP:EL and WP:SPAM are indeed guidelines, but they word out pretty much what XLinkBot is doing. If you can show that the bot is reverting links on a specific rule and is making a significant number of mistakes with that, then that would be case for concern, but until now I have not seen any proof of such (of a sample of 10 YouTube reverts, I found that it reverted 2 copyvios .. and further ago I checked 30 myspace links, of which there was 1 I would personally not have reverted - but not added either). I still believe that only a small fraction of the reverts are 'wrong', and provisions are there to catch many of them.

However, when reverting on NFCC #10c - where the rationales are not (directly or indirectly) pointing to the correct page would have a very low error rate .. but .. if the rationale is written 'acb', but 'abc' was meant, then editors say that it was obviously wrong, that it should have been repaired (it is displayed on 'abc', so of course it is a typo), not removed, and that a remover should be shot - if the rationale is 'acb', but 'acb' is actually a disambig which contains a link to 'abc', then it should have been repaired (it is displayed on 'abc (1)' - of course it belongs there then), not removed, and the remover should be shot - if the rationale is 'acb' but 'acb' is an article about the same subject as 'abc' then the rationale should be repaired (it is displayed on 'abc (1)', and even though it is not linked from 'acb', it is clear they are both about the first three letters of the alphabet), not removed, and the remover should be shot - because all these are obviously repairable cases, which are hence not wrong - even though they are absolutely NOT pointing to the correct article (if it even exists). I know, unfortunately a fraction of the images are real copyvios. And actually, it does not matter whether there is a rationale or not - the question is then if it is a correct rationale, or better, whether a correct rationale can be presented, whether it is written down or not is no issue.

With XLinkBot, editors sometimes revert, and a sub-optimal external link stays (and sometimes even a plain copyvio link, or even spam) - XLinkBot does not revert again. Here the mechanism could be the same, but one will see, that they have to revert that 'obvious mistake made by the bot' will be accompanied with a complaint .. for every edit. And editors will even revert and ignore when the bot is really correct in removing a NFCC#10c - I've seen people plainly revert cases which unmistakenly are violations of WP:NFCC, and strongly defending them, without any form of consideration that maybe, maybe the removal was right - and even continuing to defend them after they run into a block for the actions.

So yes, it is a good idea, yes, it is easy to do (I even have a framework ready - user:NFCCheckBot), but the amount of opposition I expect - both reasonable and unreasonable (and regularly backed up by significant shouting in unreasonable ways) - does not encourage me to actually make the bot work (and get it activated). --Dirk Beetstra ^{T C} 14:50, 20 October 2011 (UTC) (added something. --Dirk Beetstra ^{T C} 14:58, 20 October 2011 (UTC))

Regarding XLinkBot my apologies. I admit my analysis is completely wrong and this was a poor comparison on my part. Toshio Yamaguchi (talk) 15:05, 20 October 2011 (UTC)

No offense taken, Toshio Yamaguchi. Many bots work on the basis of guidelines (not policy), but generally it does not give much opposition, mere 'it would be better if' (and XLinkBot does get that). There there are also attempts to show the bot is massively wrong (and the bot can be wrong) - but here, even human editors get opposition which is massive. See what happens if you remove images without any form of mistaken rationale (take images without any rationale, e.g.), but where you do see that the image could be fair use. You will be reverted, and you will be asked to write the rationale, because obviously, that rationale can be written, so the image should not be removed. --Dirk Beetstra ^{T C} 15:12, 20 October 2011 (UTC)

And if that were what WP:NFCC#Enforcement said I had no problem with writing the rationales for other editors. But as it is now WP:NFCC#Enforcement prevents me from doing this by requiring me to remove the file if it lacks a rationale. Toshio Yamaguchi (talk) 15:22, 20 October 2011 (UTC)

You're hopefully understanding the social issue that we're having a problem with here. You're 100% correct: our policy says "no rationale with the exact article name (or a redirect that leads to that article) then you are completely free to remove it". But remember that we want to AGF and IAR when appropriate, and if you believe you can fix or add in the right rationale for the image use, you're also free to do that. The problem is that those that are against aggressive NFC enforcement insist that in many cases you as the potential #10c image remover must try to fix the missing or broken rationale before deletion, which is not what the policy says, but that point will be forced at you until you relent or give up. It's a social problem, not a policy problem. --MASEM (t) 15:30, 20 October 2011 (UTC)

Another problem is that the 10c violations often go hand in hand with violation of 8. And I believe many files appearing to satisfy 10c (note that in what I do I only check whether the file does have a rationale at all, but this rationale may still be invalid) in fact fail 8. The problem with 8 though is that it is unreasonable to expect from an NFCC enforcer to write new article content for the file to satisfy 8. In my opinion the 10c violations are only the tip of the iceberg. As you say, it is a social problem. In many cases the desire to include a particular file might not motivated by improvement of the article but by something that might appear to be an improvement of the article to some. Toshio Yamaguchi (talk) 15:46, 20 October 2011 (UTC)

I would ignore anything automatic with #8. That is totally subjective. If you are doing #10c enforcement, and believe you can fix the #10c issue but disagree with the rational given to support #8 and don't know if you can justify the image use, it is almost always better to treat it as a #8 issue and consider it for WP:FFD to resolve that. --MASEM (t) 16:11, 20 October 2011 (UTC)

I think that depends. I have seen some cases where a file simply sits in the article without any reference made to that file in the text. However I think the problem is within the formulation of 8 itself. There is no agreed upon definition of what "significantly increase readers' understanding of the topic" actually means, so yes the answer to the question whether a file satisfies 8 or not is in fact completely subjective and its enforcement (if at all) could only happen per a case-by-case consensus. Toshio Yamaguchi (talk) 16:28, 20 October 2011 (UTC)

Masem and Beetstra have demonstrated how this is a social problem, and I agree with them wholeheartedly. NFCC enforcers encounter one hell of a lot of opposition. It doesn't matter if the NFCC enforcer is right, by terms of policy, or not. Socially NFCC enforcement is wrong on the face of it. The amount of acrimony generated by conducting NFCC enforcement is absolutely intense. There's equivalents in real life society; think of the Confederate flag in American society. It does not matter if the flag were used or referenced in appropriate context. It is inextricably tied to the enslavement of a category of people at the hands of another category. People have deep seated, emotional reactions to that flag. So even the display of it is wrong, even if used appropriately, because of the social trauma caused by its display (I can think of more severe examples than this; trying to be polite; and you see the problem).

There is very little allegiance to NFCC. If an NFCC enforcer removes an image from ACB (nice, Beetstra) because the rationale only points to ABC, which was moved to ACB and a redirect is now at ABC, the person who comes under severe attack is the person who removed the image. Not once have I ever seen a person who conducted the article move being taken to task for not paying attention to resulting NFCC issues. Example in point; User:Fortdj33, an editor with 2 years of experience and 38,000 edits, moves an article [3], breaking the rationale for File:Moomin kuva.JPG in the process. I wish there were a report of the most prolific page movers, to help target this one aspect of the overall problem.

One of the symptoms of the overall lack of allegiance to NFCC is that editors, by and large, will do the absolute minimum to get past NFCC enforcement. Missing a rationale? Fine, slap a {{fur}} on it. Tada! Fixed! I've even seen suggestions of using a bot to add rationales. This totally, utterly misses the point. I despised the creation of templated rationales. It creates a mess which we are now in whereby images have rationales, but the rationales are ridiculously weak. So instead of someone actually taking the time to think "Hmm, do I really need this image? Yeah, I do, ok add that reason to the rationale", people just add images because they want to, don't bother applying a reasonable rationale. So let's say we live in a utopian world and ever image had a rationale for an article where it is used. Problem? We'd still have hundreds of thousands of images with insanely weak rationales. Example in point; File:Wimpy kid.jpg, with a templated rationale. What's the purpose of use? "to use". That's it. Nothing more. To use. That's all we need then, to include any image we want to use on the project? Just let everybody know that we want to use it? But, without actually reading this image description page there is no way to detect this problem. It has a rationale for one of the articles where it is used. If it had two rationales, both saying "to use" for purpose, the image wouldn't show up as orphaned, wouldn't show up as missing a rationale, wouldn't show up as being used in tons of articles, wouldn't show up as being used outside of article space. We'd never know about it. We'd never know the rationale is absurd. There is NO way to fix that without human oversight.

The underlying problem is that the masses have torn down the castle. We have this naive belief that enough people will eventually get to problems to fix them. The reality is blatantly otherwise. There is no way we will ever be even remotely in compliance with all aspects of NFCC policy. The masses don't care about NFCC. Without a fundamental shift in how we approach this problem, this will always be the case. It boils down to that our m:Mission has been compromised. We've lost our soul. There are more than 400,000 non-free images on this, the "free" encyclopedia. As I say on my userpage, if you think we're still a free encyclopedia, I'd like to buy some of what you're smoking. It's some seriously good s**t.

Fundamentally, this points back to the Foundation. I am convinced at this point that the Foundation is grossly incompetent. I have opined on this before (1, 2). They've flatly refused to step in and demonstrate any leadership on this problem. The result? The people opposed to NFCC enforcement say "Well, if they don't care why do you care?". --Hammersoft (talk) 16:40, 20 October 2011 (UTC)

If there was a bot that was ever needed, it would be one that would track page moves and either attempt rationale repair or at worst, tag images with the before/after names into a category for easy repair. This would put nearly all other (henceforth) #10c failures into ones that are typos or that are legitimately lacking the article name. --MASEM (t) 16:56, 20 October 2011 (UTC)

FairuseBot used to do that, but I got tired of constantly fighting the MediaWiki software (the last straw was the partial implementation of image redirects). --Carnildo (talk) 21:13, 21 October 2011 (UTC)

In my opinion, several significant changes are needed to improve the situation for both, the enforcers and the people adding the non-free content. I especially have a problem with two formulations at WP:NFCC#Enforcement:

A file with a valid non-free-use rationale for some (but not all) articles it is used in will not be deleted. Instead, the file should be removed from the articles for which it lacks a non-free-use rationale, or a suitable rationale added.
Note that it is the duty of users seeking to include or retain content to provide a valid rationale; those seeking to remove or delete it are not required to show that one cannot be created—see burden of proof.

Given a valid rationale can be created, why is it relevant who created that rationale? A valid rationale is valid regardless of who wrote it. Secondly, why must a file be removed, if a valid rationale can be created by the NFCC enforcer? And third, I would like to see Template:Non-free use rationale get nuked and instead have a template with fields labeled with numbers 1., 2., 3. ... 10. a. b. c. and perhaps an additional field (maybe collapsible) for the copyright tag. Toshio Yamaguchi (talk) 17:42, 20 October 2011 (UTC)

No one is saying that you, as someone looking at a #10c problem image, cannot fix or add a rationale yourself. That's completely acceptable. The social problem we have is that there are vocal editors that believe that if we (those tracking #10c violations) feel the image rationale is broken, we must fix it, not leave it to the uploader or content users. The reason we have that text in there is that there are legitimate cases where it is not obvious what the rationale should be, or where the image came from or other missing elements (eg, such as the source or copyright holder of an image), where there is no way that the #10c fixer can repair it; the uploader or those wanting to retain it must take on that duty.

And as for numbering the rationale, that's an idea that been out there before and rejected. The current FUR template hits all the major points without necessarily enumerating them. It helps us to avoid straight up mechanical implementation of rationales, requiring people to actually read the NFC and work from there. --MASEM (t) 18:01, 20 October 2011 (UTC)

My hope would be that using something like this would perhaps prevent (some) people from simply slapping a completely unspecific rationale on the file and help prevent cases such as this one? Just an idea (and not necessarily a good one). Toshio Yamaguchi (talk) 18:46, 20 October 2011 (UTC)

While it is of good intent, the fact that it *adds* more fields to be filled out means even moreso people will be disinterested in filling it out to the best they could. Additionally, and importantly, we do not require a standard rationale template, as long as the information on the file's page satisfies the 10 points of the NFCC. Yes, it's a bit easier to read in table form, but that's not a requirement nor something we can enforce. I think the current FUR maps well between been verbatim towards each of the NFCC 10 points but without making it excessive, easy-to-ignore paperwork. --MASEM (t) 19:26, 20 October 2011 (UTC)

It's already being easily ignored, as I noted above. See File:Wimpy kid.jpg. --Hammersoft (talk) 19:33, 20 October 2011 (UTC)

"The fact that it *adds* more fields to be filled out means even moreso people will be disinterested in filling it out to the best they could."

That might be true. But in this case people would be required to familiarize themselves with NFCC Policy (which would be a good thing). The current template might be easier to fill out. But the current template is also more likely to result in a rationale not representing the extent to which the file is compliant with WP:NFCC#Policy. Do we really want to have more people interested in filling out the rationale just for the sake of filling out the rationale? The rationale in itself is actually worthless if the image does not comply with that rationale. My hope simply is that this would result in more images actually compliant with the rationale. The easier we make the rationale to write, the more likely the image will not actually comply with WP:NFCC#Policy. Toshio Yamaguchi (talk) 19:53, 20 October 2011 (UTC)

Hammersoft gave the best example of why having forms doesn't work. The Wimpy Kid image rationale uses the existing template and yet half the fields are blank, the other unacceptable. There's no assurance that with a template form that is explicit to which NFCC should be met that people will spend the time to fill it correctly. Irregardless a user uploading a file is presented with so much information about the NFCC and yet we still have such cases. It's a social engineering problem, getting people to recognize the NFCC as a good metric to have, and getting there needs to be done through socializing that idea, not through beuaracry. --MASEM (t) 20:24, 20 October 2011 (UTC)

How should socializing that idea work? The only idea I have would be to get more people involved with doing what I or Hammersoft do, ie. enforcing NFCC and be willing to teach other people about NFCC in a friendly way, as I did for example here. How could we achieve this? Why would anybody want to do what Hammersoft or I are doing? Toshio Yamaguchi (talk) 21:00, 20 October 2011 (UTC)

It's an impossible question to answer. The only time there seemed to be agreement that NFCC needed to be maintained was the period of a year between when the Foundation's Resolution was enacted and the deadline to bring images into compliance with that. Nowadays, the bulk of editors often mistake "non-free" with "fair use" and assume we work under that more liberal fair use idea, or have a resentment of how NFCC is handled due to heated battles with enforcers and those trying to "undo" the damage. --MASEM (t) 21:06, 20 October 2011 (UTC)

Idea

If we want more people to provide a rationale and one that doesn't look like this one it seems we have to make the process of creating a valid rationale easier. And I can see that my rationale template proposal above is not the way to achieve that goal. We would need an easy to use template for that and personally I would be all for making an easy to use rationale template, perhaps one where people are presented with some kind of preformulated statements that can be added via a dropdown menu (which would require development efforts by our software developers). The problem with this though is that while this would perhaps help to prevent cases such as the Wimpy kid rationale, it is also more likely to produce unreasonably weak rationales. It could be at least help solving the 10c problems though. (I have to log off now and will be away until 23rd or 24th). Toshio Yamaguchi (talk) 12:24, 21 October 2011 (UTC)

A form that amounts to pre-written responses to meeting each of the 10 NFCC points is a worse solution. It also will interfere with more advanced users that do write good rationales from entering more unique qualifiers on specific parts of a rationale (for example, explaining why they have used a specifically high resolution image even though it could be a lower resolution).

I think most of us in this area know that the average layeditor on WP is lazy with NFCC. Lazy as in, they'll fill it out but they won't push to be the best possible written rationale. (but something better than the Wimpy Kid one). As long as the mechanically-required pieces are filled in, that's the best we can hope for, until such a point that the Foundation or consensus believes more stricter enforcement is needed. Furthermore, it is not that these rationales necessarily are bad excuses for using non-free, just not as tight as they could be, and certainly not a reason to say "not a good enough rationale, you can't use this image". Our processes at least are set up to filter on these over time - eg FAC and GAN to some degree with review images under a magnifying glass and will improve the rationales to a great degree - or require the images to be removed.

And to get to the key point, none of this will help with 10c rationales. They are generally broke as a result of page moves or reuse w/ rationale addition, not when an image is uploaded which would be the only reasonable time to present such a form to the user. --MASEM (t) 12:33, 21 October 2011 (UTC)

Including {{non-free with permission}} in this guideline

This template is discussed in Wikipedia:Templates for discussion/Log/2011 October 30#Template:Non-free with permission. Two administrators have concluded this template as useless and a bad encouragement because it is a mere supplemental tag that may kill images that do not have other conjuncting tags. This template has never been included in the guidelines of non-free content; there has been no rationale for including it and not including it. ~~If this were deleted, I would have the discussion reviewed in Wikipedia:Deletion review.~~ --Gh87 (talk) 10:26, 1 November 2011 (UTC)

Wikipedia:Deletion review isn't a second chance WP:TFD. As WP:DR notes, "This process should not be used simply because you disagree with a deletion debate's outcome". Threatening to use it if this TfD does not go your way is highly inappropriate. --Hammersoft (talk) 13:01, 1 November 2011 (UTC)
- I scratched that out. Well, I guess I must find the flaws in arguments to have this TfD rather than disagree enough to start the review. --Gh87 (talk) 17:24, 1 November 2011 (UTC)

FAQ

I have seen many disucssions regarding the use of pictures that have been made available to the public by large organisations. Woudl it we appropriate to add a FAQ section to this article to gove answers to specific questions? Martinvl (talk) 08:34, 5 November 2011 (UTC)

10a Identification of the source of the material

There is a lack of clarity regarding the nature of the source. Will any source suffice - even one that is using the image illegally? My assumption has always been that a source would be reliable and would be using the image appropriately, but the criteria doesn't state that. If the quality, reliability and legality of the source is not important, why are we asking people to mention the source. What exactly are we checking? If it is that the image was simply available - then that is obvious, because we have the source. If it is to check that the image is correct or accurate, then if the source is unreliable and/or illegal, then we cannot trust that source. SilkTork ^{✔Tea time} 11:29, 5 November 2011 (UTC)

Identifying the source gives us some way to check on how the original image was used, who the likely copyright owners are, and other factors in resolving issues with non-free. It demonstrates that the non-free image is not being first published at WP (that's completely inappropriate) and also can potentially give us info when it enters the PD. Now if the source is using the image "illegally" we should use a better source, the closest to the original publisher as we can. --MASEM (t) 11:49, 5 November 2011 (UTC)

If the source is using the image illegally and inappropriately, then we are getting misleading information. What would we learn from a site that said the image was owned by themselves, when it actually is owned by someone else? Unless the source is reliable, we might be better off not using it. I think it is better that we had a policy that insisted on images coming from reliable sources. SilkTork ^{✔Tea time} 13:35, 5 November 2011 (UTC)

The question, given the use of fair use outside of WP, how is one to tell if an image is being used illegally? I know there are cases of people at flickr taking obviously copyrighted work that is not their own and labelling it as CC-BY, and I've seen cases of people putting images on websites they claim their own despite the fact there's a watermark to belie that fact. Can you give an example of a image that is a problem in this manner? --MASEM (t) 13:49, 5 November 2011 (UTC)

The image that prompted my question is a publicity shot of Otis Redding - File:Otis Redding.jpg. And the discussion about it is here. The consensus is that the image can be used because a source is given, even though the source is poor. I found that odd, and so raised the issue here. We require reliable sources for challengeable material, and I would think that a non-free image is challengable - certainly such images do get challenged and deleted. The verifiability policy is talking about words, so it would be inappropriate to apply it to images, however, the reasoning behind the policy for text should also apply to images. There is the very basic principle of how do we know the image is of who or what it is purported to be unless we can rely on the source? SilkTork ^{✔Tea time} 19:08, 5 November 2011 (UTC)

Suggestions re Template:External media, extra restrictions on wire-service type photos

It might be worth adding a mention to the article that Template:External media is a decent work-around when you can't find an acceptable fair-use photo.

Also, I don't see a cross-ref to the fair-use of wire-service, Corbis or Getty (etc) images, which I understand are restricted to cases where the image itself is the subject of sourced commentary. Which is a very restrictive covenant. Did I just miss it? Best, Pete Tillman (talk) 04:32, 6 November 2011 (UTC)

NFCC 8

Could we perhaps clarify NFCC Policy 8? What does "would significantly increase readers' understanding of the topic, and its omission would be detrimental to that understanding" mean? To give a specific example: Is the use of File:Iraq National Library Destroyed.jpg in Iraq National Library and Archive compliant with 8? Toshio Yamaguchi (talk) 13:26, 31 October 2011 (UTC)

Trying to improve or change #8's language with has been argued before, and there's nothing we can do with it. It's so subjective that its best left as it and let those that want to keep or remove demonstrate the problem. The only clarity to be made is that it is an "if and only if" type scenario - you have to show it does help the reader's comprehension (usually easy) but also show that removing the image is harmful to this (tougher). There's no real metric to these beyond what consensus determines at FFD. --MASEM (t) 13:37, 31 October 2011 (UTC)

Masem's correct. Many efforts have been made to tighten up NFCC #8, to provide better metrics by which to evaluate a given image in a given context. As to this specific case, I think the claim under #8 is weak, but marginally acceptable. --Hammersoft (talk) 13:39, 31 October 2011 (UTC)

And to the specific image: Iraq has no Freedom of Panorama, so there's no way to have a free image of the building - before the destruction, as destroyed or refurished - and as there's no other apparent image to show it, the destroyed version gives much more info than either of the un-destroyed versions. Whether that's #8 worthy or not, I can't say but it is a stronger argument for it. --MASEM (t) 13:45, 31 October 2011 (UTC)

I am actually less concerned with this specific image: I have more of a concern with the project's NFCC handling as a whole. I do some 10c enforcement, but the problem is that 10c is really only the tip of the iceberg. I mean, what is the point of making all non-free files compliant with 10c, when the rationale is incredibly weak. I do not have any specific statistics, but I believe the amount of NFC with very weak rationales is large. Toshio Yamaguchi (talk) 13:54, 31 October 2011 (UTC)

I concur. It's a years long game of Whac-A-Mole. Address one problem, and another NFCC problem crops up. The overarching problem is most editors don't give two rips about the m:Mission, and instead upload images when they feel like it, with no regard to their free/non-free status. If they run into problems, they slap a band-aid on them to avoid scrutiny from the latest attempts by NFCC patrollers to encourage compliance. Giant WP:ABF leap on my part, but I've seen too much of this to naively believe otherwise. --Hammersoft (talk) 14:03, 31 October 2011 (UTC)

Unfortunatley, we're stuck with a very subjective measure of use of non-free in WP, and one that can't be aided by computer since it requires a lot of reading comprehension.
The best you can do is as follows:

If the rational given completely sucks (eg File:Wimpy kid.jpg) and beyond your capacity to fix it, I would tag it with one of the image tags like {{di-disputed fair use rationale}}.
If the rational is at least descriptive but you believe that the use still fails #8, then the only other option, short of resolving the issue with the uploader or talk page of the article it is used on, is taking the image to FFD to get consensus for its removal. --MASEM (t) 14:17, 31 October 2011 (UTC)

Agreed. Resulting problem; there's far, far more people uploading non-free images that shouldn't be here than there are people willing to tag such images. In short, we're screwed. We now have 420k non-free images on the project. --Hammersoft (talk) 14:26, 31 October 2011 (UTC)

These figures are in my opinion a strong indicator that at least the 10c enforcement should be handled by a fully automated bot. I would be willing to monitor that bots talkpage and deal with complaints left there. Otherwise we could as good simply allow as much NFCC as people wish. The whole NFCC business is really such a half-cooked meal. Either we should enforce NFCC effectively using objective criteria or we should drop the enforcement alltogether. This half done thing really isn't helping anybody. Toshio Yamaguchi (talk) 14:38, 31 October 2011 (UTC)

The only way we could be 100% objective with NFC is to disallow its use completely. That's not going to happen, barring any other statement by the Foundation to restrict image use further. That's why its not worth much effort to try to change the core NFCC points, because they reflect a medium that works for most people (enforcement, on the other hand...). We should try to enforce whatever is possible (#10c is easy, as well as #9, and yes, has been done by bots before, but met with enforcement backlash due to a complicated number of issues), but for anything else -- well, fortunately, the folks at WP:GAN and WP:FAC are quick to take up image issues, and its a reasonably strong assurance that an image in an article that passes through FAC will have been its use picked apparent finely by reviewers and fully meets NFCC. There's not many other places for such "gate reviews" beyond that if you find a bad use of an image you can try to do something about it. --MASEM (t) 15:12, 31 October 2011 (UTC)

I would think that the majority of articles containing 10c violations are not GACs or FACs, so most violations are undetectable by these two processes. And I agree that 10c is easy to enforce. It is not the difficulty of the enforcement I am concerned with, but the quantity. And here we are again: an easy task with a large number of cases - the perfect job for a bot. Toshio Yamaguchi (talk) 15:25, 31 October 2011 (UTC)

I don't think anyone's saying that a bot to run the tasks of insuring NFCC compliance with objective measures (#9, #10c) can't be run. However, if it is put in place, I would suspect that the consensus would want this bot to drop warnings to both uploaders and talk pages of articles to warn of the problems, simply to avoid communication issues that BetaCommandBot had when it ran. You'd have to go through a bot approval process, but again, this should be entirely possible. --MASEM (t) 15:32, 31 October 2011 (UTC)

Re: NFCC situation being a half cooked meal; have you read WP:VEGAN? :) Re: A bot to do 10c enforcement; we had that. But really, I think in some ways it's better to not have a bot doing it. If a person can trick a bot into ignoring their image for 10c enforcement, they'll do it. Not so easy against a human. There is a bot doing #9 enforcement, because there's no way to trick that short of licensing an image improperly. Fortunately, I haven't seen much of that (occasionally though). --Hammersoft (talk) 15:35, 31 October 2011 (UTC)

Also, isn't the specific file possibly PD per {{PD-Iraq}}, as a non-artistic photograph? We just need to determine when it was published. -- King of ♥ ♦ ♣ ♠ 08:13, 7 November 2011 (UTC)

Old, large versions of non-free images - keep or revdel?

Suppose a non-free image is uploaded at a not-quite-low resolution. A user downsamples the image and uploads it back to the same filename. Should the older, bigger image be selectively deleted? (The image provoking this question is File:Bookjacket, Luchow's Cookbook.jpg.) —C.Fred (talk) 04:10, 7 November 2011 (UTC)

Most likely yes, selectively delete the old image. In this case , since the same user did it, then there's no need to wait, but if its like the result of a non-free reduction request it might be best to wait a few days to affirm there's no edit warring over it. --MASEM (t) 04:30, 7 November 2011 (UTC)

Any high resolution non-free image still fails WP:NFCC even when a complying low resolution image has been uploaded, so the higher own should be removed. ww2censor (talk) 15:10, 7 November 2011 (UTC)

Well, I'm considering the case where user A uploads a high resolution image with the intent (but not in the rationale) that the high resolution is needed for discussion of the image; user B sees the large image and doesn't realize this use and tags it non-free reduce or does it themselves. User A may wish to revert this, so keeping the image for a brief period to counter good faith edits is appropriate, but obviously can't be kept indefinitely, otherwise, absolutely the high res should be deleted. --MASEM (t) 15:22, 7 November 2011 (UTC)

Fix link please

Wikipedia:Citing sources#Multimedia. Thanks.--♫GoP♫_T ^C _N 15:29, 11 November 2011 (UTC)

RFC: Clarifying policy on pictures of deceased persons

For the purposes of clarification, I am proposing that one of these three actions be taken:

To WP:NFCI (the acceptable use for images section) add the following numbered point: "Pictures of deceased persons, in articles about that person, provided that ever obtaining a free ~~close~~ substitute is not reasonably likely"
To WP:NFC#UUI (the unacceptable use for images section), add the following numbered point: "Pictures of deceased persons, even if no free ~~close~~ substitute can likely ever be obtained. (This does not necessarily apply if the person's physical appearance was extraordinary and a proper subject for commentary in the article, or if other elements in the picture are a proper subject for commentary in the article.)"
Neither, no change.

The use of "close" was struck through on 15:42, 9 November 2011 (UTC), on the basis of general consensus that this is better. This does not materially affect the discussion to any great extent.

The !votes sought here are "support #1" or "support #2" or "support #3". There is no reason for an "oppose" !vote since #3 is no change (unless, I suppose, an editor wished to !vote along the lines of "oppose #3, either #1 or #2 would be OK" in the interests of resolving the matter). Herostratus (talk) 02:35, 6 November 2011 (UTC)

Discussion

Option 1 is pretty much as I understand it the status quo, though we should perhaps flag that such photos should not have commercial value: it is NFCC #1 that is less sharp an issue for dead people, but that doesn't cut into NFCC #2 where such photos have particular commercial value due to their quality, rarity, or simply the fact that they are in somebody's commercial library and they are actively being marketed. Jheald (talk) 02:42, 6 November 2011 (UTC)
- It's not the status quo, see here, which is being used (for example here) as a precedence case for deleting other such photos. If it should be the status quo then we should so state, or if not, then we should state that. Herostratus (talk) 05:46, 6 November 2011 (UTC)
  - We've long held that an image of the subject of the article itself should normally satisfy NFCC #8. Those closures look to be inappropriate. Jheald (talk) 10:13, 6 November 2011 (UTC)
    - Maybe, but in these two cases the images in question were not illustrating the subject of the article in which they were used. CIreland (talk) 11:27, 6 November 2011 (UTC)
      - That's fair, given the article was called Casey Anthony trial at the time. But now the article has reverted to the more typical BLP1E title, Death of Caylee Anthony, I would have thought an image of the subject would be appropriate. Similarly for Murder of Jesse Dirkhising, which was the article title when its image was removed. Jheald (talk) 11:41, 6 November 2011 (UTC)
        Hence the subject of the article was the death and/or trial not the victim. While there will often be a strong case for including a picture of a victim, we shouldn't assume that such would automatically pass NFCC#8 as we would for a biography. CIreland (talk) 11:52, 6 November 2011 (UTC)
        I find it hard to envisage a murder article where it wouldn't. Jheald (talk) 12:19, 6 November 2011 (UTC)
SUPPORT Option 1, to clarify that this really is our policy. --Pete Tillman (talk) 07:17, 6 November 2011 (UTC)
Neither 1 nor 2 in the present form. Version 1 is indeed close to present practice, but not quite: the legitimate interest in showing a likeness of a deceased article subject does not always override other prohibiting factors, most notably that of NFCC#2 in the case of commercial news agency pictures. Fut.Perf. ☼ 07:31, 6 November 2011 (UTC)
Support #1 with any appropriate tweaks to address FPAS's objections. Jclemens (talk) 07:36, 6 November 2011 (UTC)
Oppose #1 while the phrase "free close substitute" is in it. We don't need people saying "no, we should use the non-free image, as the free image is not a close substitute". J Milburn (talk) 12:02, 6 November 2011 (UTC)
Oppose #1 per JMilburn, absolutely no need to put another phrase into NFCC that people can wikilawyer over. If the free version is completely worthless as a substitute it shouldn't be contentious anyway.. Black Kite (t) 12:06, 6 November 2011 (UTC)
support #1 though I can live with a change to address NFCC#2 as needed. Hobit (talk) 03:55, 7 November 2011 (UTC)
Support #1 on the basis that we are an encyclopedia, and all other rules are subservient to that. We must obey copyright legislation, but the encyclopedia is located in the US, where fair use would permit almost all such uses of images in an educational resource like ours. We must obey even the resolutions of the Board, but we are allowed to interpret them. and should interpret them in the way that we think best balances the need for freedom with the need for information. The wording of the title of an article is not the limitation on what the article is actually about, and I interpret "about" to mean, a substantial subject critical and central to the article. DGG ( talk ) 04:21, 7 November 2011 (UTC)
Comment; If we can tweak this to take NFCC#2 into account, and remove the word "close", I think you'd probably have very quick consensus here. Black Kite (t) 07:29, 7 November 2011 (UTC)
Support #1 per DGG. Beyond My Ken (talk) 07:48, 7 November 2011 (UTC)
I support #3. I oppose #1 as written but would support if the word "close" were removed (in which case I would see #1 as merely a clarification of #3).—S Marshall T/C 17:37, 7 November 2011 (UTC)
- I'm largely fine with that suggestion too. I do think that, for example, a free baby picture wouldn't cut it. But I'm not sure how close "close" is. Hobit (talk) 19:11, 7 November 2011 (UTC)
  - Yeah, that's the problem; it's too wikilawyerable. If we have a free image of someone aged 20, and a non-free image of them aged 60 (which is when they were murdered) then that's one thing. But even if we have a fairly poor image of a murder victim that is free, it should usually be preferred. Black Kite (t) 19:22, 7 November 2011 (UTC)
    - I agree. Hobit (talk) 02:52, 8 November 2011 (UTC)
Support #1, which is the status quo ante - the recent deletions of these images are a new and wholly inappropriate phenomenon. Remember the reason we require free images is to encourage the creation of more free images, which is patently impossible when the subject is deceased. --Philosopher ^{Let us reason together.} 02:47, 8 November 2011 (UTC)
- Note: The word "close" should be removed. --Philosopher ^{Let us reason together.} 02:49, 8 November 2011 (UTC)
Comment from RfC initiation: would it be OK to strikethrough the word "close" at this point? It's not usually OK to change proposals while in progress, but given the thrust of commentary here it might be OK I think? At any rate, removing "close" is fine with me and I think that that 1) the person closing the RfC should remove the word "close" and 2) everyone should assume that word "close" won't be in the final text (if it's accepted). (I was only trying to avoid situations where an editor would claim that a picture of person's house or whatever was a "substitute" for a portrait, but apparently this was not a good choice of words.) Herostratus (talk) 07:23, 8 November 2011 (UTC)
- I think it would be fine. Also mark the discussion showing at what point the strike occurred perhaps? Hobit (talk) 20:59, 8 November 2011 (UTC)
  - OK then, I struck through the term at 15:42, 9 November 2011 (UTC). All comments above this point referred to the non-struckthrough version, all the ones below to the struckthrough version. I'll add a double line to emphasize this. Comments above to the effect "OK, but only if the word 'close' is removed" should be considered to have been addressed. Herostratus (talk) 15:42, 9 November 2011 (UTC)

Support #1. (Proposer, but hadn't expressed an opinion before now.) My reasons are twofold:
- Lots of times it's good to avoid having specific rules, but in this case I think it's worthwhile to clarify this and avoid future contention, so I can't support #3. #1 seems to be general usual practice, and it's best when rules codify usual practice, if this can be supported.
- To the question of whether a picture "significantly increase[s] readers' understanding". Humans are attuned to the human face above all other objects in the universe. This is true across all cultures and is seen in newborns. Rightly or wrongly, we believe that the merest variation in features and expression carries meaning. I suppose a lot of this is bogus when it comes to features (e.g. a receding chin supposedly indicating weak character etc.) but no so much when it comes to expression. The forward-thrust chin, the fey slight turning of the head, the grim-set jaw, the furrowed brow, the shy half-smile, the guileless grin. Setting and clothing also. We believe that these things may reveal something of the person, and there's something to that, I think. So yes I guess that pictures that show the person reasonably clearly do meet WP:NFCC#8.

As to WP:NFCC#2 (Non-free content not to usurp items with market value), note that "Acceptable use -- images" (where this text is proposed to be added) says "All non-free images must still meet each non-free content criteria; failure to meet those overrides any acceptable allowance here". So I don't see this as an issue. If use of an item fails to meet WP:NFCC#2 it's out, period, and you never even get to this text. Appending "provided it does not fail the non-free content criteria regarding respect for commercial opportunities" (or just more broadly "provided it does not fail any criteria of the non-free content criteria" is perfectly acceptable to me, but redundant and inelegant. But if it'll get the proposition accepted, then fine. Herostratus (talk) 03:51, 12 November 2011 (UTC)

Oppose #1. This goes too far. The parallel examples in this section (#s 1-7) all deal with subjects under copyright (not simply images), where no free equivalents can exist (absent a release of copyright), and where general rules can be reliably stated. The Wikimedia Foundation's governing resolution makes special provision for use of nonfree content with regard to those subjects. Images of deceased persons involve rather different considerations, and case-by-case analysis retains its particular importance. It is not at all reasonable to presume that images of deceased persons will meet all of the necessary NFCC criteria (while it is reasonable to presume that for most of the examples currently listed); the experiences we've had with disputes over the uses of TV episode screenshots should serve as a caution against including overly broad classes of examples. Fundamentally, it's a very bad idea to change the examples in order to influence discussions of the application of NFCC policy. If the policy itself remains unchanged, no case is made for modifying the policy language, and no evidence is presented of widespread error in its application, then there's no reason for making such a change. If we we have to add the disclaimer proposed, that's a strong signal that the class of images is really not generally acceptable under NFV policy. (And yes, that means I don't think example #8 was well-selected.) Also oppose #2. which goes too far in the opposite direction. Hullaballoo Wolfowitz (talk) 23:03, 17 November 2011 (UTC)
Support #1, Oppose #2, largely per DGG. The needs of the encyclopaedia should be paramount, as that is the most important aim of the project. Supporting and encouraging the creation of free images and other free media is a very important goal too, but it has to be secondary to the needs of the encyclopaedia. For living people the balance between free and non-free images is nearly right to prohibit non-free media in almost all circumstances (I think it needs relaxing very slightly in certain circumstances, but that's a different discussion). For deceased people the creation of free images is not possible, and so we don't need to put a great emphasis on encouraging it. As such we should use non-free media if no free media is know to exist that represents the subject within a reasonable timeperiod of their notability (e.g. if they're notable for something they did as a 60-year old a photo of them in their teens isn't representative (nor vice-versa). Thryduulf (talk) 15:53, 18 November 2011 (UTC)
Neither 1 nor 2 in the present form. as per User:Future Perfect at Sunrise - and oppose 1 as per User:Hullaballoo Wolfowitz - Off2riorob (talk) 10:52, 19 November 2011 (UTC)

Comments

I guess the basic question is: do portraits and similar pictures fail NFCC#8, or not? (When no free substitute is likely possible.) In other words, does a portait "significantly increase readers' understanding" of a biographical article, or not?

In certain cases it probably does: Cyrano de Bergerac for instance (that is the reason for the "physical appearance was extraordinary" clause). Most certainly a photo of an architect that also shows a important destroyed building that is discussed in the article. Arguably a baseball player showing his characteristic stance, or a rock musician showing a characteristic style (Pete Townhshend windmilling for instance), or an inventor shown using his unique apparatus, and so on, providing that these are discussed in the article or reasonably could be (that is the reason for the "other elements in the picture" clause).

But if it's just a portrait or portrait-like picture? I don't know, but it seems like something that there ought to be general guidance on.

It's frustrating to me if editors are in favor of clarifying this but are opposed to the precise wording. @Fut.Perf.: well of course if the image fails other criteria beyond NFCC#8 it's not allowed. @J Milburn: The point of "close substitute" is to precisely to deny Wikilawyering arguments on the line of "It's reasonably likely that crowd shot with him in it might surface" or something. As it stands, NFCC#8 is open to endless contention (Wikilawyering if you will) over whether or not a portrait " "significantly increase[s] readers' understanding". It's not possible short of an overly-long exposition to detail every case of what or is not an acceptable substitute. Perhaps it should say "substitute" instead of "close substitute", but then you have "well, I have a free picture of his house, and since portraits are essentially decorative, that's a reasonable substitute" or whatever. And after all, this is just a change to the "Guideline examples" section which says "These examples are not meant to be exhaustive, and depending on the situation there are exceptions".

The point raised re Jess Dirkhising was "Well, why haven't the pictures of Matthew Shepard and Larry Forbes King and so forth also been deleted"? and with no clear answer given I infer that the answer is "Well, we just haven't gotten to those yet".

One other thing -- I guess "Oppose #1" means "Support #3"; if editors support #2 they should make this clear. Supporting #3 is OK, to leave the answer to question unspecified (or, if you will, vague). The advantage of that is it allows greater leeway for persons to interpret NFCC#8 re portraits of dead people. The disadvantage of that is it allows greater leeway for persons to interpret NFCC#8re portraits of dead people.

I note that the Jess Dirkhising discussion was not heavily populated (this is typical for IFD discussion), and FWIW the sole !vote was "Keep" and the image was deleted anyway. So this leaves a lot of leeway for the closing admin to apply his personal opinion on whether or not portraits of dead people are OK or not, and thus these two conditions seem to appy:

Some are in and some are out, depending basically on chance.
And all are subject to endless discussion on a case-by-case basis even though all the cases are basically similar.

Neither of these is healthy, I don't think. Herostratus (talk) 18:56, 6 November 2011 (UTC)

There's a simple resolution here: On articles of a deceased notable person X, we allow a non-free image of that person if there is no reasonable free equivalent. If it is on any other article that is not solely about that person, then, no, the use of a non-free image is not immediately allowed, though there may be some justification. You're trying to argue the case for using an image of person X on any other article but X. --MASEM (t) 19:11, 6 November 2011 (UTC)

I'm still waiting two days later for an explanation as to why the non-free images of Shepard and King remain. Those two images are not free so why are they still here? User:SchuminWeb nominated the image of Jesse Dirkhising for deletion (it was deleted by User:Fastily without consensus) claiming the murder victim's image wasnt free. Yet Schumin has no issues with the non-free images of Shepard and King? It makes zero sense to me. It appears as a double standard. As long as the victim represents a special interest group their pictures remain. If not, they are deleted. That's how I see it. Otherwise the images would have been deleted long ago. Caden ^cool 21:37, 6 November 2011 (UTC)

You are doing yourself absolutely no favours whatsoever by suggesting that the images of King and Shepard remain because one was gay and one was black. There are hundreds of images of white heterosexual murder victims that haven't been deleted, as a quick trawl through Category:Murder victims would confirm. It would be far better to concentrate on why the Dirkhising one was deleted rather than why others weren't. Black Kite (t) 23:50, 6 November 2011 (UTC)

To a point we can't and shouldn't use WP:OTHERSTUFFEXISTS to complain about why some images are deleted and some are kept. However, given what we would normally say, the image of King should be deleted because the article is about the shooting. The problem with the Shepard article is that the editors have made it a bio focusing on the victim rather than the crime; that article probably fails WP:BLP1E and should be renamed to be on the crime, thus nullifying the picture itself. --MASEM (t) 23:59, 6 November 2011 (UTC)

I'd agree about King. Not so much Shepard; that particular story has spread so much from the original murder that Cultural depictions of Matthew Shepard had to be split out of the article, and there's the Matthew Shepard Foundation and the Matthew Shepard and James Byrd, Jr. Hate Crimes Prevention Act, that I think it would probably be pointless to try to rename it back to the crime. Black Kite (t) 00:17, 7 November 2011 (UTC)

Ok, I'd agree to that point, but argue that the article still is approaching the crime which likely should be a separate article to keep the bio appropriate. (so yes, not a BLP1E issue). --MASEM (t) 00:20, 7 November 2011 (UTC)

The precise choice for name of an article is normally chosen not as much to indicate the subject, but to avoid BLP implications--or BLP-inspired technical notability limitations in the case of the deceased. An article tiled "Murder of X" is about X in every meaningful sense of the word, just as an split from a very large article about X titled "Death of X" would be DGG ( talk ) 04:33, 7 November 2011 (UTC)
Actually, it's rather important. Most of the time, we are using non-free images of deceased persons in the same vein as cover art - for identification without commentary of the image itself. On an article that is strictly about that person, this is putting it in context of discussion of the person itself. If the person is only discussed in context of a larger topic, and the image of that person is not discussed at all, that's failing NFCC#8. I can understand the empathic approach of having images of victims in articles about the crime, but I'm positive that past FFDs have always fallen in favor of removing non-frees of victims that are otherwise non-notable outside of being the victim. (Black Kite points out the Sheppard case above as an exception where the victim became notable after the crime due to the motive and impact). --MASEM (t) 14:47, 7 November 2011 (UTC)

I'll say something; when flicking through our articles for examples today, there's a hell of a lot of non-notable biographies of murder victims out there - have a look at Category:British murder victims for a start. Being murdered doesn't make you notable. Black Kite (t) 19:13, 7 November 2011 (UTC)

George Francis (suspected mobster) .... arrrgh. Black Kite (t) 19:14, 7 November 2011 (UTC)

Whereas ironically Murder of Lesley Whittle is an easily notable case (it was utterly notorious), yet Whittle is a redirect to her murderer. Anyway, this is nothing to do with NFC, so I'll stop there. Black Kite (t) 19:19, 7 November 2011 (UTC)

To an extent there is some NFC overlap here, but its more in the court of BLP/BLP1E. We don't want bios created on murder victims just so that we can have a non-free image of that person per NFCC/NFCI. I have a feeling that if we starting pushing this, people would try to go against BLP/BLP1E to make this happen. There are other reasons besides the NFC aspects, but that's just one point that's showing a larger problem. Still, to that end to the RFC above, this is why the NFCI as given is fine and no clarification is needed, because it's not here that is the problem. --MASEM (t) 20:06, 7 November 2011 (UTC)

If we have a "Murder of X" article, I would think in general an image of X is appropriate, since they are so central to the story, and (per what you've just written) this should usually be the substitute for a separate bio article on them (BLP1E minus the L). Jheald (talk) 20:53, 7 November 2011 (UTC)

There is a very slippery slope that follows that if you say a non-free picture of X on "Murder of X" is appropriate even if X is only simply identified by name, age, and profession, and not given much other discussion, that we should have the same on multiple murders, or serial killer pages, or mass murder events (eg Columbine) for every victim. You can argue that this slope shouldn't go in that direction, but it can and will be argued it you make the allowance that is being asked for above. At somepoint, too, we have to worry about WP:NOTMEMORIAL; unless the person is more than just the victim, showing the photo of the victim without other discussion would seem to fall into line of trying to gain sympathy for victim, what NOTMEMORIAL cautions against. For most crimes, it is not the victims that are of encyclopedic manner beyond the facts that they were the victims; its the resolution of the trials and appeals and following changes in law enforcement or the like that result.
I'm not dismissing that in some very select cases that the murder victim becomes a very recognizable figure even if we don't have a page specifically dedicated to that person. Matthew Sheppard is one such case and I'm sure there are others. But really, I think it's pretty obvious that if a victim X falls into the case of BLP1E and as such can only be discussed in a larger article about the crime, then using a non-free image of that person on that crime article simply isn't appropriate. (This does not nullify any free image use, of course) --MASEM (t) 21:58, 7 November 2011 (UTC)

I'm sorry, but I don't think that's obvious at all. If you were to try to remove the image of Meredith Kercher from Murder of Meredith Kercher you would run into a firestorm. In very many cases a sense of the victim's style and appearance very much helps anchor the crime and add to reader understanding. Jheald (talk) 22:42, 7 November 2011 (UTC)

That's exactly the type of case where we don't need the non-free image of the victim. The victim as a person is described all in about 3 sentences in the first section in the body, and certainly is not notable prior to the event, and this clearly falls into BLP1E territory for justify why there's no article about her. So as for the picture does it aid in any understanding of the event? Nope, once you've identified it as a young adult Caucasian female, that's sufficiently descriptive for the rest of the article which is highly detailed on the events of the crime and trial. Of course, I'm not going to be pointy and do anything about it, but its clear that the picture is not meeting NFCC#8 here. --MASEM (t) 01:21, 8 November 2011 (UTC)

We'd have to agree to disagree on that; I'm very dubious the community would agree with you. Jheald (talk) 02:08, 8 November 2011 (UTC)

What is the notability criterion for murders / murder victims ? (And yes, I've looked at WP:CRIME). You say that there are a "hell of a lot" of non-notable bios; but there is (and long has been) a thriving interest in the true crime book-genre, which may cite very many of these. I see that you have now redirected George Francis (suspected mobster) to Brink's-MAT robbery, but if you're going to do that, shouldn't you transfer over some of the content too? What we've got as it stands now is a redirect to an article which doesn't even mention the subject of the redirect. Jheald (talk) 20:45, 7 November 2011 (UTC)

Likely WP:NEVENT and WP:BLP1E are primary guidance. Eg, I know there are murders in my local area on a weekly basis, but I wouldn't consider most of them even notable for WP. Instead, cases that attract national attention (and moreso beyond tabloid-like coverage) even if they are a local event are the types that lend to notable crimes and/or victims. And for a victim to be notable, there needs to be significant change or influence in law or the like following said even that is directed tied to the victim; otherwise, again, it's just part of the crime. Redirects from victim names to crimes, are, of course, completely acceptable. --MASEM (t) 22:01, 7 November 2011 (UTC)

RFC: Clarifying policy on pictures of deceased persons

For the purposes of clarification, I am proposing that one of these three actions be taken:

To WP:NFCI (the acceptable use for images section) add the following numbered point: "Pictures of deceased persons, in articles about that person, provided that ever obtaining a free ~~close~~ substitute is not reasonably likely"
To WP:NFC#UUI (the unacceptable use for images section), add the following numbered point: "Pictures of deceased persons, even if no free ~~close~~ substitute can likely ever be obtained. (This does not necessarily apply if the person's physical appearance was extraordinary and a proper subject for commentary in the article, or if other elements in the picture are a proper subject for commentary in the article.)"
Neither, no change.

The use of "close" was struck through on 15:42, 9 November 2011 (UTC), on the basis of general consensus that this is better. This does not materially affect the discussion to any great extent.

The !votes sought here are "support #1" or "support #2" or "support #3". There is no reason for an "oppose" !vote since #3 is no change (unless, I suppose, an editor wished to !vote along the lines of "oppose #3, either #1 or #2 would be OK" in the interests of resolving the matter). Herostratus (talk) 02:35, 6 November 2011 (UTC)

Discussion

Option 1 is pretty much as I understand it the status quo, though we should perhaps flag that such photos should not have commercial value: it is NFCC #1 that is less sharp an issue for dead people, but that doesn't cut into NFCC #2 where such photos have particular commercial value due to their quality, rarity, or simply the fact that they are in somebody's commercial library and they are actively being marketed. Jheald (talk) 02:42, 6 November 2011 (UTC)
- It's not the status quo, see here, which is being used (for example here) as a precedence case for deleting other such photos. If it should be the status quo then we should so state, or if not, then we should state that. Herostratus (talk) 05:46, 6 November 2011 (UTC)
  - We've long held that an image of the subject of the article itself should normally satisfy NFCC #8. Those closures look to be inappropriate. Jheald (talk) 10:13, 6 November 2011 (UTC)
    - Maybe, but in these two cases the images in question were not illustrating the subject of the article in which they were used. CIreland (talk) 11:27, 6 November 2011 (UTC)
      - That's fair, given the article was called Casey Anthony trial at the time. But now the article has reverted to the more typical BLP1E title, Death of Caylee Anthony, I would have thought an image of the subject would be appropriate. Similarly for Murder of Jesse Dirkhising, which was the article title when its image was removed. Jheald (talk) 11:41, 6 November 2011 (UTC)
        Hence the subject of the article was the death and/or trial not the victim. While there will often be a strong case for including a picture of a victim, we shouldn't assume that such would automatically pass NFCC#8 as we would for a biography. CIreland (talk) 11:52, 6 November 2011 (UTC)
        I find it hard to envisage a murder article where it wouldn't. Jheald (talk) 12:19, 6 November 2011 (UTC)
SUPPORT Option 1, to clarify that this really is our policy. --Pete Tillman (talk) 07:17, 6 November 2011 (UTC)
Neither 1 nor 2 in the present form. Version 1 is indeed close to present practice, but not quite: the legitimate interest in showing a likeness of a deceased article subject does not always override other prohibiting factors, most notably that of NFCC#2 in the case of commercial news agency pictures. Fut.Perf. ☼ 07:31, 6 November 2011 (UTC)
Support #1 with any appropriate tweaks to address FPAS's objections. Jclemens (talk) 07:36, 6 November 2011 (UTC)
Oppose #1 while the phrase "free close substitute" is in it. We don't need people saying "no, we should use the non-free image, as the free image is not a close substitute". J Milburn (talk) 12:02, 6 November 2011 (UTC)
Oppose #1 per JMilburn, absolutely no need to put another phrase into NFCC that people can wikilawyer over. If the free version is completely worthless as a substitute it shouldn't be contentious anyway.. Black Kite (t) 12:06, 6 November 2011 (UTC)
support #1 though I can live with a change to address NFCC#2 as needed. Hobit (talk) 03:55, 7 November 2011 (UTC)
Support #1 on the basis that we are an encyclopedia, and all other rules are subservient to that. We must obey copyright legislation, but the encyclopedia is located in the US, where fair use would permit almost all such uses of images in an educational resource like ours. We must obey even the resolutions of the Board, but we are allowed to interpret them. and should interpret them in the way that we think best balances the need for freedom with the need for information. The wording of the title of an article is not the limitation on what the article is actually about, and I interpret "about" to mean, a substantial subject critical and central to the article. DGG ( talk ) 04:21, 7 November 2011 (UTC)
Comment; If we can tweak this to take NFCC#2 into account, and remove the word "close", I think you'd probably have very quick consensus here. Black Kite (t) 07:29, 7 November 2011 (UTC)
Support #1 per DGG. Beyond My Ken (talk) 07:48, 7 November 2011 (UTC)
I support #3. I oppose #1 as written but would support if the word "close" were removed (in which case I would see #1 as merely a clarification of #3).—S Marshall T/C 17:37, 7 November 2011 (UTC)
- I'm largely fine with that suggestion too. I do think that, for example, a free baby picture wouldn't cut it. But I'm not sure how close "close" is. Hobit (talk) 19:11, 7 November 2011 (UTC)
  - Yeah, that's the problem; it's too wikilawyerable. If we have a free image of someone aged 20, and a non-free image of them aged 60 (which is when they were murdered) then that's one thing. But even if we have a fairly poor image of a murder victim that is free, it should usually be preferred. Black Kite (t) 19:22, 7 November 2011 (UTC)
    - I agree. Hobit (talk) 02:52, 8 November 2011 (UTC)
Support #1, which is the status quo ante - the recent deletions of these images are a new and wholly inappropriate phenomenon. Remember the reason we require free images is to encourage the creation of more free images, which is patently impossible when the subject is deceased. --Philosopher ^{Let us reason together.} 02:47, 8 November 2011 (UTC)
- Note: The word "close" should be removed. --Philosopher ^{Let us reason together.} 02:49, 8 November 2011 (UTC)
Comment from RfC initiation: would it be OK to strikethrough the word "close" at this point? It's not usually OK to change proposals while in progress, but given the thrust of commentary here it might be OK I think? At any rate, removing "close" is fine with me and I think that that 1) the person closing the RfC should remove the word "close" and 2) everyone should assume that word "close" won't be in the final text (if it's accepted). (I was only trying to avoid situations where an editor would claim that a picture of person's house or whatever was a "substitute" for a portrait, but apparently this was not a good choice of words.) Herostratus (talk) 07:23, 8 November 2011 (UTC)
- I think it would be fine. Also mark the discussion showing at what point the strike occurred perhaps? Hobit (talk) 20:59, 8 November 2011 (UTC)
  - OK then, I struck through the term at 15:42, 9 November 2011 (UTC). All comments above this point referred to the non-struckthrough version, all the ones below to the struckthrough version. I'll add a double line to emphasize this. Comments above to the effect "OK, but only if the word 'close' is removed" should be considered to have been addressed. Herostratus (talk) 15:42, 9 November 2011 (UTC)

Support #1. (Proposer, but hadn't expressed an opinion before now.) My reasons are twofold:
- Lots of times it's good to avoid having specific rules, but in this case I think it's worthwhile to clarify this and avoid future contention, so I can't support #3. #1 seems to be general usual practice, and it's best when rules codify usual practice, if this can be supported.
- To the question of whether a picture "significantly increase[s] readers' understanding". Humans are attuned to the human face above all other objects in the universe. This is true across all cultures and is seen in newborns. Rightly or wrongly, we believe that the merest variation in features and expression carries meaning. I suppose a lot of this is bogus when it comes to features (e.g. a receding chin supposedly indicating weak character etc.) but no so much when it comes to expression. The forward-thrust chin, the fey slight turning of the head, the grim-set jaw, the furrowed brow, the shy half-smile, the guileless grin. Setting and clothing also. We believe that these things may reveal something of the person, and there's something to that, I think. So yes I guess that pictures that show the person reasonably clearly do meet WP:NFCC#8.

As to WP:NFCC#2 (Non-free content not to usurp items with market value), note that "Acceptable use -- images" (where this text is proposed to be added) says "All non-free images must still meet each non-free content criteria; failure to meet those overrides any acceptable allowance here". So I don't see this as an issue. If use of an item fails to meet WP:NFCC#2 it's out, period, and you never even get to this text. Appending "provided it does not fail the non-free content criteria regarding respect for commercial opportunities" (or just more broadly "provided it does not fail any criteria of the non-free content criteria" is perfectly acceptable to me, but redundant and inelegant. But if it'll get the proposition accepted, then fine. Herostratus (talk) 03:51, 12 November 2011 (UTC)

Oppose #1. This goes too far. The parallel examples in this section (#s 1-7) all deal with subjects under copyright (not simply images), where no free equivalents can exist (absent a release of copyright), and where general rules can be reliably stated. The Wikimedia Foundation's governing resolution makes special provision for use of nonfree content with regard to those subjects. Images of deceased persons involve rather different considerations, and case-by-case analysis retains its particular importance. It is not at all reasonable to presume that images of deceased persons will meet all of the necessary NFCC criteria (while it is reasonable to presume that for most of the examples currently listed); the experiences we've had with disputes over the uses of TV episode screenshots should serve as a caution against including overly broad classes of examples. Fundamentally, it's a very bad idea to change the examples in order to influence discussions of the application of NFCC policy. If the policy itself remains unchanged, no case is made for modifying the policy language, and no evidence is presented of widespread error in its application, then there's no reason for making such a change. If we we have to add the disclaimer proposed, that's a strong signal that the class of images is really not generally acceptable under NFV policy. (And yes, that means I don't think example #8 was well-selected.) Also oppose #2. which goes too far in the opposite direction. Hullaballoo Wolfowitz (talk) 23:03, 17 November 2011 (UTC)
Support #1, Oppose #2, largely per DGG. The needs of the encyclopaedia should be paramount, as that is the most important aim of the project. Supporting and encouraging the creation of free images and other free media is a very important goal too, but it has to be secondary to the needs of the encyclopaedia. For living people the balance between free and non-free images is nearly right to prohibit non-free media in almost all circumstances (I think it needs relaxing very slightly in certain circumstances, but that's a different discussion). For deceased people the creation of free images is not possible, and so we don't need to put a great emphasis on encouraging it. As such we should use non-free media if no free media is know to exist that represents the subject within a reasonable timeperiod of their notability (e.g. if they're notable for something they did as a 60-year old a photo of them in their teens isn't representative (nor vice-versa). Thryduulf (talk) 15:53, 18 November 2011 (UTC)
Neither 1 nor 2 in the present form. as per User:Future Perfect at Sunrise - and oppose 1 as per User:Hullaballoo Wolfowitz - Off2riorob (talk) 10:52, 19 November 2011 (UTC)

Comments

I guess the basic question is: do portraits and similar pictures fail NFCC#8, or not? (When no free substitute is likely possible.) In other words, does a portait "significantly increase readers' understanding" of a biographical article, or not?

In certain cases it probably does: Cyrano de Bergerac for instance (that is the reason for the "physical appearance was extraordinary" clause). Most certainly a photo of an architect that also shows a important destroyed building that is discussed in the article. Arguably a baseball player showing his characteristic stance, or a rock musician showing a characteristic style (Pete Townhshend windmilling for instance), or an inventor shown using his unique apparatus, and so on, providing that these are discussed in the article or reasonably could be (that is the reason for the "other elements in the picture" clause).

But if it's just a portrait or portrait-like picture? I don't know, but it seems like something that there ought to be general guidance on.

It's frustrating to me if editors are in favor of clarifying this but are opposed to the precise wording. @Fut.Perf.: well of course if the image fails other criteria beyond NFCC#8 it's not allowed. @J Milburn: The point of "close substitute" is to precisely to deny Wikilawyering arguments on the line of "It's reasonably likely that crowd shot with him in it might surface" or something. As it stands, NFCC#8 is open to endless contention (Wikilawyering if you will) over whether or not a portrait " "significantly increase[s] readers' understanding". It's not possible short of an overly-long exposition to detail every case of what or is not an acceptable substitute. Perhaps it should say "substitute" instead of "close substitute", but then you have "well, I have a free picture of his house, and since portraits are essentially decorative, that's a reasonable substitute" or whatever. And after all, this is just a change to the "Guideline examples" section which says "These examples are not meant to be exhaustive, and depending on the situation there are exceptions".

The point raised re Jess Dirkhising was "Well, why haven't the pictures of Matthew Shepard and Larry Forbes King and so forth also been deleted"? and with no clear answer given I infer that the answer is "Well, we just haven't gotten to those yet".

One other thing -- I guess "Oppose #1" means "Support #3"; if editors support #2 they should make this clear. Supporting #3 is OK, to leave the answer to question unspecified (or, if you will, vague). The advantage of that is it allows greater leeway for persons to interpret NFCC#8 re portraits of dead people. The disadvantage of that is it allows greater leeway for persons to interpret NFCC#8re portraits of dead people.

I note that the Jess Dirkhising discussion was not heavily populated (this is typical for IFD discussion), and FWIW the sole !vote was "Keep" and the image was deleted anyway. So this leaves a lot of leeway for the closing admin to apply his personal opinion on whether or not portraits of dead people are OK or not, and thus these two conditions seem to appy:

Some are in and some are out, depending basically on chance.
And all are subject to endless discussion on a case-by-case basis even though all the cases are basically similar.

Neither of these is healthy, I don't think. Herostratus (talk) 18:56, 6 November 2011 (UTC)

There's a simple resolution here: On articles of a deceased notable person X, we allow a non-free image of that person if there is no reasonable free equivalent. If it is on any other article that is not solely about that person, then, no, the use of a non-free image is not immediately allowed, though there may be some justification. You're trying to argue the case for using an image of person X on any other article but X. --MASEM (t) 19:11, 6 November 2011 (UTC)

I'm still waiting two days later for an explanation as to why the non-free images of Shepard and King remain. Those two images are not free so why are they still here? User:SchuminWeb nominated the image of Jesse Dirkhising for deletion (it was deleted by User:Fastily without consensus) claiming the murder victim's image wasnt free. Yet Schumin has no issues with the non-free images of Shepard and King? It makes zero sense to me. It appears as a double standard. As long as the victim represents a special interest group their pictures remain. If not, they are deleted. That's how I see it. Otherwise the images would have been deleted long ago. Caden ^cool 21:37, 6 November 2011 (UTC)

You are doing yourself absolutely no favours whatsoever by suggesting that the images of King and Shepard remain because one was gay and one was black. There are hundreds of images of white heterosexual murder victims that haven't been deleted, as a quick trawl through Category:Murder victims would confirm. It would be far better to concentrate on why the Dirkhising one was deleted rather than why others weren't. Black Kite (t) 23:50, 6 November 2011 (UTC)

To a point we can't and shouldn't use WP:OTHERSTUFFEXISTS to complain about why some images are deleted and some are kept. However, given what we would normally say, the image of King should be deleted because the article is about the shooting. The problem with the Shepard article is that the editors have made it a bio focusing on the victim rather than the crime; that article probably fails WP:BLP1E and should be renamed to be on the crime, thus nullifying the picture itself. --MASEM (t) 23:59, 6 November 2011 (UTC)

I'd agree about King. Not so much Shepard; that particular story has spread so much from the original murder that Cultural depictions of Matthew Shepard had to be split out of the article, and there's the Matthew Shepard Foundation and the Matthew Shepard and James Byrd, Jr. Hate Crimes Prevention Act, that I think it would probably be pointless to try to rename it back to the crime. Black Kite (t) 00:17, 7 November 2011 (UTC)

Ok, I'd agree to that point, but argue that the article still is approaching the crime which likely should be a separate article to keep the bio appropriate. (so yes, not a BLP1E issue). --MASEM (t) 00:20, 7 November 2011 (UTC)

The precise choice for name of an article is normally chosen not as much to indicate the subject, but to avoid BLP implications--or BLP-inspired technical notability limitations in the case of the deceased. An article tiled "Murder of X" is about X in every meaningful sense of the word, just as an split from a very large article about X titled "Death of X" would be DGG ( talk ) 04:33, 7 November 2011 (UTC)
Actually, it's rather important. Most of the time, we are using non-free images of deceased persons in the same vein as cover art - for identification without commentary of the image itself. On an article that is strictly about that person, this is putting it in context of discussion of the person itself. If the person is only discussed in context of a larger topic, and the image of that person is not discussed at all, that's failing NFCC#8. I can understand the empathic approach of having images of victims in articles about the crime, but I'm positive that past FFDs have always fallen in favor of removing non-frees of victims that are otherwise non-notable outside of being the victim. (Black Kite points out the Sheppard case above as an exception where the victim became notable after the crime due to the motive and impact). --MASEM (t) 14:47, 7 November 2011 (UTC)

I'll say something; when flicking through our articles for examples today, there's a hell of a lot of non-notable biographies of murder victims out there - have a look at Category:British murder victims for a start. Being murdered doesn't make you notable. Black Kite (t) 19:13, 7 November 2011 (UTC)

George Francis (suspected mobster) .... arrrgh. Black Kite (t) 19:14, 7 November 2011 (UTC)

Whereas ironically Murder of Lesley Whittle is an easily notable case (it was utterly notorious), yet Whittle is a redirect to her murderer. Anyway, this is nothing to do with NFC, so I'll stop there. Black Kite (t) 19:19, 7 November 2011 (UTC)

To an extent there is some NFC overlap here, but its more in the court of BLP/BLP1E. We don't want bios created on murder victims just so that we can have a non-free image of that person per NFCC/NFCI. I have a feeling that if we starting pushing this, people would try to go against BLP/BLP1E to make this happen. There are other reasons besides the NFC aspects, but that's just one point that's showing a larger problem. Still, to that end to the RFC above, this is why the NFCI as given is fine and no clarification is needed, because it's not here that is the problem. --MASEM (t) 20:06, 7 November 2011 (UTC)

If we have a "Murder of X" article, I would think in general an image of X is appropriate, since they are so central to the story, and (per what you've just written) this should usually be the substitute for a separate bio article on them (BLP1E minus the L). Jheald (talk) 20:53, 7 November 2011 (UTC)

There is a very slippery slope that follows that if you say a non-free picture of X on "Murder of X" is appropriate even if X is only simply identified by name, age, and profession, and not given much other discussion, that we should have the same on multiple murders, or serial killer pages, or mass murder events (eg Columbine) for every victim. You can argue that this slope shouldn't go in that direction, but it can and will be argued it you make the allowance that is being asked for above. At somepoint, too, we have to worry about WP:NOTMEMORIAL; unless the person is more than just the victim, showing the photo of the victim without other discussion would seem to fall into line of trying to gain sympathy for victim, what NOTMEMORIAL cautions against. For most crimes, it is not the victims that are of encyclopedic manner beyond the facts that they were the victims; its the resolution of the trials and appeals and following changes in law enforcement or the like that result.
I'm not dismissing that in some very select cases that the murder victim becomes a very recognizable figure even if we don't have a page specifically dedicated to that person. Matthew Sheppard is one such case and I'm sure there are others. But really, I think it's pretty obvious that if a victim X falls into the case of BLP1E and as such can only be discussed in a larger article about the crime, then using a non-free image of that person on that crime article simply isn't appropriate. (This does not nullify any free image use, of course) --MASEM (t) 21:58, 7 November 2011 (UTC)

I'm sorry, but I don't think that's obvious at all. If you were to try to remove the image of Meredith Kercher from Murder of Meredith Kercher you would run into a firestorm. In very many cases a sense of the victim's style and appearance very much helps anchor the crime and add to reader understanding. Jheald (talk) 22:42, 7 November 2011 (UTC)

That's exactly the type of case where we don't need the non-free image of the victim. The victim as a person is described all in about 3 sentences in the first section in the body, and certainly is not notable prior to the event, and this clearly falls into BLP1E territory for justify why there's no article about her. So as for the picture does it aid in any understanding of the event? Nope, once you've identified it as a young adult Caucasian female, that's sufficiently descriptive for the rest of the article which is highly detailed on the events of the crime and trial. Of course, I'm not going to be pointy and do anything about it, but its clear that the picture is not meeting NFCC#8 here. --MASEM (t) 01:21, 8 November 2011 (UTC)

We'd have to agree to disagree on that; I'm very dubious the community would agree with you. Jheald (talk) 02:08, 8 November 2011 (UTC)

What is the notability criterion for murders / murder victims ? (And yes, I've looked at WP:CRIME). You say that there are a "hell of a lot" of non-notable bios; but there is (and long has been) a thriving interest in the true crime book-genre, which may cite very many of these. I see that you have now redirected George Francis (suspected mobster) to Brink's-MAT robbery, but if you're going to do that, shouldn't you transfer over some of the content too? What we've got as it stands now is a redirect to an article which doesn't even mention the subject of the redirect. Jheald (talk) 20:45, 7 November 2011 (UTC)

Likely WP:NEVENT and WP:BLP1E are primary guidance. Eg, I know there are murders in my local area on a weekly basis, but I wouldn't consider most of them even notable for WP. Instead, cases that attract national attention (and moreso beyond tabloid-like coverage) even if they are a local event are the types that lend to notable crimes and/or victims. And for a victim to be notable, there needs to be significant change or influence in law or the like following said even that is directed tied to the victim; otherwise, again, it's just part of the crime. Redirects from victim names to crimes, are, of course, completely acceptable. --MASEM (t) 22:01, 7 November 2011 (UTC)

Notification of removing rationales

User:Hammersoft removed the fair use rationale for the use of file:Kubica crash.jpg in 2007 Canadian Grand Prix, without placing any indication he had done so at the article talk page. I posted on his talk page about this, and his response was basically that it is not required to do this.

While the removal of the rationale may or may not have been correct or the best way of doing it, that isn't the issue. The file was used in that article, with what the editors there had every reason to believe was a valid rationale, for four years until that rationale was removed with no notice. With no rationale, the image was (correctly) subject to removal from the article at any time for not meeting the NFCC. In this specific case, there is a very strong case for using the picture, if a correct rationale is provided.

My thinking is that if, for any reason, a fair use rationale is edited or changed in such a way that its use on an article is no longer covered by a rationale, that a message should be posted to the talk page of the relevant article informing editors there of this. This need not be more than a template with a link to the image, an optional reason for editing/removing the rationale, a link to instructions about FURs and a link to a good place to discuss it if they want to.

I don't think this requirement would be unreasonable or unduly onerous on NFCC patrollers, and in the best interests of openness, encouraging complience with the NFCC and avoiding acrimony and drama down the line. Thryduulf (talk) 15:20, 16 November 2011 (UTC)

A rationale as here violates WP:NFCC Policy 10c. If you want to include the image in a number of articles, it is your responsibility to ensure the file has a separate rationale for each article where the file is used (see WP:NFCC#Enforcement). Toshio Yamaguchi (talk) 15:31, 16 November 2011 (UTC)

Yes, but that's not my point. Where there is a rationale for use in an article (whether that rationale is valid or not), and that rationale is removed or changed for any reason (rightly or wrongly), then a note should be placed on the talk page of the relevant article informing the users there that there is no longer a rationale for the use of that image in the article. Thryduulf (talk) 15:41, 16 November 2011 (UTC)

(1) My removing the group rationale was correct, as borne out by prior discussion regarding blanket rationales. (2) The editors did not have every reason to believe it was a valid rationale, as the rationale explicitly violated WP:NFCC #10c which requires a separate, specific rationale for each use. (3) As I noted [4] at WP:NFCR, there isn't a case for using the picture at all, at least not as supported by secondary sources. (4) The requirement being requested would double the work load of an NFCC patroller removing group rationales. (6) There's been no acrimony in the dozens of group rationale removals I've done since February of this year until today, when you accused me [5] of being lazy. I assert the source of the acrimony isn't the group rationales, but your decision to become uncivil in approaching me about this issue. Thank you, --Hammersoft (talk) 15:52, 16 November 2011 (UTC)

"The file was used in that article, with what the editors there had every reason to believe was a valid rationale, for four years until that rationale was removed with no notice." Had the editors of the article actually taken the time to familiarize themselves with WP:NFCC#Policy, they would have known the rationale was not valid and could have fixed it, but obviously that did not happen. Toshio Yamaguchi (talk) 16:04, 16 November 2011 (UTC)

By the way; a related suggestion to add to NFCC policy (under enforcement) to say "A file may not be removed for failing NFCC unless there has been prior notice of the issue on the relevant article talk page for 7 days" failed this past July. See failed proposal. --Hammersoft (talk) 16:11, 16 November 2011 (UTC)

Without having yet read that proposal, I have to say that surprises me. As for your other point, why should every editor of an article be intimately familiar with the NFCC requirements? I would expect the person uploading the image and anyone adding to the article to familiarise themselves enough to do their best at getting the rationale correct, but beyond that the way Wikipedia normally works in my long experience is that if someone finds a problem they either fix it themselves or, and/or post about it somewhere so that it can be fixed by someone who is able to.

I honestly don't understand why simple acts of communication seem to be so opposed by NFCC folk. I'm not speaking about anyone specifically here, but in almost all my dealings with NFCC enforcement I'm left with the distinct impression of an attitude of "Everybody should be intimately familiar with the entirety of all the rules, policies and guidelines relating to non-free content, and if you can't be bothered to learn this then why on earth should we bother talking to you at all?" I also get a sense of complete bafflement that anyone (other than possibly the uploaded, maybe) might want to know about potential problems with an image. Thryduulf (talk) 17:07, 16 November 2011 (UTC)

"Why should every editor of an article be intimately familiar with the NFCC requirements?" Nobody says all editors of an article must be familiar with NFCC policy, but an editor adding a non-free file to an article has the responsibility to make sure the file is compliant with all of the criteria listed there as otherwise it is likely to result in the files removal. Toshio Yamaguchi (talk) 18:06, 16 November 2011 (UTC)

It's not really in the uploader's court four years later. This obsession with only notifying the uploader, and holding the uploader "responsible," doesn't make sense beyond the initial moment when the image is first uploaded, nor is it consistent with WP:OWN. If your notification quota only allows for one notice per image, I think we'd be better off giving article talk page notices and not even bothering to notify the uploader in such stale cases. And an article's current editors are in the best position to understand the informational value of the image to the article and so either fix the NFUR if it can be, or find a replacement that can be compliant. postdlf (talk) 18:11, 16 November 2011 (UTC)

(edit conflict)Yes they should, but NFCC is not the simplest thing about Wikipedia and historically the enforcement of it has got more stringent over time. So some invalid rationales have (and maybe still do) slip through the net and get used in articles in good faith. The problem can be easily rectified if only people know about it, so why not tell them about it? Thryduulf (talk) 18:15, 16 November 2011 (UTC)

@Thryduulf; As noted in the failed proposal linked above, we had a bot that used to do notifications of missing rationales for articles where an image was used. The result was little activity to fix the problems. The notifications don't work. As to perceived attitudes of NFCC enforcers, I understand that perception. I really do. I don't hold that attitude, though it might come across that way. The simple fact is there are more than 420,000 non-free images on this project. There's more than a thousand added every week. There's more than 8,000 uses of non-free files for which there is no accompanying rationale. There's roughly 3,000 non-free files that are completely missing rationales. In contrast, the number of people who regularly do NFCC enforcement work can be counted on two hands. Yet, these people are routinely attacked and asked to do MORE work in order to conduct NFCC enforcement. You yourself did this; you attacked me as lazy, and insisted that out of common courtesy I should have posted notices in three different places [6] when I removed a group rationale. I think you can see the problem. --Hammersoft (talk) 18:25, 16 November 2011 (UTC)

I'd rather have you process fewer images per day and give more complete notices. It's not a race, and you're not a Wikimedia employee with a supervisor breathing down your neck, so you can't use those numbers to justify your own voluntary choices. NFCC needs to be a collaborative environment just like the rest of Wikipedia, which means communication needs to be valued, not just some bottom line of how many images you can remove. As a boss of mine used to say at a real paying job I had, "you should never be in too much hurry to do the job right." postdlf (talk) 18:29, 16 November 2011 (UTC)

I'd be happy to do so, as I've previously noted, if such was required by policy or guideline. Every proposal to date to have something like this has failed. "Doing the job right" is defined by policy and guideline here. I am doing the job right. I'm sorry your personal opinion of how I should be doing the job is not equivalent to that of policy and guideline. If you can generate a good enough argument to get policy and guideline changed to reflect your opinion, I'd be more than happy to abide by it. --Hammersoft (talk) 18:33, 16 November 2011 (UTC)

As long as some problems were fixed the notifications were worth it. If your using any sort of automated process, the notification should take all of a couple of seconds per image - not at all significant. It will also reduce the amount of time you have to spend in discussions like this. I suspect as well that more and better communication would make more people aware of the work you are doing and so encourage more people to help you. As Postdlf says above, there is no deadline, and no requirement for you to put yourselves under so much pressure - artificial time constraints are no excuse. Thryduulf (talk) 18:42, 16 November 2011 (UTC)

As I noted, I'll be happy to provide (the ineffective, as proven by response to prior notices) the notices if policy and/or guideline were modified to require such. I do not use any automated process, nor should I be required to do so in order to do this work. I am a volunteer, just like you. I'm happy to spend the time in discussions like this, so that others can learn the NFCC issues and understand them. I feel this helps the project. I welcome such discussions, and don't consider them a waste of time. I don't feel myself under any particular pressure, nor did I suggest that I was. I highlighted the problem with NFCC enforcement, to counter your argument that NFCC patrollers should be doing even more work to 'help' (when it's been shown that such notices don't help) the project. So to summarize; you're asking me to double my work when removing group rationales, insisting I do something that we already know does not work, calling me lazy for not doing it, and still finding fault with me because I see no reason to do it. I think at this point the best we can hope for is to agree to disagree. --Hammersoft (talk) 18:45, 16 November 2011 (UTC)

@Thryduulf This "no deadline" argument is completely misplaced here. To be more precise, there should not be even one non-free file NOT compliant with NFCC Policy, as all files MUST be compliant. Toshio Yamaguchi (talk) 18:50, 16 November 2011 (UTC)

I agree there should be no files that do not comply. However I completely disagree that it's not worth taking the small extra bit of time to do things in a away that encourages civility and good communication. We're coming at this from completely different perspectives and I'm just not understanding your reasoning and it doesn't look like you're understanding why I'm taking the position I am. It's increasingly feeling like the entire NFCC system is broken in some key way if there is just so much work involved that simple things that are so much a part of other aspects of Wikipedia just cannot be done here. Thryduulf (talk) 19:48, 16 November 2011 (UTC)

Encourages civility? Such as calling me lazy perhaps? I concur the entire NFCC system is broken. I could go one for a while about this. There are a dizzying array of factors to be concerned with in that regard. For example, the incredible disconnect between our being a free content resource juxtaposed with us being one of if not the largest repository of non-free content in the world. The mere fact that we permit non-free content means we will always be dancing around a line in the sand that moves every time the wind blows (read: someone comments about NFCC issues). There is no ready way of fixing that. I've said before on my userpage that I feel the only way to stop the never ending NFCC disputes is to permit non-free content up to the maximum level allowed by U.S. fair use law. Yes, the people opposed to non-free content would not be happy, but it would be a firm line in the sand with not much in the way of questionable content, especially since we have a liberal hand being an educational resource. Of course, this will never happen. Equally impossible is banning non-free content, as non-free content is culturally entrenched here (contrast the German wikipedia, among other languages too that do not allow non-free content at all). Proof of that is in the 420k+ non-free images we have here. So, the polar ends of the problem are impossible to attain. Therefore, what we're left with is the middle ground, which will always (and I do mean always) be contentious. It is what it is. --Hammersoft (talk) 20:31, 16 November 2011 (UTC)

There is no policy against group rationales, as there is no requirement for a separate handcrafted exercise in drafting prose for each image use. That old notion was never accepted, and more or less died several years ago. If an editor separately considers two image uses and the exact same description fits both, they can use the exact same description for both. That's dubious here, as it looks like this particular rationale is inadequate for both... What is important, though, is that any bots or automated tools dealing with images are going to get confused if you try to stuff two different articles in the same field of the template. So it's reasonable to insist on either one article per template, or else switch to a different template that doesn't mess up the system. There's no particular requirement to inform editors every time you alter their contributions - as pointed out above that's not in the spirit of WP:OWN and that's what watch lists are for. Hammersoft edited a single image that day so it isn't a big deal, but simply blanking informational content because you don't agree with the way it's organized isn't terribly helpful. - Wikidemon (talk) 20:47, 16 November 2011 (UTC)

NFCC#10c: The name of each article (a link to each article is also recommended) in which fair use is claimed for the item, and a separate, specific non-free use rationale for each use of the item, as explained at Wikipedia:Non-free use rationale guideline. The rationale is presented in clear, plain language and is relevant to each use. has been in place for years. It is policy. Ergo: group rationales are not allowed. Technically, we do allow a copy-and-paste duplication (save for article name) for repeated, near-equivalent uses, and tolerate to some extent cookie-cutter rationales for things like logos. --MASEM (t) 21:05, 16 November 2011 (UTC)

Let's get back to reality for a minute. Here we have an attempt to create a rationale for two images that is both inadequately vague, and also in a form that is technically (but not as a matter of policy, as policy is not about syntactic technicalities) inadequate. The best response would be to fix the rationale for both if they can be fixed, as we are trying to improve an encyclopedia rather than to run around excising things based on technicalities. It's also an acceptable response to question or challenge both, if it looks like the rationale is questionable. Deleting one and keeping the other because they're not supposed to be in the same template is, again, unhelpful. But also again, it's just one image and that's not worth a lot of discussion. If this were a widespread problem or yet another salvo in a mass deletion campaign, that would be something else. - Wikidemon (talk) 21:21, 16 November 2011 (UTC)

I never said that Hammersoft was in the right to delete. Just that he is correct that separate rationales are needed. I'm still trying to compose my thoughts on the larger issue which is not in alignment with what Hammersoft's actually doing. There needs to be a better way to handle NFCC#8 issues when they compound with other NFCC problems, since NFCC#8 is very objective. --MASEM (t) 21:35, 16 November 2011 (UTC)

@Wikidemon; Thank you for characterizing my edits as running around excising things based on technicalities. It is approbation like this that makes my editing comfortable here. Further, implying that this is some effort on my part to launch "another salvo in a mass deletion campaign" you are exceptionally off base. As to your other points, improving the encyclopedia comes in many forms. Sometimes it's by deletion. Sometimes it's by addition. I removed the group rationale, leaving the only rationale in existence intact. I did not write a second rationale in part because I am not required to do so and in part because I do not feel that the non-free use is justified in any case where the image is used. I am not going to write a rationale to support the usage of something that I don't think should be used in the first place. That's ridiculous, and insisting I should have done so is flat wrong. I am aware the image is under review at WP:NFCR, and I did not remove it from any use while that discussion is pending, nor did I contest Sarek adding [7] a new rationale to justify one of the uses. I note that you take me to task for not fixing the problem when you think I should have, but you didn't take Sarek to task for not supplying the required-by-policy rationale for the third use, where there is still no rationale for the use. When do you plan on taking him to task for not complying with policy and not fixing the problem as you say I should have done? --Hammersoft (talk) 23:19, 16 November 2011 (UTC)

Hammer, you do often come across as rather inflexible and didactic on NFCC enforcement. I put this down to 1) you having been involved in the area for so long and having seen every possible counter-argument and personal attack possible while doing it; and 2) you are completely right that no existing policy or guideline says you have to go fix things yourself. Also, your views on NFCC are rather more nuanced than they first appear, I've learned that over time. Myself, I nuke text copyvios all the time and yeah, the option is mentioned that I could also rewrite the material myself. Sometimes I do, but much more often I'm simply not familiar with the town the enthusiastic new editor has copied a blurb from the tourist website to flesh out an article - so I remove the material. But there is a valid point here about where best to place a notification to let other active editors of an article know there is a problem, which they can either fix or ignore. I personally do read the various (mostly bot) notificatios I see in articles on my watchlist, and almost always let them go as I can't really conjure up a FUR, i.e. it's probably not a justifiable use. The big problem I see is that when your (justifiable under policy, or even questionable) image use removals orphan an image, it starts the 7-day clock to that image disappearing altogether. If that image could have a valid FUR and it gets deleted, then that is a loss to the encyclopedia. So is there a middle way here? Twinkle seems to help RC patrollers double their edit count by placing warning templates onto user pages for every instance of detected vandalism. Could/would you use such a facility if it was extended to NF image removals? For instance, a tick box to generate an article talk page notice as well as user page notice? I'm not suggesting you switch over to automated tools (which I've never used in my wiki-life) and I'm not suggesting that I have the Twinkle code in hand to do this. But if you're not doing talk notices anyway (which I agree you are not strictly required to do), would you be willing to explore a technical solution to meet the objections half way? Franamax (talk) 02:19, 23 November 2011 (UTC)

Pic of a magazine

hello,

am I allowed to upload a picture from a Jet magazine as fair use? It says "A magazine or book cover, to illustrate the article on the person whose photograph is on the cover. However, if the cover itself is the subject of sourced discussion in the article, it may be appropriate if placed inline next to the commentary. Similarly, a photo of a copyrighted statue (assuming there is no freedom of panorama in the country where the statue is) can only be used to discuss the statue itself, not the subject of it."; I want to use it in Otis Redding (it is sourced), so it would be ok? Thanks.--♫GoP♫_T ^C _N 13:57, 24 November 2011 (UTC)

The question to ask is if there is enough sourced discussion about that specific picture to meet NFCC#8 to qualify for its use. It shows Otis (already pictured in the article), his body in rigor mortis in a airplane seat laid out on the floor in a morgue. Visually, there's nothing that cannot be described by text, and the section on his death does not go into much detail of this aspect. In other words, this would fail for significance in the article. (One can also note that being a death photo, there's probably some sensibilities to consider, so another reason to avoid it). --MASEM (t) 16:13, 24 November 2011 (UTC)

Thanks. Just to clarify, I meant the other pic on page 59 ;P, where his body is rescued by the divers. I think I could use it as reference his body did not look that bad like some could think, but ok I think it is not very notable anyway. Just wanted to bring more pictures to the article.--♫GoP♫_T ^C _N 16:38, 24 November 2011 (UTC)

Burger King, restaurant logos, and NFCC #1 No free equivalent

Apparently the Burger King talk page isn't the proper avenue since all I am getting in response to concerns of fair use images being used are a guideline, WP:LOGO, as a reasoning for inclusion and an essay, Wikipedia:BOLD, revert, discuss cycle, as a reason to reinsert contested material.

My concern is articles like Burger King and similar restaurants use of logos. Obviously, we do permit the use of logos, and any other non-free content, when there is "no free equivalent" to using the logo. However, the problem is that we have an extensive category of free images on the Commons related to Burger King, such as old, generic logos and (more useful) some restaurants themselves. I tried to insert File:BK no. 5500, Colma.JPG (the image to the right) under the assumption that free images were more valued than non-free. Apparently I'm being accused to cherry-picking the policy by actually even suggesting that logos not be the best choice of a lead image. The entire process of determining and giving rationales to non-free images is "cherry-picking" in its very nature, though. If an image doesn't meet one criteria, then it is usually nominated for deletion, and considering this is criteria number one, I think it personally an important one. The discussion provided on Talk:Burger King resulted in the following reasons it should be kept:

WP:LOGO - a guideline, which ironically says logos are permitted, but if non-free content is disputed, it is the responsibility of the editors who want to include it actually present a valid case
"It has a rationale" - a rationale which says there is no free alternative, which is, 1) the very thing being contested and 2) in my personal opinion, would be a lie given the amount of content on the Commons
"It's more professional" - a completely irrelevant point to non-free content
"Buildings look different depending on location" - a fair complaint which can be discussed and fixed, however that doesn't justify a non-free image either.
"Identification purposes" - see below

If you're skipping down to find the question, here it is. Some topics aren't so notable and a non-free image is warranted, fully, and I upload non-free images and logos for this purpose as well. However, we have an article that has significant commentary on the topic, Burger King, it certainly can not be confused with any other restaurant by visiting the Burger King article, and we have a Commons category full of free images that would more than suffice for the lead image related to the topic. How exactly can we claim there is "no free alternative", and more to the point, how does the logo make it fully more comprehensible that Burger King is not the article we are looking for? Does the image to the right, a free image, not fully make anyone understand that this is Burger King any less than the logo would? — Moe ε 03:07, 25 November 2011 (UTC)

I think the logo by itself is more capable of quickly and unequivocally identifying the subject in this case than a picture of a particular restaurant building. The style and type of building housing a Burger King will vary quite a bit from location to location, as you can see from the variety of images in the article, and it's likely that other fast food restaurants are also housed in similar buildings. The logo, however, is used by the company not only on every building sign, but also on every burger wrapper, bag, billboard, and television advertisement, and is of course not used by any other company. And I suppose that's the main reason why logos are preferred to identify a company to other images. There's a reason why companies use them: they are a more powerful visual identifier than mere unadorned words. Sometimes companies will treat their restaurants almost as logos ("trade dress" is the technical term) by making sure they all follow some common visual format (White Castle is probably one of the best examples of this), but Burger King does not.

Regardless, this is really one of those completely pointless non-free content fights, given that using a logo to identify a product or service is completely standardized across Wikipedia and about the least controversial use you can make of nonfree content, and the logo is arguably not even copyrightable given that it's basically some simple shapes and some text. Most important, your "free" restaurant photo actually contains the same nonfree logo, so really what would replacing the standalone logo accomplish? We treat your photo as free because we likely think the logo is "incidental" to photographing the building and so permitted on Commons as "de minimis" use of nonfree content, the same as a photo of Times Square can't help but contain some copyrighted billboards in the background. postdlf (talk) 04:18, 25 November 2011 (UTC)

The logo includes a simple graphic that goes well beyond stylized text, which would be uncopyrightable. What is free in the commons image is the photograph, not every element in it. Similarly, if we copy, scan, or make an .svg out of a flat logo, the copy, scan, or .svg itself has no copyright but the logo that is pictured does. As a nonfree image, the logo is nonfree whether it is standing alone or embedded as it is here in an otherwise free image on commons. So using an exterior shot doesn't get around the copyright question. Our normal practice is to use a .svg company logo in the infobox of articles about a brand name company, fast food restaurant chains being no exception. If the storefront or architecture of the store is iconic, it might make some encyclopedic sense to include that photo as well at the discretion of the article editors. - Wikidemon (talk) 04:54, 25 November 2011 (UTC)

Agree that even though there is free imagery of what a BK restaurant would be, we would still include the non-free logo as the primary identifier of the company since that is the best recognized symbol (and their trademark) for them. --MASEM (t) 06:35, 25 November 2011 (UTC)

Agree. The logo is simply more universal, both in recognition by readers already familiar with the franchise and as a starting point for readers unfamiliar with the franchise. --Philosopher ^{Let us reason together.} 19:58, 25 November 2011 (UTC)

I would suggest that one or more images of restaurants are also valid (and good) content, but I would not remove the logo. Rich Farmbrough, 17:30, 26 November 2011 (UTC).

Well, since enough of you feel it does meet NFCC, I won't remove it, even though I never got a proper reply to how no free equivalent exists when we have free images. I'm not debating that the restaurant image is more iconic, representative or otherwise than the logo, my concern lies solely in the definition of "no free equivalent" being treated as "not the best we can do". Specifically the policy saying Non-free content is used only where no free equivalent is available, or could be created, that would serve the same encyclopedic purpose. Is the logo a better representative and more iconic? Yes. Is there no way to convey that it is Burger King without it? Hardly. Regards, — Moe ε 00:14, 27 November 2011 (UTC)

I suspect the thinking is along the lines of that the logo is the best identifier for the company. Is the logo Free? No. Can it be replaced by a Free alternative? No. Is there a Free alternative that will identify the company just as well as the logo? No. Therefore the free use is allowed. Thryduulf (talk) 00:26, 27 November 2011 (UTC)

The logo not being free and the logo not being replaceable, individually, is always going to be "no." It's precisely "Is there a free alternative that will identify the company just as well as the logo?" that I am contesting. I understand the points being brought, that restaurant buildings change and the logo is iconic and a better representative. However, it is my argument that a generic, free restaurant building can do a sufficient job of identifying and since that is the case, that would make the logo replaceable. It's almost like us using a photograph an editor took for an actor's profile over a copyrighted photo by famous photographer, to me. Is the famous photographer's image free? No. Would our image produce a better quality or be more iconic than the copyrighted one? No. Does our free image from one of our editors suffice? Yes, and that is why we don't have non-free images of living people, because we can take pictures of them and it would do just as good identifying them as a copyrighted one would. Obviously this is different, I'm replacing a logo with a building, however, I think the same principle of it being enough to suffice for being an identifier applies the same. — Moe ε 01:05, 27 November 2011 (UTC)

Again, I refer to the current consensus for most cover art in the info box. Even in cases of being obviously replaceable art, the non-free version of covers is considered by consensus to be appropriate. Same would apply to logos, but only strictly on the page about the company itself. Now, for example if there was "List of menu items at Burger King" (putting the issue if that was appropriate or not), the logo would NOT be appropriate being non-free, but the photo of the restaurant would be. But not specifically on the company page. --MASEM (t) 01:08, 27 November 2011 (UTC)

Well, I'm not going to debate it further for the simple point this is more a contradiction to our own policy and the spirit of a free content encyclopedia to keep replaceable non-free images, rather than a violation of fair use itself. Regards, — Moe ε 01:58, 27 November 2011 (UTC)

Update to the image resolution section

The guideline appears to be outdated, and I'd like to modify it to reflect the current reality of non-free file sizes. It reads as follows now:

"Ideally, most common image uses can likely be represented in an image containing no more than 0.1 megapixels (obtained by multiplying the horizontal and vertical pixel count in an image). This allows, for example, images with 4:3 aspect ratios, to be show at 320 x 240 pixels, generally good for screenshots from television, movies, and video games, while allowing for the cover art for most published works to be shown at 250 x 400 pixels."

However in the past year, a slightly larger number has been used, both by DASHBot, the bot that handles non-free size reduction requests, and by file workers like me, who do the resizings manually either when we catch them during our other work, or when DASHBot goes down and the category starts to get into the quadruple digits. I propose the following change:

"Ideally, most common image uses can likely be represented in an image containing no more than 160,000 megapixels (obtained by multiplying the horizontal and vertical pixel count in an image). This allows, for example, images with 4:3 aspect ratios, to be show at 460 x 345 pixels, generally good for screenshots from television, movies, and video games, while allowing for the cover art for most published works to be shown at 300 x 480 pixels."

Yes, I realize that this seems like a rather drastic change, but

a) it really isn't, 160,000 pixels is still far too small to do anything with, especially since about 90% of the non-free files I've seen, including any that have ever been resized, have 72 or 96 DPI (meaing they're useless for printing).

b) DASHBot has been doing 160,000 pixels for a while. If we decide to enforce the lower number we'll have to resize somewhere around 41,000 files, probably by hand.

Thoughts? Sven Manguard Wha? 14:42, 19 November 2011 (UTC)

I don't seeing that being a big problem, as long as this is not the type of thing that's going to be bumped up more and more over time. (I don't think this is of course). Preferably, I'd suggest a max size range around 0.1 to 0.16 MPs to keep the flavor of the original. --MASEM (t) 14:46, 19 November 2011 (UTC)

I'd say leave it as is. The whole point is that the advice here is setting a rough ball-park target figure rather than a hard limit. The point that DashBot doesn't flag anything until that target is exceeded by at least 60% is entirely compatible with that.

Is there any evidence that screenshots are being routinely uploaded that are larger than 320 x 240, or cover art larger than 250 x 400, (or album covers larger than 300 x 300), with any suggestion that that lower resolution would be insufficient or inconvenient? I suspect these numbers do not need revision; but if they were to be revised, I would prefer that to be based on a sound argument of value, rather than just a number pulled out of the air.

But it seems to me the main point here is that it is a good thing that there is a buffer betweeen the size we state as a usually reasonable target, compared to the size at which a bot necessarily kicks in. Jheald (talk) 15:48, 19 November 2011 (UTC)

I agree with Jheald. The bot has to be conservative to avoid damaging those few images which do need a slightly higher resolution. The ballpark figures we have at the moment seem sufficient. J Milburn (talk) 16:45, 19 November 2011 (UTC)

Should we then add language that states explicitly that while 100,000 is preferred, 160,000 is acceptable and does not need to be resized? I say this because to me, this seems like common sense, but to other users, this is a point of contention (which is why I started this RfC in the first place). Sven Manguard Wha? 03:19, 20 November 2011 (UTC)

Speaking of ballpark figures, the Bill Ripken baseball card has been an example of an allowed image in the Unacceptable use of images section for over 5 years.[8] It is 384 by 534 pixels. The image must be large enough to see the information that the "non-free use rationale" justifies. A bot resizing 500x500 images to 300x300 jpg images may just be creating unintelligible blurs. -- SWTPC6800 (talk) 05:01, 20 November 2011 (UTC)

This is why we don't have bots going around to automatically resize images. If they are humanly tagged with a reduce tag, that's fine, but yea, we're not about to have a bot readily enforce it. --MASEM (t) 05:20, 20 November 2011 (UTC)

So as for "160,000 is acceptable and does not need to be resized": no, it's a case-by-case basis; the existing wording—"ideally", "likely"—is sufficient. Uniplex (talk) 08:46, 20 November 2011 (UTC)

If we want to change it to be slightly less rigid, but still keep the healthy buffer we currently have then perhaps adding the word "around" will be all we need - "Ideally, most common image uses can likely be represented in an image containing no more than around 0.1 megapixels". I do though prefer the size expression in pixels (as opposed to megapixels) in the original proposal. Thryduulf (talk) 12:44, 20 November 2011 (UTC)

You could have an otherwise free picture with a non-free item in it. For example, two actors standing on either side of a movie poster. Is the pixel count on the whole image or just the poster? -- SWTPC6800 (talk) 20:16, 20 November 2011 (UTC)

Depends. If the poster is incidental, the image is free under de minimis. If the poster is the focus, the image is copyrighted, and the pixel count is for the whole image. Sven Manguard Wha? 14:05, 27 November 2011 (UTC)

This is why we don't set an absolute limit, but a guideline for most cases (which a non-free element in a free picture is not). Basically the spirit of the rule is that the image must not be bigger than is required to see in reasonable quality the relevant parts of the image. The guideline is just saying that in most (but not all) cases 0.1 megapixels will be sufficiently large. Thryduulf (talk) 23:10, 20 November 2011 (UTC)

Notification of a image size tagging bot

I just noticed that today BAG approved Fbot 9 to tag images greater than 160k pixels for non-free reduce, which I'm pretty sure is counter to past approaches with NFC (DASHbot 9 is ok to reduce images that are tagged with non-free reduce but we have always were of the approach that only humans can really judge if an image is oversized). The above discussion initiated by Sven appears related to that bot approval, but as to the actual size and method, I'm concerned that we weren't informed on this page; I don't want to nullify the bot task, but the bounds and method it works by needs to jive with how we actually enforce #3a. I've started discussion at [9] for this. --MASEM (t) 06:06, 29 November 2011 (UTC)