Wikipedia talk:Non-free content/Archive 27

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Minimal use requirement

I propose adding to item 3a a line like: Non-free content should only be used in the main article concerning what is depicted in the image and not in ancillary or broader topics in which a subset of the main article is included.

Editors are taking a copyrighted image and sticking in multiple articles, which violates the concept of minimal use. -Nv8200p talk 04:44, 25 August 2007 (UTC)

What stance would that take on the subject of: images of fictional characters used on both the characters' pages and the actors' pages? — pd_THOR | =/\= | 04:47, 25 August 2007 (UTC)
To me, images of fictional characters have no business on the on the actor's page. -Nv8200p talk 12:20, 25 August 2007 (UTC)
Criterion 3 is about using as small a portion of the original as one needs, not about using it as infrequently as possible. The latter concern is already addressed in Criterion 8, which says use it only where it is significant. There are indeed a number of proper uses where the article is about a subject other than what is depicted in the image, e.g. an album cover in an article about the album (not about what is shown in the album art), a logo in an article about the company that uses the logo (not whatever image the logo depicts), etc. Similarly, a fair use image may well be significant to more than one article. A photo of a contemporary artwork may be relevant to the article about the work, as well as an article about the genre and an article about the technique used, among other things. Wikidemo 08:31, 25 August 2007 (UTC)

Proposed rewording of WP:NFCC#3b

Note: This conversation originated at Wikipedia talk:Non-free content#Need guidance on WP:NFCC#3b.

Current wording:

(b) Resolution/fidelity. Low- rather than high-resolution/fidelity is used (especially where the original is of such high resolution/fidelity that it could be used for piracy). This rule includes the copy in the Image: namespace.

Proposed wording:

(b) Resolution/fidelity. Low- rather than high-resolution/fidelity is used. The copy in the Image: namespace will not be significantly larger than that needed for display in the Article: namespace. A need for higher resolution must be justified in the fair use rationale for the image.

Hopefully an agreement on the basic principle will allow to develop better/more specific guidelines, but I think an agreement on whether or not the Image space copy needs to be larger than the Article space copy is an important first step. I understand there will be exceptions, I just invite comment on the general rule. Videmus Omnia Talk 14:17, 25 August 2007 (UTC)

In a word, no to this proposed change. #3 is about all media files. Please keep in ming that the types of media files that are the most problematic in present-day copyright law are audio and other non-still files. Low resolution still images need only be adequately low resolution that the image copied off of a Wikipedia page cannot be used to produce a counterfeit version of the image that could be sold by someone or otherwise used in a way that would interfere with or dilute the creator/publisher's ability to make money off of it. Generally, with a book, magazine, comic book cover, any resolution equal to or less than what' already widely put on people's browsers by, say, Amazon and the numerous mirror sites, or Barnes and Noble, Borders and other booksellers, is adeuqate to meet this standard. If someone wants to undertake a project to copy and reduce the jpeg resolution of images en masse, then that's a separate discussion that should under no circumstances be part of NFCC #3. If anything should require discussion under #3(b) it would rightly only be resolutions that are greater than that already widely disseminated by booksellers, article sellers, audio/CD/DVD sellers to offer their products for sale to the public. ... Kenosis 14:29, 25 August 2007 (UTC)

Hmmm, this is the first I've heard that #3 is about media files. I've been digging and can't find the previous conversation on this - can you point me that way? The thinking behind the proposed change is the statement in the licensing resolution that says "...EDPs must be minimal." If a 300px image is all that is required for illustration, then a 700px image in the Image: namespace is not "minimal". Videmus Omnia Talk 16:07, 25 August 2007 (UTC)
VO, sorry, I thought this was understood that the Board, in its Resolution, was giving an (er, pardon me) "illustration" of the principle, using an "image" to give WP a (pardon me again) "picture" of what they were referring to. But #3 is about NFC in general, most relevantly NFC media files, which of course includes still images. Minimal use is also a term of art in copyright law, but in the NFCC it refers to all kinds of media files. Thus the use of "high-resolution/fidelity" and similar terms. ... Kenosis 16:22, 25 August 2007 (UTC)
Ah, sorry. I've gotten accustomed to thinking of "media" in terms of audio/video, my bad. Speaking in general terms, why would Image space copies need to be significantly larger than Article space copies? Videmus Omnia Talk 16:39, 25 August 2007 (UTC)
Perhaps this proposed rewording actually belongs as supplementary clarification in the guideline page? --Pekaje 16:27, 25 August 2007 (UTC)

Wikipedia:Image_use_policy#Displayed_image_size states: In articles, if you wish to have a photo beside the text, you should generally use the "thumbnail" option available in the "Image markup" (this results in 180 pixels wide display in standard preferences default setting). This is a small image, which gives an impression, but not much detail, which is why a bigger image would need to be available in Image: namespace. Images on the web with a maximum dimension of 500 or 600 pixels are generally regarded as low resolution, so it might simplify matters to pick a maximum dimension of 550 pixels and define that as low res for our purposes, as the image use policy states this size "can comfortably be displayed on 800x600 monitors". Tyrenius 17:02, 25 August 2007 (UTC)

In all fairness, that is for the size used in the actual article, and is meant for all images. The NFCC policy is additional requirements for non-free content, and can (and IMO should) certainly impose stricter requirements. --Pekaje 17:09, 25 August 2007 (UTC)
(ec)I see where you're coming from - however, that section doesn't differentiate between free images (the bigger the better) and non-free images ("must be minimal"). Videmus Omnia Talk 17:10, 25 August 2007 (UTC)
The thumbnail option in articles, defaulting to 180 pixels wide, is recommended for all images, free or not. The current proposal wording ties the Image namespace size in with that, or else encourages people to needlessly make the article image bigger, so the Image namespace size can be justified as being bigger. It makes life simpler for everyone and prevents numerous arguments, unexpected deletions and bad feelings, to simply set a size for non-free images. I am suggesting 550 pixels maximum dimension, with maybe a preference for a smaller size where possible, say 350 pixels maximum dimension (but failing to meet this would not be a cause for deletion). It makes sense to base the protocol on the Image namespace size, which will be a constant, whereas the article image size is liable to fluctuation through editing. Tyrenius 17:46, 25 August 2007 (UTC)

We're getting away from use rationales as it is. I wouldn't support making them a dumping ground for more information or creating any more busywork, particularly not the obvious. Moreover, there are already rules and a procedure for reducing image sizes and challenging images, and I see no sign of a significant problem that warrants adding new things to policy. Finally, policy violations are enforceable by deletion. Overly large images should not be deleted, they should just be shrunk.Wikidemo 18:07, 25 August 2007 (UTC)

That was always my understanding until recently, where there have been disputes over image size in the Image space. A policy/guideline specifying a threshold would be extremely valuable in preventing disputes/wasted time. I really like Tyrenius' suggestion, above, though I would prefer the common-law 0.1 megapixels (which is unfortunately confusing to some people). A 550px image would be pretty narrow on the other axis to fit within this. Videmus Omnia Talk 19:19, 25 August 2007 (UTC)
I guess basically what I am asking for is an image equivalent of Wikipedia:Music samples, which specifies length and sampling rate for non-free audio. Videmus Omnia Talk 19:28, 25 August 2007 (UTC)
It would be very unwise to say that no images can be wider than 550px. For one thing, this would quickly become unfair in the case of very tall images. Also, software screenshots are often 800x600 pixels wide, and any reduction in resolution would make them difficult or impossible to read, thus defeating their purpose. —Remember the dot (talk) 19:27, 25 August 2007 (UTC)
I said max dimension, i.e. height or width. Neither to be larger than 550 pixels. If that's correct about screenshots, then it can be listed as an exception. There can be exceptions. Tyrenius 20:06, 25 August 2007 (UTC)
Are there any articles where images that large are displayed? Videmus Omnia Talk 19:29, 25 August 2007 (UTC)
Few, if any, because of basic layout concerns. Some images require 500-600px resolution to maintain a proper, decipherable presentation, e.g. Dali's Hallucinogenic Toreador. I'd say that a max dimension of 550 is reasonable, this would roughly translate to a .3 megapixel limit. Exceptions could be made for certain classes where print resolution doesn't matter, like software screenshots. Fair use claims on sound files are traditionally limited to 30 sec. or less. Anyone know what to do with PDFs? :) ˉˉanetode╦╩ 21:28, 25 August 2007 (UTC)

For 0.1 megapixels you have elsewhere cited Image:John Cleese.jpg, which is w x h, 246 x 365 pixels, which seems a bit on the small side for anything with much detail. It also seems restrictive compared with the music standard of 10%, as on this reckoning the original image would be 1 megapixel. It depends on what is meant by "original image", but even family digital cameras would far exceed this to generate an image. Tyrenius 20:23, 25 August 2007 (UTC)

Professional cameras today are well over 10 mega pixels. A check of a typical image (42-18102696) on the Corbis site shows that 427 by 640 is lowest resolution offered. The highest resolution for this image is 3400 by 5100 (17 mega pixels). Corbis recommends the low resolution image for a 1 inch by 2 inch print.
I don't think a pixel limit is the best way to test the appropriateness of an image size. 300 by 300 may be too generous for a postage stamp and too small for a movie poster. A pixel per inch limit may be a better metric. If the text on a magazine cover is significant to the article it must be readable. Shrinking the image so the text is illegible is vandalism. (I am talking about reasonable fair use image sizes. For example: A 400 by 500 existing image resized down to 240 by 300 to meet a 300 pixel rule.) -- SWTPC6800 22:19, 25 August 2007 (UTC)
A max dimension (height or width) of 550 pixels is less than the lowest resolution offered by Corbis, then — and even that is only considered suitable for a 1 x 2" print. I suggest a yardstick of a maximum size to start with (with stated exceptions). Maybe it would be a good idea to look at different subjects and image sizes already on Wiki/Commons to see what works. Tyrenius 22:41, 25 August 2007 (UTC)

Non-free image on multiple pages

The artwork Gravitation by M.C. Escher has a somewhat iconic status among us afficionados. Page Kepler-Poinsot polyhedron has a thumbnail of it next to a link, while the Small stellated dodecahedron page has only a link (but IMHO would benefit from a thumbnail too). Question: is the thumbnail Fair use? I have been assuming that its iconic status justifies its presence, since many people familiar with the appearance of the work and the name of the artist will not know the name of the work. But another person has been arguing that it is unnecessary, just a link is fine. Maybe this particular judgement hinges on how "iconic" it really is? Thoughts and advice greatly appreciated. -- Steelpillow 09:56, 25 August 2007 (UTC)

Justified for the article about the artwork, no question. No real need to use it on any other pages though. I would just link to the artwork article from the small stellated dodecahedron page. Carcharoth 10:24, 25 August 2007 (UTC)
I've removed the image from small stellated dodecahedron, and removed the non free use rationale for that article from the image page. In these cases, I do think we should leave it to the readers to follow the links. They only need to see the image when they get to the article about the image. Carcharoth 10:31, 25 August 2007 (UTC)
Well, that was my point about the iconic nature of the work. WP:NONFREE#Unacceptable_uses item 4 gives, A work of art, not sufficiently well known to be recognized by a large percentage of casual readers, whose theme happens to be the Spanish Civil War, to illustrate an article on the war. (However, because of its iconic status, it is presumably "fair use" where there is a small image of Picasso's Guernica in the article Bombing of Guernica.) With the thumbnail there, people who visit the geometry page can see it and think for example, "well, I never realised that famous picture was one of these". If someone does not know the name of the work, they may never know what to look for unless the thumbnail is there to guide them. -- Steelpillow 11:42, 25 August 2007 (UTC)
If we have an article about the image, then minimal use should suggest that it the image is only used on that article. Everything else can link to that article. In cases where there is no article about the image to link to, a non free image can be used on several articles if they have a strong claim for use of the image. It makes sense to me, but is hard to put into words. Maybe "if a non free picture can be replaced by a link to an article about the image, then this is nearly always preferable"? Carcharoth 00:36, 26 August 2007 (UTC)

Imagine a non-free image with no article, used on three different articles. Then imagine that someone writes a well-written, informative article about that non-free image, and uses the image there. The three other articles using it can now replace their uses of the image with links to the article. Linkage and information is kept, and non free use is minimised. A win-win situation. Carcharoth 00:38, 26 August 2007 (UTC)

Sounds logical to me. -- Ned Scott 01:13, 26 August 2007 (UTC)
Thanks for the replies. But again I refer you back to that quote, "However, because of its iconic status, it is presumably "fair use" where there is a small image of Picasso's Guernica in the article Bombing of Guernica." In other words, there is a getout clause where the work is "sufficiently well known to be recognized by a large percentage of casual readers". Nobody's reply has yet made reference to this, to say either "Yes, this getout clause does after all take precedence over what we just said", or "No, it does not apply in the present case because ...". This is the ruling (if that's the right word) that I am seeking. -- Steelpillow 09:27, 26 August 2007 (UTC)
I disagree with the Guernica example. Have a look at the Bombing of Guernica article without the image. It doesn't suffer at all, in my opinion, from a lack of the image. If people are unsure what this 'Guernica' painting is, they should click on the link. I'm normally all for articles being self-contained, and for explaining context in articles, rather than leaving the reader to click a link, but in the case of minimising use of non-free images, it makes sense to send people off to the article on the image instead, if one exists. Carcharoth 16:03, 26 August 2007 (UTC)
I've been bold and made changes to the policy and the guideline. Probably best to discuss in the new section below. See Wikipedia talk:Non-free content#Articles with their own images. Carcharoth 16:26, 26 August 2007 (UTC)

New class of acceptable images

I've added a new class of acceptable images. See my edits here. The new text is: "Iconic and historic images: Some images attain iconic status (eg. Che Guevara (photo) and Raising the Flag on Iwo Jima)." Does anyone have any problems with that? Carcharoth 16:12, 26 August 2007 (UTC)

I would add that these images should only be used in the context of commentary about the images themselves, but recent ifd discussion are implying that now we can use any good image for any purpose (even when in direct competition with the copyright holder) as long as it's useful for us.
Personally, I'm not sure I still care. Maybe we could remove all but item #4 in WP:NFCC. It would save a lot of time wasted on deletion discussion while just marginally increasing the number of non-free material uploaded daily.--Abu badali (talk) 02:25, 27 August 2007 (UTC)
Are you referring to discussions I participated in, or other ones? I don't follow IfD that closely. I did notice a lot of effort being expended on that evolution image. Is that an example of what you were talking about, or do you mean other ones? If you mean historical ones (I saw the Nobel photos debate as well), then I seriously doubt that old black-and-white photos account for most of what is uploaded every day, though I'd welcome being corrected on that. Carcharoth 03:04, 27 August 2007 (UTC)

Images that mainly serve to illustrate a copyrighted item

I am interested in opinions on Image:Seagate_160_GB_hard_drive_box.jpg. Our longstanding practice is that this is a nonfree image because it is a derivative work of its subject. Recently there is a discussion on commons about such images, including a comment by our lawyer that I don't understand [1]. For our purposes, is this image "non-free"? — Carl (CBM · talk) 18:13, 25 August 2007 (UTC)

From what I understand, it isn't, though I could be wrong. 17Drew 19:09, 25 August 2007 (UTC)
My understanding of what he (Godwin) is saying there, is that we shouldn't delete such images, or overly restrict their usage, unless we (ie. the Foundation or the OTRS people) receive a complaint. But best to check with him directly, obviously. My view would be to tag the image as copyright, or use a tag that states that it is of a trademarked object or copyrighted design, and then provide a non free use rationale. And then show Godwin's comment to 'those who have a narrow interpretation of Wikipedia's non free use criteria and argue for the deletion of the image', while you respond by politely suggesting that it really wouldn't do any harm to keep the image. Then, if the image still gets deleted, take it to deletion review, quoting Godwin at each stage if needed, though don't rely too much on Godwin's quote. If he gets involved, all the better. If not, you've made a principled stand. Obviously, you won't be able to tag with a copyright tag on Commons, so upload it to Wikipedia. Carcharoth 00:07, 26 August 2007 (UTC)
I think I may be the one with the narrow view here. If we have to put a nonfree use rationale on this, then it isn't a free image, and for the use it has in the article it's replaceable. The question is whether we can just put a GFDL tag on it and nothing else (which is what I believe the uploader would like to do). — Carl (CBM · talk) 00:11, 26 August 2007 (UTC)
(edit conflict) Hang on. I thought the image was being used in the article about the product. I was wondering why someone took a photo of the packaging and not the product inside. Then I realised it was being used to show misleading labelling. Scrub the above. Godwin's comment doesn't apply here. The labelling does not need to be illustrated. It can be described by text, and needs to be sourced to reliable sources, not 'proven' by a series of photographs. Wikipedia is an encyclopedia, not a place for exposes or original research or investigative journalism. Carcharoth 00:12, 26 August 2007 (UTC)
Godwin's comment still applies: "a photograph that happens to include all or part of a copyrighted image or a trademark does not raise significant intellectual property issues." That is regardless of any article content. The objection, "not a place for exposes or original research or investigative journalism" applies to the article per WP:NOR, but is not relevant to the photo copyright issue. Tyrenius 01:29, 26 August 2007 (UTC)
This photo doesn't "happen" to include the copyrighted work - the entire purpose of the photo is to reproduce it. A particular reading of Godwin's comment might suggest that a photo of an album cover serving only to reproduce the cover could be a free image. That's why I said I don't understand Godwin's comment. — Carl (CBM · talk) 01:37, 26 August 2007 (UTC)
This is the image given as an example at the beginning of the thread Godwin commented on: Image:CornNuts Dscn0266 crop.jpg. It seems to be the same situation as the example you are querying. A possible rationale is that the photo is of a package (cardboard boxes aren't copyright) which happens to have a design on it. An album cover reproduction is usually just of the design as the substantive part of the image. Maybe if the album cover was at an angle in a larger setting it would just "happen" to have a copyright work included. That is only speculative and it really needs further definition from him. Tyrenius 02:02, 26 August 2007 (UTC)
It depends on how significantly featured the copyrighted/trademarked work is. If I take a photo of a street scene, and included in that photo happens to be a Starbucks logo on a visible store and a discarded Coke can in the street, Starbucks or Coke can't come after me—their inclusion is incidental. On the other hand, if I obviously set up a Coke can with a totally white background and nothing else in the shot, and it's the obvious subject of the photo, Coke can probably claim it as a derivative work. Seraphimblade Talk to me 23:19, 26 August 2007 (UTC)
What I'm looking for is not a general theory, which is more difficult anyway. I'm looking for opinions about whether th eimage named above is free or nonfree. That's a specific question that should be simpler to answer. — Carl (CBM · talk) 12:10, 27 August 2007 (UTC)

Articles with their own images

Articles about non free images provide a convenient way to minimise non free use. I've edited NFCC#3a with this edit to say "If an image has its own article, other articles should link to that rather than use the image." Does this seem like a logical and non-destructive way to minimise non free use? I've also modified the Guernica example here. Carcharoth 16:18, 26 August 2007 (UTC)

(ec) I don't think that's appropriate in every instance, so I'd rather not see it in the policy. I'd rather have the guideline strongly encourage this kind of minimization of use. Not sure how to phrase it, though. --Pekaje 16:27, 26 August 2007 (UTC)
Encourage rather than set in stone, I'd agree. The problem with putting stuff in the guideline and not the policy is that the policy doesn't clearly say "see guideline below" it relies on people to carry reading past "the fold". Also, the guideline doesn't currently refer back to the policy, and specifically the numbers. Carcharoth 23:27, 26 August 2007 (UTC)
Also, I can't find where was this (rather substantial) change was discussed. Since this talk page and its archives are getting quite long, could you point me to it? --Pekaje 21:47, 26 August 2007 (UTC)
It was discussed at Wikipedia talk:Non-free_content#Non-free image on multiple pages. But that was really a preliminary discussion, and I gauged that the response (or lack of responses) was enough for me to be bold and get more responses by actually making the changes. So the real discussion is taking place now. It is dispiriting to make suggestions and comments (read the many talk page threads above) and get little response. I've seen too many discussions bog down and not achieve anything, so I wanted to try something different here. I note my other bold change above (the historical one) hasn't generated any opposition so far. Carcharoth 23:27, 26 August 2007 (UTC)
The change gives an invalid commentary on NFCC#3a. The third criteria makes no mention of multiple uses of the same item. It is about multiple items where one item will suffice. The commentary that has been added gets this wrong.
# A work of art, not sufficiently well known to be recognized by a large percentage of casual readers, whose theme happens to be the Spanish Civil War, to illustrate an article on the war. (However, because of its iconic status, it is presumably "fair use" where there is a small image of Picasso's Guernica in the article Bombing of Guernica.)
Carcharoth's revision
# A work of art, not sufficiently well known to be recognized by a large percentage of casual readers, whose theme happens to be the Spanish Civil War, to illustrate an article on the war. (Because of its iconic status, Guernica has its own article, but as detailed in NFCC#3a, Bombing of Guernica should only link to the article.)
The proper basis for omitting the image from Bombing of Guernica would be lack of critical commentary in that article. The policy does not include a limit to the number of articles, but rather general restriction on any use in any article, which are sufficiently onerous in their own right that use will limited to a small number of articles where it is of significant relevance. I think the policy gives the appropriate way to manage the balance between high-quality and non-free content; and that this change to the guideline gets the plain meaning of NFCC#3 wrong. Duae Quartunciae (talk · cont) 22:38, 26 August 2007 (UTC)
You are right. I've confused excessive non free use in a single article with using the same image in multiple articles. I do think that something along the lines of what I said in the earlier section should go in though. "Imagine a non-free image with no article, used on three different articles. Then imagine that someone writes a well-written, informative article about that non-free image, and uses the image there. The three other articles using it can now replace their uses of the image with links to the article. Linkage and information is kept, and non free use is minimised. A win-win situation." - I now recognise that this is a separate issue, but would you support something like this being added somewhere? Carcharoth 23:27, 26 August 2007 (UTC)
This proposal seems rather unhelpful for someone who can't click to another article (such as someone reading a paper version of an article). 17Drew 22:47, 26 August 2007 (UTC)
I suspect it is unlikely that paper versions would use non free images. I may be wrong though. In any case, people reading a paper version would see a bold name, and turn to the page with that article on it. This is quite normal in print encyclopedias, who have less room for pictures and so tend to keep them where they are needed. Carcharoth 23:27, 26 August 2007 (UTC)
In general, I think that if there is a consensus that an image helps to give an increase in understanding to an article, and if there is commentary specifically on the image, then it should be permitted for use in that article. But if all critical commentary on the image is deferred to another article dealing with the topic more thoroughly, then the image should be limited to the other article as well. Duae Quartunciae (talk · cont) 23:41, 26 August 2007 (UTC)
This sounds very close to what I've just put here. I also removed the Guernica example (while reverting back to the previous wording), as it is not clear enough, in my opinion. Would you be able to change my edit to add your exception ("if there is a consensus that an image helps to give an increase in understanding to an article, and if there is commentary specifically on the image, then it should be permitted for use in that article")? Carcharoth 23:44, 26 August 2007 (UTC)
Duae Quartunciae, I agree with the version you wrote, but indented it, as I'm unsure if we need to preserve numbering or not. People don't generally refer to these examples by numbers do they? They should probably be rearranged and organised thematically anyway. Carcharoth 00:07, 27 August 2007 (UTC)

Whoopsie... edit conflict. Try again. Here is my attempt, with indent shown also.
# A non free work of art whose theme happens to be a war, to illustrate an article on the war, unless the work of art itself becomes the subject of critical commentary.
:* A non free work of art should not be used in an article if all critical commentary is given in a separate article on that artwork. (For example, an image of Picasso's famous painting Guernica may be used to illustrate an article about the painting. The article Bombing of Guernica, if it has minimal commentary on the painting itself, is should omit the image also. A link takes readers to the image and the commentary.)
I've also replaced "extensive" with "critical". The policy does not require "extensive" commentary, and I don't think that should be implied. I've gone ahead with the change... Duae Quartunciae (talk · cont) 00:11, 27 August 2007 (UTC)
I've replaced all "extensive"s with "critical"s, and added some more "critical"s in other places. Carcharoth 00:43, 27 August 2007 (UTC)
I wasn't referring to print versions of the entire encyclopedia. I'm referring to off-line versions of articles or mirrors where there aren't any links used. The whole idea behind summary style is that the article can stand and be understood on its own. If the article needs an image to properly discuss a topic, then readers should not need to go to another article to understand what's being said. They should need to go to the other article if they want further details. 17Drew 08:55, 27 August 2007 (UTC)

Related: A new numbered point has been added as follows:

# If a non free image has its own article, commentary from reliable sources about the image and its history should be placed in the article about the image, and other articles should, in general, summarise this and link to that article rather than re-use the image.

As worded, this addition goes beyond limits on non-free content, and actually expresses a limit on content in general, apparently in an effort to ensure that any argument for using an image is removed. I oppose this addition to the guideline. Duae Quartunciae (talk · cont) 00:26, 27 August 2007 (UTC)

I modified that with "in general", but I see it merely as a rewording of what you wrote: "A non free work of art should not be used in an article if all critical commentary is given in a separate article on that artwork. [...] A link takes readers to the image and the commentary." How is that any different from what I wrote? Carcharoth 00:42, 27 August 2007 (UTC)
The wording given here actually advises limits on the critical commentary on other pages. Duae Quartunciae (talk · cont) 01:17, 27 August 2007 (UTC)
You are absolutely right. I'll retract that example. OK, I've now done that, but generalised your wording to cover all images, not just artworks. Carcharoth 01:34, 27 August 2007 (UTC)
Good catch; yes, I should have said images. Thanks Duae Quartunciae (talk · cont) 02:31, 27 August 2007 (UTC)

A change by User:Wikidemo has the same problems as I identify above.

#* A non-free image should not be used in an article if all critical commentary from reliable sources is given in a separate article on that image. (For example, an image of Picasso's famous painting Guernica may be used to illustrate an article about the painting. The article Bombing of Guernica, if it has minimal commentary on the painting itself, should omit the non-free image also. A link takes readers to the image and the commentary.)
Wikidemo's revision
# A non-free image to illustrate an article section about the image, if the image has its own article (in which event the section should instead contain a very brief summary as necessary and a link to the article about the image)

This change goes beyond rules about non-free content, and makes a general restriction on content in an article itself. It should not be presumed that discussion in the context of one article should be deferred to an another article specifically on the image. There are many conceivable cases in which an article on a particular topic might consider issues related to the image which are properly discussed in the context of the special topic, rather than in the article about the image itself. We should not be attempting to constrain the content of articles as a way of controlling non-free images. Duae Quartunciae (talk · cont) 09:02, 27 August 2007 (UTC)

Large percentage of casual readers

I've just tided up unacceptable uses 4 and 5 with this edit. I hope the shorter wording still preserves the spirit of the examples. The main change was to remove the hopelessly subjective "not sufficiently well known to be recognized by a large percentage of casual readers" clause. Those wanting to keep an image will say it is sufficiently well known, and those wanting to remove an image will say the reverse. The "large percentage" and "casual readers" bits are hopelessly vague. I've replaced all that with "unless the [image] itself becomes the subject of extensive commentary." This is far more objective and easier to prove. Does this look OK? Carcharoth 23:56, 26 August 2007 (UTC)

That's much better, and you're correct it's much more objective. It's also in line with the general principle that we decide by sourcing (or lack thereof), not personal viewpoints or guesses. Though we might want to define that the extensive commentary should be in reliable sources, otherwise we'll get people trying to use blog chatter or the like. Seraphimblade Talk to me 00:00, 27 August 2007 (UTC)
I have a concern with the word "extensive". The phrase in policy is "critical", which I think is better. Duae Quartunciae (talk · cont) 00:19, 27 August 2007 (UTC)
(edit conflict)I've added lots of "from reliable sources" throughout, plus some other minor changes. See here. That also introduced one big change, which I'll discuss in a new section, but could someone do a sanity check on that edit? The section above (and Duae Quartunciae in the comment directly above) also suggests that extensive should be replaced with critical, which I'd be happy with. I don't care either way as far as that goes. Carcharoth 00:21, 27 August 2007 (UTC)

Changes made to guideline examples

I'm going through some of these trying to keep the tone and format consistent from one example to the next, avoiding unnecessarily specific examples, keeping things short and to the point. Also, "from reliable sources" is not a relevant issue here. Everything on Wikipedia must be sourced. However, the issue is that there must be commentary (or whatever else the issue is) in the article about the image; whether there is commentary about the images in reliable sources is not directly relevant. The commentary in the article is not from reliable sources exactly (it is from a Wikipedia editor, technically). Rather, it is sourced to reliable sources. But that goes without saying because under WP:V everything must be. So no need to repeat it throughout the examples. I'm concerned we might be going a little fast....if it works, fine, but an effort to change too many examples too quickly could crash. Despite / in addition to these changes we'll still need to go through all of the examples methodically and clean them up. That might be safer one section at a time on a "proposal" page or section. Wikidemo 08:51, 27 August 2007 (UTC)

Update - in going through these some more I found I could not constructively deal with them except by reverting (which I didn't want to do) or going all the way to edit them as they should be. I hope that's not too big of a chunk all at once but there really is no middle ground. I've tried not to change the meaning or purpose of any examples. I eliminated quite a bit of redundancy and extraneous detail so if you see something specific missing, it's probably there in a different form. Also, generally attempted to iron out inconsistencies in terminology, tone, format, numbering, etc. A few things that came up --
  • Vis-a-vis my last comment "sourced xxxxxxx" is probably the best phrasing, not "xxxxx from reliable sources." Phrased that way it's accurate and brief.
  • "critical commentary" should in nearly all cases be "commentary." The word "critical" comes from the copyright statute but is not properly used as an adjective to describe commentary - in the statute it is "criticism or commentary." We don't do criticism here on Wikipedia, we just do commentary.
  • For similar reasons "analysis" is probably not a good word to use and should be changed to "commentary" too. Analysis implies original editorial contributions. We don't analyze, we just comment.
-- Wikidemo 09:48, 27 August 2007 (UTC)
I think it will all come out in the wash. I probably did make too many changes too fast, but if you go through each diff, you should be able to see how the text evolved. I'm going to look through the changes, but from what you say above, they look fine. Carcharoth 17:08, 27 August 2007 (UTC)

OK, I've looked through the recent changes, and they look fine. In fact, the re-organisation by theme has made things much clearer. A few points:

I'm also going to raise the text issue in more detail below. Carcharoth 17:34, 27 August 2007 (UTC)

Retired people

I added "Care should be taken over retired groups and people, as images of them now may not serve the same purpose as images of them during their career." to suggest the possibility of some exceptions for the living people rule. This could be controversial, so I'm raising it here (along with nearly every edit I've been making to the page). The argument goes that pictures of an actor or sportsperson or musician during their career are the encyclopedic ideal, and that pictures of them long after they have retired, maybe when they are very old, may be less than ideal. The equivalent for buildings, which I didn't mention, is that a building may get partly demolished or rebuilt or extended. In other words, the appearance might change significantly. What's the verdict on this? My guess is it comes down to whether the change in appearance is discussed in the article or not, or whether the mere act of illustrating a unique historical moment is enough. Carcharoth 00:29, 27 August 2007 (UTC)

Yeah, I think this is already covered. The key here is wheter or not a free licensed image can be created to serve the same ensyclopedic use. If the image is spesificaly used to discuss how someone looked while they where active 40 years ago then obviously a random photo of them today won't do the same job. That said it obviously needs to be used in that way in the article and explain why it's use is significant in it's rationale. It may well be non-replacable, but if it's not significant either it will still fail the criteria, albeit a different part of it. --Sherool (talk) 08:30, 27 August 2007 (UTC)

Images probably in the public domain

I have raised a question here about how to handle images which are probably, but not certainly, in the public domain. I'd appreciate any feedback there. – Quadell (talk) (random) 14:17, 27 August 2007 (UTC)

This may be relevant: Save Orphan Works. --Abu badali (talk) 16:49, 27 August 2007 (UTC)
Are we discussing this here or there? Can someone consolidate the discussion and relocate everything to one location? I think a note on DRV pointing here is best. For what it is worth, DRV shouldn't be taken as precedent. The people at DRV are no more or less competent than the people at IfD. It is entirely possible to renominate an image that was overturned at DRV, and get a new IfD consensus to delete. Of course, what is really needed is authoritative information. Carcharoth 17:14, 27 August 2007 (UTC)

Excerpts of copyrighted text

I'd like to consider the acceptable use of text:

"Brief quotations of copyrighted text may be used to illustrate a point, establish context, or attribute a point of view or idea. Copyrighted text must be attributed and used verbatim. Any alterations must be clearly marked, i.e. [brackets] for added text, an elipses (...) for removed text, and emphasis noted after the quotation as "(emphasis added") or "(emphasis in the original)". Extensive quotation of copyrighted text is prohibited."

Some examples, taken from articles on works by J. R. R. Tolkien because I know where to find them, though I am sure more representative examples can be found:

  • A Elbereth Gilthoniel - three versions given, but no commentary to justify such extensive quoting.
  • The Road Goes Ever On (song) - again, different versions given, some comparison and commentary, but probably not enough to justify the extensive quoting, though unlike the previous example, only the first poem is given in full. The later stanzas are just excerpts, though still fairly long ones.
  • Oath of Fëanor - again, different versions of a poem. The thing that struck me here was that these versions are scattered through different books, and I was only aware of a few of them. Though I own copies of all the books (The History of Middle-earth series), I haven't read them all. In effect, the gathering together of the poems here has made it unnecessary for the reader to read them in the posthumous books, though some (many?) will follow this article up in the posthumous books.
  • Namárië used to be only an extract, but now has changed beyond all recognition, with a full quote and translation and previous version.
  • Errantry and The Man in the Moon Stayed Up Too Late, are examples of extracts only.
  • Secret Fire is an example of quotes giving examples of a phrase.
  • Malbeth the Seer, and All that is Gold Does Not Glitter are examples where the quoting may be excessive.
  • Bilbo's Last Song, Fastitocalon and The Mewlips, are examples where there are no quotes at all.
  • The opening section of Mirkwood is a good example of a justified quote for the "beyond Myrcwudu" bit.
  • The One Ring is an interesting example. It used to have a template that quoted the Ring Verse, but that got deleted. I then made these changes to "reduce quote and add commentary and explication". I wasn't terribly surprised to find, while writing this, that someone had later 'helpfully' extended the quote with this edit.
  • Finally, J. R. R. Tolkien itself has quotes from his letters and some quotes from his essays. An example of more extensive quoting from his letters can be seen at The Lord of the Rings (1955 radio series).

Anyway, this has been discussed before at Wikipedia:WikiProject Middle-earth, along with the need to consolidate and merge some of the more obscure articles on minor stuff that only die-hard Tolkien fans are interested in. A different, and even more extensive approach is seen on another wiki here. Not all those entries on that other wiki quote the whole poems, but some do, such as this. Some quote excerpts, such as this. An extremely extensive quote is here. Those are all examples of early stuff, but examples from The Lord of the Rings include this and this and this. I realise that others doing this is no defence, but wanted to give examples outside Wikipedia. Another example is the approach taken at The Encyclopedia of Arda (which is not a wiki), such as here, where they give a short quote for "atmosphere". Of course, Wikipedia is aiming for encyclopedic content, rather than atmosphere, and explicitly avoids literary analysis. Anyway, getting away from wikis and other sites, and back to this one, without commenting overly much on the fannish aspects and some instances of original research, I'd appreciate any guidance on when quoting something becomes overly extensive, and on what is the best way to handle this, especially with regards to literary analysis. There is a growing body of secondary literature on Tolkien's works, and much stuff can be reliably sourced on the literary analysis angle. How much is it acceptable to quote to help illustrate a particular point of literary analysis that is being made? Carcharoth 19:01, 27 August 2007 (UTC)

How much text to quote is a complex subject that depends on so many factors that the subject seems to resist any rule of thumb or quick summary. The Tolkein gateway goes well beyond what we would allow and includes some things that in my opinion are clearly outside of fair use and are therefore copyright infringements. But as an example of how complex this is, including an entire copyrighted poem, whether in original English, or one of Tolkein's invented languages (or in somebody's translation of them), seems wrong because it replaces the original in its entirety and goes far beyond the minimal amount necessary for commentary or analysis. Even an entire stanza of a long poem is unnecessarily wrong unless the whole stanza is needed for the commentary. If a poem is only one line or one stanza then it might be too much to quote the whole thing even if a quotation of equal length would be reasonable if it's only a small part of a longer work. Moreover, most links to Tolkein poems are going to be links to copyright violations and therefore disallowed. We might just be stuck here, with no practical way to do certain things. On the other hand, it's equally clear (in my opinion again) that it's okay to quote the entire "One ring to rule them all" poem in its entirety, if that's used for analysis. The reason? It is not a freestanding poem. Nobody buys a book of poetry just to read that poem. It is a tiny part of a much long novel, and quoting the poem is transformative and in now way replaces the commercial value of novel. There are many, many articles, essays, tutorials, etc., about this in the context of fair use.
I wouldn't put any of that in these guidelines because anything we could say would either be too short to be helpful as guidance, or if complete it would be so long it would bloat an already-long page here. But it might make sense to have a separate sub-page, guideline, or essay, on how to quote text, when it is okay, and how much is okay. Also, how to link to sources when quoting text. Do we have something like that already? If not that would be a good project. There are lots of examples on the net of people doing this in the context of fair use. Of course our rules are a little different than fair use but it might be useful to review some of these to see the different issues and situations that arise and how they're handled. Wikidemo 19:24, 27 August 2007 (UTC)
Thanks. That's pretty much what I thought. I think quotes of single lines, or several lines from poem, with ellipses, might be the way to go. Would I be right to say that separate quotes, broken up, put out of order, and each quote relating to a point raised in some independent souce, would be OK? For example: "The poem in question starts with line 1. In 1990, Professor X said this about line 5 of the poem. In 1994, line 3 of the poem was used as the title of a music album. In 1996, literary critic Y interpreted line 4 to mean this. In 2003, a book published on Sindarin translated line 2 as this." If the whole poem was 5 lines long, with the first line provided for identification purposes, then would quoting all the other lines, out of order as described above, be acceptable? Theoretically, the whole poem could be reconstructed in order, but I think the above is OK. I agree that some of the examples in our articles go too far. That will be something that needs to be looked at in the not too distant future. Carcharoth 19:40, 27 August 2007 (UTC)

Historical images and NFCC#2

My more detailed post above got no response, though a later thread did cover the issue, so I'm raising the issue of "respect for commercial opportunities" (NFCC#2) again here. The basic issue, already covered as CSD, is a need to insert a clause into NFCC#2 saying that the mere presence of a historical image (or indeed any image) on a stock image library, or an archive image library, doesn't automatically mean that NFCC#2 should apply. Free images are sold by image libraries, though they apply the standard copyright tag to all their images regardless. These free images may be ones that were always free (eg. NASA pics) or ones that have become free (expired copyright, public domain by age). What I want to see is whether there is a consensus to explicitly acknowledge this in the NFCC#2 section.

An example of this is the following:

"The Hulton Getty Picture Collection (formerly the Hulton Deutsch Archive) scarcely needs an introduction. Based in London, this collection is universally acknowledged as the greastest library of photojournalism in the world. The collection comprises in excess of 15 million photographs, prints and engravings, including the work of such famous names as Keystone, Picture Post, Fox and Central Press. More recently, the HGPC became responsible for the management of Mirror Syndication International and for the digital archiving of the Reuters News Picture Service; it has published several CD-ROMs covering evocative images of a selection of decades from the twentieth century. HGPC is also co-operating in RACE project called MEDIATOR, which involves the production of a digital newspaper. From its vast archival collection, the HGPC is contributing 15,000 images for the main HELIX database and another 7,000 images for a module covering the Social and Political History of Britain from 1859 to the present day (SPHB)."[2]

This and other image libraries perform a valuable service in archiving and preserving historical images, and a lot of works goes into finding, restoring and scanning old images, but I'd really like more guidance on when the images fall out of copyright, or whether historical archives can and do renew the copyrights? Note that even when images have fallen out of copyright, the image libraries will still mark them as copyright and carry on selling them. The mere presence of a historical image on an image library's website shouldn't lead to automatic cries of NFCC#2 (respect commercial opportunities). For a start, low-resolution images impact commercial opportunities to a very small extent. Image libraries provide extremely high-resolution scans for print purposes, and if we wanted high-resolution scans, we would need to do those ourselves.

I'm not sure how NFCC#2 needs adjusting for this, but how about:

"Respect for commercial opportunities. Non-free content should not be used in a manner that is likely to replace the original market role of the original copyrighted media. Please note that image libraries and other organisations often mark all their images as copyrighted, including ones that they have obtained from free or public domain sources, or that are now free or public domain by age. This means that the presence of an image on an image library's commercial sales website does not automatically mean this criterion applies. What also needs to be determined is the actual copyright status of the image."

Possibly all this is implied in the phrase "original market role of the original copyrighted media", but I am not sure people are actually understanding what that means. If there is a more elegant way to phrase this, please suggest it. Carcharoth 11:13, 25 August 2007 (UTC)

One thing I forgot. Sometimes old image archives include unpublished old negatives, and the image library is effectively publishing the images for the first time. In these cases, NFCC#4 applies, so care should be taken there. Carcharoth 11:15, 25 August 2007 (UTC)
It is true that many libraries and museums maintain that they hold copyright over their collections when they in fact do not. I think, however, that there are more appropriate places, such as a copyright tutorial or essay, to detail false claim of copyright than in the non-free content criteria. --Iamunknown 18:14, 25 August 2007 (UTC)
Agreed. But such an essay is long overdue. I am willing to make a start, but what would be the best title? Carcharoth 00:32, 26 August 2007 (UTC)

An issue I worry about is that we have no way to "determine is the actual copyright status of the image". Especially for historic images, we often have no publication information, no information on the creator, etc. As an ordinary editor, I see no way to comply with the last sentence of the policy paragraph. — Carl (CBM · talk) 18:40, 25 August 2007 (UTC)

That is a problem. However, it doesn't stop people trying to delete historical images under NFCC#10. Maybe change to "What also needs to be determined, as far as is reasonably possible,' is the actual copyright status of the image." We should have a template for "historical images of uncertain status". We used to have {{non-free unsure}}, but for some reason (I suspect copyright paranoia) it got deprecated. Carcharoth 00:32, 26 August 2007 (UTC)
How do you plan to differentiate "images with unknown copyright status" from "images the uploader didn't find the copyright status in a google search"?. As a side note, do you have any idea how unhealthy for a discussion it is to classify opinions you disagree with as "paranoia"? --Abu badali (talk) 15:38, 28 August 2007 (UTC)
As I said elsewhere, sorry if the paranoia word offended anyone. That wasn't my intention. I was referring more to the substance of the argument, rather than the connotations of the word. As for differentiating, I personally don't trust anything any uploader says until I've checked it. We should really start a verification method for people to sign off on images as confirmed PD, etc. Carcharoth 17:40, 28 August 2007 (UTC)

Examples of NFCC#8 (image significance for article)

I wondered if it is worth comparing four different approaches to commentary on an image:

  • Image:Time evolution wars.jpg in Intelligent design: "The public controversy was given widespread media coverage in the United States, particularly during the Kitzmiller v. Dover trial in 2005. Prominent coverage of the public controversy was given on the front page of Time magazine with a story on Evolution Wars, on 15 August, 2005. The cover shows God facing off against a chimpanzee"
  • Image:Time-magazine-neville-chamberlain.jpg in Neville Chamberlain: "During his tenure as Chancellor, Chamberlain emerged as the most active minister of the government. In successive budgets he sought to undo the harsh budget cuts of 1931; he also took a lead in ending war debts, which were finally cancelled at a conference at Lausanne in 1932. In June 1933, Britain hosted the World Monetary and Economic Conference. Describing the event as the "most crucial gathering since Versailles," top U.S. newsmagazine Time featured Chamberlain on its cover, referring to him as "that mighty mover behind British Cabinet scenes, lean, taciturn, iron-willed... [I]t is no secret that Scot MacDonald remains Prime Minister by Prime Mover Chamberlain's leave."[2]"
  • Image:Burningmonk.jpg in Thich Quang Duc: "Browne's photographs quickly spread across the world wire services and leapt off the front pages of newspapers worldwide. [...] no news picture in history has generated so much emotion around the world as that one [...] In Europe the photos were hawked on the streets, and Communist China distributed millions of copies of the photo throughout Asia and Africa as evidence of “US imperialism”. One of Browne's photos remains affixed to the sedan in which Thich Quang Duc drove to his self-immolation and is part of a tourist attraction in what is now Ho Chi Minh City commemorating the event."
  • Image:WW2 Iwo Jima flag raising.jpg in Raising the Flag on Iwo Jima: "The photograph was extremely popular, being reprinted in thousands of publications. Later, it became the only photograph to win the Pulitzer Prize for Photography in the same year as its publication, and ultimately came to be regarded as one of the most significant and recognizable images of the war, and possibly the most reproduced photograph of all time." (from the lead section), plus Raising the Flag on Iwo Jima#Legacy and other sections.

Given the examples above, is it possible to come to an objective decision on NFCC#8 cases? There is a whole spectrum of way to add commentary, from a range of sources. What should be assessed? The relevance of the visual information to the commentary, the impact of the actual image, the lasting legacy, or what? Many more examples can, and should be provided, to guide people on this.

On a separate, but related issue, I think some consideration should be given to the educational nature of historical images. I think the simplest way to put this is that sometimes old images are needed in history articles purely to increase the reader's understanding of the history. Text descriptions can only go so far, and the educational value of showing what things looked like at the time cannot be overestimated, in my opinion. This would apply even when the particular picture was not iconic or remembered as a picture. Thoughts? Carcharoth 09:52, 28 August 2007 (UTC)

I think this is a really random result, that gets the policy badly wrong. Of these images, the one that has the best cause for actually increasing understanding of the topic itself is the one that was deleted. It reflects the fact, meaning no offense, that people are actually being guided more by the iconic nature of the image than by real contribution to understanding as provided for in the policy. The other images are more forceful and memorable, I grant. This seems to be the real basis for decisions, I guess. IMO Duae Quartunciae (talk · cont) 10:29, 28 August 2007 (UTC)
The second one might get deleted as well. A parallel discussion is going on at User talk:Neil#Interpretations of NFCC#8. I'm going to try and end that and ask him to copy his thoughts over here. I'll bring my comments over as well. Carcharoth 11:36, 28 August 2007 (UTC)
(copied across) Well, of the four examples you give, only the latter two contain critical commentary on the image (note that critical is not used in the sense of criticism, rather in the sense of discussing the value/impact/legacy/etc of the image). I don't, personally, see how showing Image:Time-magazine-neville-chamberlain.jpg adds to the current text when the current fair use criteria are applied. It is possible to come to a subjective decision - the important question as regards the NFCC#8 criterion is "is the image itself (not what the image portrays, or the fact the image exists) discussed within the article"? If yes, NFCC#8 is met. I agree that educational value of "what things looked like at the time" would be really useful; this is not, at present, within the criteria - and in the present climate, I don't forsee any attempt to make the less stringent or more inclusive being succesful. Neil  11:51, 28 August 2007 (UTC)
Maybe sustained impact is a better way to phrase all this? The two TIME images certainly had an impact at the time, but it needs distance in time to ascertain the historical legacy of something. My gripe is with 70+ years old images which are old enough, in my opinion, to qualify as genuinely historically educational (eg. Image:1heldeplatz.jpg), without the image itself having its own legacy. The problem seems to be people applying the letter of NFCC#8, rather than the spirit of it. Historical images that are educational due to their age are definitely in the spirit of NFCC#8, in my opinion. My evidence for the spirit comes from {{non-free historic image}}, which seems to have been meant for images of historic moments regardless of whether the image was itself notable with its own impact and legacy, and also, surprisingly to some, from the Foundation licensing policy itself: [3]: "Their use, with limited exception, should be to illustrate historically significant events..." I can't interpret that any other way than to say that the WMF is saying that it is OK to use images just to illustrate historically significant event. NFCC#8, which is part of Wikipedia's response to the WMF policy, seems to go beyond what was intended. What do people here think? Carcharoth 11:37, 28 August 2007 (UTC)
Some of the oldest photographs (those older than the date of the author's death plus 70 years) are immediately public domain under US, EU, Canadia and most other laws (most others are less restrictive than the 70 years). It's the historical images that don't fall into that criterion that the problem applies to. People are (rightly or wrongly) very stringent about applying the letter of policy with regards to image copyright (a lot more so than text, I think because images are far easier to deal with). The Foundation policy and the en.Wikipedia policy differ, yes, and so the Wikipedia fair use criteria could be changed, but a change in the criteria itself, even if it's just in the phrasing, would be the only way to change it so people were happy that images solely used to illustrate historical events could get onto Wikipedia. Neil  11:51, 28 August 2007 (UTC)

Non-free images in "list of..." articles

Hi! I've got a question regarding the non-free policy, part 3(a). For a "list of [anything] in [some media]" article, how much leeway does 3(a) provide? As a specific example: List of characters in Neon Genesis Evangelion was tagged today with the {{non-free}} template. There are currently 25 non-free images in the article. Of the corresponding character entries, only 10 have their own separate articles, and the rest do not have enough information to justify splitting them into their own articles, which to me would justify the use of the images for those entries even if the images for the articled characters were removed. What should be done? Willbyr (talk | contribs) 17:03, 24 August 2007 (UTC)

My understanding of that part of the policy is that they're only disallowed for galleries and trivial lists, such as discographies and TV show season lists. In the above article there is a non-trivial amount of coverage for most of them, so I think it might be acceptable. Just because an article is called a list, does not (IMO) immediately disqualify it from having non-free images. Particularly a fictional character list does seem to need to show what the character looks like. --Pekaje 10:41, 25 August 2007 (UTC)
Thanks for the input. Angel (Neon Genesis Evangelion) was tagged today in the same fashion. Willbyr (talk | contribs) 12:41, 28 August 2007 (UTC)
Angel (Neon Genesis Evangelion) certainly has too many images - it has multiple images per character for most of the characters. Although it might be nice to be able to illustrate our articles in this way, nonfree images are not permitted to be used so liberally. List of characters in Neon Genesis Evangelion is more reasonable; it uses one image per character to identify the characters. There is some debate about whether this is acceptable, but since one image would be allowed per article if these were split into separate articles, I think it can be defended. — Carl (CBM · talk) 12:53, 28 August 2007 (UTC)
I both agree and disagree with the above statement. For some of the entries in Angel (Neon Genesis Evangelion), I wouldn't have a problem with cutting the number of images down (especially the ones with images from the trailers for the new movie, as their source can't be readily viewed by non-Japan-based web access), but for a few of the entries, the multiple images are used to depict either drastic changes in the appearance or nature of the character (especially here, since the images are taken from two separate works) or illustrate plot points of the episode of the series in which the character appears. If the entries have to be cut down regardless of this, are links to the image source(s) permitted? Willbyr (talk | contribs) 14:14, 28 August 2007 (UTC)

I think there should be some latitude for non free image use to illustrate the different ways a character is portrayed in different adaptations. For example, have a look at Frodo Baggins#Adaptations. Most people throw their hands up in horror at seeing a gallery of non free images (and one free image), and fail to stop and think about the purpose of the gallery. The gallery is not showcasing different images of the same thing, but it is putting images of different things together, and the accompanying text is given below (or above) the gallery. It is more a layout choice, than ZOMG! GALLERY! When a gallery is informative, can this over-ride the non free content criteria? Another example would be a gallery of the changing appearance of a building over time, with some free and some non-free pictures, or other historical timelines. These would seem, to me, to be justifiable collections of non free content. Normally, the images would be scattered throughout the text, but sometimes a gallery layout looks better, and is easier for the reader to follow. Can we come up with examples where galleries of non free content are acceptable? Carcharoth 13:51, 28 August 2007 (UTC)

I disagree that Frodo Baggins#Adaptations is acceptable. There isn't nearly enough text about the various adaptations to justify four nonfree images. This isn't a matter of the images being inside <gallery tags: it's just nonminimal use of nonfree content. Nonfree images use to illustrate the appearance of each adaptation belong in the articles about the adaptations themselves.
What is needed to support the nonfree images in that article are reference to reliable sources who have commented on the appearance of Frodo. Then we can use a nonfree image to support our commentary. But just including four images to show that the character is different in each adaptation isn't right. Of course he looks different in each adaptation. Consider a hypothetical case in which Death of a Salesman includes four nonfree images to illustrate four different actors all portraying Willy Loman in different productions. — Carl (CBM · talk) 14:15, 28 August 2007 (UTC)
Earlier this year, the Foundation took a strong stance against Fair Use images, stating that "All projects are expected to host only content which is under a Free Content License, or which is otherwise free as recognized by the 'Definition of Free Cultural Works' as referenced above" (Foundation:Resolution:Licensing policy) We're expected to remove excessive fair use images per the Foundation's policy and WP:NFCC. Rockstar (T/C) 18:51, 28 August 2007 (UTC)
This is also a good thing to read about image removals in lists. Rockstar (T/C) 19:17, 28 August 2007 (UTC)
Responding to your first point: but what about this bit of the policy? "Such EDPs must be minimal. Their use, with limited exception, should be to illustrate historically significant events..." - it says right there: "to illustrate historically significant events". How do you interpret that, remembering that you can't refer to Wikipedia's EDP (WP:NFC) to help you interpret the Foundation's Licensing Policy. Also, I've seen people say that there is a deadline of "March 23, 2008" to sort all this out, but in fact that deadline is currently written as a subclause of point 6. ie. It only applies to projects without an EDP. "For the projects which currently do not have an EDP in place, the following action shall be taken [...] By March 23, 2008, all existing files under an unacceptable license as per the above must either be accepted under an EDP, or shall be deleted." This may just be a mistake in the layout of the Licensing Policy, but if this is so, it needs to be changed and the change widely advertised. Carcharoth 19:20, 28 August 2007 (UTC)
Look at bullet 5 of the Foundation policy. — Carl (CBM · talk) 19:33, 28 August 2007 (UTC)
I have read bullet 5. There is no deadline for how long the discussions should take over the images uploaded before the March 2007 date. There probably should be a deadline. I agree there should be one. I'm just saying that the current placement of the March 2008 deadline makes it look like it only applies to bullet 6. Carcharoth 19:47, 28 August 2007 (UTC)
  • So far in this thread an important point is being overlooked. The image usages in the cited examples are fair use images being used for depiction purposes only. That's an extremely weak fair use claim. If the images were being used to support commentary on the appearance of the character, the claim would be better. But, that's not what is happening here. The images simply aren't necessary to the content of the article. Such uses as thesee are clear failure of WP:NFCC #8. The images simply aren't significant. You guys cites List of characters in Neon Genesis Evangelion as defensible with 15 fair use images. Good grief. 15! If a character is soooo minor that we are incapable of generating anything more than a stub on the character, then why on God's green Earth MUST we have an image? It's a minor character!
  • It is important to all people to understand that Wikipedia is a free content encyclopedia. Copyrighted fair use content detracts from this. Copyrighted content has to jump through pretty big hoops before it is acceptable on Wikipedia. It can't just be because it makes the article look better, or for depiction purposes. If depiction alone were a reasonable rationale for inclusion of fair use images, than we would have millions of non-free images. But, depiction alone is not enough in the majority of cases.
  • There's been so many arguments about these points that it becomes tiring to rehash it over and over and over and over and over again. I fully recognize that with each debate there's a new crop of editors who fail to grasp our core principles, who are unaware of these debates, and in many cases simply do not care what our mission is. I'm quite willing to educate people. I've done so on innumerable occasions. But, everyone has their limits. I sorely wish there was a coherent, single location to point people to as to stop these endless debates. The resolution and the policy, despite being in essence the word of God, are not enough. Jimbo himself has spoken on this, and his position is even more restrictive than the policy is and it is STILL not enough to stop these endless debates.
  • All of you need to start with the premise How can I best make this article without using any copyrighted imagery? Work towards that goal. Work hard at it. Only in a tiny minority of cases should we ever get to the point of asking How many non-free images should I have on this article to support the article? and only in extremely rare cases should we be adding more than an extremely small number to any article. ANYthing else directly violates the very premise of what Wikipedia is. --Durin 23:41, 28 August 2007 (UTC)
    • I tend to agree. Though I've been thinking for some time that maybe a strategy of fair-use for a group of articles might work. To take the Tolkien example, because I have an extensive knowledge of that, it is fairly easy for those with an in-depth knowledge of an area to draw up a wish-list of images. Would it be possible to select a limited number from such a wishlist to cover a group of articles? In the case of the Tolkien articles, 2-3 images to show examples of artworks inspired by Tolkien, 2-3 images to show examples of films based on Tolkien's works, 2-3 images of Tolkien himself, 2-3 images of the book covers, and so on. In each class, there would be a great number of possible images to chose from, but only 2-3 would actually get selected for use under fair use. The basis of the selection would be to provide as much visual information as possible, while still minimising non free material. This would reduce the number of non free images from, say, 50+ to around 15 images, covering around 300 articles (that many will probably remain even after the stubs and minor stuff are merged). There would only be about 1 non free image for every 20 articles, and many articles would appear to be missing images, but editors would find a notice directing them to the "allowed list", and asking them not to add any more images.
    • Though I still don't understand why Wikipedia doesn't just strip non-free images out of the database dumps. Downstream users obtaining the content in different ways, and reusing non-free images, are not a concern, as the legal liability falls on them, not us. We should clearly tell all re-users that they should not re-use the non-free images unless they are happy that their use of the material is appropriate (eg. commercial uses would be inappropriate). If the downstream re-users think "free" means "we don't have to worry about the content we are re-using, as Wikipedia will make sure it is squeaky clean". Free doesn't mean "hassle-free for re-users".
    • And while you are here, Durin, what do you think of all the historical images stuff? The Wikimedia Foundation Licensing Policy clearly states that non free images can be used to illustrate historically significant events. Doesn't that directly contradict your "It can't just be because it makes the article look better, or for depiction purposes."? Carcharoth 00:27, 29 August 2007 (UTC)
A. Speaking only of the main series, all but two of those characters aren't important enough to have images, the others either play constant prominent roles throughout the series or were of great importance to the plot, and most of them have their own articles because of that.
B. If "depiction" isn't enough, go right ahead and remove all album, movie, video game and book cover images from Wikipedia. What're they there for? Some underwhelming art and the name/makers of the subject, all of which can be described immediately in a sentence or two? Pictures of fictional characters serve more purpose than covers do and covers are a huge part of Wikipedia.
C. Links to said arguments plx. If this subject has been done before please provide evidence so people here can address it formally.
D. I'm not seeing the point of NOT including fair-use material if it makes Wikipedia a better and more easily accessible place. Our definition of what the premise of Wikipedia is seem to be very different. In a related note though, what happens if a user himself creates a group image of characters/whathaveyou using official art, makes it low-res and uploads it to Wiki? Would it put an end to all arguments or what? - Egamm 00:24, 29 August 2007 (UTC)
That won't work; derivative works of copyrighted characters are not allowed on the Wikimedia Commons and can only be considered as fair use on en.wikipedia. User:Zscout370 (Return Fire) 02:08, 29 August 2007 (UTC)
Not what I was referring to, I was was speaking of splicing multiple images together into a singular image to bypass the one image per article restriction. - Egamm 01:02, 30 August 2007 (UTC)
  • That would be rapidly deleted. It's not the number of images in question here. The number is actually fairly meaningless. Someone attempted this with Magic sets, and it was rapidly deleted. --Durin 12:59, 30 August 2007 (UTC)
I wouldn't say the number is "fairly meaningless", as you've been pretty stoic on not allowing more than one Fair Use image per article. Oh and was the image rapidly deleted because of its content being meaningless or because pasteing images together violates a rule? Because under the same reason this is in its article, wouldn't an image roughly like this be allowed at the top of "List of Major Konoha Teams"? - The Norse 22:47, 30 August 2007 (UTC)

A rare sketch

When researching the article on the ship wreck Vasa I came across information about a painting found on the front of the foremast depicting an unknown kingly figure. When the mast was salvaged, the painting of the king disappeared, probably because the outer layer of the mast was washed off, and all that was left was some paint residue. The painting was filmed before the mast was salvaged and a pencil sketch of the painting was made by the head of the diving team, Per Edvin Fälting (1911-1995). The sketch is included in Vasa I, the most comprehensive and most definitive book on the archaeology of Vasa published so far. As far as I know, the picture has no commercial value and was done by Fälting while he was engaged in the investigation and salvaging of Vasa, which more or less qualifies it as scientific, rather than artistic work, even if Fälting himself was not a scholar or archaeologist. I asked the director of the Vasa Museum, Klas Helmersson, about the picture, but he told me that the film was not preserved and that there seems to be no other depictions of the mast painting other than Fälting's sketch.

I've scanned the picture from Vasa I, but before uploading it with a fair use rationale, I would like some feedback to see if others think it would be reasonable to do so.

Peter Isotalo 17:23, 30 August 2007 (UTC)

This sounds very reasonable to me. Mention the above in your fair use rationale, and there should be no problems. Carcharoth 17:42, 30 August 2007 (UTC)
That sounds like the very definition of irreplaceable to me. As long as it's accompanied by commentary, I see no trouble at all. Seraphimblade Talk to me 06:09, 31 August 2007 (UTC)

Nonfree use rationale requrement weakened?

I know that the possibility of removing the need for nonfree use rationales for certain categories of images (album covers, say) has been discussed. Was there an outcome to those discussions? It would be helpful for those who resolve the image deletion backlogs to know the current consensus on this issue. — Carl (CBM · talk) 15:58, 28 August 2007 (UTC)

It is my understanding that they cannot be completely removed because the Foundation resolution requires them. I think the current discussion involves using standardized rationales for certain categories. (But if I'm wrong, feel free to call me a happy idiot!) -- But|seriously|folks  16:30, 28 August 2007 (UTC)
No, just happy. But Kat Walsh recently clarified that the foundation resolution is not intended to impose a requirement for explicit written rationales for each image or image use. Following that, the requirement was removed from the policy page (10(c)) for a few days, then the change was reverted. Also, deletions for lack of rationale seem to have ground to a halt in the case of legacy images but they are continuing at a clip of 100-200 per day for new images added without a rationale. In short, we're in a bit of a holding pattern right now. Wikidemo 16:50, 28 August 2007 (UTC)
  • My understanding is that every image needs a "rationale", but not all rationales need be hand-written (i.e. some--logos, for example--can be transcluded with the non-free logo template). I suppose we just need to come up with a wording that clarifies this. --Iamunknown 19:01, 28 August 2007 (UTC)
If enough users would come out and support it, we could go back to only requiring a fair use tag, which explains the claim of fair use adequately enough in most cases. —Remember the dot (talk) 19:16, 28 August 2007 (UTC)
  • I disagree that the tag adequately explains it in most cases ... in some cases, certainly. Consider a logo for example: Using one in an infobox is generally accepted as okay, but using one elsewhere? Not necessarily. --Iamunknown 19:18, 28 August 2007 (UTC)

I should be more clear. The Foundation:Resolution:Licensing_policy requires a "rationale" only in the theoretical sense: there has to be a reason for using each image. This was clarified by Kat Walsh at some point. But English Wikipedia has historically required more: we require a "nonfree use rationale" which by definition is a written statement in the image description page, separate from the license tag, explaining why the image is needed. WP:CSD#I6 and WP:CSD#I7 permit deletion of any image that does not have a nonfree use rationale after being tagged for a few days. This is independent of whether they have a rationale in the sense of the Foundation resolution. Whatever the consensus is here about the relationship between nonfree use rationales and rationales in the sense of the Foundation needs to be reflected at WP:CSD. — Carl (CBM · talk) 19:19, 28 August 2007 (UTC)

In other words, it is OK to have an image if someone can provide a rationale if challenged. But I too prefer to request rationales of people up front. Carcharoth 19:54, 28 August 2007 (UTC)
"In other words, it is OK to have an image if someone can provide a rationale if challenged." That agrees with the Foundation resolution but not with CSD. — Carl (CBM · talk) 20:00, 28 August 2007 (UTC)
Actualy that's pretty much exactly how the CSD currently works. Someone challenge the use of the image, if no one steps up and provide a rationale within a week then and only then is it deleted. So even if we changed the policy to say that rationales are only required if someone ask for one it wouldn't rely change anyting as I see it. Besides it would still be in everyone's best interest to just provide the rationale up front. It would make it less likely that someone would challenge it in the first place, and if you where to become less active down the road you don't risk the image getting deleted just because you didn't check in often enough to catch the challenge before the alloted deadline. --Sherool (talk) 21:27, 28 August 2007 (UTC)
The biggest problem has been the wholesale deletion of legacy images that are perfectly appropriate as per the licensing guideline and Wikipedia's substantive rules for non-free use. Tens of thousands were deleted already on this procedural ground and about 150,000 more would be deleted - about half the non-free images now on Wikipedia - if the policy were aggressively enforced. The image deletions, and the so-called consensus on the matter, was kept in place for the past several months by arguments that the Foundation policy required written rationales. That was sometimes by reasoned appeal, other times by simply shutting down the discussion or reverting things. It turns out they were wrong. That revelation seems to have dampened people's enthusiasm for purging Wikipedia of its old images so, as I said, we seem to be in a bit of a holding pattern while this is resolved.
Personally, I've urged both sides to go slow. People who think use rationales should be optional ought to wait until we have a consensus before changing the policy page. More than a consensus, we need something like absolute certainty that the consensus is here to stay because as soon as we open the floodgates to new images without written rationales it would be very sloppy to reverse that decision. On the flip side of that, I see no harm in continuing to tag and delete new images for lack of written rationales while we sort that out. Just a little extra effort for the uploaders. At the same time it would be hasty to continue deleting old images. Wikidemo 21:45, 28 August 2007 (UTC)
There's no harm in insisting that new images have a use rationale. —Preceding unsigned comment added by Wikidemo (talkcontribs) 21:45, 28 August 2007 (UTC)
I am trying to pause before continuing to delete images for lack of rationale. There seems to be a lack of effort put into writing rationales for images; almost anyone could open up a category such as Category:Images_with_no_fair_use_rationale_as_of_20_August_2007 and start adding written rationales. Whenever I look at those categories, I'm surprised nobody has done so. — Carl (CBM · talk) 22:03, 28 August 2007 (UTC)
I have, actually, probably 20-40 images this month. But mostly the obvious cases like logos and album covers, and mostly the old images. I dont' want to see a good article weakened by deletion of a good image that merely lacks a rationale. But that's for the good of the project, not as a favor to people who can't be bothered to fix their own uploads. Also, now that there are fewer deletion proposals I find that many if not most of the images lacking fair use rationales ought to be deleted, rationale or no rationale. Many of them are either clear policy violations, or of trivial importance to the encyclopedia. That kind of thing tends to make me think we should be insisting on use rationales whatever the foundation says. Not making people write everything out in long-hand, mind you, but at least getting them to say in one way or another why they uploaded the image. That could at least help sort the valid images from the invalid ones. Wikidemo 22:54, 28 August 2007 (UTC)
(ec) I try not to delete certain types of images with no UR, like an album cover used in an article about the album or a logo used in an article about the company that is identified thereby. (Unless I'm off on one of my unwarranted deletion sprees, of course.) It seems like a waste of time, even though it's justified by our current policies. So would writing individual URs for all of those images. I personally think that for these types of images, we should transclude the URs into the NFC tags and be done with it. -- But|seriously|folks  22:18, 28 August 2007 (UTC)
What I see is two fold. Yes some images have the same the same use, and templates are an ok method of doing this. BUT that template rationale is explicitly for the use of that image in a certain article. IE logos logos for company page, album covers for album pages. BUT those two are the only uses of templates that I can see. If a logo is not used in the companies article then that template rationale shouldnt apply. we can have template rationales for explicit reasons. But even those templates should include the pagename of article where its used. βcommand 22:15, 28 August 2007 (UTC)

My take on it is this: if a user wants to upload a non-free image then they need to explain why they need the image. This could be done via a checkbox, a drop down menu, multiple choice, a paragraph, an essay, who knows what. Many methods, with many pros and cons. Some work for some uses / images, but maybe not others. I got really wrapped up in this argument before, then I had to tell myself to step back and say, wait, as long as we are still asking the uploader to explain, and in a way that is satisfactory, then that is our use-rationale. But, as said before, we need to now look at what methods will work for what images/uses. -- Ned Scott 00:06, 29 August 2007 (UTC)

And said examples of checkboxes and such are simply examples. I tend to lean towards a more thoughtful process of explanation. But hey, who knows. -- Ned Scott 00:08, 29 August 2007 (UTC)
In brief, the current system is confusing and off-putting for new users, and experienced users are simply copy-pasting templates nine times in ten. What exactly is all this red tape accomplishing? – Luna Santin (talk) 17:36, 29 August 2007 (UTC)
No, it's worse than that. It's a religious following (aka cult) of a flawed and worthless policy. I am kind of waiting for enough editors to stick around who want to change policy. There really has to be enough of us to do it. But I forget, wikipedia is run by a cult. -Nodekeeper 23:23, 29 August 2007 (UTC)
I'd be happy to codify some exceptions to NFCC#10c, or make templatized fair use rationales for these articles, so that I don't have to waste time with them. I don't know whether I'm part of the cult, though. — Carl (CBM · talk) 23:37, 29 August 2007 (UTC)
That you are complaining of a cult, when such editors as myself, Ned Scott, Butseriouslyfolks, etc. who normally comment in favour of a strict non-free content policy instead commented favourably in response to removing some of the restrictions, is amusing. --Iamunknown 00:03, 30 August 2007 (UTC)
Some of the stuff being put up for deletion on IfD should never be there. I know I'm not the only one thinking this. As I said, I'm just waiting for enough people to get ticked off. The deletionists are always quoting NFCC#8 fantically, adhereing to it above reason. I'd call that a cult. -Nodekeeper 01:48, 30 August 2007 (UTC)
I am in favor of keeping the requirement for individual justification. Blanket "nonfree images are always alright in..." classifications are by definition excessive use, as they will not be necessary for understanding in all those cases. Rationales should be neither templatized nor standardized, each use should require an individual justification, not "Well we always do this." Seraphimblade Talk to me 00:08, 30 August 2007 (UTC)
Here we go again. Reminder. The policy does not contain any requirement, either implicit of explicit, that non-free images should be "necessary" for understanding. The policy requirement is "significant increase in understanding". Understanding is not a binary yes/no proposition, where you either have it or you don't. In most topics, there is ample scope for ongoing increasing understanding, as someone studies the topic and learns more about it. As an encyclopedia, wikipedia cannot hope to give full understanding of complex topics, but rather a basic introductory level of understanding. Images are a particularly good and efficient way to give further increased understanding for the general reader. So the question addressed in policy is — how do you manage the tension between high-quality and free-content? There IS a tension, and policy allows that a certain amount of non-free content can be used subject to certain limits... one of which is that it gives a significant increase in understanding. For some reason this is repeatedly being rephrased into much stronger terms, which if given official recognition would constitute a major change in policy. No offense intended by this. It just seems to be something that needs to be emphasized from time to time, to keep things on track. Duae Quartunciae (talk · cont) 03:18, 30 August 2007 (UTC)
Use rationales can already be standardized and templated as it is, and I don't see that changing, especially with the clarification from the Wikimedia Foundation. There are several broad categories encompassing many image uses where the rationale is entirely identical from instance to instance, two of the most obvious being company logos and record albums in their respective infoboxes. Other common cases that for one reason are another are slightly less standard are book covers, film posters, product packaging and labels, images of contemporary sculptures, DVD and game box art, and video caps of anime characters. These are all completely legit images to use on Wikipedia, and I see little purpose in making busywork just for the sake of making it harder. If you were to make people type out by hand an imperfect and often erroneous facsimile of the reason why they are allowed, you not only waste people's time but you make the images harder to manage. Far better to categories them and deal with the categories than it is to deal with one at a time freeform text descriptions, or unconstrained template use for that matter.
I oppose doing away with use rationales entirely, but I favor a requirement that the standard cases be flagged as such, with as much data as possible parameterized so we can work with it in an organized way. One thing we can do is require the standard uses to all be put into infoboxes, with infobox fields reflecting the various things we need to know about each image. For example, if you use a company logo in the company infobox, you can require the infobox to include the various information in image size, source, owner, what it is a logo for (the company, a product, a brand?), etc. Users can then type all that in once when they use the image in the infobox, and find a way to transclude that onto the image page. That's even more relevant to the less standard cases such as book covers, where the infobox / image box can include things like the publisher, edition, whether it's a front cover or some other cover, the artist who created it if known, etc. Wikidemo 00:32, 30 August 2007 (UTC)
Well, you and I still disagree that blanket use of logos, or DVD covers, or whatever the case may be, is acceptable or minimal. If we don't want to say "necessary" for understanding, we should at least say "essential". Setting the bar simply at "increases understanding" allows nonfree images in pretty much every scenario, so long as they're even tangentially relevant, they could be said to some degree to "increase understanding". Obviously, this would be maximal use, not minimal. Seraphimblade Talk to me 04:18, 30 August 2007 (UTC)
The word in policy is not "necessary". It is not "essential". It is "significant". That is the word we should be using if we want to stay with policy. The word "essential" has all the same flaws as "necessary". It suggests that understanding is something you either have or you don't; whereas in real life we have many degrees of understanding. The policy as worded is that if a non-free image gives a significant increase in understanding, then it meets WP:NFCC#8. It is the same yardstick I would use in contexts where I am writing with a strong need for brevity, which is sometimes the case; such as writing for a journal with a page limit. In such circumstances, an equation, an image, a paragraph, an example, all have the same requirement. They have to make a significant contribution.
In wikipedia, we have lots of disk space; it is not a paper encyclopedia. But we do have this non-free ideal. So the restriction is not a general one; it applies especially for non-free content.
Some wikipedias have a different policy, like the German wikipedia. No non-free images are allowed at all. It's simpler, and you can still understand the topics. Nothing is really "essential" when you come right down to it. But in my opinion, the end result is inferior. I prefer the English standard, which is that if a non-free image gives a significant increase in understanding, not obtained by text or free images, then it is permitted. Duae Quartunciae (talk · cont) 04:34, 30 August 2007 (UTC)
I'm entirely willing to go for "significant", if it were actually adhered to. I fail, however, to see how placing an image of a CD cover, when it's not a critical or iconic part of that CD's history, in a CD article, is a significant increase in understanding. I fail to see how many corporate logos could give a significant increase in understanding (unless they're an icon in and of themselves, such as the logos of Nike or Coca-Cola or McDonald's, or the logos themselves are a significant subject of discussion or controversy). Every time I ask that question, the answer is generally in the vein of "They just do" or "They identify the subject" (which the name of the subject also does). If someone could answer that for me, how in all cases those create a significant increase in understanding, even in CD articles which are simply a track listing and a nonfree image of the cover, I might change my position. Seraphimblade Talk to me 06:51, 30 August 2007 (UTC)
Significance is a subjective term. Some folks seem to take it as meaning "pivotal", so that this contribution stands out from all others. Other folks consider a contribution significant if it makes a credible difference, as opposed to a difference that can be ignored as pretty much irrelevant or decorative.
My cards on the table; I take the latter meaning. It may be partly my background; in sciences something is "significant" if you need to take it into account, as opposed to "negligible" if it can be safely ignored. A significant contribution can still be a small contribution, but if it makes a credible difference, so that at least some readers would see that item and have a kind of "aha" reaction, then it is significant.
The notion of "identification" is recognized as a "significant" contribution. That's how I'd use it. I imagine someone looking for a CD in their collection. Their friend describes it to them over the phone, but in a big collection of CDs you really want a better understanding than this. So you go to wikipedia as a source of useful information. There's the cover shown for you. This lets you flip rapidly through the collection to pick out the CD. For identification, a cover is comparable with an ASIN identifier for a DvD. The image is the information you need when picking it out from a physical stack. The ASIN is the information you need to find it in a database. Both are significant useful information to be included; unless you have some strong reason for degrading the quality of information available for the sake of some other principle.
That's just one example. Other cases include an image with significant iconography, or special visual impact that need the image to be appreciated. And so on.
One concern I have is that the review process is not giving nearly enough consideration to the consensus view of those working on a page. This is related to the idea that understanding is not an either/or proposition, but a case of something that can be at many levels; and which continues to grow as you study a topic — even as you work on the article. The understanding available through an image, or an equation, or a quote, is something that also can grow and deepen with more appreciation of the context. I am afraid that we have a situation right now where there is a major campaign to get rid of images that don't meet up with the standards; but that the significance is being judged by people who don't actually know all that much about the topic.
This is a really bad idea, and it leads to poor decisions. The idea that you either understand or not, fosters the idea that anyone can glance quickly over a page and see if the image increases their understanding or not. That's a fallacy; unfortunately one to which the internet contributes generally. With an effort to study and learn, understanding grows over time. Consider a physics article. An equation might be black magic to a total novice; but a crucial step in understanding for someone with a bit of background. Most people are wise enough to appreciate that those writing the page are in the best position to give informed views on significance. Significance of equations does get debated but not everyone grasps the significance or lack thereof equally well. Sadly, this does not occur nearly so much with images; with the result that important images have been tossed out based on flawed analysis.
I guess you can't write this into the guidelines, but I think and administrator should be very cautious indeed about overriding a strong consensus on significance by people working on the page where it is used. I am thinking still, of course, of the Times Evolution Wars cover, which I think was a dreadfully flawed decision that is a significant loss of available understanding in the article on Intelligent design. Duae Quartunciae (talk · cont) 07:28, 30 August 2007 (UTC)
The thing is here, we do have a strong reason for "degrading the information", as you put it (and as I disagree that it does, in most cases, I don't think most people consult Wikipedia before flipping through their CD collection, and even if they did, CD location is really not one of our critical tasks). We're not just looking to write an encyclopedia. If that were the case, we'd push the boundaries of fair use as far as we could go, we'd happily and readily accept "Wikipedia-only", "noncommercial only", "no modifications", so on so on. But that's not the idea here. We're looking to write a free-content encyclopedia. If I had it to do, we'd adapt the German solution, and say "Hey, free content means free content, and don't bring anything else around here!" However, the Foundation does allow some nonfree image use, and it seems for now that is going to happen. At the same time, it requires that the use must be minimal. Your interpretation seems to be that any increase in understanding is enough, that that's the definition of significant. I disagree. That's not minimal use. Minimal use is where, if the article would still convey substantially the same information ("CD X is an album by Band Y", "Time Magazine ran a cover story on the evolution conflict entitled 'Evolution Wars'"), we skip the nonfree image and use only the free text. And that's why I think Evolution Wars was a good decision. One can understand that Time ran a cover story without seeing the cover. The image in itself is not iconic or a subject of significant discussion or controversy. It's not needed. I hope that most admins looking at image deletions will do the same, and ignore the "Keep, PRETTY! (I mean, uh, critical to understanding!)" if there's a well-founded "Delete, here's why it's unnecessary." Seraphimblade Talk to me 08:34, 30 August 2007 (UTC)
I object strongly to this as a failure to assume my good faith. The remark above "Keep, PRETTY! (I mean, uh, critical to understanding!)" is unfair, insulting and inaccurate.
I can respect a desire to change policy. I can engage robustly but with respect with those I disagree with. My main point here has been to raise a concern with policy changes occurring "under the radar" by using a rephrasing of policy without first getting a consensus for a new wording to policy. I'd rather you work at getting your preferred wording into policy; but that if that fails to have consensus, than I'd like you to use the wording of policy as given.
As a minor point on the one example of the Times cover, and an apparent refusal to even admit I might really mean what I say, let me emphasize. I really am in good faith saying that the cover image gives a significant contribution to understanding, and in fact it IS iconic. That is the reason for keeping the cover. The only reason. Not decoration. Not prettiness. Understanding. I've lost this debate, unfortunately, but my position has been totally sincere.
Recent decisions seem have given overmuch consideration to whether an image is well known, or "famous"; and the phrase "iconic" is often used in that sense. This actually has little directly to do with giving understanding. The Time cover was certainly iconic in the sense of using graphic iconic imagery giving visual cues for a strong message of direct relevance to the whole debate. Appreciating the full depth of this requires seeing the image, and the background of the court case and the fight to try and portray it as independent of religion. The iconography of the Time cover exploded that defense right out in the public arena. It pisses me off that I can explain this repeatedly, and STILL I get an insinuation that I'm being insincere and actually more concerned about maintaining a decoration. Blech. Duae Quartunciae (talk · cont) 09:42, 30 August 2007 (UTC)
Logos, record albums, and other identifying graphics are obviously about more than pleasing the eye as well. If logos failed to capture more than a mere word, companies would not spend so many billions of dollars per year promoting them. If "this is album X by band Y" were as good as the picture, that is what you would see on the album cover. There is no way in which these two in particular harm the free mission of Wikipedia. They don't make it any less possible to re-use our content, don't make it more likely that anyone will be sued, and don't discourage people from creating free equivalents. All of this is just a matter of free content theorists being so dogmatic that common sense flies out the window. In fact, the coined notion that anything primarily visual in nature is "decorative" use is a misunderstanding of the role of graphics in communication, utterly irrelevant, and unhelpful in analyzing either free content or fair use.Wikidemo 12:34, 30 August 2007 (UTC)
Luna, Nodekeeper, and others: Wait a second here, I'm giving my support to fixing our current situation and reducing the silly red tape. In my mind, we can still give a rationale statement, but do so in a more efficient, effective way, and making the entire process of uploading a non-free image easier to the new user (and the experienced user). For a while I was dead on just keeping the same requirement we have now, because I thought that we might be going off into a slippery slope, but I no longer feel that way. It took a while, but a lot of ideas, such as the ones brought up by Wikidemo, have really started to grow on me. I think we can make things easier, but still ask for an explanation from the uploader in some way. In any case, even if it's a selection from a drop down menu, you are technically receiving a rationale from the uploader. What I would like to see us do now is explore what images/uses will work with other methods of generating rationales, since I don't think one size will fit all. -- Ned Scott 04:24, 30 August 2007 (UTC)
Well it would probably require some software changes (or at least some javascript hacks), but having some kind of checkbox or drop down list for broadly accepted uses for scertain types of images on the upload page doesn't sound like that bad of an idea. How about something like this: If you choose a non-free option on the upload page we pop up a (inline to avoid triggering popup blockers) context sensitive "rationale wizard", if you choose for example albumcover it will prompt you for an article name and then let you either fill in one of the pre-written rationales for acceptable use(s) of that kind of image via one or more radio buttons or something, or just write something manualy. Maybe throw in an "add rationale for a second article" button too and have the script fill in the rationale template(s) and put it all in the upload summary box ready to submit once you hit "done". The boilerplate options offered should only be the most obvius uncontroversial uses naturaly (logos in main company article, cover art in the article about the album/book/DVD/whatever etc), using for example a logo in a different place would still require a hand written ratoinale. --Sherool (talk) 08:07, 30 August 2007 (UTC)
Thanks, Ned. Sounds like your approach would probably have to involve developer attention at some point, but it sounds like we might be heading in the same direction. I can definitely sympathize with the concern on slippery slope; the sort of paradigm change I'm hoping for, I guess, is to move to a more helpful stance, making it clear that fair (ab)use needs to be limited, but hopefully directing the bulk of our efforts toward helping people who can be helped, and enforcing NFCC in cases where it's being abused. – Luna Santin (talk) 21:30, 30 August 2007 (UTC)

Confusion about the NFCC and Board's language

Some of the confusion about the Board's Licensing Policy Resolution, it seems to me, arises out of lack of clarity about the word "media". Wikidemo pointed out above in this section that "Kat Walsh recently clarified that the foundation resolution is not intended to impose a requirement for explicit written rationales for each image or image use." The main copyright problem these days is the widespread dissemination of audio and video clips under "fair use", a practice which if not kept to a minimum can run right into the teeth of modern copyright law. Also, a vitally important issue with these audio and video media is faithful representation of the source where the uploader got the image from. (See, e.g. this section of the US Code, which deals with the need to diligently protect copyright management information (CMI).) The most recent series of changes in modern copyright law were, in fact, primarily driven by the increasing proliferation of reproductions of whole songs, movies, etc. on the web. Still images, particularly low resolution images that have faithfully represented where the image was found, are a very substantially different issue. Thus, in my opinion it may be necessary to more extensively discuss the specific language of the NFCCs, which fail to effectively differentiate between reasonable use of low-resolution still images that do not detract from the copyright holder's ability to make money from the image, and indeed often enhance the ability of the copyright holder to make money from the work from which the image is drawn.

This issue, along with a number of others, in my opinion, speaks to the need to continue the discussion about the language of the NFCCs, particularly now that several months worth of practical experience allows an assessment of where some of the practical problems arise. The failure to differentiate between low resolution still images and audio/video media surely is one of the problems that we've had an opportunity to identify, and IMO should begin to address more specifically.

Among the other issues that I think will need to be revisited and addressed is that the Board, in its March 2007 Resolution, never used words like "as little as possible", "necessary" (or "essential"), etc., words that have proven to be extremely divisive on the wiki and which indeed appear to have contributed to unnecessary arguments that in many instances have been somewhat corrosive to the collective good will among users, and which might not occur to anywhere nearly the same extent if these issues were better clarified. I recognize that there is a significant contingent which believes that no fair-use should be permitted on the wiki, and who will seek the strongest possible language weighted against fair use of anything other than explicitly "fee-license" images or other media. But nonetheless, these issues about differentiating low-res still images (album covers, book and magazine covers, photos of non living persons, iconic images, historically important persons, etc.) from the comparatively dangerous minefield of audio-video media, I should think, really needs to be discussed and clarified and the NFCCs such as #1, #3, and as mentioned, #10, rewritten to whatever extent may be consensused to be appropriate in light of the experiences of the past five months since the Board resolution. ... Kenosis 18:03, 30 August 2007 (UTC)

They may not have used the words "as little as possible", "necessary" or "essential", but they did use the words "minimal" and "within narrow limits". (Both in paragraph 3) -- But|seriously|folks  18:59, 30 August 2007 (UTC)
Yes, definitely true BSF. It appears they may have been directing those words primarily at the audio-video and other "media" files. It's a potential source of great confusion. For instance, Videmus Omnia and I had to clarify what's meant in a recent conversation with one another. If I remember our conversation correctly, he'd become accustomed to thinking of "media" more-or-less what I think the Board was referring to, i.e., the audio, video, audio-video files, and moving gifs and such (correct me if I'm wrong, VO). And I was using "media" to refer to all media including even the low-res still images. Now, in light of the clarification from Kat, I begin to sense Omnia's use was more in line with the intent of the Board in its resolution. This issue, IMO, does need better clarity and, I think, further clarifications from the Board in seeking better clarity in the NFCCs. ... Kenosis 19:09, 30 August 2007 (UTC)
I agree that it is unfortunately ambiguous. Clearer guidance from the Foundation and its legal counsel in this area would help save us all a lot of time and energy. -- But|seriously|folks  19:17, 30 August 2007 (UTC)
The resolution uses "media" to refer to anything contributed to the projects that isn't text. Images are clearly included. There isn't much point in rephrasing the policy to be more permissive when we are already sinking in non-free content. --Abu badali (talk) 19:39, 30 August 2007 (UTC)
(EC) Granted "media" can have several meansing, but given the context I hardly think it's ambibous (we are not talking about the mass media, or a person who communicate with the dead or something in between two extremes and so on), a media is simply any format used to communicate or presnet information, as far as I know we have always used it simply as a "catch all" for stuff in the image namespace to ensure that things that are not actualy images don't get left out. --Sherool (talk) 19:55, 30 August 2007 (UTC)
I thought it was clear that "media" included images. Oh well . . . -- But|seriously|folks  19:56, 30 August 2007 (UTC)
It's pretty clear to me that "media" includes images. -- Ned Scott 20:19, 30 August 2007 (UTC)
Indeed, it certainly does. There's a lot that's commonly misunderstood in that statement, but that it includes images isn't part of that. WilyD 20:27, 30 August 2007 (UTC)

I did not anticipate that discussions of the needed clarifications would be without controversy. But a very strong indicator that the Board was aiming its sights more strongly at the audio-video media is the following passage from their resolution: "Non-free content used under an EDP must be identified in a machine-readable format so that it can be easily identified by users of the site as well as re-users." This is a passage that appears to refer to the practice of encoding copyright management information (CMI) into the file itself, as is increasingly being done by the copyright holder with music and video files. Jpeg's of still photographic images generally do not involve such encoding, unless someone has added it. We increasingly see this on the web, e.g. here, where the words "" identify the fact that the image passed through their hands, so to speak. And if the image is uploaded from a source page that does have CMI or other identifications encoded into it, we're expected to leave that CMI or other encoding in the image, not use, say photoshop, to remove the info. At least that's my take on it. This leads me to think that the Board's emphasis was more on the audio-video and other files commonly encoded with copyright management information than it was to still images. And, of course, with still images that do not have source info embedded into it, we're expected to faithfully represent the source from which we obtained the image and any additional relevant information on its progeny. At least that's my take on it. As I was trying to say, I think we do need clarification on this. ... Kenosis 20:38, 30 August 2007 (UTC)

The entire nonfree media policy applies to all media, including images. For images (and indeed other media files), the "machine-readable" requirement is satisfied by having a nonfree license template on the image page, since a program or script can easily look for such templates. It's not required that the information be encoded into the file itself. Seraphimblade Talk to me 23:43, 30 August 2007 (UTC)
Surely, Seraphimblade is correct here. Machine-readable means it has a distinct template so the non-free content can be identified and filtered out automatically, such as by using special tags for non-free images (as we do). -- But|seriously|folks  23:48, 30 August 2007 (UTC)
Ahh, yes. OK, that particular language becomes clearer now. Thanks. So we can agree that media refers to all media including still photographic images. ... Kenosis 00:42, 31 August 2007 (UTC)
I think there's a consensus on that, anyhow. WilyD 18:02, 31 August 2007 (UTC)

Album covers in band articles, again

I feel sure this has been discussed before but I cannot find the relevant archive. My understanding of the licensing of nonfree cover art is that its use is only normally permitted on the article about the album itself, and not on the one about the band (or any other articles such as discographies).


3. (a) Minimal use. As little non-free content as possible is used in an article. Short rather than long video and audio excerpts are used. Multiple items are not used if one will suffice; one is used only if necessary.
8. Significance. Non-free content is used only if its presence would significantly increase readers' understanding of the topic, and its omission would be detrimental to that understanding. Non-free media files are not used if they can be replaced by text that serves a similar function.
    • The focus of both of these items of policy is to limit the use of copyrighted, fair use imagery as much as possible.
  • "It is believed that the use of low-resolution images of such covers solely to illustrate the audio recording in question..." (from the boilerplate nonfree licensing text displayed on album cover images).


I have previously used this rationale to remove nonfree album covers from many band articles, including Pink Floyd. It has come up again because another user was advised by User:Moe Epsilon (in relation to the Rolling Stones article) that "Just don't add every album cover to every bit of text and it shouldn't be an issue. :)". Moe is very firmly maintaining that retaining the images on the Spice Girls article is vital; I don't see it that way and I thought I would invite wider comment from the community. Quite apart from violating our own policy on nonfree images, and the directive from the foundation, I think using images this way in band articles is lazy and unencyclopedic. Unless there is genuine critical commentary in the article on the album cover (as opposed to the album) I cannot see how it can ever be justified. What do others think? --John 16:09, 31 August 2007 (UTC)

Is that what you think? That commentary on the physical cover needs to be provided? Please read this page carefully:
Cover art: Cover art from various items, for identification only in the context of critical commentary of that item
Emphasis mine. of that item doesn't have to refer to commentary of the image in which it represents. The cover art of a CD isn't what needs to be discussed in the Spice Girls article, it's commentary of the actual albums. How many articles do you see that even remotely go into detail about the physical cover art. Most, if any, don't. Album covers wouldn't be existant at Wikipedia if we were to write commentary on every image or the significance of that image is. — Moe ε 16:24, 31 August 2007 (UTC)
The position is advocated by one or two people that the actual cover needs to be discussed, rather than the album. This is transparently wrong, and doesn't happen in practice, but it still gets argued from time to time. WilyD 16:41, 31 August 2007 (UTC)
Thanks for the comments. I'd be more interested in an opinion that refutes the minimal use policy of Wikipedia:Non-free content criteria (though I don't really see how that is possible). I started out thinking as you guys do; unfortunately (in my view), it seems the project has a clear policy on this issue which we are duty bound to follow. The existence of articles which breach a particular policy or guideline has never been a good argument against any such policy or guideline. --John 16:48, 31 August 2007 (UTC)
It's not clear what you're looking for. There's no need to "refute" the miminal use criterion, but realise that the minimal use criterion can be interpreted to mean that album covers can be used in articles about albums, sections about albums, et cetera, without the need for specific discussion of the cover art specifically. The foundation has not clarified the position, although Jimbo has said that cover art is appropriate and sensible in articles about albums, for instance.
In conclusion, ues of unfree media should be minimal. Minimal does not mean zero, and what it does mean is essentially an opinion. Some cases are very clearcut, others are not, and even in fairly clearcut cases there'll be people with two or three sigma opinions. I don't think minimal use as a rule is up for discussion, just what it means in implimentation. WilyD 16:57, 31 August 2007 (UTC)
My interpretation is that an album has to be discussed rather significantly in order to include the album image in the artist article. Certainly, a mere mention or sentence is insufficient, unless it relates to the art itself. Of course, if the album is discussed rather significantly, it probably deserves its own separate article, where the cover would more properly appear. YMMV. -- But|seriously|folks  17:39, 31 August 2007 (UTC)
In your opinion, are the images being properly used at the Spice Girls article? — Moe ε 17:47, 31 August 2007 (UTC)
Definitely not for the Greatest Hits, and probably not for the others. There are two or three paragraphs of text in each album section, but only two or three sentences pertain to each album and those generally review sales figures, which does not consitute the critical discussion necessary to use the images. -- But|seriously|folks  18:12, 31 August 2007 (UTC)
I would agree with that statement. The Spice Girls article has two or three large paragraphs of text per album. WilyD 18:00, 31 August 2007 (UTC)
(←) I don't think that the images of album covers at Spice Girls contribute to my understanding of the band or their music. Surely the appearance of an album cover, in all but a few cases, is of only marginal importance for a group whose purpose is to produce music? Even nonfree images of the band performing would contribute more to the article. — Carl (CBM · talk) 18:04, 31 August 2007 (UTC)
Aye, they would. But you're talking hypothetically. Surely we don't remove images based on hypothetical situations. — Moe ε 18:21, 31 August 2007 (UTC)
The images in the article are not hypothetical and don't significantly contribute to an understanding of the topic of the article. Ergo they ought to be removed. — Carl (CBM · talk) 18:25, 31 August 2007 (UTC)
No, no. The hypothetical images of the band performing your alluding to. You said "Even nonfree images of the band performing would contribute more to the article.", so we remove images based on free images that haven't been made yet? — Moe ε 18:27, 31 August 2007 (UTC)
We remove nonfree images all the time if they could be replaced by free images that don't exist yet, but that isn't what I'm saying here. I'm not claiming there are any images to replace these, but I still don't think they impart any significant understanding to the reader about the topic of the article - the Spice Girls. In other words, I'm looking at WP:NFCC#8, not #1. — Carl (CBM · talk) 18:32, 31 August 2007 (UTC)

(deindent) I agree with CBM here. These images are basically being used as decoration here. Quite apart from the (I believe highly dubious) legal status of our using these images this way under fair use, I also think this is a lazy way to illustrate a band article. Especially for a recent band like the Spice Girls, there must be free images out there that we could use. The continued existence of the album cover images on the article may well discourage people from adding better pictures. Just another reason for taking them down. --John 20:44, 31 August 2007 (UTC)

Here is an archived discussion on this from January, the last time I remember serioously discussing this matter. Here is User:Durin agreeing with my point of view on this issue. Here is the Foundation decision I mentioned earlier. --John 20:52, 31 August 2007 (UTC)

Checking Commons first!

Just a timely reminder for those participating in IfD debates to remember to check Commons for free replacement images. See this edit here, where, after a long, long debate, it was eventually discovered, just after the debate closed, that the free replacement image someone found late in the discussion, had been on Commons since July 2006! Copied to WT:IFD. Carcharoth 00:09, 1 September 2007 (UTC)

Image used in multiple articles

I disagree with the following clause currently present in the guideline as an example of inappropriate use:

This seems to go against the policy, in which it appears to be expected that an image can appear in more than one article, as long as it meets the criteria in all instances. This clause also runs the risk of going against the minimal use clause WP:NFCC#3. The intent of the minimal clause is to limit the number of different images. It is desirable to have one image used in different articles, rather than have different images for each case. Duae Quartunciae (talk · cont) 11:13, 27 August 2007 (UTC)

I don't see how using the image in only one article rater than two would go against any of the non-free policies... Sure the same image can be used in more than one article provided it meets all the criteria (for example a picture of two fictional characters could be used in each of theyr articles), but one of the criteria is minimal use, and if there is an article spesificaly about the image there is no need to inline it as illustration in other articles that mention that image, just link to it's article. If there are significant discussion about the image in two different articles most of it should probably be merged to the image article anyway. Using an the image in only one article rater than two sounds like it's entierly in line with mininal use to me. --Sherool (talk) 14:18, 27 August 2007 (UTC)
It does not go against the policy, so much as add an additional criterion not in the policy, and unreasonably limiting on non-free images that have been recognized as acceptable. The minimal use criterion is all about limiting the amount of a particular non-free item; and makes no mention of limiting the number of articles in which it may be used. The policy already has tough requirements, which must be satisfied for every article in which an item is used. The irony is that restricting an image from an otherwise well justified use, simply because of its use in a different article, might actually work against minimal use, in that there becomes a motive to look for a second image that is not used elsewhere.
Minimal use, as described in policy, is best satisfied by ensuring that if there is good case for using non-free content under the policy criteria, then the image is best to be one that is already accepted for use in another context. Duae Quartunciae (talk · cont) 14:59, 27 August 2007 (UTC)
From a copyright perspective, each I believe each individual placement of the image counts as a separate infringement for which a fair use defense has to be provided. It doesn't matter whether you have three infringements of picture A or one infringements each of pictures A, B, and C. Hence if a disgruntled editor replaces the removed picture with something else, it's the same situation, not worse. Therefore, this rule can potentially only improve the situation, not make it worse. (Though, I'm still not sure whether I support it.) nadav (talk) 11:27, 28 August 2007 (UTC)

I was kind of proud of the way I phrased that :(. The example dovetails with the prohibition against lists and galleries, and also a style issue, that it is better to link to the main article about a topic than to have a mirror or alternative discussion about the topic outside of its main page. Thus, in an article about peanuts you wouldn't have a section complete with pictures of Snoopy, Charlie Brown, and Lucy van Pelt (Note, how the real article does some of that, though). Rather, if you do have a list or section on the characters you keep the list short and link to the main articles. By contrast, it is okay to use the image of the painting No. 5, 1948 in the article on abstract expressionism because it's not used in the context of a redundant section on the painting, but rather to illustrate the characteristic style of the movement. Please note the intent is for users to omit non-free images entirely in favor of a link, not to encourage them to hunt for an alternate non-free image. Indeed, using an image more than one place when one place will do, or an image when a link will do, both run against the spirit of minimal use (even if they're not a clear violation of that policy). However, I don't have a strong feeling one way or another. Wikidemo 17:13, 27 August 2007 (UTC)

If our goal is to generate reuseable content, then images should appear everywhere they're needed. A CD version of Wikipedia's content, or a print version will almost certainly drop a lot of "minor" pages, which probably includes most specific image pages. WilyD 15:47, 28 August 2007 (UTC)
Yes, they should appear everywhere they are needed. But if there is an article on a particular person, there's no need to put a nonfree image of the person on every article that mentions the person, since we expect readers to follow links when they want to learn more. The same applies to fictional characters, works of art, etc. That's what was meant by "article about the image". — Carl (CBM · talk) 15:54, 28 August 2007 (UTC)
There is another possibility, that a work may be iconic and instantly recognisable to a large number of visitors, but that its name might not be, for example Gravitation (M. C. Escher). In such a case, and where another page has a text link to the main critical discussion, it is sensible for the link to be accompanied by a thumbnail image. (OK that's not what paper encyclopedias tend to do, but paper editors are grouchy about space and print costs). -- Steelpillow 20:30, 28 August 2007 (UTC)
Nobody has responded to my last comment, so unless someone does so soon I will take it as acceptable, and modify the guideline to something like:
-- Steelpillow 10:50, 1 September 2007 (UTC)
I think this would cause more difficulty than it prevents. One issue is that although you feel that that image is better known than its name, others may not. Establishing that an image is well known, but its name is not, would be difficult. The entire section on "acceptable images" is meant to describe images that are acceptable to be used even once; an image about an iconic piece of art can be used in the article about that piece of art. But the minimal use provisions of the policy at the top still apply. — Carl (CBM · talk) 12:03, 1 September 2007 (UTC)
How do we judge whether an image is better known than its name? Well, how do we judge whether an image "has achieved iconic status as a representation ... or is historically important in the context..."? I would suggest that whatever manner we judge the second instance (which remains in the guidelines, despite efforts to remove it) should also be how we judge the first (which I wish to put in the guidelines). As for minimal use, it has been pointed out before that the policy statement refers to the use of many different images on a single page, which is not relevant to the use of a single image on multiple pages. I can find no policy on this other than the unspoken default of taking each individual usage on its own merits - which is precisely how I wish to continue. So I really do not see the problem. -- Steelpillow 18:57, 1 September 2007 (UTC)


Galleries of non free images not significant for an article are, for good reasons, not allowed. But what would happen if several images, all with acceptable non free use rationales were used in an article, and the editors decided instead of having the images scattered throughout the article, it would make more sense to present them as a small gallery? Presenting them this way would be purely a layout decision, not a non free use decision. Considerations like this make me think that the automatic ban on galleries misses the point. If the number of images in the gallery is excessive, or the images in the galleries are not discussed in any way in the article, or not needed for identification, then there are perfectly adequate reasons to cite for removing the images, instead of crying "no galleries". I think the presumption is that galleries are used to add more images to an article, over and above the normal ones, but this misses the rare possibility that the images used in an article might be presented in gallery format, instead of at various points throughout the article. My feeling is that galleries should, in limited circumstances, be allowed if the gallery is accompanied by text commenting on, discussing and supporting the images. Is this feasible? Carcharoth 17:54, 30 August 2007 (UTC)

The images would ideally each be supported by accompanying text. Putting each of the images beside its text makes more sense than gathering them at the bottom. Do you have an example of an article that has valid nonfree images but puts them in a gallery? — Carl (CBM · talk) 18:02, 30 August 2007 (UTC)
Actually, Padmé Amidala uses two non-free image in a gallery where the gallery is beneath the text rather than beside it for what is basically a formatting issue. You could reformat the images to be on the side of the section rather than the bottom, but it'd inhibit readability. WilyD 20:02, 30 August 2007 (UTC)
Thank you. A perfect example. Now, how can this point be made to people who seem programmed to say "non free gallery = bad". What they probably mean to say is "non free images not contributing to the article" (NFCC#8) or "excessive use of non free images" (NFCC#3). In most cases, non free image galleries will be examples of this, but not all cases. So instead of saying "it's a gallery, it must be bad", people should be removing the images per NFCC#8 and NFCC#3. It seems to me that "removing gallery, non free image galleries are not allowed" is a misleading shorthand. People could easily think that the same images could be re-inserted into the article, spread throughout the article, when this is rarely the case. The guideline already says "The use of non-free media in lists, galleries, discographies, and navigational and user-interface elements normally fails the test for significance (criterion #8), and is thus unacceptable", but people should judge each gallery on its NFCC#8 basis, rather than blindly assuming a gallery is bad. Carcharoth 22:39, 30 August 2007 (UTC)
  • If that can be clarified in the policy, then great. Problem is, we get into mysterious definitions of what is and is not significant commentary. We're having constant battles about that. --Durin 22:51, 30 August 2007 (UTC)
  • As in any case, rules have exceptions. I don't see any particular problems in the Amidala article, as the images are genuinely contributory to significant commentary and add greatly to the understanding of it. I don't see any particular problem with a gallery, in that instance. However, 99.9% of galleries are just "Let's throw together 20 screenshots of this (video game|movie|etc.)!" It's understood that all policies have exceptions in a few cases, but in the vast majority of cases it's a good rule. Seraphimblade Talk to me 23:40, 30 August 2007 (UTC)
In the Amidala article, the gallery is OK because the images relate to the text in that very section. If the discussion of fashion was higher on the page, it would make less sense to have a gallery there. Similarly, it would make less sense to have an image of her fighting style in the fashion gallery. — Carl (CBM · talk) 00:00, 31 August 2007 (UTC)
Would you believe that I thought to myself, while writing the original post, "I bet someone will say 'there's always the exception that proves the rule'"? Anyway, I'm not talking about changing the rule, merely making clear that it is not an iron-cast rule. In fact the current wording already does this, so Seraphimblade and I obviously both failed to read the current wording: "The use of non-free media in lists, galleries, discographies, and navigational and user-interface elements normally fails the test for significance (criterion #8), and is thus unacceptable.". This kind of wording is useful when trying to politely reason with people who are convinced that what they are demanding is beyond reproach and is policy. Carcharoth 23:39, 31 August 2007 (UTC)

Following the above discussion, I've made a very slight tweak to the wording, adding "nearly always", as see here. Hopefully that will be OK. Carcharoth 23:43, 31 August 2007 (UTC)

The change looks okay, though I thought that was implied by the use of "normally", i.e. normally fails #8 and is thus unacceptable means normally (fails 8 and is thus unacceptable), as opposed to (normally fails 8) and is thus unacceptable. I see no harm to making that clear but I usually don't like seeing words like "never" and "always" in the policy and guideline pages, even with a qualifier. Does "nearly always" mean fewer exceptions than a typical guideline rule? Wikidemo 16:50, 1 September 2007 (UTC)
Well, I agree that "always" and "never" are not good words to use on guideline and policy pages, but in this case, what I found unacceptable was the word "unacceptable"! But this is moot because I've just realised that "normally" should be "usually" - how can a failure be "normal"?? Carcharoth 01:36, 2 September 2007 (UTC)

Kat Walsh's clarification

Can you point me to where the original clarification by Kat Walsh, mentioned by Wikidemo, concerning individual images each having specific "fair-use rationales" is located? Thank you. ... Kenosis 03:32, 2 September 2007 (UTC)

I believe that it was at the foundation-l mailing list. --Iamunknown 04:07, 2 September 2007 (UTC)
Yes, that's it! It was first mentioned on this page in the archive here. Wikidemo 04:14, 2 September 2007 (UTC)
Appreciate it, Iamunknown. Wikidemo? Who's that? Am I on the right page? Iamunkown, I thought we weren't speaking with each other for at least a couple more days.  ;-) [wink and a smile. Thanks for the link] ... Kenosis 04:19, 2 September 2007 (UTC)

Promo images as de facto free images

Promotional images are de facto free images. If an artist or company releases a certain image as promotional, it means the media is fully authorized to use said image to illustrate articles about said artist or company. Thus, I believe they should be allowed in Wikipedia without restriction. -- Stormwatch 06:27, 21 August 2007 (UTC)

That does not meet the criteria of free images. Free images must be free for anyone to redistribute, modify, and use for any purpose, commercial or noncommercial. Promotional images would generally not be allowed for use in commercial purposes, and it's also generally forbidden to modify them. Seraphimblade Talk to me 06:34, 21 August 2007 (UTC)
People suggest this from time to time, but it's not going to happen. We need an explicit release under terms compatable with a free license in order to consider something free. What you believe theyr intention is doesn't rely matter. Sometimes people are in fact prepated to release promotional or simmilar material under a true free license if you just ask them nicely (especialy independent artists who don't have layer upon layer of corporate legal beurocracy to deal with), they may simply be unaware that it is an option, but unless we have that comfirmation from the copyright holder directly we can't simply make asumptions about how freely they intended to allow theyr material to be used. --Sherool (talk) 07:14, 21 August 2007 (UTC)
The problem with this sentiment is that copyright lasts too long. There is publicity material from old TV shows and movies that will not fall into the public domain for another 50? years. So does that mean that wikipedia has a barren article page so that it might hold up its 'ideal' of completely free content? No, it doesn't, because that would mean that Wikipedia's purpose is not to build an encylopedia, but to build up the 'GFDL'. Building the 'free' is wonderful where it can be done, but not at the expense of building an 'encyclopedia'. The two are not mutually exlcusive. -Nodekeeper 03:50, 23 August 2007 (UTC)
If we are successful at building a free encyclopedia, people will release free images for us to use, because they want to have them in the articles on their shows. If we use nonfree images too often, people have no incentive to release free ones for us. — Carl (CBM · talk) 03:55, 23 August 2007 (UTC)
This rationale is flawed for a number of reasons. But I'm not going to restate something that has most likely been argued a mumber of times anyway. -Nodekeeper 08:31, 24 August 2007 (UTC)
Some people actively seek to be anal-retentive, sticking to the minutiae of formality and not bending to reasonability. Most are lawyers, but some are Wikipedia admins. Hope you'll never have to meet one in person! -- 23:09, 24 August 2007 (UTC)
Note that by definition if you are stating people are not going to release their images under free licenses for a variety of reasons you are agreeing that promo images are not 'free' as expected by our free content goal. If you are trying to argue that we should allow promo images even though they are not 'free', that is a completely different argument from the claims of the original poster that promo images are defacto free Nil Einne 15:20, 1 September 2007 (UTC)
Will someone explain to me... why is there a "Non-free promotional" tag if promotional images are not supposed to be used?! --Stormwatch 23:45, 24 August 2007 (UTC)
Many editors such as you and myself were uniformed about the policy changes that a large cabal of editors have put into effect (following a policy statement from above). The only way that it changes is if a large number of editors stick around to change it. I am not hopeful. BTW the 'non-free promotional' tag will most likely eventually be deleted. Somebody is welcome to correct me on this. -Nodekeeper 00:02, 25 August 2007 (UTC)
In some cases, nonfree promotional images are acceptable, just like any nonfree images. But they still are nonfree images, and must meet all the standard requirements (minimal use, irreplaceable, non-decorative, etc. etc.) The template just illustrates the copyright status of such an image. I doubt if it will get deleted. Seraphimblade Talk to me 23:38, 26 August 2007 (UTC)
Try modifying a promotional image in some degratory way or if it's a person perhaps use it to make a nude image and find out for yourself whether it's really a free image. Or perhaps try printing some mugs, tshirts and stuff with that image. You'll likely quickly find out it probably isn't and it'll in fact very likely be considered a violation of the copyright in addition to any publicity rights that may be applicable in the jurisdiction if it's a person. Indeed a simple read through of the conditions release with such an image should be enough. Also, if promo images are 'free' then why is it so hard to get people to release them under free licenses? Clearly they are not free. Nil Einne 08:40, 1 September 2007 (UTC)

User Seraphimblade, above, is absolutely right. There is no justifible reason for admins and editors to go around "cleansing" Wikipedia of images, especially ones that are used for illustration and/or review purposes, such as album covers (see the ludicrous discussion below about this) and publicity photos, which are both used in print and broadcast media all the time without any problem at all. I think, as was suggested above, that some editors and admins have lost sight of the purpose of this site - to build an encyclopedia. Building up the "free image" cult is not its goal, or wasn't, anyway. Further, pointing out that these images aren't free to alter in any way one chooses misses the point, and doesn't prove anything, except the opposite - That no band or artist is going to release images into this mythical GFDL universe. If they do, they lose ALL rights to the image. Does that make any sense? But publicity photos, album cover images, etc, used solely to illustrate an article, are indeed allowed. Confusing these two issues is a full-time job for some editors, and I'm hoping people eventually wake up before these busybodies denude this site entirely of images. - Nhprman 17:55, 2 September 2007 (UTC)

You may want to check Wikipedia:WikiProject Free images for a WikiProject that tries to contact copyright holder like agencies and discographies to release promotional material under a free license. -- ReyBrujo 18:44, 2 September 2007 (UTC)

Tagging for review?

Can we please get rid of this sub-section: WP:NONFREE#Tagging_for_review?

It's been listed as a "proposed" addition for some time and it seems to be inactive at this point. If we do adopt it, we should link it to a separate page because it's a voluntary process that only some people will use, not something everyone needs to see when they want to read our non-free guideline rules. And if it's still under discussion we should put it on a proposal page instead letting it clutter up the guideline page.

In case anyone wonders why I care I'm slowly copy-editing the non-free guideline page in order to try to make it more readable and bring it down to manageable length. Slowly, so that if I make a horrible mistake or anyone has a big objection they can stop me before I mess up the whole page. I just took a pass at the "non-free image tagging" section, and the "tagging for review" section is next. It seems to add nothing, so rather than editing I would like to move it off the page. Thx. Wikidemo 13:34, 30 August 2007 (UTC)

Anyone? Remember, silence=consent!Wikidemo 16:43, 1 September 2007 (UTC)
Okay, removed. If anyone wants to save this content beyond the edit history, please put it on some auxiliary page. Wikidemo 20:18, 2 September 2007 (UTC)

Living people and replaceable images (Bobby Fisher)

Concerning living people and replaceable images, I recently came across Bobby Fischer. We have a picture of him from 2005, complete with the beard. Any reputable encyclopedia would look to include a picture of him from the famous chess world championship match in 1972. It would seem to me that a fair use case could be made for including a picture of him playing in that match, such as the one seen here. What do people think? I was going to be protesting this article as an example of a free pic being misleading, with the "old man photo" presenting a misleading POV, and how including a fair-use pic from 1972 would redress the imbalance and restore NPOV, but I'm unsure as to whether NPOV applies to pictures. Howerever, rather ironically, while looking for a picture of Fischer from 1972, I came across the image we have (Image:Robert Fischer - March 2005.jpg - marked under a Flickr CC license), on a BBC article marked as an AP picture: here (different cropping).

This leads to several questions:

  • (1) Does anyone agree with me that our picture (from Flickr) is the same as the AP picture?
  • (2) How rigorous is the Commons process to check Flickr pics? I followed the link to the Flickr page given as a source - see here, and while the Commons tag only said that the licence was verified, what about checking whether the person uploading took the picture? As far as I can tell, that Flickr photo was uploaded by ChrisL_AK/Chris Lott, an Instructional Developer from Fairbanks. My guess is that he followed the news coverage in March 2005, grabbed the pic from a news website, and uploaded with the comment "Sad". I've dropped a note to User:Nilfanion, asking him to comment here. See User talk:Nilfanion#Commons image from Flickr - possible problem.
  • (3) If we had a free pic of Bobby Fischer, would the "his significance derives mainly from 1972 event" argument be good enough to justify using a fair-use pic of him from that event even if we had a free pic from earlier or later? What would happen if, for example, the only free pic of him was as a 10-year-old child?

Any answers? Carcharoth 01:29, 3 September 2007 (UTC)

(1) Our photograph (from Flickr) does indeed appear to be the same AP photograph. (2) The Commons Flickr-checking process is not often rigorous enough to determine if the Flickr uploader can legally license the image under a free license. (Such images are occasionally caught, but it is usually only by chance.) (3) I don't know enough about Bobby Fischer to offer an opinion. --Iamunknown 02:33, 3 September 2007 (UTC)
According to World Chess Championship 1972, it was considered the Chess Match of the Century. --tjstrf talk 02:54, 3 September 2007 (UTC)
(1) Thanks. I will remove from the article and nominate for deletion at Commons. (2) "usually only by chance"? It took me all of two seconds to click on the source link and think to myself "hang on, this is not a picture that this Flickr user is likely to have taken" (I would have expected the Flickr user to have it amongst a collection of pics taken while on holiday to Iceland). A few minutes of investigation would have revealed that the picture was a copyvio from a news photo agency. (3) I'm of the opinion that even if someone took a picture right now of Bobby Fischer in Iceland, a picture of him playing at the match in 1972 would still be OK under our NFCC. Carcharoth 03:20, 3 September 2007 (UTC)
By chance I meant that, since a bot verifies the licensing status of many images uploaded from Flickr, a great many of them are likely incorrectly licensed on Flickr. Many images are never (or, at least, not originally) verified by a person knowledgeable in copyright law or even by a personal knowledgeable in common sense.  ;-) --Iamunknown 03:27, 3 September 2007 (UTC)
Yes, agreed as to both points -- that many are incorrectly licensed, and that many were never verified in the first place. I surmise that there may in the end be more work to do verifying free-licensed images than there is work to do on the sum total of all the NFC used on the wiki, both presently and in the future. Just a speculation at this point. ... Kenosis 03:45, 3 September 2007 (UTC)
The primary purpose of Flickr review was, and always will be, an assurance against Flickr-license rot. Detecting copyvios is a bonus, not the objective - which is why the review tag says "it was available on flickr with this license" as opposed to "it was available with this license". As it is a simple test, that is why a bot is ideally suited to the task. Incidentally when the bot confirms, that means it found valid EXIF data; which isn't likely on copyvio images.
Kenosis is likely right in his above comment - I wouldn't be suprised to hear there are more non-free pictures tagged with {{PD-self}} on Commons than {{Restricted use}} ones on WP... Regarding this image, I was just fixing an invalid change-of-license (by the uploader). Best approach for an image on this subject would be to write to him...--Nilfanion (talk) 10:26, 3 September 2007 (UTC)

Regarding the issue of illustrating the 1972 match: I don't see how a picture aids the reader's understanding that Fischer played chess or won the match. If we had a free photo, since it would do no harm, I wouldn't argue against it. But if we only have a nonfree photo, I don't see a strong justification for using it given our principle of minimal use of nonfree media. The same applies to photographs of sporting events, as far as I can see - a photograph of someone crossing a finish line doesn't increase the readers' understanding of the claim that the person competed and won. — Carl (CBM · talk) 12:29, 3 September 2007 (UTC)

I'm more in the camp here that if we are to use any non-free images to illustrate chess and chess matches, then the 1972 match is the one to choose. It would feel strange if all the others had an image and not this one. It would feel like the coverage was unbalanced. Similarly, if we only had free pics of Fischer as a 10-year-old child (theoretically these 'young' pics will fall out of copyright first) or as an old, bearded recluse, then using just those images would similarly feel unbalanced. I wonder if Wikipedia's "free content mission" might end up impeding the "NPOV" pillar (the requirement for balanced coverage)? Carcharoth 14:41, 3 September 2007 (UTC)
This image isn't illustrating chess, though; it's illustrating the appearance of Bobby Fischer in his biographical article. I don't think we need any nonfree images in the article about chess. That article has an excessive number of free images, as well. We don't need pictures of a person whenever the person is mentioned; they should generally be kept to biographical articles. — Carl (CBM · talk) 15:06, 3 September 2007 (UTC)
Sorry, that was sloppy wording. Of course it is not needed in chess. I meant the specific chess match article. World Chess Championship 1972 - quite probably the most famous chess match ever. Carcharoth 15:13, 3 September 2007 (UTC)
I guess I'm failing to see how the "kid picture" or "old man picture" would be "misleading"? Indeed, Fischer has been at least as famous and controversial for his life after chess than he was during it, and the article goes into significant detail about his later life and the controversy surrounding him. It's not like he dropped out of sight as that old man. (Good work on catching the Flickr copyvio though!) Seraphimblade Talk to me 17:00, 3 September 2007 (UTC)
I would have to see a proposed picture, but a picture of someone competing in an important contest is certainly appropriate to illustrate the contest, if either the image of the event is iconic (e.g. the picture of Brandi Chastain in her bra) or it illustrates something significant about the way the contest took place. If it looks like two stiff-looking and generic chess players, no need. If something about the image gets to the heart of what the contest was about, Fischer's state of mind, or what happened, that can't be described suitably in words, then of course the image is appropriate. That's a tall order but sports and other event photos try to get to that essence. If the result were the only thing relevant we would all do away with coverage of contests sporting events and just print a table of winners in the paper every day. Wikidemo 18:07, 3 September 2007 (UTC)
Note: There was no iconic Brandi Chastain picture in her article so I added one. Wikidemo 18:52, 3 September 2007 (UTC)
The "1972 match picture" I was referring to was this one. It's not particularly memorable or iconic though. There were appearances on the cover of Life and Newsweek. See here and here. Carcharoth 19:56, 3 September 2007 (UTC)
I'm thinking that those pictures aren't really necessary / don't contribute substantially to the material in the article. The last one in particualar is an excellent evocative documentary photo that shows his state of mind, but not in a way that's truly important to the contents of the articles. So in this encyclopedia given its rules, my opinion is that it's surplus and thus inappropriate as a non-free image. Wikidemo 20:18, 3 September 2007 (UTC)
That's fair enough. But what do you think of my point that once you start to introduce peripheral free pictures to an article (eg. pictures of Fischer as a child, or as a bearded recluse in the present era), the article starts to become unbalanced? Just as you would expect the text to be balanced and to cover all parts of his life, including his childhood and his current life, as well as his moment of fame as challenger and then World Chess Champion, surely a biographical article starts to become unbalanced once you introduce pictures from only part of his life, and not other parts of his life? My question can be best summed up long the lines of "how much to the pictures in an article contribute to the overall NPOV of the article"? I'm going to raise this over at WP:NPOV. Carcharoth 20:31, 3 September 2007 (UTC)


Why does the article have to be written in such arcane terminology and language? What exactly is a free content license (I tried clicking the link and it brought me to a page of links, do I have to read them all?! I think a lot of pages on Wikipedia need the "too technical tag" including this one. You shouldn't have to be a law school graduate! The snare 07:19, 3 September 2007 (UTC)

A free content license is a license under which content (i.e. works of art, literature, software, etc) is released with the freedom to do lots of things that are usually forbidden by "non-free" licenses. The Free content license page starts with a link to the free content page, which explains in more depth. I have just made that intro a little longer and emphasised the link. Hope it helps. Sadly, there are shedloads of different licenses, each with legal differences to keep the gobblygook professionals employed for a lifetime. If you ever find a way to simplify these pages - be our guest! -- Steelpillow 19:33, 3 September 2007 (UTC)

Well, I hope this one could be simplified so I could understand it better, as it's what some strict admins delete pictures based on, and he redirects people here. The snare 06:04, 4 September 2007 (UTC)

General overview is, a free content license must permit the image to be copied, modified, and reused by anyone for any purpose without asking permission. It may require that anyone who uses the image attribute the copyright holder (for example, "Photo by John Doe"), and it may also require that any future copies remain under the same license. It may not disallow commercial/for-profit reuse, modification, or require permission to be given by the copyright holder for use. Some common examples of free licenses are the GNU Free Documentation License, which Wikipedia is licensed under, and the Creative Commons Attribution and Attribution/ShareAlike licenses. Hopefully that clears it up a bit? Seraphimblade Talk to me 06:16, 4 September 2007 (UTC)

Image used in multiple articles (continued)

Nobody seemed to notice, so here is the recent bit again, edited for readability: -- Steelpillow 19:13, 4 September 2007 (UTC)

There is another possibility, that a work may be famous or even iconic, and instantly recognisable to a large number of visitors, but that its name might not be, for example Gravitation (M. C. Escher). In such a case, and where another page has a text link to the main critical discussion, it is sensible for the link to be accompanied by a thumbnail image. (OK that's not what paper encyclopedias tend to do, but paper editors are grouchy about space and print costs). -- Steelpillow 20:30, 28 August 2007 (UTC)

I propose modifying the guideline to something like:

-- Steelpillow 10:50, 1 September 2007 (UTC)

I think this would cause more difficulty than it prevents. One issue is that although you feel that that image is better known than its name, others may not. Establishing that an image is well known, but its name is not, would be difficult. The entire section on "acceptable images" is meant to describe images that are acceptable to be used even once; an image about an iconic piece of art can be used in the article about that piece of art. But the minimal use provisions of the policy at the top still apply. — Carl (CBM · talk) 12:03, 1 September 2007 (UTC)
How do we judge whether an image is better known than its name? Well, how do we judge whether an image "has achieved iconic status as a representation ... or is historically important in the context..."? I would suggest that whatever manner we judge the second instance (which remains in the guidelines, despite efforts to remove it) should also be how we judge the first (which I wish to put in the guidelines).
As for minimal use, it has been pointed out before that the policy statement refers to the use of many different images on a single page, which is not relevant to the use of a single image on multiple pages. I can find no policy on this other than the unspoken default of taking each individual usage on its own merits - which is precisely how I wish to continue.
So I really do not see the problem. -- Steelpillow 18:57, 1 September 2007 (UTC)
Indeed, if an image is so iconic in context it has it's own article, it's probably very reasonable to include it in the parent article. Think about downstream use - reusers of Wikipedia who want to distribute content on a CD or DVD (or print) won't include a lot of minor articles, so parent articles need to be as "complete" as possible. This is still a minimal use ... WilyD 19:26, 4 September 2007 (UTC)
It's not a minimal use in any sense. The "minimal use" provision (WP:NFCC#3) says "As little non-free content as possible is used in an article." If it's possible to link to another article where the image is included, then it isn't necessary to use it in the parent article. Our mission is to provide downstream users with free content whenever possible. — Carl (CBM · talk) 19:49, 4 September 2007 (UTC)
Actually, Steelpillow's comment wasn't 100% true at the time he/she made it but I've since changed the policy page so that it is, by moving prohibition on long audio/video samples from 3a to 3b. Now 3a is solely about limiting the # of non-free uses on each page, and 3b is solely about limiting the extent of each non-free use. The comment about it being a page-by-page issue, not a policy against use on multiple pages, remains true. Carl's right too. Minimal use means zero is better than one. As a note, the phrase "minimal use" is used in a much different sense than "extent of use" under the fair use law. Wikidemo 20:01, 4 September 2007 (UTC)
It's transparently minimal use - we want to generate free content for downstream users, but not "free" to the exclusion of "content". An article that needs an unfree image is justified in using it, regardless of what appears in other articles. Zero is more minimal than one, but where an article needs unfree media to explain a concept or whatever, directing the reader to another article is not an acceptable substitute. We cannot rely on subarticles being available. Articles need to stand alone. WilyD 20:57, 4 September 2007 (UTC)
(EC) I would say minimal use refeer to all use across the site. If people can follow a link to get more details on a subject then that's where we should generaly keep any non-free images who's purpose is to simply identify the subject. Yes there can be acceptable use other places too, but "visual recognition" for navigational purposes have been explicitly rejected as a valid use of non-free images in cases such as album covers and tv-show episodes and such, and I don't rely see the fundamental difference here. If you can justify how adding the image near the text link will greatly improve the understanding of the topic in the image rationale that might be acceptable in some rare cases. Generaly speaking though I don't think we should put anyting into the guideline that suggests that we generaly accept the use of non-free images whenever the subject is linked to as long as the editor believe the image will be more widely known than the target's name or title (that tends to be how it works, any number of editrs will line up and tell you how iconic the cover of theyr favourite album is, and how people will know it better than the actual title and so forth as a justification to sneak a dozen non-free images back into a discography list and simmilar things). --Sherool (talk) 20:49, 4 September 2007 (UTC)
Visual recognition for navigation purposes isn't a valid usage, I agree. But if every M.C. Escher work had it's own article, M C Escher would still require an example of two of his work to explain what he's all about. WilyD 20:57, 4 September 2007 (UTC)
Yes, a couple of examples as long as they can be said to increase the articles understanding is fine. I was mostly commenting on the "proposal" Steelpillow had, wich was not about giving examples of his work, it was explicitly about useing the image next to a text link of it's article, not because the image would be used as an example of the artist's work, but because a user believe it's name alone is not well enough known for people to know what painting the linked to article is about without having a tumbnail of it next to the link. IMHO "minimal use" defenently trumph that kind of use, and should continue to do so in the vast majority of cases (exceptions may always be possible in some cases if acompanied by a suitable rationale, but not as a general clause in the guideline). --Sherool (talk) 21:26, 4 September 2007 (UTC)
I'm not sure we understand Steelpillow the same way. His proposal specifically refers to a "passage about the image", which I visualise as a substantial piece of text about the image (or other media, it might well apply to including a sample of American Pie both in American Pie and Don McClean.
Actually, looking back at it now, his proposed language is really unclear. Maybe I misunderstand. Roughly speaking, I'll alledge the policy should say basically that each article can make its own case for unfree media, regardless of the existence of daughter passages. This is mostly for downstream use, but also readability. No special considerations of any kind (either inclusion for idenfication for navigation, or exclusion due to availability in daughter articles) is supportable under this policy. WilyD 22:15, 4 September 2007 (UTC)
NFCC dovetails with rules about article content. It's not good to create a satellite discussion about a subject outside of its main article because you get all kinds of forking problems like redundancy, wasted effort, divergent statements about the same thing, etc. Note how there's only limited disccussion of Guernica (painting) in the articles about Pablo Picasso, Guernica (town), or Bombing of Guernica. It's interesting that the Picasso has a second copy of the image but the others don't. Is that added use inappropriate? Non-minimal? More likely to lead to copyright problems? Bad for Wikipedia's free mission? I'm not sure. Wikidemo 22:58, 4 September 2007 (UTC)

I think we should not be trying to read too much into the words "minimal use". Basically, "fair use" is "minimal use" that will not interfere with the copyright holder's ability to make money from the copyrighted work. Is the ability of a book, magazine or music publisher to make money off a book or magazine or CD album diminished by the use of the image of cover because it's plastered all over Amazon and other vendors' websites? Quite the contrary, those images are put out into the public to help market the copyrighted product (which may raise the opposite question from certain quarters of Wikipedia -- specifically, Should we avoid helping commercial entities in their financial pursuits. Using a book, magazine or album cover image, particularly a low-resolution image, is fair use, minimal use, not because it's only in one article, but because it only represents a small portiion of the work, the most visible portion that the publsher puts out there to help sell the product. Use it in as many articles as you care to -- it's still minimal use. Two very notable cases are used in the US to illustrate the principle of accepted fair use, one with respect to text and the other with respect to images.

*With repsect to images, the following was considered to be fair use. A search engine's practice of creating small reproductions ("thumbnails") of images and placing them on its own website (known as "inlining") did not undermine the potential market for the sale or licensing of those images. Important factors were that the thumbnails were much smaller and of much poorer quality than the original photos and served to index the images and help the public access them. (Kelly v. Arriba-Soft, 03 C.D.O.S. 5888 (9th Cir. 2003).)

*With respect to text, the following was considered to be fair use. Publisher Larry Flynt made disparaging statements about the Reverend Jerry Falwell on one page of Hustler magazine. Rev. Falwell made several hundred thousand copies of the page and distributed them as part of a fund-raising effort. Important factors were that Rev. Falwell's copying did not diminish the sales of the magazine (since it was already off the market) and would not adversely affect the marketability of back issues. (Hustler Magazine, Inc. v. Moral Majority, Inc., 606 F. Supp. 1526 (C.D. Cal. 1985).)

... Kenosis 21:31, 4 September 2007 (UTC)

Do not confuse our non-free content policy with the legal requirements for fair use (this is why we changed the name from Wikipedia:Fair use. When the policy says minimal use it's to be taken quite literaly. We are a free content project and want non-free material to be used as little as possible, and only when not using it would make an article harder to understand. Just because something is legal doesn't mean Wikipedia policy allow it. --Sherool (talk) 21:58, 4 September 2007 (UTC)

I prefer the use of "the" to refer to NFCC, rather than "our" NFCC -- sounds too much like WP:OWN. My primary point is that fair-use is minimal use. The fact that the NFCC are in some ways more restrictive than the four fair-use criteria doesn't affect this part. NFCC #3, in saying "minimal use", is using a virtual synonym for "fair use", except that "minimal use" is not a common-law term of legal art in the same way that "fair use" is. The "minimal" part is fulfilled by using only a minimal portion of the copyrighted work, and is not affected by how many articles it's used in. The use of the words "minimal use" as a heading in NFCC #3 is, well, let's say it reflects the prior lack of knowledge about copyright considerations. Hopefully, this will pass as more regular participants become more and more familiar. ... Kenosis 22:06, 4 September 2007 (UTC)

As Sherool says, NFCC 3a is about both the size of each nonfree image and the number of nonfree images used. The second sentence of 3a makes this clear. Our nonfree use requirements are stricter than fair use would require. — Carl (CBM · talk) 22:25, 4 September 2007 (UTC)
I just fixed that, I think. 3a is about number of uses, 3b is about the extent of each use. Same result, just trying to keep things clear. I also changed the headers to make it sound less legalistic and avoid confusing people into thinking we're referencing the fair use standard. Wikipedia is using this in a plain English way: simply, use as little non-free content, as few times, as possible. Fair use does not have the same rules. Under fair use Google can create a hundred million thumbnails and each is legal. We choose not to go that way.Wikidemo 22:58, 4 September 2007 (UTC)

Sure, Wikidemo. In the meantime, plainly well consensused, well thought out, judicious use of images in, say two or three relevant articles instead of one article are being trampled upon, with immense damage to the goodwill of participants. Typically, with, say, a book cover image, you have the article on the book, the article on the author, and the article on the topic-- still very much in keeping with "minimal use" in both the numeric and substantive senses of the words. Yet, this notion that "minimal" means "only on the article about the book, magazine or particular album continues to persist. So, what will the participants in the project proceed to do about that very misguided situation? ... Kenosis 23:07, 4 September 2007 (UTC)

Yes, that is indeed the present dabate. And I fully understand that it will not stop presently, but rather will go on for some time until WP participants get a better understanding of the relevant concerns, rather than trying to divine the reasoning behind the NFCC. It seems to me it would be sensible to understand what reliable sources outside of wikipedia mean by the words "minimal use", I should think. Otherwise, the rules just get, well, sort of made up without any real reference point, it seems to me. So, since WP participants have made up the rules here, why use the words "minimal use"? (Rhetorical question -- IMO, the likely source of the words is that one or more WP users heard or read the words "minimal use" in the context of copyright law, along with the Board Resolution #3, "Such EDPs must be minimal". Kind of a juxtaposition,so to speak.) ... Kenosis 22:52, 4 September 2007 (UTC)

As I was also attempting to say, this is a learning process, and the NFCCs have given evidence that the way they were written can create great damage to interpersonal relationships on the Wiki. I would hope that the NFCCs are not quite yet believed to be forever written in stone, but rather, that the growth process will continue and that the NFCCs will be adapted with the experiences of the past several months (since late-March, 2007) kept in mind. ... Kenosis 23:01, 4 September 2007 (UTC)

Can a quoted sentence be used in multiple articles? Or should it be limited to one article and the other articles have links to the quote? This would be "minimal use" of a non-free text passage. -- SWTPC6800 03:06, 5 September 2007 (UTC)
Direct quotes should also be kept to a minimum and only used when they add significantly to the article. But there are significant differences between quotes and images (such as ease of linking) that lead us to handle them differently. — Carl (CBM · talk) 03:13, 5 September 2007 (UTC)
IMHO each article should be taken on its merits. Does it benefit the reader significantly to have the text quoted here? Sometimes we can say yes, other times we can say no - let's just link to a relevant page which does have the quote and an intelligent discussion leading off it. I think this makes sense for images too. -- Steelpillow 20:53, 5 September 2007 (UTC)

Somebody made a point about people abusing an 'iconic' getout to plaster copies of their favorite album cover or whatever all over the place. IMHO this would be quickly picked up on, on the basis that it did not contribute significantly to the individual articles. There already seems to be a steady flow of people thinking that some non-free thumbnail is acceptable and others tearing them down as fast as they go up - would moving the barrier a little to the liberal end make any difference to the overall level of policing activity? -- Steelpillow 20:53, 5 September 2007 (UTC)

"Minimal use", in that context, means that the barrier is never at the liberal end, and that the default is "do not use". Seraphimblade Talk to me 04:39, 6 September 2007 (UTC)


I just noticed {{Biorationale}} and am unsure whether it satisfies WP:NFC's rationale requirements. Any thoughts? -- But|seriously|folks  19:46, 5 September 2007 (UTC)

In no way does it pass as a valid rationale. Someone needs to delete it. βcommand 23:09, 5 September 2007 (UTC)
I'll remove it. --Yamla 23:11, 5 September 2007 (UTC)

Could you please restore it while we consider? You can leave a tag on the template if you want to say it's been disputed. But five hours is not a proper discussion and that has the effect of invalidating a bunch of images that may or may not be legitimate, without even leaving an edit history so anyone can know what happened.Wikidemo 00:17, 6 September 2007 (UTC)

Nevermind. I have just restored the content but with a warning. Now, exactly what is the problem / issue, and how would you deal with the appx. 30-50 images that this has been used on? —Preceding unsigned comment added by Wikidemo (talkcontribs) 00:24, 6 September 2007 (UTC)
Only one of those rationales is valid (that the person is deceased, rendering replaceability impossible). In any other case, using nonfree images of a living person fails #1, as the image is replaceable. Seraphimblade Talk to me 04:33, 6 September 2007 (UTC)
It looks as if this is only meant to be used for those who are not living. -- Ned Scott 04:53, 6 September 2007 (UTC)
I can see situations where all parts of the rationale will be true, but I feel very reluctant to have these kinds of templates floating around. In the very least, having a template be configurable and requiring some form of input would be better, even for simple situations, as well as to encourage a more detailed rationale (if one is needed). -- Ned Scott 04:58, 6 September 2007 (UTC)
I'd like to suggest moving this to {{deceasedbiorationale}} and including instructions on proper usage in the /doc. While rationales for images of deceased persons are generally uniform, there are special cases where any usage compromises commercial value. ˉˉanetode╦╩ 08:16, 6 September 2007 (UTC)

Temporarily suspending deletions based on lack of fair use rationale

I have made a proposal, by edit to WP:CSD, that we temporarily suspend I6 because Betacommandbot, the bot that was tagging images for lack of a fair use rationale, attempts to have been restarted and malfunctioned, tagging many images that actually did have rationales. In addition, there has been some serious discussion on this page regarding whether a written rationale should be required and whether it is appropriate to speedily delete legacy images for lack of rationale, given that the prior arguments that the Wikimedia Foundation was requiring us to do that turned out to be untrue.

Can we please hold off on further tagging and deletions until we resolve this issue? Re-starting the large scale deletions would be a provocation at this point. This is a matter that ought to be addressed in a calm, sensible way. Thx, Wikidemo 03:12, 6 September 2007 (UTC)

We should be able to do tagging by hand still, but we can wait for the deletions. User:Zscout370 (Return Fire) 03:17, 6 September 2007 (UTC)
the bot as been rolled back. Wikidemo, 10c is a sense able requirement even if we have a template for a few cases that have a similar use, IE logos and album covers, along with a specific page reference were we are allowing that image to be used. but wholesale removal is just asking for non-free images to be abused. we cannot also say it is a logo we can use it where ever if it is a logo we can clearly state that and the companies page that it should be on. βcommand 04:16, 6 September 2007 (UTC)
This isn't a matter of a few valid images. The vast majority of tags and deletions of legacy images in the past, thousands to tens of thousands of them, have been otherwise appropriate images that lack an explicit rationale. Hand tagging and deleting in contested cases, and policing to make sure new images have tags until we resolve the matter, are both okay. Massive purging of Wikipedia images is not, and that is a provocation if it's done before this issue gets a real hearing. If that is starting up we need to stop that, now. Glad to hear the latest tags have been rolled back. It would be good to hear some promise that this isn't going to start up again. Wikidemo 05:03, 6 September 2007 (UTC)
Images that violate policy need to be tagged and fixed, and if people are un willing to fix then deleted. I will start it up again. Im sorry but we need to get wikipedia to be compliant. I would be in support of a very limited use of a template style rationale for a few specific uses. I see only 3 current examples Book covers on the book pages, album covers on the album page, and company logos on the companies main article. βcommand 05:10, 6 September 2007 (UTC)
So you're saying you're going to start the bot up again to purge Wikimedia of its old images? That would be a big mistake, and will probably result in a lot of bad will and upheaval at this point. There is no Wikimedia Foundation mandate. We now know that was a mistake. There is only a Wikipedia policy, supported lately by some faulty reasoning. We don't purge half the non-free images by bot on a technicality over that kind of thing. Please, don't start that kind of disruption. Decisions are made by editors, not bots. We can push the issue and get some consensus on the right way to go about it. \Wikidemo 05:19, 6 September 2007 (UTC)
yet again, you completely fail to understand why Im going this. In a perfect world BCbot IDs a image with a problem, notifies the uploader(s) and the article(s) where the image is being used, that there is a problem. Once people know there is a problem they fix it. But for the most part users dont care about fixing the problems they just want to ignore the issue. Ive postponed image tagging now several times, in june by request, mid-July hoping that the previous taggings had gotten people motivated, and I did a little in early August, But for the most part I have been hoping people would fix the problem, (I guess Im an idealist because it doesnt seem to be happening). Your proposal to remove rationales will go no where as it is a VERY bad idea and will lead to a massive abuse of NFC, which will be 10x harder to clean up. βcommand 05:36, 6 September 2007 (UTC)
I understand the issue pretty well, thank you, and you misrepresent my position. Some people do care about the issue. You could help fix the problem too, you know, by helping to add rationales instead of just deleting images. Using the fact that the person who uploaded an image a year or two ago does not fix it in the one week you give them as some kind of proof that people on Wikipedia don't care about the issue is strange. One of the things that makes the whole process uneasy is when a person who takes sides on the issue also calls the shots on running the bot. It would be nice if we all had bots to make several thousand edits a night to back up my position, Wikipedia would start looking like battle-bots. Wikidemo 18:56, 6 September 2007 (UTC)
Like I said on WT:CSD, the bot is just tagging. What we need to do is ask for a grace period from admins instead of blaming the bot. -- Ned Scott 05:22, 6 September 2007 (UTC)
There already is one, I6's have to be tagged at least one week before being deleted. Seraphimblade Talk to me 05:25, 6 September 2007 (UTC)
I mean if one honestly does need more time, admins are generally nice about granting that. -- Ned Scott 05:26, 6 September 2007 (UTC)
Oh sure. And if someone's to come to me and say "Hey, I want to write a rationale for one you deleted", and they're prepared to do it, I absolutely will undelete it for them. Regardless of what some think, the aim here is not to be unreasonable. Seraphimblade Talk to me 05:29, 6 September 2007 (UTC)
How about this one. "Hey, I want to discuss the need for rationales and what our procedure should be for dealing with images that have rationales that are not explicitly written on the image page. This is a contentious issue and there is no support for the requirement that there be a written rationale. Please undelete any images you deleted on this basis while we discuss thie issue. If we decide that implicit, template, or remote rationales are appropriate I and others need some time to write them. We need more time, a week or two, or however long it takes to make sure we have consensus on what to do about these images."
"Just tagging" is only the triggering event in the inevitable 7-day deletion process unless admins agree to forgo deletions. If they agree not to delete legacy images that are appropriate and salvageable, fine. But why add a bunch of tags in the first place and provoke people that way before we resolve the issue? Other bots have limited themselves to tagging new images. That would be an easy step. We've let the issue lie for a few weeks after Kat Walsh's statement. We can have that discussion now and pursue it if we need. Doing so under the cloud of newly tagged images, where many of the tags may turn out to be invalid depending on the outcome of the discussion, doesn't help anything. Wikidemo 05:41, 6 September 2007 (UTC)
Ok that is the problem We need more time, a week or two, or however long it takes to make sure we have consensus on what to do about these images you have had three months to do this, why has it not been fixed? two: rationales should be per use, to that length Ive stated that some simi templates would be good, and they would fall under the policy as it is. As it stands consensus is against you, our policy states that rationales are needed. Until such a time that policy changes I will continue to enforce it. βcommand 05:50, 6 September 2007 (UTC)
One more thing to add to the quagmire here. If BCbot is doing automated notifications, it ought be giving notice to participants in the article(s) and/or other page(s) on the relevant talk page(s) of where the image presently is being used, and perhaps even wherever the image has previously been used. Often the uploader has no working relationship with the people invovled in the page(s) using the image. ... Kenosis 13:39, 6 September 2007 (UTC)
Apparently, wide notification leads to complaints the bot is spamming. It would be a ood idea to add images in articles you frequent to your watchlist. — Carl (CBM · talk) 14:04, 6 September 2007 (UTC)
Interesting argument, particularly since complaints from local participants do not appear to deter the circular arguments by certain regulars in this part of the wiki (e.g. "we made the rule as presently written, and therefore we must follow it because it's the rule"). Indeed those users appear energized by the complaints. So the argument Carl is making here is that proper notification is trumped by ill-founded notions of spamming?!! ... Kenosis 14:20, 6 September 2007 (UTC)
And once again, "minimal use" indicates that we err on the side of not using in the event of no consensus. I see no clear consensus that boilerplate tags are acceptable rationales (especially given that the policy said they were not well before the Foundation made its resolution), so we err on the side that rationales must be individualized. Seraphimblade Talk to me 06:13, 6 September 2007 (UTC)
What consensus? Saying "I will continue to enforce it" (via unapproved, non-consensus use of a bot to enforce one person's reading of policy) is silly and contentious. *I* have not had three months to do anything. We (Wikipedia as a project) face a question of what to do about half the fair-use images on the project. It's unfair and ridiculous to suggest that I, a single editor using the "edit this page" feature am supposed to compete with a bot that tags thousands of images a night and a team of zealous deleters using bot-assisted application to delete 5-10 images per minute each. And why should I? We're talking policy for Wikipedia, not a competition of editor versus the bot. Why haven't the tagging system and bot been fixed in the last three months to function properly, operate in an orderly way, and do something more sensible? "Minimal use" and erring on the side of caution does not mean wily-nilly purging of half of our images on a technicality on a random schedule. Erring on the side of caution does not mean embarking on such a program. There is no rush. Kat Walsh's statement that belied the whole point of the deletion program came only a few weeks ago. Why force the issue now? Wikidemo 07:14, 6 September 2007 (UTC)
There is a rush only in the sense that there is a deadline next year that we are supposed to meet. I think Betacommand can give stats about how many images need attention by then.
I think it would be very reasonable to pause deletions for a short period of time, and I have taken a break from dealing with the CSD categories. But my attempt to clarify the issue of written versus implicit rationales and Kat Walsh's statement was not successful (see WT:NONFREE#Nonfree_use_rationale_requrement_weakened? above). How may weeks should we stand around waiting for consensus to develop? It seems to me that those who are in favor of loosening the requirements should take more initiative in obtaining consensus to change the policies, if they want to see the change happen. — Carl (CBM · talk) 14:01, 6 September 2007 (UTC)
The Board Resolution only requires that NFC be tagged so they are "machine readable", that is, identifiable as NFC by a bot. This is a completely separate issue from each image having an explicit written fair-use rationale for each particular use, as Kat Walsh has already made clear. ... Kenosis 14:23, 6 September 2007 (UTC)
As I pointed out in the previous thread, there is a difference between the foundation resolution and what is required here on Our policy is stricter than some interpretations of the foundation resolution. I tried to see if there is consensus to weaken our requirements here somewhat (which would be fine with me) but it didn't appear there was consensus to do so. — Carl (CBM · talk) 14:41, 6 September 2007 (UTC)
Weaken what? These NFCCs were initially formed very much on the fly, so to speak. It's also been pointed out in the previous thread that there are confusing and highly divisive aspects in the NFCCs. Some of them are based upon mistaken perceptions of what "fair-use" means in the real world. And importance of the "real world" issues, contrary to assertions that it's irrelevent, is evidenced by the numerous comments along the way referencing the "real world", such as , e.g. "Omigosh, what if copyright holder X sued the foundation?" and such. I respect the need to be conservative. But, the notion that "minimal use" means "only in the article about the book or magazine" is an arbitrary decision made without understanding what "minimal" means. The proposals to revisit this issue are not threatening to weaken anything that has anything to do with reality. Defining, e.g., "minimal" as "only in direct relationship to a relevant topic" is very enforceable and not divisive except to a user who's truly irrational. Similarly, the notion that each NFC use must have an explicit rationale is an approach that can quite readily be replaced by machine readable boilerplate. Just like a library, there are three uses for, e.g., a cover image (1) Title of book, (2) Author of book, and (3) Subject matter of book. Any bot designer worth her or his salt can handle that, in plenty of time for March 2008, if that's the chosen route. ... Kenosis 15:06, 6 September 2007 (UTC)

Another bot problem

Another bot, User:OsamaKBOT, just started going through the logos on Wikipedia and systematically tagging them for lack of source, about 900 in the last 3 days. It's going in alphabetical order and it is up to AR. See

A few problems with this. These are the same class of legacy images that we have use rationale concerns for. It looks like yet another image deletion campaign that's going to get all these deleted, in the case of logos probably most of the company logos from Wikipedia. These aren't fly-by-night uploads by problem editors. These are the articles about the most important companies on Wikipedia: Apple, Archer Daniels Midland, American Express, etc., with logos uploaded as long ago as 2004 from what I've seen. As with the use rationales, the answer to sourcing in the case of logos is to add a source, not delete the images. Invariably, the source of the logo is the company for which the logo is used. There is no copyright held by anyone else. So simply adding the name of the company as the source, in the logo template rationale, fixed the image. Deleting images here is disruptive to the project and will create a lot more work for all concerned. Moreover, many (perhaps most) of these logos are not really copyrighted. So that is a limitation of the logo copyright tag.

As with Betacommandbot, and other mass tagging and deletion campaigns, I ask that we pause this kind of thing until we sort out the issues and see if we can get a realistic program for fixing the logos (see below). Wikidemo 02:06, 8 September 2007 (UTC)

What is there to "sort out"? All images require sources. They always have. If people don't provide sources for their images, the only way to enforce that requirement is to delete the images (secure in the knowledge that if someone later provides a source, the image can be undeleted). What other possible approach can there be? --bainer (talk) 12:57, 8 September 2007 (UTC)
While I agree in principle we do have a long standing understanding that scertain images are practicaly "self sourced". I mean would you realy delete the Microsoft logo just because the image page doesn't explicitly say that it was in fact from Microsoft? I'm all for getting images sourced, but bot tagging logos as unsourced does more harm than good by fortifying the impression that image maintainance is beeing done in a completely mindles fashon. Not everything tagged as a logo is in fact a logo, but you can't rely detect that with a bot. I would strongly reccoment skipping over most forms of "identifying works" (logos, cover art etc) when using a bot to tag unsourced images. There is rarely any confution about where those images orignaly came from or who hold the copyrght to them, wich is the point of having sources in the first place after all. --Sherool (talk) 13:49, 8 September 2007 (UTC)
Indeed. The approach is to simply note that the logo is owned by the company. Is anyone really going to want to undelete 10,000 images so I can get a bot to add that note? That seems a backwards way to go about things. This goes for album covers, film posters, paintings, and other 2D art where the copyright is in the work, not the photograph. Wikidemo 15:05, 8 September 2007 (UTC)

The mass tagging of images has long been a problem. Look at what happenned to Betacommand and his bot. The bot operators are too often not receptive to legit criticism of mistakes. Quality control of their code needs to be better.Rlevse 16:54, 8 September 2007 (UTC)

You don't shoot the messenger. OsamaKBOT is doing what should be done to un-sourced files, tagging them. Users upload all sorts of logos that they assume represent something, when it is actually identifying something else. I can talk for other wikis, where they had the wrong logos for Disney and Sony. One wiki had Microsoft logo with the colors mixed up. Some even had the CIA logo placed in M5 article. And is no exception, it has its share. The logo source is important to make sure you got the real/accurate thing.
An alarming note was written above: album covers, film posters, paintings, and other 2D art where the copyright is in the work, not the photograph, a statement that I think (imho) is not accurate; when it comes to copyright, everything is copyrighted, album covers, film posters, paintings, and other 2D art, all copyrighted. Fairuse means a source/owner is a must; these logos belong to someone, assuming others will know the owner is not accurate.
Ignoring the problem will not solve it, images either have a valid source or get tagged. Now, the tag it self or the actual action after the tag process, might need to be worked on. Aain, don't shoot the messenger. --Tarawneh 01:23, 9 September 2007 (UTC)
No, the statement of copyright is absolutely correct as a matter of US law. The only copyright in an album cover, logo, film poster, painting, etc., is in the source artwork, not in the image file. The only relevant sourcing information is who owns that copyright. That differs from photographs of 3D objects, where there is typically a copyright to the photo in addition to possible copyrights in the subject of the photo. If these images lack use rationales or sourcing information we can add them in short order rather than deleting the images. But first we have to agree on which images that is appropriate for, how to notate them, etc. That is why we're urging a halt to the bot-initiated tag-and-delete process with respect to older images (Betacommand proposes a January 1, 2007 date below) until we get this sorted out.Wikidemo 03:40, 9 September 2007 (UTC)
Uh, Osamabot is also taggin images as no source when in fact they have source. He's also tagged thing witht the wrong tag. See his WP:AN thread too.Rlevse 02:17, 10 September 2007 (UTC)
You know Rlevse, I have fixed most them.--{{subst:#ifeq:{{subst:NAMESPACE}}|User talk|{{subst:#ifeq:{{subst:PAGENAME}}|OsamaK|OsamaK|OsamaKReply? on my talk page, please}}|OsamaK}} 11:03, 10 September 2007 (UTC)

What to do about legacy images lacking written rationales

Half of our old non-free images lack written use rationales, and people keep adding new images without a written statement. We want to do something, even if the Foundation is not requiring them. But there is no urgency, as we discovered from Kat Walsh's statement a few weeks ago. The premise of the Bot and the program to delete 170,000 images, that we needed all the images on Wikipedia to have written rationales by next March or else we are violating a resolution by our Board, is faulty. When we learned that we started discussing what kind of rationales we really need, but the discussion as usual degenerated into revert warring at the hands of a few people who said any change to the status quo is premature and "not consensus." We noticed that the bot had stopped, which took the pressure off. To start up yet again, unannounced, with no willingness to discuss, is pretty unfortunate. For the people on one side of the debate to turn the bot on and off, while opposing efforts to change the requirements, is unreasonable and unfair. A more reasonable process is to decide what we should do about use rationales first, and then do it, without ultimatums and without the false preconception that the current state of affairs represents some kind of consensus. We should limit the discussion specifically to the question of rationales, and avoid opening the wider policy of what kinds of non-free images are okay to use. The steps would be:

  • Decide what kind of use rationales are needed on images - what they consist of, how made, what they say, etc.
  • Assuming this leaves a large class of images without adequate rationales, decide how we are going to add those rationales to images that are appropriate, delete images that are clearly not appropriate, and how to handle the disputable cases.
  • Agree on an implementation program and enforcement procedure, including a schedule for enforcement, notifications, etc.
  • Build and test automated and human-assisted bots for all aspects of the program - identifying and sorting images, notification, adding or modifying rationales, deletion, and reporting.
  • Appoint someone, a group, or a project not on one side of the debate or the other to oversee the bots and other enforcement procedures

Until we're done, it is needlessly confrontational to continue on the present course of image deletions. The most neutral and least disruptive course is to (1) keep insisting that new images have written rationales as presently required, and do everything we can to prevent adding new images without rationales, and (2) leave the old images alone until we decide what to do about them, subject to the normal case-by-case review system we have in place.

Wikidemo 18:56, 6 September 2007 (UTC)

Makes sense to me. -- But|seriously|folks  19:34, 6 September 2007 (UTC)
If we can set a date that is considered "non legacy" I can have BCbot ignore those images. Like Ive said I dont care if the images are fixed or deleted. (Im neutral about that) Ive proposed a few ideas but they are just ignored. I like most of your ideas but I do disagree with some. βcommand 20:04, 6 September 2007 (UTC)
Good, thanks. And sorry if I have ignored or missed your ideas. I'll keep my eyes open and please feel free to re-post or mention them on my talk page. Three logical dates that come to mind for a distinction between legacy and current images are: March 23, 2007 (the day of the new Foundation policy, on theory that things changed then), mid-June, 2007 (when the bot first started and people started getting serious, on theory that people should have known by then), and Aug 1, 2007 (on theory that if you haven't enforced a missing rationale within a month, a large-scale mop-up is harsh). I think July 2006 is too early for a cutoff because even though the rule was in place then and people should have known, the fact is they didn't and this is not an action to punish uploaders for breaking rules but rather to shape up images across the project. So I would favor a March or June, 2007 date. Once we do clean up legacy images there should not be a lingering problem that needs cut-off dates. Wikidemo 22:00, 6 September 2007 (UTC)
Just a quick note. I'm going to see if we can get people to come together to resolve these issues, starting in the next few days. Take the weekend off if you want, but we ought to get moving! Wikidemo 02:06, 8 September 2007 (UTC)
how about January 1, 2007? that would have given 6 months for the new policy to be put into effect.
as for my ideas, Ive suggested a very very limited use of template style rationales for 3 classes of images. I say limited because I see only 3 areas that such a style of rationale would be concrete: Album covers, logos, book covers. And even those three classes would only be covered for a single use of said image, on one specific page: Album covers -> the album article; Logos -> for use solely on the main companies page; Book covers -> the article about the book. Those template style rationales are not blanket use, you would still require that those templates have a parameter for the article that it is used in. (you cant just add {{logo rationale}} you would need to specify what article that it applies to) One of the features of the template would be a tag for deletion if the parameter is blank. If A user wants to use the image else ware we would require a hand written rationale.
addendum to that statement, any and all such templates MUST be approved here and there MUST be a very solid claim that every use is the EXACT same, and is truly needed in said articles. Ive got a few other ideas but we are not yet there. βcommand 02:24, 8 September 2007 (UTC)
January 1 2007 sounds good to me. If it looks like people are really address this we should ask other bot operators to follow suit and/or write this in as a temporary matter of enforcement policy. I agree that we should have a template approval process and some other protections. I think we'll end up with a number of specialized categories, but we can start with those first three (logos, album covers, and book covers) and wait to see what other proposals people make. It would be great if we could have a bot or some tool to fill in the "article used" template parameter for the old images as we go through them because that's just simple data lookup from the image page...but that's getting ahead of things. First we should figure out what's acceptable, then we should figure out how to get there.
I might take a rest for a day or so but I'll try to have some more concrete ideas after that. For this to go anywhere we'll need broader participation. Perhaps a proposal sub-page just for discussing the issue of reforming (and/or tagging and deleting) the legacy images? Wikidemo 07:24, 9 September 2007 (UTC)

Question about Image:Darwinsblackbox.jpg

Is the use of Image:Darwinsblackbox.jpg on Intelligent design in violation of WP:NFCC#3 and WP:NFCC#8? The image is already on the article about the book, which suggests this use fails #3, and I can see little significance to the reader from including a picture of a book to support the claim that the book is important, which suggests ift fails #8. There is a long discussion about it on the image talk page Image talk:Darwinsblackbox.jpg. To keep everything concrete, I am interested only in the use on Intelligent design. — Carl (CBM · talk) 12:52, 7 September 2007 (UTC)

It's valid in the Darwin's Black Box article. I'd remove it from the other two, though. Borisblue 14:10, 7 September 2007 (UTC)
I do not believe it is in violation. The imagery used in the cover bears particular relevance to the substance of the whole framing of the intelligent design debate, and so it has good cause to be included in that article in its own right. It is not merely decoration. It is not merely a pointer to the book. It is a major part of the framing of the ID debate and properly belongs in the ID article. People will disagree on that, to be sure. I understand that, and I don't object to a fair and open discussion of the matter.
I DO object to that debate being fragmented over as many pages as possible. The proper place to discuss particular images is the fair use review. Are you confused about the best place for that debate, or are you just trying to bring it up in as many different places as possible? Is it some kind of tactical ploy? In particular, are you TRYING to find a venue where people don't really have much background in the content of the article itself? If so, it really sucks. Theres an open fair use review. It is a place where people hang out who are familiar with the fair use guidelines, and where people with the content background to explain the perspective on significance hang out as well.
The manner in which people are pushing the delete program is disruptive, and shows a cavalier disregard for the basic principles of consensus and discussion on which wikipedia is based. Please take this to the proper forums. Duae Quartunciae (talk · cont) 14:22, 7 September 2007 (UTC)
Please AGF and consider that maybe the reason why so many editors bring this up is that it is in fact in violation of our policy. You should also know that IFD debates routinely handle the debate of which pages a non-free image may be used in. While it should technically just be at FUR, a decision handed out at IFD (which in this case was to remove it) is just as qualified. Keeping the debate where it started is the best way to not fragment it. --Pekaje 14:30, 7 September 2007 (UTC)
I am assuming good faith. That's entirely consistent with concluding poor judgment. THIS page here is for discussing the guidelines; not particular images. The point about "technically" it should be an FUR matter is quite right; and there is a still open FUR for the page. I'm begging here. You guys KNOW this is a contentious subject. There have been arbitration cases already in which advocates for deletion of images are urged to be a bit sensitive. So please, try and keep the discussion where it belongs, and let the fair use review do its proper job. Duae Quartunciae (talk · cont) 15:02, 7 September 2007 (UTC)
The issue does appear to have already been settled at IFD. I'd be surprised if FUR gave a different outcome, at least with the arguments presented. I don't think the image should be deleted, but it does need to be removed from the articles where it is used in violation of policy. It would appear that OP was using it as a specific example on what can and can't be used, according to these policies, so this page was not an entirely inappropriate place to ask. --Pekaje 15:27, 7 September 2007 (UTC)
I'm not sure that the fair use review is the right place to discuss the image, but I am certain it isn't the only place. This page is followed by many people familiar with the image policy; my hope was to get some new people into the conversation. Just as WT:V is used to discuss possible problems with particular articles, it is appropriate to use this page to discuss particular situations. — Carl (CBM · talk) 15:12, 7 September 2007 (UTC)
When there is already an open discussion in more appropriate forums, and when it is known to be a contentious subject where existing arbitration rulings have urged a bit of sensitivity, I think this is very poor judgment indeed. There is a principle in wikipedia that policy overrides consensus. That is quite true. But it does not mean anyone can unlaterally decide which way policy goes. That is itself contentious... just look at the boxes at the top of this page! One of the major points in contention is whether or not the content is significant. That is a matter for which background in the content is directly pertinent. I'm not proposing that you should just roll over and let the editors of an article do what they like. But I do think that when there is a strong consensus about significance from editors, there should be a bit more willingness to recognize that we might really mean that in good faith. In any case, it is a debate for which people involved in the content have a stake, and part of the reason that this has been so disruptive, and the focus of arbitration cases and edit wars and whatnot, is a widespread perception that the advocates for deletion are just riding roughshod over anyone else without stopping for adequate consideration of the roll of editors in making the case for rationale and significance.
If there's a fair use review open, then you HAVE a more appropriate forum available. Use it, please! Duae Quartunciae (talk · cont) 15:39, 7 September 2007 (UTC)

<unindent>Obviously the reason this is here is because it's viewed as a test case of sorts. Fine -- unfortunate, but, fine. Since it's here, I should call attention to the fact that this debate about images in the article on intelligent design has gone across a number of userspaces, four IfDs, a fair-use review, various templating and other strategies, administrators' noticeboard, and an attempt to short-circuit standard mediation processes by filing a Request for Arbitration -- all initiated by deletion advocates. In legal parlance, this is known as forum shopping. It's a shame, IMO, how divisive to the community these NFCCs #1, #3 and #8 have turned out to be. For the IfD w.r.t. this particular image, see the Images for Deletion proceeding, where a strong consensus to keep this image in the article on intelligent design was unilaterally overruled by the closing adminitrator with the words "Images removed from articles per Wikipedia policy on non-free content and image copyright tag requirements. -Nv8200p" For the fair use review, see Wikipedia:Fair_use_review#Intelligent_design. See also, e.g., the very recently archived discussion at Administrators' noticeboard. As well, I should call attention to the Request for arbitration. And, in addition to the increasingly lengthy discussion at the image talk page, there have been the various other strategies in efforts to delete, including the arguments brought to bear in this section here. In short, once they set their sights on something, this group wants their way. I'll get back to y'all with some substantive comments a little later, as my time permits. ... Kenosis 16:23, 7 September 2007 (UTC)

NFCC Discussion

Speaking of criterion 8, shouldn't someone start a discussion about it? I believe a lot of images have been listed on IfD recently on the grounds of IFD 8, and I read a lot of discontent about it. So, if nobody minds, can a discussion on that begin again, and others be informed of it through the Village Pump?--Alasdair 02:41, 8 September 2007 (UTC)
IMO, if these competing tensions are not resolved in some reasonable, sensible way, the previously publicized goals of Wikipedia are in some jeopardy. Among the competing tensions at the present time are: 1) a free encyclopedia (what does this mean?), 2) the best achieveable quality (what does this mean and is the consensus process dead except for throwing a bone to contributors at the local, individual article level to resolve petty disputes at that level?), 3) an encyclopedia as safe as possible from cheap-shot lawsuits, 4) compliance with the law. There certainly are other relavant tensions, but IMO the ones I just mentioned are more than adequately serious to deserve broader discussion. As to the way the NFCC are presently worded and how they are presently being applied, these appear to me to deserve continuing discussion on this page until they are resolved in a way that effectiively reflects a sustainable approach to what WP participants have, by consensus, chosen to term "Non-free content". ... Kenosis 03:31, 8 September 2007 (UTC)
Also, I noted the new NFC #1. As long as someone's writing skill is good enough, anything could be described by words. Which means thoretically, screenshots should more or less NEVER be used in articles about film. But what quality will the article be of? If anything, I'd prefer tension 2, as you state. Now then, should we broaden the discussion (by posting at the Village Pump) and ask others for their opinions? Since I think #1 and #8 are far too restrictive.--Alasdair 06:27, 8 September 2007 (UTC)
I think that while a slackening of the rules would be popular (I certainly wouldn't mind), it is unlikely to pass. The foundation resolution is extremely strict as it is, and any further local loosening than we already have is likely to be overruled. As for the specific case of screenshots, well some do pass #8 and some don't. Often they fail #3. Use it solely for when there is some amount of non-trivial coverage of what is shown in that screenshot. If it's important enough to be included, it's important enough for someone to write at least a couple of paragraphs about. Anything else comes across as decorative use. --Pekaje 08:26, 8 September 2007 (UTC)
Having read the resolution, I find this phrase to be a key one: "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." The question is, how would replacing album covers, screenshots, posters and such with text serve the same educational purpose? Such images are more than just used to show the thing exists. More can be seen from those images. For instance, for album covers, you may get to comment on its design, and why the group chose this cover and so on. For film screenshots, besides showing a scene in the film, it gives the reader an idea into the setting of the film, what the characters are experiencing, their current emotions during that situation and such. In fact, the images may contain far more information than you think compared to just text. And even if you can express such complicated designs in text, the reader is going to have a hard time to understanding it. Hence I feel for those cases, text can almost never "serve the same educational purpose". Therefore, if the uploader can explain it clearly with a very detailed and thoughtful rationale as to why "no free equivalent exists" and that it's significant.
As for minimal use, I agree in principle since those articles aren't supposed to be fansites or whatnot. However, it may help to give an idea how much is considered minimal, as in terms of quantity per length or something like that.--Alasdair 09:03, 8 September 2007 (UTC)
Just so it's clear, I agree with the points made above. There are a whole heap of non-free images that could never be replaced by text. What I object to is the same non-free images (which have valid rationales for the article about the subject they depict) being used decoratively in other articles with little or no coverage of the image. That includes album covers in discographies, posters in an article about the history of musical theater, and book covers in an article discussing some of the same issues as those books. --Pekaje 09:18, 8 September 2007 (UTC)
I see. Either way, I think the wording of several criteria should be reworded to prevent misunderstandings. Take an article I've been working on, for example: Rob-B-Hood. In the article about a film, I've added a screenshot to show the inclusion of one of the characters in the film, which also shows him acting in his typical slapstick comedy ways as he did with the main actor in the past. (The latter cannot be explained by words, can it?) Then, I've a picture showing main actor Jackie Chan standing in the set. That particular image shows the setting for the film, the costume design and the equipment used during filming (such as the make of the camera and such). Then, there is a picture showing his stunt in film. The next image shows a picture of the baby that stars in the film. After that, we have 2 movie screenshots so the readers can have a better understanding of what is going on. And yet in the FAC, a reviewer states that many of these pics are unnecessary and can be expressed by words due to the NFCC, while there is more information than what you think in those shots.
This is why I suggest a rewording, in a sense that encourages uploaders to provide a thoughtful rationale on the image's contents, to explain why all the image increases people's understanding in a way words can't.--Alasdair 09:45, 8 September 2007 (UTC)
That's a great idea. The problem is that the handful of editors that are controlling this policy page are not listening. -Nodekeeper 11:37, 8 September 2007 (UTC)
How so? I don't think you'll find any of the "regulars" here that would object to encouraging people to write more detailed and thoughtfull rationales for the use of non-free images -- or any rationale at all for that matter. That's sort of the main aspect of the policy in the first place, only include non-free images if you have a good reason for doing so and explain what that reason is in each instance. Indiscriminately putting in screenshots of every character in a movie is bad, using a couple of screenshots of illustrate particular points of acting, visual style or such is ok. Sometimes it's hard to tell the difference if the (required) rationale is not present. So anyting that makes it more clear to people that they have to provide an "explanation" for why each image is significant to the article has my support. Sometimes we will disagree on the merits of the provided rationales, but at least then we'll have a starting point for a real debate rater than it getting lost in the backlog of tens of thousands other images with no rationale. --Sherool (talk) 11:53, 8 September 2007 (UTC)
(←) Let me emphasize a point to back up Sherool here. The important part of the written rationale (especially in light of Kat Walsh's statement) is the plain English explanation of why the article benefits from the inclusion of the image. Things like "small size" and "no commercial harm" are completely generic and don't actually explain why the image should be used. A written rationale consisting entirely of generic statements is of no benefit in evaluating whether the image should be used in the article. — Carl (CBM · talk) 12:08, 8 September 2007 (UTC)
Yup, in fact, I've taken efforts to write more detail on the movie screenshots these days, explaining the content of interest in it, and why it can't be replaced by words. I'm sure that "Minimal use" already is enough to prevent people from indiscriminately adding screenshot. My proposal is to add a section to WP:NFCC (and maybe WP:NFC) encouraging people to write detailed rationales to explain each detail on a non-free image, and why it's significant. I'll come back to the discussion tmr. Be right back.--Alasdair 13:30, 8 September 2007 (UTC)
I have run into an interesting little problem when it comes to explaining why an image is significant.
An encyclopedia is not the place for original research. However there is a role for expertise and judgment in being able to identify the most appropriate sources and organize a clear description of using those resources. The use of images, whether free or not, is an example of applied judgment. But suppose now that a dispute arises over whether or not an image is significant or not. In my view, this is a meta-discussion; and not the meat of the article where the image is used. It is usually distracting and undue weight to give a long argument for the significance of an image in an article. That is, in effect, explaining to someone why the article is written as it is.
I attempted to give a rationale for some images, and then someone immediately slapped a "citation" rag on it. Effectively, I was being told that it was not enough to give a good argument on my own account for significance; I actually have to find some other authority who independently identifies that image as significant. This is really hard to do; because even if the image is used in other contexts, they are not necessarily going to go into a meta-discussion of significance, as they don't have the extra burden which is imposed at wikipedia for formally justifying any non-free content.
Frankly, at that point I gave up. I just took it as yet another in an endless series of hoops and demands being arbitrarily imposed by advocates for the deletion of as much non-free content as possible. Trying to make my rationale both sufficiently detailed to explain the significance of the image, and also cited to others who have made the same argument independently, is too much. If that is what it is going to take, I'd just be inclined to throw in the towel entirely.
If you are going to advise longer rationales explaining significance in more detail for novices to the subject area, I think you had better recognize that this is a request for an independent analysis; and may take a form that would look like original research if transferred into an article.
One consequence of the delete wars going on is that images tend to be kept if the image is seen as notable. But what the policy actually requires is that image contribute to understanding. The former is easier for a newcomer to the subject to grasp. If the image is famous, then it gets the benefit of the doubt. But if an image is not particularly famous, then by default it is presumed to be insignificant; but this is actually not the same thing at all as saying it gives no significant contribution to understanding. The two notions are independent of each other. Duae Quartunciae (talk · cont) 13:58, 8 September 2007 (UTC)
Not sure what you are saying here. If you are trying to make the article say that an image is significant you'll naturaly need to cite a source for that claim, but we are talking about the non-free use rationales here right? All you need there is to explain why you believe the image adds significantly to the understanding of the article, and how it otherwise meet the non-free criteria. Demanding citations from 3. party sources for an image use rationale seems odd. I mean anyone can dispute the rationale naturaly, but bringing original research into it seems odd unless the rationale tries to make some kind of auhtorative claim as part of it... --Sherool (talk) 16:59, 8 September 2007 (UTC)
Exactly. The rationale explains the editor's own reasons for including the article image. Duae Quartunciae (talk · cont) 20:30, 8 September 2007 (UTC)
(EC) This may be true, and I think in several cases that does make a difference. But the number of images which are historic and iconic enough to stand on that alone is very small. Also, I think people need to learn the meaning of the word "significant". It is not enough for a nonfree image to add some degree of understanding to an article, as that could be true of even decorative, tangentially-related images. The article would have to really suffer without it. So here's a simple version for newer editors. A good nonfree rationale should state how it would be impossible (not hard) to convey a significant amount (not any amount) of information without use of the image.[citation needed] If you can't do that, you shouldn't be using the image. As to the original research bit, well, image pages aren't technically articles, but anyone is still welcome to challenge any assertion you make (such as "This image is so historically iconic that we need to use it"). At that point, you do need to back it somehow. If the image really is iconic, that shouldn't be hard, there will be dozens of sources about that specific image. Seraphimblade Talk to me 17:05, 8 September 2007 (UTC)
That is a somewhat extreme interpretation of the policy. :-) SWTPC6800 17:25, 8 September 2007 (UTC)
Very much so. In fact, it doesn't seem to have any basis in the policy. The policy phrase, which was arrived at deliberately and with quite a bit of discussion, is that an image "would significantly increase readers' understanding of the topic." I don't see "impossible without" in there at all. Wikidemo 17:29, 8 September 2007 (UTC)
One thing that's apparent to me, at the moment, is that the hatred by some users of "fair-use" and/or "non-free content," or whatever it's chosen to be called, along with the perception of some that images do nothing significant to improve readers' understanding, is holding this tagging project up a great deal, hardly even to mention the effects on community morale and good-will across the wiki. It seems to me that with an appropriately programmed bot it ought take no more than a month or two to speedy-delete all orphaned NFCs whether they're tagged or not, as well as to delete all images lacking tags. Presumably, Deletion Review would be quite busy during this period, but so what? After that, it seems to me we all can readily continue to argue the merits, or lack thereof, of the various specific kinds of NFC in particular applications. ... Kenosis 21:02, 8 September 2007 (UTC)
You mean something like User:Betacommandbot? But anyway, having looked at WP:IFD, I don't see as much nominations of images on the grounds of criterion 1 and 8 as a few weeks ago. If you want to minimise the chance of images deleted inappropriately, participate more at IFD and state your reasonings so to why it increases understanding and whatnot.--Alasdair 03:47, 9 September 2007 (UTC)
Yes, I do mean "something like User:Betacommandbot". The March, 2007 Licensing policy Resolution demanded three very important things that can be expediently done by a competent bot desisnger such as User:Betacommand. After defining what the Wikimedia Board meant by "free-content license", the Resolution said: "2. Non-free content used under an EDP must be identified in a machine-readable format so that it can be easily identified by users of the site as well as re-users." It was very recently made clear that the words "machine readable" mean, essentially, able to be dealt with by a 'bot', or otherwise automatically categorized. Board member Kat Walsh's clarification has further made clear that an individually hand-typed rationale is not necessary for every particular use of a "non-free" image. The Board also said: "4. Media used under EDPs are subject to deletion if they lack an applicable rationale." The Resolution also said: "5. ... As of March 23, 2007, all new media uploaded under unacceptable licenses (as defined above) and lacking an exemption rationale should be deleted, and existing media under such licenses should go through a discussion process where it is determined whether such a rationale exists; if not, they should be deleted as well."' Like I was just trying to say, #2 and #4 of the Resolution, and part of #5, are achievable by the use of a bot. Let's proceed to get it done please. IMO, once that's set into motion, we'all can continue to argue about the rest of #5 of the Board's resolution. I'm advocating that a bot should proceed to delete anything that doesn't have a rationale, a template on the image page (which includes all images without a template, and also includes NFC files that are "orphaned", not presently in use, as well as NFC files that to not have a template on the file page). The rest, per the Board resolution, must be sorted out by hand, so to speak. I'm advocating to at least get the "machine-readable" requirement done as quickly as possible. If there are too many Deletion Reviews to deal with, the pace of the bot can readily be notched back somewhat, and if the pace is deemed inadequately rapid, multiple bots can be set into motion with identical methods or the pace of the agreed bot can be increased, depending on which might be deemed more feasible. Seems to me, let's get the automated part out of the way, as pronto as possible. ... Kenosis 05:48, 9 September 2007 (UTC)
Well, you and I agree on something. As to the {{fact}} tag in my comment above, I actually do have a source—the definition of the word "minimal". I don't believe anyone would consider "the least possible" to be an unreasonable definition of that word. Well, to use the least possible number of nonfree images, we must make sure the ones we do use are absolutely necessary—not just nice, not just convenient, not just difficult to replace textually or with free images, but actually indispensable. That will result in a small, minimal use of nonfree content, and that is what is required. Seraphimblade Talk to me 06:43, 9 September 2007 (UTC)
The argument is duly noted. Now, about that bot to get the major business out of the way, such that "we're never going to get this done on time" can be removed from the mix of arguments about precisely how much and what type of NFC will be used in precisely what article. Can the participants here at least form a workable consensus on that much for the time being? . ... Kenosis 06:51, 9 September 2007 (UTC)
Seaphimblade is advocating an unaccepted minority position regarding what kind of are appropriate to use in Wikipedia, and I think it confuses the issue to mix that with the question of how we enforce the technical requirements regarding image meta-data. I'm not clear on what you (Kenosis) are advocating. As far as I know we've consistently deleted all the images that are orphaned or lack a copyright tag and there is no serious backlog, hence no issue there. What's left are 170,000 legacy images as per Betacommand's estimate, about half of all the non-free images on Wikipedia, that lack written use rationales or statements of source. The Foundation is not requiring us to remove them; that was a misunderstanding. Further, many and probably the vast majority are perfectly appropriate images for Wikipedia, they just don't have the information written out. Summary deletion of 170,000 perfectly good images would be a huge disruption, and completely unnecessary. Modestly assuming that each took just over 10 minutes to find, format, prepare, upload, and add to the articles, that's 30,000 editor hours worth of work that would be trashed and have to be redone, fifteen person-years at a full time work schedule. By contrast, it would be a simple matter to just add this information for many of the common cases. Even for cases that are not so easy we can do a lot better than summarily deleting them. The use policy as it now stands (which might be changed) merely requires the images to have a rationale or else be deleted. That is still two outcomes. Focusing on deleting them, without making a serious effort fix them, is a serious mistake. There are plenty of ways we can fix them in an orderly fashion. That's what I'm hoping to get people to plan rather than an aggressive purge of images. Wikidemo 07:12, 9 September 2007 (UTC)
Understood. (This, incidentally, is why I posed the question on your talk page about where the 170,000 figure came from.) And it's already been made clear that a custom hand-typed rationale is not necessary for each image, but only that they be placed in a "machine-readable" class so they can be sorted and dealt with automatically in the future. So, among that 170,000, there are a bunch in which the source does not appear to be disclosed on the image page. These would seem to be a priority. If there's no disclosure of the source of the image, that is arguably a lapse in the legally required custodianship of proper copyright management information (CMI, per [ §1202 of the US Copyright laws). Anybody know roughly how many or what percentage of that estimated 170,000 appear to have no source information? .... Kenosis 15:51, 9 September 2007 (UTC)
CMI laws are specialized and I'm pretty certain they don't apply here. Many to most images without rationales lack explicit source information b/c source is one of the fields in the rationale. The way Wikipedia implements that is pretty messed up but that's another story. However, it's also a case where the source is implicit. When you say that the death star (an image that would be summarily deleted) is AT&T's logo, you're saying that the source is AT&T. Perhaps you're thinking of cases where someone is re-using a historic photo or publicity photo. There the source is not clear so it's important to know where the photo came from. Perhaps for that reason, a lot of these images do say where the uploader got them from even if they lack a fair use rationale. Wikidemo 19:40, 9 September 2007 (UTC)
Wikidemo, what you've just said here readily raises about a dozen or two dozen issues that I don't have adequate time or space in the middle of this thread to sort out and effectively discuss. Suffice it to say at the moment that "copyright management information (CMI)" is merely the focus of the title of the section of Title 17 of the US Code (Copyright), and that it's not any more "specialized" than most other sections of Title 17. The issue of expectation that the uploader identify the source from which that uploader got the image is, at present, mixed in with a lot of other issues that are far less important in terms of complying with any relevant law relating to putting up images in Wikipedia aritlcles. What I'm advocating at present is that the importance of an explicit identification of the source of the material is vastly more important than many, or even most, perhaps all, of the various other very confusing issues presently presented to the uploader at Special:Upload. And the present "simplified version", Wikipedia:Upload is, after the prospective uploader clicks on what s/he thinks may be the proper option, is not exactly easy either. Both of these approaches presently lack a demand that the source of the image be clearly identified, or if unknown, to state "unknown" in the source field and in its place identify the copyright holder or other pertinent information relating to copyright or public domain status. All of this is a separate issue from arguments about what particular "fair-uses" or NFCC uses are appropriate in what particular articles. I'm advocating raising the importance of an explicit "machine readable" standard "form field" that must be filled in with either an http IP address or other source where the uploader got the image. Presently, that's definitely not the way it's organized. ... Kenosis 03:33, 10 September 2007 (UTC)
We haven't given a whole lot of thought lately to the question of sourcing image files. It's not directly related to the question of which images are appropriate here or not, or establishing the legality of use, but it's a good way to verify the accuracy of information in the image file. Knowing the source, as opposed to who owns the copyright, doesn't seem vitally important in the case of book covers, film posters, stamps, corporate logos, album covers, paintings, etc. In some cases people pay far too much attention to where the image comes from and not enough to the underlying copyright. For example, take a look at the article Henri Moore. Every single image there claims it's free because the photo is creative commons, public domain, etc. Yet every single one ignores the second issue, the copyright in the sculpture itself. I'm not sure what you would say about the "source" of this copyright. Most people don't exactly know where the sculpture comes from. I suppose you could describe the location and name of the sculpture. But I do agree, the source information should be required at the time of upload. It's weird that it is often added as part of the use rationale template, not a source field related to the image. BTW, I don't see that Chapter 12 of the copyright law requires Wikipedia to do anything with sourcing information or anything else.Wikidemo 07:54, 10 September 2007 (UTC)
On what basis are these conclusions about §1202 drawn? Without getting into details about how I know this, I can tell you that Wikipedia's interests are served by requiring the uploader to provide a statement of the source from which they got the image. This section of USC Title 17 (copyrights) requires that nobody mess with the copyright information. The easiest way to cover the arse of an individual or an organization is to state where you got the image from. It's not the only way, but it's the easiest and safer way. The expectations that a statement by the uploader on the image page of what the source of an image is should be part of the standard expectation, i.e. mandatory to put some information into a field specifically titled "Source from which you obtained the image:__________"
....... And as to the relative importance of knowing who holds copyright, if something is fair use, it's not essential to know precisely who holds the copyright. ... Kenosis 14:40, 10 September 2007 (UTC)
I re-read the text of Chapter 12 a few times and some commentaries on it. 1202 is basically an anti-piracy law. It forbids tampering with Copyright Management Information knowingly and with the intent to facilitate copyright infringement. It's not outlawing anything new, just adding criminal penalties to acts that already constitute direct or contributory infringement. Nowhere does it require anyone to actually have or provide CMI, nor does it apply to good faith attempts (like those of Wikipedia and its editors) to provide CMI. Just knowingly false CMI. The only possible Chapter 12 violation I can see is if an editor makes a bad faith misrepresentation or alterations of the copyright information on image pages, knowing that the images are infringing and that the tampering facilitates the infringement. But the culpability does not rub off on anyone else, just the person who makes the knowing misrepresentation.Wikidemo 16:12, 10 September 2007 (UTC)
Absolutely correct. And part of our job here is to do our best to make sure the info doesn't just mysteriously disappear. Thus, all uploaders should be required to provide a source from which they got the image or other copyrighted file, as opposed to "I dunno". It puts WP in a better position to help prevent piracy. Fairly straigtforward. ... Kenosis 17:43, 10 September 2007 (UTC)
Just want to add a voice that Seraphimblade's reading of the word "minimal" is not sensible in the context of the policy. Minimal, as a word taken in isolation, means none at all. In normal English, the word is almost always used, as it is in this policy, in the context of some requirement. It is the minimal for some purpose. Other wikipedias manage non non-free content at all, which blows out the water the idea that anything is ever strictly necessary. That is not the demands of policy. The English wikipedia adopted an exemption policy that explicitly recognizes a balance between quality and exclusively free content. The word "minimal" in the context of this policy means minimal use in line with the recognized rationale. Hence, if an image is used for a certain purpose; don't take that as an excuse for using it in another context without a solid rationale. Don't use more than is required for the nominated purpose. The policy emphatically does not demand "necessary" or "indispensable". Nothing is truly indispensable and that is not any part of policy. Sorry; I think it is worth repeatedly emphasizing this point. The policy does not demand that images have to be necessary, and it is simply incorrect to present policy in those terms. Duae Quartunciae (talk · cont) 07:40, 9 September 2007 (UTC)
Just to clarify, I'm not against deleting images that lack rationale, I'm just against the deletion of those with proper rationale written on the description (to be replaced by text) on the grounds that "We are a free encyclopedia".--Alasdair 11:14, 9 September 2007 (UTC)
Judging by the discussion, it seems those fair use images could be used when it's difficult (but not impossible) for the info to be conveyed by text, is that right? Also, compare the article Star Wars Episode III: Revenge of the Sith to its German counterpart, and you'll see how dull the latter looks.--Alasdair 01:14, 10 September 2007 (UTC)
Yes, although the information they convey also needs to be non-trivial and relevant for understanding the article better. Merely making it look less dull is obviously not enough. --Sherool (talk) 07:04, 10 September 2007 (UTC)

<Unindented> Of course, I'm referring to the production photos, and one or two screenshots of the film to be used on the plot to highlight the vital battles, climax etc. These are things that words alone are hard to describe, and even if words are used, the average person may still find it difficult to picture the scene in his head.--Alasdair 08:48, 10 September 2007 (UTC)

I've just compared the de and en versions of Revenge of the Sith, and I have to agree that the en one, using fair-use pictures, is much better. The cinematic poster does a good identifying job at the beginning, the green screen pic vividly demonstrates what is meant by green-screen technology. The two plot screenshots are probably not needed, but the cast pic and the 'Odessa Steps' screenshot really add value to the article. The most damning indictment of the de version though is how the picture they have used make them look desperate. A Star Wars logo and a really bad picture of Ewan McGregor at a film premiere. Typical, amateur, low-quality, fan-taken photograph with McGregor's mouth half-open and some photographers in the background. That sort of thing is what gives free content a bad name. Similar to the "nearly-but-not-quite-relevant" pictures used in Main Page TFA blurbs in a desperate scramble to avoid using non free images. In such cases, I'm firmly in the "better to have no pics rather than a poor-quality or irrelevant free pic" camp. Carcharoth 10:48, 10 September 2007 (UTC)
Another gripe I have is that there are websites out there that allow you to freely share and distribute their work, but are of no help to Wikipedia due to the NonCommercial or NoDerivs CC license. Just how willing are they to let others distribute it? If I write an email to the webmaster, will he be willing to distribute it with a CC license that is simply Attribution or Sharealike?--Alasdair 06:27, 11 September 2007 (UTC)