A general reminder: Please stop uploading images where permission is granted for non-commercial use only, effective immediately. Under official Wikipedia policy, these images are no longer accepted. [1]. It is anticipated that existing images with the {{noncommercial}} tag will be deleted at some point in the future (possibly after a new upload form is in place), except for images whose use can be justified on other grounds. --Michael Snow 16:36, 11 Aug 2004 (UTC)
At the risk of making a lot of extra work for myself, I would be willing to accept requests for creating GFDL replacements for noncommercial-use illustrations. See my user page for a list of the sort of things I have illustrated. -- Wapcaplet 16:48, 11 Aug 2004 (UTC)
I can help out as well with diagrams. (Some of my diagrams can be found here) theresa knott 18:21, 12 Aug 2004 (UTC)
I guess this means we shouldn't be featuring these images on the front page? [2]anthony(see warning) 16:51, 11 Aug 2004 (UTC)
Yes. Middle-earth is actually what called my attention to the problem, but by the time I noticed it was already on the front page, and I didn't think the issue warranted taking it down once it had gotten there. We are not yet to the point of removing all of these images from articles and deleting them, but I agree that they should not be used on the front page. I regret that your objection wasn't acted on while featuring this article was still in the planning stages. --Michael Snow 18:19, 11 Aug 2004 (UTC)
This is a great blow to the ongoing process of illustrating Wikipedia. There are a lot of people out there taking a lot of good photos which understandably they don't want other people making money out of. However they are quite happy, even honoured, to allow use of their images for noble projects like Wikipedia. I have been uploading a few of these non-commercial images recently to illustrate articles on towns. There is absolutely no reason why they should not be used. Downstream reproducers of Wikipedia content should simply not incorporate the images into their content if they intend to put it to commercial use. This can be achieved very easily with the tagging of images with their licensing status. What this policy is doing is allowing downstream commercial users to dictate to us here at the main project what we can and can't include. Can somebody please offer a decent explanation as to why non-commercial images shouldn't be included so that we can all come to an informed consensus on the matter instead of having policy decided by a small clique on the mailing list and announced to the rest of us from on high. If this policy is adopted then we are pointlessly preventing ourselves from using images which their creators are quite happy for us to use. A far greater problem Wikipedians should be devoting their time to is the lack of any licensing information whatsoever on the vast majority of uploaded images. — Trilobite(Talk) 20:55, 11 Aug 2004 (UTC)
IANAL, but it appears that restrictions on re-distribution directly conflict with the GFDL, our license of choice. - jredmond 21:58, 11 Aug 2004 (UTC)
The issue is not the GFDL — the GFDL is chosen because one wants to allow commercial use, not vice versa. —Steven G. Johnson 22:22, Aug 11, 2004 (UTC)
Our text is GFDL - images are not (which is why we have the image pages). →Raul654 22:07, Aug 11, 2004 (UTC)
The trick comes in when we place images on articles, though. Is the image a part of the article? If so, what license applies to the compilation of GFDL text plus non-GFDL images? - jredmond 22:12, 11 Aug 2004 (UTC)
But then there is an inconsistency. Mr Wales writes, "For the time being, I think we should rely on fair use, because it's a good thing, but cautiously so." We certainly cannot grant licences for images, but we use them nevertheless. Another problem is that the restriction on non-commercial images can be easily evaded with the fair use doctrine. -- Emsworth 22:18, 11 Aug 2004 (UTC)
In the present (US) legal environment, "fair use" is a pretty flimsy crutch to lean on. Besides which, Wikipedians seem to think "fair use" means "we can use any image we like as long as we really really want to." —Steven G. Johnson 22:32, Aug 11, 2004 (UTC)
Jimbo'll have to speak for himself, but I read that sentence to mean "Until we can get new, more libre images, fair use will have to do". This is consistent with the bits on fair-use content in Wikipedia:Copyrights. - jredmond 22:36, 11 Aug 2004 (UTC)
Yes, it's pretty clear what Jimbo means here. He says that non-free content "should be removed from Wikipedia with reasonable haste." Then he says "This decree is only about non-free licenses _as a justification_ for images being on Wikipedia, and does not comment on, nor affect, evolving doctrine on 'fair use'." "For the time being, I think we should rely on fair use, because it's a good thing, but cautiously so." He is saying that we should get rid of content that is used under a non-free license, but that this doesn't apply to free use images. It's an interesting statement, because you could technically say since these images are copyrighted, they can be used under fair use. But IANAL. — マイケル₪ 00:10, Aug 12, 2004 (UTC)
In addition, the statement at the bottom of each page reads, "All text is available under the terms of the GNU Free Documentation License" (emphasis added).
Non-commercial-use images are problematic for the same reason that non-commercial-use text is — there is no reason to allow one and not the other. We allow commercial use because we want to allow things like Wikipedia being distributed on CD by CheapBytes for a few dollars, being included with future Linux DVDs as a built-in OS resource, being bundled with every PalmPilot sold... as long as the encyclopedia material itself is never made proprietary. This is the same as the free-software/open-source philosophy (both of which movements require that commercial use be allowed). —Steven G. Johnson 22:22, Aug 11, 2004 (UTC)
From a contributor's standpoint, why not just contribute the image under the GFDL? Although the GFDL does not prohibit "commercial" use per se, it prohibits most uses that people ordinarily think of as "commercial" — for example, usage in a typical magazine or newspaper — because it prohibits proprietary use (all derived works need to be under the GFDL as well). (Indeed, just as companies do with GPL software, you could imagine a professional photographer contributing GFDL images as a promotion, and then selling the right to use a non-GFDL, proprietary license to magazines etc. that want to use the image.) —Steven G. Johnson 22:28, Aug 11, 2004 (UTC)
I think what Trilobite is talking about is images that have been copied from elsewhere under non-commercial use permissions. In that case, you don't have the ability to contribute the image under the GFDL yourself. --Michael Snow 22:31, 11 Aug 2004 (UTC)
I think he was talking about photos taken by individuals...unlike companies, individuals can sometimes be persuaded, and you just need to convince them that the GFDL prohibits most of the uses that they want to prohibit with a noncommercial restriction. —Steven G. Johnson 22:35, Aug 11, 2004 (UTC)
Sure they can be persuaded, and if so, great. But it does take a little more work, and as you note about fair use, some people are just dying to contribute this lovely image they found "right now", without caring about the implications of copyleft. --Michael Snow 22:52, 11 Aug 2004 (UTC)
I meant photos taken by anyone who has nothing to do with Wikipedia, so they can't just say, "I'll make things easier and license my images under the GFDL," as I would (and have done) with my own images I want to put on Wikipedia. There are a lot of people making their very useful photos available for non-commercial use which Wikipedia should be able to take advantage of. — Trilobite(Talk) 23:04, 11 Aug 2004 (UTC)
Wikipedia is committed to being as free (libre) a project as possible, as part of the open content community that relies on copyleft licenses. This is a core part of our mission. We define ourselves as an open-content encyclopedia on the Main Page. This principle has been policy since the beginning of the project.
There are a lot of people out there writing a lot of good text which understandably they don't want other people making money out of. This text is not allowed on Wikipedia, because it is not open content. There are plenty of people who might let us use their text, or their images, as long as it can only be used on Wikipedia. Because we're a noble project, because they're honoured to have it published, because they want publicity, motives may vary. We can't accept it on those terms, because it's not open content. The policy against non-commercial-use-only images reflects that commitment. --Michael Snow 22:31, 11 Aug 2004 (UTC)
I quite agree with you where text is concerned. However, text which someone wanted to contribute on a non-commercial basis would of course make things very difficult and complex, as text is added to and taken away from, edited mercilessly etc. It would be absurd to have different portions and fragments of text under different licenses, but images are a very different matter. They are discrete entities instead of something that can be mixed up with new contributions until it's impossible to extricate the original. They are also, as Raul654 pointed out, on seperate image pages which are simply referenced to in the Wiki markup. By tagging those images which are not available for commercial use, downstream reproducers, or future commercial applications of Wikipedia such as those which have been mentioned, can remove them automatically. This makes things a little bit more complicated, but is greatly preferable to purging Wikipedia of vast swathes of perfectly good graphical content. Am I the only one who still isn't persuaded that this policy makes sense? — Trilobite(Talk) 23:04, 11 Aug 2004 (UTC)
A license to use something only for noncommercial use is not free, and it's inconsistent with our underlying philosophy. The fact that you can separate the images out from text doesn't matter. What we would be doing is flatly saying no, you can't use this content if it's for commercial purposes. In other words, the content is definitely not open, even though we claim that we are.
If you can claim fair use for an image, that has a slightly better shot at working in an open-content world, because commerciality is only one issue considered in fair use analysis. And with fair use, we're not telling people "you can't use this stuff", but we're tagging it so they can separate it if necessary. What we're really telling them is to figure out for yourself if what you're doing is still fair use. --Michael Snow 23:27, 11 Aug 2004 (UTC)
The mainstay of our content, which is text, would be very much open and free. Images not available for commercial use would serve as an embellishment on the Wikipedia website itself, as this is not a commercial use. In some other applications of our content, those images would not be available. It's as simple as that. Fair use, as I understand it (and I am by no means an expert), is a phenomenon of US copyright law of dubious international applicability. I have always thought it best avoided as it is often far from clear where the line between fair use and unauthorised copying lies. Non-commercial permission however is clear and unequivocal — we can use it on this website and any other non-commercial application, and we simply blank it out from anything commercial. This is easily achieved by putting all such images into a category. This is the Wiki equivalent of the sort of machine-readable metadata Creative Commons encourages the use of along with their licenses, so that computers can be used to selectively do things with content according to how it's licensed. Technologically this is very simple for Wikipedia and need not contradict the philosophy of the project at all, as long as we remain a text-based encyclopedia with images as non-essential extras. After all, we should already be aiming at this if only for accessibility reasons. I would appreciate some input into this debate from others as I think it's one worth having. — Trilobite(Talk) 23:58, 11 Aug 2004 (UTC)
Noncommercial use is technically legal on Wikipedia itself right now, but would not be if we decided to add advertisements to the site (I'm not suggesting this is planned, but it has been contemplated). But anyway, such images are clearly not open and free, and I don't see why we should stray from our commitment to open content in order to embellish the website.
Fair use is specifically US, but other countries have fair dealing, and for a more international basis, the Berne Convention has fair practice. Determining what's "fair" tends to be case-by-case analysis, and the US may well be the most liberal jurisdiction in that regard, but the principle is internationally available.
Incidentally, if images are "non-essential extras", why exactly is it so important to allow images under noncommercial-use permissions? That philosophy seems to negate all of the arguments raised for including them. --Michael Snow 00:20, 12 Aug 2004 (UTC)
Personally, I'm glad to have non-commercial images go. The goal of Wikipedia is to create a free resource which anybody can easily take material from. Moving to the GFDL will not remove any credit from you – and you aren't losing any money anyway, unless you're rich enough to distribute the picture, etc. What's so wrong about letting a company use your image? As long as they credit you, there's nothing you're losing. As for fair use, I consider images commonly seen (i.e. a particularly famous image of a celebrity), or images distributed publicly (i.e. broadcast on television, published in major publications) to be valid fair use material. Anything else is dubious. So, for example, an image of the cover of the Yesterday single would be fair use, but not a copyrighted image of the Beatles (or anyone else) performing it, unless licensed under the GFDL and/or published in several major publications. Johnleemk | Talk 10:53, 12 Aug 2004 (UTC)
Have you considered this from the contributors point of view? If you have an individual who donates his or her time to write an article for which there are none of the traditional benefits of commercial reward or peer recognition and that individual also prepares images to support and embellish the article then it should be entirely up to them if they do not want to see their work profited from commercially unless they do so too. prometheus1 20:51, 12 Aug 2004 (UTC)
I am not contesting that right. What I am contesting is that we allow that to happen on Wikipedia, and that as a contributor, IMO, there's little reason for it. Unless you have a way to make loads of money from it, there's no reason for not licensing the image under the GFDL or some other free license, unless you're one of those anti-corporate...um, "girlie men". (Don't take the comment seriously.) The decision is up to the image's owner, but we really shouldn't be using these images on Wikipedia. Jimbo's posts on the mailing list say it much better than I ever could. Johnleemk | Talk 05:12, 13 Aug 2004 (UTC)
I'm a big fan of Arnold and I don't think you would call me a "girlie man" if you saw me ;). But I am not a fan of having others sell and make money out of a contribution that I made in the interests of sharing knowledge. From both an ideological and economic perspective if others wish to make money out of it then I should be entitled to my fair share in the great corporate tradition! If you go to pubmed books online you will find that some images cannot be shown in the online version. If it's good enough for them to compromise it should be good enough for wikipedia. I think placing such a restriction in the event that wikpedia want to change the business model so that they can sell CD's or use banner advertising to generate income is illogical. It is easy to tag the images such that if an ad appears or if the content is going to CD that it be not included. Otherwise pay a percentage back to the contributor - or better still stay non-profit like everyone believes wikipedia is, then there's no problem of using non-commercial use. prometheus1 06:42, 13 Aug 2004 (UTC)
A non-profit organization can make commercial use of images (or other things), right? Dan Gardner 19:20, 14 Aug 2004 (UTC)
But if you really had a strong interest in sharing knowledge, you wouldn't mind allowing commercial uses of your contributions. After all, don't you know that every piece of text on Wikipedia can be taken and resold as long as we are credited, and nobody can say a thing? Jimbo makes the point for why we shouldn't be allowing non-commercial use only uploads on Wikipedia brilliantly – it's supposed to be free. If we have to rely on restrictive licensing, it goes against our original goal of an encyclopedia anybody can take and reuse. I can't wait to see how you're going to ask for a cut when some company decides to lift text from DNA repair, by the way. ;-) Johnleemk | Talk 07:30, 16 Aug 2004 (UTC)
I think this decision has the priorities wrong. At least non-commercial-use images have licences and at least we can use them. Our first priority should be to remove images with no source; our second should be to remove copyrighted images unless the "fair use" defence is really solid (e.g., corporate logos). Non-commecical-use images should come a distant third. Gdr 15:35, 2004 Aug 12 (UTC)
The first priority is quite correct. People shouldn't be uploading images without source information any more than they should be uploading non-commercial-use images. In both cases, actual removal is waiting until we have a new upload form that will improve compliance. Weeding out improper claims of fair use would be good, too, but you have to debate those individually. Keep in mind that some of the images used under non-free licenses will also need to be considered for possible fair use claims. --Michael Snow 17:47, 12 Aug 2004 (UTC)
First off, I agree with Trilobite: "non-commercial use only" images can – if tagged – be easily removed by commercial re-users of Wikipedia content. Wikipedia itself is a non-commercial endeavour, and I can't see anything wrong with the use of such images here. Removing them from here would only needlessly deprive Wikipedia of many great pictures.
It seems to me that downstream republishers of Wikipedia content will find it far more difficult to properly deal with "fair use" claims: they will in effect have to re-evaluate each and every of these images to check whether the fair use claim made by Wikipedia also applies in the jurisdiction they're under.
I think Gdr has the priorities exactly right. First deal with copyvios (we already do), then images without source and licensing info, then verify those "fair use" claims. Deal with problems that could affect Wikipedia itself first. "Non-commercial use only" images pose no legal problems for Wikipedia, and as I wrote above, commercial dowstream re-users can remove them. We have more important things to do.
How many "non-commercial use only" images do we have, anyway? Category:Non-commercial use only images currently lists 81 images, but I know that this number is far too low, maybe due to some corrupt link table in the database. A search of the "Image" namespace for "non-commercial" lists more than 500 results. (I tried to find out how many exactly by playing around with "&limit=" & "&offset=" in the URL, but queries invariably timed out for me.)
I don't know. Any image which is fair use for Wikipedia is probably fair use for just about any noncommercial encyclopedia, so non-commercial only images are probably more restrictive. That said, both issues need to be addressed. We shouldn't have many images here which we can't put in the print version, as having them here will just make us lazy about replacing them with free ones. anthony(see warning) 20:31, 12 Aug 2004 (UTC)
A lot of people are confused to as to what Jimbo meant. I think the text of from his latest e-mail states much more clearly what he means:
> We have a lot of images with "no commercial use" caveats. Does that
>mean these images have to be removed?
My position is that yes, eventually, these do need to be removed.
There can be exceptions, though...
If an image meets our fair use/fair dealing guidelines, which
basically means that it is easily fair use for us, and likely fair use
for most contemplated reusers, then we can keep it (because it is free
in the relevant sense) *even if* we are *also* able to obtain a
license of some sort. It can be wise for us to have licenses for
content that we could use without a license, just to make things more
clear.
An example of a "fair use" that would likely be fine for just about
any contemplated reusers would be a quotation from a book that an
article is discussing. Another example would be a screen shot from a
movie in an article about that movie.
If the _only_ way we can use a particular image is through a non-free
license, and we believe that a fair use defense would be unavailable
to us, or to most contemplated reusers, then it should be avoided.
--Jimbo
I hope that clears up any confusion anyone had. — マイケル₪ 20:37, Aug 12, 2004 (UTC)
Có ai hiểu họ noí gì không? dịch em nghe vơí! thank!
Images which include an advertisement for the photographer watermarked on them
(apologies for duplicating this on several talk pages that don't see a lot of traffic, because I don't know which one is most likely to get a response)
This picture has a watermark on it that advertises the photographer and his website. I didn't delete it because it IS a really great picture, but I'm asking if that watermark is an acceptable element when it's on the Venice article. Alternately, would it be permissible to crop that top 3% of the photo and use that one instead? Fred Zepelin (talk) 16:31, 9 March 2022 (UTC)[reply]
Photos and scanned images should be in JPEG format, though a PNG may be useful as well, especially for software screenshots when only a raster image is available (JPEGs are a lossy image format, and PNGs allow further editing without degrading the image).
– to –
Photos and scanned images should be in JPEG format, though a PNG may be useful as well for simple subjects (where PNG would result in a smaller file without degrading quality).
This was originally added, apparently without any discussion, back in 2015. I think it's important to qualify the reasons one might choose PNG for a photograph or scanned image. Note also that the bullet directly above the one I edited already says this:
So the PNG for screenshots situation is already explicitly allowed (making the old version somewhat redundant). One of the other things mentioned in the previous version was the concern about image degradation for images that are repeatedly edited. While the concern is definitely valid (see Digital generation loss), whether or not it is an issue depends largely on the type of image under discussion. For free content that may undergo future revisions (text that is updated, etc), it is a valid concern. For non-free content, we should not be making edits that would introduce that kind of problem. If anyone has any suggestions or questions, please let me know, but I wanted to put my reasoning on the talk page in the event I'm reverted. —Locke Cole • t • c05:43, 24 May 2022 (UTC)[reply]
I can't seem to find any policies involving obstructions in images of people.
I think certain obstructions in images of people should be prohibited (e.g., Mizkif's baseball cap covering part of his face + Sodapoppin wearing sunglasses), while others shouldn't be a problem (e.g., Jacksepticeye holding a microphone not too close to his face + TommyInnit's headset microphone). L33tm4n (talk) 21:51, 19 July 2022 (UTC)[reply]
You need to get consensus here - perhaps with a wider discussion as this is a major change to policy and seems unnecessary to add to the policy itself as it's not that big of a deal and shouldn't be spelled out as policy. Do not restore until you can get a clear consensus. PRAXIDICAE🌈04:24, 18 August 2022 (UTC)[reply]
I don't think aesthetic judgements on photos, like how clearly they depict the subject, belong as hard and fast rules in image policy. And the heavily pop-culture-based examples, written as if with the incorrect assumption that all editors would recognize those names and those images, are right out. —David Eppstein (talk) 04:32, 18 August 2022 (UTC)[reply]
That's not how this works or how policy changes work. This makes no sense in the context of any policy as this is a preference. If you think it should be changed, you need to start an WP:RFC. PRAXIDICAE🌈19:55, 19 August 2022 (UTC)[reply]
@Praxidicae:WP:PGBOLD encourages you to take risks in discussions like these. Also, I created this discussion a month ago, but only got a response yesterday. Also, WP:EDITCON allows you to edit these pages directly, and, in some cases, let the content remain until a consensus is reached. L33tm4n (talk) 19:59, 19 August 2022 (UTC)[reply]
Please re-read that which you link. Bold changes to policy are not the same as the rest, not to mention your previous edit summary about 1RR isn't close to being applicable here. You boldly changed it and were reverted, that means you discuss and get consensus, not reinstate your edit.
Regardless, this was a really bad change to policy and if you actually want to change it you need to create a proper RFC. Discussing it alone in this single section on this single talk page isn't going to fly. PRAXIDICAE🌈20:01, 19 August 2022 (UTC)[reply]
Request for comment on obstructions in images of people
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
I can't seem to find any policies involving obstructions in images of people.
I think certain obstructions in images of people should be prohibited (e.g., Mizkif's baseball cap covering part of his face + Sodapoppin wearing his sunglasses) unless the person is notable for them (e.g., Andrew Tate wearing his sunglasses), while others shouldn't be a problem (e.g., Jacksepticeye holding a microphone not too close to his face + TommyInnit's headset microphone). L33tm4n (talk) 20:12, 19 August 2022 (UTC)[reply]
Oppose obviously. This is WP:CREEP but also unnecessary and there is no adequate explanation of how or why this is a problem, nor that it actually exists as a major problem in need of changing. I would also vehemently oppose the change to wording as proposer previously added as it's unduly focused on random individuals and confusing to both readers and editors. PRAXIDICAE🌈20:26, 19 August 2022 (UTC)[reply]
Oppose, mostly per WP:CREEP. The proposer has repeatedly used this bizarre wording focusing on three specific images that they find problematic, something completely inappropriate for a general policy. —David Eppstein (talk) 20:53, 19 August 2022 (UTC)[reply]
Oppose, I don't think we need an extra rule but why would we prefer no image to one with an "obstruction" if that is all we have? Also, your examples are very unclear (or do I have to know these people to understand what you are saying?) What kind of images would be allowed or prohibited? —Kusma (talk) 20:56, 19 August 2022 (UTC)[reply]
Oppose We should be using the best images, not artifically limiting ourselves about which images we van use. In addition, in some cases, obstructions are an important part of a person's image - think of Sia, who spent much of her career deliberately obscuring her face.Nigel Ish (talk) 22:13, 19 August 2022 (UTC)[reply]
Oppose. Counterproductive, pointless rule creep that doesn't address any actual issue. The sections on selecting images and Wikipedia:Manual of Style/Images already instruct people to use photos where the subject is clear and unobstructed. Since we require images of living people to be under a free use licence we are limited as to what images are available. Discarding otherwise good quality images of people because they happen to be wearing glasses or a hat seems at best to be counterproductive; I certainly fail to see how they present a big enough issue to require prohibition. For images where the subject is obscured to such a degree that they they aren't actually visible "The subject of the photograph should be visible in the photograph" is a common sense statement that does not require legislating. I agree with my fellow editors above that the wording added to the policy was inappropriate - a policy applicable to every article on the site should not be written in terms of English speaking YouTube personalities - how is this policy supposed to work with people who wear religious headwear for example? 86.23.109.101 (talk) 22:24, 19 August 2022 (UTC)[reply]
Oppose. I think, all other things equal, an unobstructed image should be preferred if available and properly licensed. But if the partially obscured image is all there is available, we should use that until and unless something better comes along. I don't see any usefulness in the proposed rule. SeraphimbladeTalk to me22:47, 19 August 2022 (UTC)[reply]
Oppose. As a policy prohibition, absolutely not. Now, should we use the best (editorially subjective) image we can get with an open license - sure, but that determination of which openly licensed image can be made by editors case by case. — xaosfluxTalk23:16, 19 August 2022 (UTC)[reply]
Support This is kind of wild but sure, let's do it. If there is an obstruction in the image then the article is in violation and someone should report it on the talk page. The image can stay until replaced, but from the time of notice the article is on alert. During that time editors should 1) seek alternative images 2) contact the subject of the article to ask them for better images and appropriate licensing or 3) draw representative replacement images using art skills or AI drawing. Thanks for taking initiative to make a proposal. Bluerasberry (talk)23:32, 19 August 2022 (UTC)[reply]
In violation of what?! Are you seriously suggesting if no freely licensed articles of living people are available we should allow what is effectively fanart? PRAXIDICAE🌈23:39, 19 August 2022 (UTC)[reply]
@Praxidicae: In violation of this proposed policy.
The second part you raise about fanart is a different and more interesting issue: Yes! I am seriously suggesting that if no freely licensed images are available then we should use fanart. I commissioned fanart for Jashodaben Modi for example, and Satyajit Ray had fan art for the lede until someone identified a photo. One of the Wikimedia Foundation's major financial investments in the United States this year was meta:Communications/Wiki Unseen, where they commissioned fan art. Fan art is on the table, and more importantly, so is AI generated art. Thoughts on that? Bluerasberry (talk)20:41, 20 August 2022 (UTC)[reply]
Re using fan art for articles on living people: see WP:BLPIMAGE "Images of living persons should not be used ... to present a person in a false or disparaging light". If the fanart is a trace of a copyrighted photo (which they often are) then it is a copyvio, and otherwise it is of a fictional and therefore false scene. See also WP:PORTRAIT "editors should be reluctant to resort to such depictions given the absence of any concrete encyclopedic information value and the danger of covert POV overtones". —David Eppstein (talk) 21:02, 20 August 2022 (UTC)[reply]
Oppose per WP:CREEP. Any partially obstructed photo would become a debate, I would add. No actual problem has been presented that this would "fix". Dennis Brown - 2¢08:24, 20 August 2022 (UTC)[reply]
Oppose. The image on TommyInnit is the only free image of the person. I hadn't even noticed that he had an earpiece microphone until you pointed it out. I think that, naturally, "better" images implicitly make a remark on the obstruction of images. No hard rule. SWinxy (talk) 22:15, 20 August 2022 (UTC)[reply]
Oppose per WP:CREEP. I am also quite confused by the examples that User:L33tm4n has given. The pictures for Mizkif and Sodapoppin don't display the obstructions that were mentioned (and I can find no evidence of such obstructed photos having been in the article recently) and TommyInnit uses a picture that would be objectionable for its general blurriness regardless of the microphone. --User:Khajidha (talk) (contributions) 08:56, 21 August 2022 (UTC)[reply]
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
An editor on Talk:Mahsa Amini protests has asked to blur all faces of user-uploaded images of protesting crowds, given that Iranian authorities have reportedly used social media photos on other sites to identify and arrest protesters. My interpretation of Wikipedia policy is that (presuming we agree that blurring the photos and facial expressions could reduce their documentary value) we have no discretion to blur the photos even if we page editors decided we want to. Is this interpretation correct? If an editor disagrees with this policy, has "personal safety" or similar been raised as an issue before, and if not, what would be the best forum for such an editor to raise it in? Rolf H Nelson (talk) 01:40, 13 November 2022 (UTC)[reply]
If the photos are taken in a public location, we really cannot take steps from showing the faces that were involved. We do consider this for photos in a clearly private setting, where those photograph would have a reasonably degree of privacy. Masem (t) 01:47, 13 November 2022 (UTC)[reply]
Whether to use the photos in the Wikipedia at all could be something decided through discussion on the article's talk page; however, blurring out faces would probably need to be discussed over on Commons with respect to any issue realted to c:COM:BLP. It does seem though that the photos were taken by some who was present at the time, who then decided to upload them willingly to Commons under a license that makes them not only easier to use on Wikipedia, but also easier to use by others in other ways that have nothing do with Wikipedia and nothing to do with Commons. Of course, "others" also includes the Iranian Government and "other ways" also includes identifying individual protestors, I guess the uploader could request that the files be deleted (e.g. a courtesy deletion), but they should do that asap. Even if the files are deleted from Commons, though, it won't really stop anyone who's already downloaded them from continuing to do so. I guess it's possible that the uploader didn't fully realize what it means to upload a photo to Commons, but they would need to be one that has to correct that by requesting its deletion. FWIW, this kind of thing happens fairly often, including some recent well-publicized examples like the January 6 United States Capitol attack, and individuals find the photos they taken and posted online are subsequently used by authorities to identify and prosecute themselves or other individuals. So, it really would be kind of hard these days for the uploader to claim ignorance and to have not fully realized the possibility of the Iranian government using their photos to somehow identify these protestors. -- Marchjuly (talk) 05:50, 13 November 2022 (UTC)[reply]
This very same subject has also been raised here: Talk:2022 COVID-19 protests in China#Images with identifiable faces. It seems that a broader conversation about ethics and safety related to image use is perhaps warranted, especially given the prevalence of facial recognition technologies and better understanding Wikipedia's role in all of this. While it is easy enough to say that a photograph has been taken in the public sphere, and that the photographer has granted permission to use the image, this does not imply consent from any or all of those who were photographed. I understand it is each individual's responsibility to protect themselves and their identity while in public, but Wikipedia could also balance risks to personal safety, especially when images have originated from exceptionally repressive regions of the world, where repercussions of protest sometimes involve not just arrest, but indefinite detention, torture, or worse. Thanks! 98.155.8.5 (talk) 19:37, 4 December 2022 (UTC)[reply]
The need for subject consent seems to depend upon the copyright laws of where a photograph is taken per c:COM:NCR. The consent of someone protesting in a public place isn't needed under US copyright law per c:COM:BLP and I don't believe it's typically considered a violation of their privacy to photograph them unless they've got a legitimate expectation for privacy. As for the "balancing of risks", I don't think there's any way for Wikipedia to do such a thing except on a case-by-case basis through article talk page discussion. You would have to somehow establish that it's a WP:BLPIMAGE issue and then figure out how to resolve it. Trying to create a project-wide policy change that suits your particular perspective on this seems like a very hard (if not impossible thing to do) and would almost certainly need a well-thought out and well-participate WP:RFC to achieve. -- Marchjuly (talk) 00:00, 5 December 2022 (UTC)[reply]
Interesting, thanks for your reply. I didn't know anything about the Country specific consent requirements, that's useful for consideration (but both Iran and China, in these examples, do not require individual consent for image use). And yeah, I realize it would be a major uphill battle to push for a project-wide change of policy. That isn't in my interest, unless there were others who are super passionate about trying to draft an RfC that could take into consideration some of the nuances of privacy, facial recognition, and specific threats of repressive regimes etc. Cheers! 98.155.8.5 (talk) 01:34, 5 December 2022 (UTC)[reply]
Somewhere there used to be explicit good useful discussion that images may be used in Wikipedia as an exception to wp:NOR. We are allowed to go to a historic site, to a current event, anywhere, and take photos, and bring them into wikipedia and give sensible straightforward interpretation. "there were at least 10 people at the February 9, 2023 rally". "since listing on the historic register in 2010, the building has collapsed". Say it out loud: this is effectively an end-run around "no original research" rule. It is, and that is fine.
Currently, the "GNU Free Documentation License" section reads, in part:
The GNU Free Documentation License (GFDL) is not permitted as the only acceptable license where all of the following are true:
I'm confused as to what exactly this sentence is saying. Does this mean that works licensed under the GFDL alone are not considered free use if those conditions are all true, or does it mean that works licensed under the GFDL at all are not considered free use if those conditions are all true (regardless of whether other licenses hold)? Or does it mean that works licensed under the GFDL are not considered free use at all (because the GFDL is the only acceptable license where those conditions are all true)? And why call it an "acceptable" license if it's not permitted? I think the wording of this section could really use some clarification/rewriting. Edderiofer (talk) 14:28, 15 November 2022 (UTC)[reply]
Thus, it is preferable to publish the work with a dual license, adding to the GFDL a license that permits use of the photo or text easily
Ah, I see, I was completely missing this context. I think this would be a very helpful sentence to add to the policy for clarification. Should I WP:BEBOLD here and just add it, hoping that someone else will fix it if it's unclear, then; or should I seek consensus first? (I'm hesitant because this is a Wikipedia policy, not a standard article.) Edderiofer (talk) 11:41, 24 February 2023 (UTC)[reply]
Many images on the Commons use the PD-US-no notice template, which states that works published in the United States between 1928 and 1977, inclusive, without a copyright notice are in public domain. For example, this image of the Beatles photographed by Henry Grossman appeared in a July1967 issue of Billboard magazine without the appropriate copyright notice, and so it is in the public domain. But what about later higher resolution instances of the same image?
A cleaner scan and higher resolution instance of Grossman's photograph appeared in The New York Times in January2023. Miklogfeather uploaded the image to the Commons here, but I am concerned because I see that in the Times article, it includes a notice below the image which states: "Henry Grossman/Grossman Enterprises. All rights reserved." Thoughts? Tkbrett (✉)00:00, 25 June 2023 (UTC)[reply]
I think the relevant rule here is Reuse of PD-Art photographs and the Wikimedia Foundation state that "faithful reproductions of two-dimensional public domain works of art are public domain", so higher resolution images of PD images are acceptable. Endoftalk10:18, 29 July 2023 (UTC)[reply]
It might be a dumb question and I apologize beforehand.
But what exactly is the ruling on photos in "historical books" that have been luckily archived online; can they be used and posted to Wiki? And also, books that are potentially hidden in the public and not archived? Thanks Yeahimaboss413 (talk) 03:00, 3 September 2023 (UTC)[reply]
In this case, I would think we'd have a hard time of justifying that image under WP:NFCC#1, in that, within prose, stating "an image of two dark-skinned people labeled as 'Gorillas'", would be equivalent to the non-free Twitter post, as well as avoid issues with photos of people that could be clearly identified and whom are not within the public eye. Masem (t) 03:22, 23 October 2023 (UTC)[reply]
Concur with Masem. If that image garnered enough attention to warrant a stand alone article about the image, it would be appropriate as fair use there. To illustrate a case of AI failing at something, no. There's a decent chance that a free license example could be found or created. As Masem notes, that generates a failure of WP:NFCC#1. --Hammersoft (talk) 05:45, 23 October 2023 (UTC)[reply]
I searched this page for disgusting, offensive, gross etc.. and can't find any guidance. Wikipedia has a tendency to settle on extreme images particularly with medical topics ie. dealing with the human body. IMO the image guidance should be for a typical case, not an extreme one, with a sensitivity to distraction from reading or even wanting to read the text, when your eye is literally forced towards a repellent image. I brought this up at Talk:Hematochezia#Photo_extreme, but perhaps unsurprisingly, this is not an article where anyone is hanging out much. Maybe we could get more readers and editors if the image wasn't so repellent. The image doesn't need to be so extreme to get the idea across. -- GreenC15:18, 19 October 2023 (UTC)[reply]
Second paragraph under WP:IMGCONTENT is what you are looking for. We so want to operate under the principle if least surprise, but at the same time we are not censored. That type if image on that page seems a reasonable selection given it a medicinal condition. Masem (t) 16:52, 19 October 2023 (UTC)[reply]
From your link, I found MOS:OMIMG which says "Avoid images that contain irrelevant or extraneous elements that might seem offensive or harassing to readers". So the terminology is "extraneously offensive" "offensively extraneous" images.
Say I replaced that image with one of soiled toilet paper (red colored). That would also be reasonable, to get the idea across. Someone else would say hey, let's make this "better", they add an image of a single turd with some blood in the water. Then someone says we can improve this by adding more turds and more blood, the image in the article now. This is how Wikipedia works. There is a race to the bottom, so to speak, ending at an image that is offensive in the extraneous amount of detail. -- GreenC17:58, 19 October 2023 (UTC)[reply]
In the article about Hematochezia, it is hard to say that a bloody stool is extraneous or irrelevant, it's exactly what the medical condition is about. Masem (t) 11:46, 20 October 2023 (UTC)[reply]
I'm pointing out that the image on that page is not extraneous nor offensive given the topic matter. Now, you may have a point if the image was taken to the extreme as you suggest (a full bowl, for example), because we don't need that degree to show it. There is a tastefulness factor, but in an article about a medical condition related to bloody stools, its hard to avoid actually showing a simple representative image of that, knowing that anywhere else on WP it would be tasteless to include. Masem (t) 15:39, 21 October 2023 (UTC)[reply]
Right, the same could be depicted in a less voluminous manner. That's the test, can a less disgusting image get the same idea across, even when the less disgusting image is also disgusting. Wikipedia tends to go in the direction of maximum disgusting, because it's a factual depiction afterall. Disgusting maximalist vs. minimalist. Another way of saying offensively extraneous. -- GreenC18:37, 21 October 2023 (UTC)[reply]
That is an important factor, that what we show is limited by what free work is out there. If there was a better free image (possibly a diagram of the human body that shows why this happens), by all means we should use that and move the photo down.
We gave taken action when people tried to upload highly pormographic images for articles on sexuality, reproductive organs, and so forth, so the hypothetical example of a full bowl given above would be handled the same way - that easily seen as purely offensive or lacking educational content. But as pointed out below, there is a point where what may be disgusting is subjective, and as long as a consensus of editors see no problem, we can't really censor that. Masem (t) 23:16, 23 October 2023 (UTC)[reply]
If you were looking for "rules" (i.e., not trying to make a common-sense decision), then the archives of Talk:Smallpox probably list every rule that anyone could possibly use to force an image out. (The photo of the last known victim of smallpox apparently horrifies people.) Attempts at systematically addressing such photos pretty much died in 2011 with the m:Image filter referendum/en.
If you were looking for a compromise, then the solution used for the naked pregnant woman at Pregnancy might be a useful model. Move the WP:ASTONISHING image to an obviously relevant section, with a very specific caption, and then put something else at the top. WhatamIdoing (talk) 20:29, 21 October 2023 (UTC)[reply]
Thank you for these historical precedents. In terms of rules, we have the MOS which says "Avoid images that contain .. extraneous elements that might seem offensive .. ", because it's true we need to have some disgusting images, they don't need be extraneously disgusting. Such a position can mitigate the tendency towards maximalist disgusting, done in good faith to get the point across, but arguably unnecessarily disgusting. This image for example lacks the extraneous details such as a toilet bowl, it's not very offensive (the image is copyright, as an example). -- GreenC14:41, 23 October 2023 (UTC)[reply]
Keep in mind also that "disgust" is entirely subjective, and for some people barely exists (doctors could not do what they do if they found diseased tissue "disgusting") while for others it encompasses far more than probably anyone here is willing to define that way (most of my redneck relatives are homophobes, and an image of two men holding hands would "disgust" them). See also a bunch of articles on research showing that a propensity for feeling easily disgusted correlates strongly with conservatism.[3] — SMcCandlish☏¢ 😼 22:50, 23 October 2023 (UTC)[reply]
What is the policy of uploading medical images which are own work? There is an essay WP:MEDPIC but it refers journals and commons. Is uploading of own work medical images allowed to use them in wikipedia medical imaging? In particular, I'm asking about uploading an image of electrocardiogram record to the section Paroxysmal supraventricular tachycardia, similar to [[4]]? --Maxim Masiutin (talk) 14:28, 6 November 2023 (UTC)[reply]
@Maxim Masiutin, I think the editor with the most experience in this area is Doc James. You have to consider the patient privacy rights as well as the copyright issues. My impression is that the US copyright law considers images that are produced by machine in a routine manner (e.g., X-rays of a broken bone) to be ineligible for copyright protection, but that won't be true for every country. WhatamIdoing (talk) 22:25, 6 November 2023 (UTC)[reply]
In the USA technically X-rays and ECGs are not copyrightable. Unless of course you draw an arrow on it and than as creative input was applied it becomes copyrightable. Many medical images however have been copyright anyway and my practice is to respect the copyright which is present.
With respect to consent in a healthcare environment, this is required for images by peoples licensing bodies and I keep a file of signed consents for all images that are identifiable as per instructions from my licensing body. For non identifiable images verbal consent was deemed sufficient.
We here however should not get into managing patient consents directly. It is reasonable to have people mention that consent was obtained. But there are very well funded bodies that manage this in various jurisdictions.
However, my question on Image use policy was more on general terms rather than my particular case. Please consider updating the article on Image use policy with the general information that may help other users like me in the future on how to handle own works with medical images better, as it now only covers medical journals, not own work. Maxim Masiutin (talk) 12:27, 7 November 2023 (UTC)[reply]
The following was boldy added to the guideline a while back, then recently reverted as needing discussion, so let's discuss:
AI-generated images
With new processes like Stable Diffusion that can generate high-quality art from a large database of art, the question of copyright remains open, and at the present time, such images should not be used on Wikipedia unless there is assurance that all images used in the AI system are from free or public domain sources. The U.S. Copyright Office has ruled that AI-generated images cannot be copyrighted,[1] but there remain questions if these are potentially derivative works of the images used in the AI's database.[2][3]
For my part (and I saw that original edit above and just let it sit, and thought about it for a long time), I don't agree with this "copyright paranoia". Knowing what I do about prior legislative and case law (in the US, anyway) involving copyright, trademark, "look and feel", databases and processing thereof, and related concepts, plus how these intersect with the public interest and individual rights, including fair use and freedom of expression. I do not think there is a snowball's chance in hell of AI-generated works (at least of the nature currently being generated by LLMs and similar systems) being treated as works subject to copyright, much less to someone else's copyright for having rough similarities to prior works, visual or textual, unless a specific case of direct plagiarism is undeniable (e.g. when the original author's watermark is still discernably present on a face that an AI pretty much pasted into to something it otherwise generated more randomly). And that kind of glitch probably won't be around for long. Even the sources cited above lean in this skeptical direction, and there are many more. [I am not a lawyer, but I worked with digital media int-prop lawyers for a decade or so on a daily basis, and much of my job was summarizing legal arguments and case results for a lay audience.] Wikipedia is not in a position to try to imagine everything that some day may become impermissible in this or any other jurisdiction and try to avoid doing any of it now just in case. We'd simply close up shop. That which is not forbidden is permitted, and where we have a good reason to use an AI-generated image, e.g. to illustrate articles about AI "art", we must not fear to do so. This is not SelfCensorshipPedia. And there are other uses of AI we already engage in, like for translation, metadata processing, etc. (see summary at WP:AI).That said, we probably do need to address much more burning questions of AI-generated content and its use, arguably misuse, for questionably encyclopedic purposes. This takes (so far) primarily three forms: AI-processed image "enhancement" which is literally guesswork by an AI, and often produces utter crap results, especially when working with human faces. 2) AI-created new works that purport to illustrate something, e.g. what a particular type of nebula might look like, or a dinosaur of a particular type on the hunt, or a supposedly representative example of an artistic style such as Impressionism. Much of this will also apply to other AI-generated content, such as audio ("music" if you want to call something AI-made by that name). 3) AI-generated text that purports to summarize and/or research a topic or aspect of one for us, often populated with "hallucinations" and even outright falsification of sourcing (both making up sources that don't exist and falsely citing ones that do as supporting claims they do not). In my own off-site experients getting ChatGPT to do something as simple as generate a timeline of key points in a country's history, almost every single line item was faulty in one way or another, from misleading oversimplification to outright falsehood.These are not all image-related concerns, of course, and all three of these issues (probably others I have not thought of) need to be hashed out at Wikipedia talk:Artificial intelligence (or even a broader venue like WP:VPPOL), and some base-principles guidance/policy created from it. I think it's not going to be very constructive to try to gin up image-specific new policy here without addressing the issues more broadly at WP:AI or elsewhere first, or very least drafting image-specific stuff with a particular eye to also integrating conforming material at that page in more generalized vein, while also trying to draw in more eyes and minds to the question, including from WT:MOSIMAGES, WP:WPAI, etc. In particular, any drafting of rules that relies on making legal arguments needs to be "advertised" at places like WT:WPLAW, WT:COPYRIGHT, and other on-site venues where actual-lawyer editors are likely concentrated, because this is an area that is going to require expert imput. WP:AI is presently marked an information page, but it seems obvious to me than an actual guideline or policy is going to have to develop, and pretty quickly. — SMcCandlish☏¢ 😼 20:23, 11 November 2023 (UTC)[reply]
We absolutely need to be ahead of the copyright curve, to speak, in dealing with AI-generated images. Right now, it may appear that you should tag these ineligible for copyright due to the US Copyright office's ruling , but if it turns out that they are taken as derivative works (where current lawsuits are sitting), we would have inappropriately labeled them as freely licensed and thus create a problem. thus we need to take the conservative stance that there may be copyright on these. Masem (t) 20:37, 11 November 2023 (UTC)[reply]
Then we'd just un-label them that way and in most cases delete them (if there wasn't some compelling fair-use or other WP:NFC rationale to keep and use one in particular for something). Commons all day every day, and Wikipedia probably at least a few times every day, has to deal with essentially the same theoretical issue but in a very non-theoretical, right-now-for-a-fact way, when people upload images that they claim are their own but which turn out to have been ripped from someone else's website. — SMcCandlish☏¢ 😼 21:09, 11 November 2023 (UTC)[reply]
But, the genie would already be out of the bottle. At the time they were declared derivative works, there could be thousands of them, and they could have already spread to the 4 winds because of WP. Those images, posted with an open license, could continue to spread under that license, causing harm to the copyright holders of the works that the images were derived from. — The Transhumanist22:30, 11 November 2023 (UTC)[reply]
My entire point is that this is already the case with all sorts of images that eventually get deleted. Commons is actually very slow to act on such matters (sometimes it takes months), meanwhile the bad images still continue to appear with open licenses that aren't actually applicable and to get used by off-site people, yet the sky has not fallen down. But there's also a legal faith distinction here; uploading someone else's copyrighted work on purpose to Commons or WP and falsely claiming an open license on it is very different (a bad-faith action) from uploading an AI image to which no derivative-work copyright [presently] applies (something that a court or legislation would have to impose), and to which your own copyright could conceivably apply in some jurisdictions as the prompt-using "creator" of the image, and putting an open license on it (a good-faith action). — SMcCandlish☏¢ 😼 16:59, 12 November 2023 (UTC)[reply]
Yes, people upload (intentionally or not) images with less-strict license requirements, and some of these take years to find. But that's less a problem with copyright law and more just that our processes, being volunteer driven, are slow.
With AI made images, we do not know well enough about how to evaluate their copyright, as whether they are derivative works of the images in the training database is currently in debate in several lawsuits. Should these courts say that there is no derivative work aspects to AI generated images, then as we know that the US Copyright Office has already said images made by automated processes cannot be copyrights, and AI images will be great fair game for free license images. But until we have a strong answer from these court cases, if we assumed they would be free and be proven wrong, we would have definitely encouraged that copyright violations if we don't warn against AI images at this time (outside where the article is about AI generated images like at Stable Diffussion). Masem (t) 21:38, 12 November 2023 (UTC)[reply]
Also as the one that added that section, I remember my reasoning for it was prompted by a discussion elsewhere on WP about AI images and content, and I cannot find where that was, but in general, I have found that the general consensus is that WP should stay far away from AI-based content or at least with heavy user oversight to fact-check and the like. Masem (t) 21:44, 12 November 2023 (UTC)[reply]
The law doesn't work that way. You aren't guilty of something criminal or unlawful if it is not yet criminal/unlawful. There is no "retroactive guilt" principle. Presently, these images are not criminal or unlawful (in any jurisdiction that matters to WP/WMF, anyway). To put this another way, unless WP:OFFICE imposes changes at WP:COPYRIGHT and/or WP:IMAGES policy based on the advice of WMF's own in-house or external counsel, it is not our job to try to play some kind of "Nostradumus, Attorney at Law" game. Our copyright-related policies are based on WMF's legal concerns, not our own WP:OR qausi- or pseudo-legal hypothesizing about potential future what-ifs. As for "WP should stay far away from AI-based content", I strongly agree, as outlined in three points above, but that's not because of imagining hypothetical legal scenarios very few of us are anywhere near competent to be making decisions about, but because the AI output is poor quality, factually wrong, misleading in various ways, or otherwise unencyclopedic. And it's probably not something that's going to get addressed properly at this particular page anyway. — SMcCandlish☏¢ 😼 21:54, 12 November 2023 (UTC)[reply]
@User579987, it sounds like you are concerned about someone mistakenly uploading an image that is non-free, because they saw it in a larger publication (e.g., a book or article) that is free, so they assumed that everything inside the publication was free. Do I understand your concern correctly? WhatamIdoing (talk) 05:41, 17 January 2024 (UTC)[reply]
@WhatamIdoing Exactly; that's what I meant. I believe it's an important exception that should be looked out for. I admit that some of the words I used are a bit complicated and difficult to understand. Perhaps we can rephrase it using simpler language. User579987 (talk) 14:01, 17 January 2024 (UTC)[reply]
I think this is pretty good, and I wonder if we can make it even shorter. Perhaps as short as "Watch out for sources using multiple licenses. Sometimes an article will have a compatible free license for the text but use a non-free license for specific components (e.g., a photo or diagram)." WhatamIdoing (talk) 02:13, 19 January 2024 (UTC)[reply]
@WhatamIdoing That's great as well, but I think clarification is slightly better as the paragraph I sent isn't very long and to avoid templates like Clarify since this article is normally read by beginners and in my opinion needs to be as clear as possible. I also think it's worth clarifying by mentioning examples like "Reprinted with permission from", because some articles clearly mention this exception by saying, "The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material", while others use more vague statements like "Image: Oxford University Press" for example to state that images are copyrighted. Perhaps some clarification won't do much harm. User579987 (talk) 04:28, 19 January 2024 (UTC)[reply]
I would reasonably argue that if work by group A that is published CC, but includes an image or other element that they credit, without mention of copyright, to group B, it is best to assume that group B's work is copyrighted, unless specifically stated by the credit that it is used under some type of free license. So the "Image:Oxford University Press" is where we should assume that it is copyrighted unless proven otherwise, and included in the CC work via fair use. Masem (t) 03:50, 20 January 2024 (UTC)[reply]
I think it'll be better than what we have, so you should do that.
(The problem with credit lines is that "Image: Oxford University Press" is probably copyrighted but "Image: Wikimedia Commons" is not, and the difference may not be immediately obvious.) WhatamIdoing (talk) 22:22, 20 January 2024 (UTC)[reply]
Are we really uploading screenshots from Youtube now? That almost feels like cheating. It was funner when we had to take the pictures ourselves or scour flickr for scraps. That was more like the wiki-way. Anyways, thoughts on this image? ~WikiOriginal-9~ (talk) 14:18, 9 February 2024 (UTC)[reply]
Not really, beyond the restrictions on all images. But see Commons:Commons:AI-generated media for some useful guidance. There are probably some circumstances in which AI-generated images are useful, but I think it would be a bad idea to try to use a photorealistic AI image without making clear its artificial nature, and my experience is that AI image generators can be ok for clip art like images but are bad at technical illustration. —David Eppstein (talk) 01:06, 26 March 2024 (UTC)[reply]
For clarity, I'm curating a film and TV portal and wanted to confirm if Film/TV posters which have been uploaded under fair use can be used alongside the excerpts on their entries. Thank you. Princess of Ara14:44, 28 April 2024 (UTC)[reply]
I have observed a recent spate of editors adding collages to infoboxes for military conflict articles. These might consist of four, six and sometimes more images. In my view, these are too noisy. If they do not significantly increase the footprint of the infobox, they are too small to be viewable. Furthermore, detailed captions add to and bloat the infobox. Infobox size is a particular issue for mobile devices. It is my view that collages as a lead image are generally inappropriate and contrary to P&G on several points.
Per WP:COLLAGE, collages are single images that illustrate multiple closely related concepts, where overlapping or similar careful placement of component images is necessary to illustrate a point in an encyclopedic way [emphasis added]. Such collages do not satisfy image use policy. The rationale for use appears to be largely decorative.
Per MOS:PERTINENCE: Images must be significant and relevant in the topic's context, not primarily decorative. They are often an important illustrative aid to understanding. This is consistent with WP:IMGCONTENT (policy): The purpose of an image is to increase readers' understanding of the article's subject matter, usually by directly depicting people, things, activities, and concepts described in the article. The relevant aspect of the image should be clear and central [emphasis added].
WP:INFOBOXPURPOSE would tell us (in essence) not to try to write the article in the infobox and that, [t]he less information it contains, the more effectively it serves that purpose ... Collages would appear to be inconsistent with this.
Cognative theory and good presentation practice would tell us that too much visual information in one place is counterproductive (ie a sensory overload) and hence, my view that such images are too noisy.
Editors adding such collages would apply a justification of other stuff. However, this is only a reasonable justification if it represents best practice represented by our best quality articles. Very few (if any) of our best quality articles use collages for a lead image.
My reading of P&G and best practice is that the use of collages as lead images/in infoboxes generally and for military conflicts more specifically, should be exceptional.
On one hand, I understand that you can't cover the breadth and scope of many wars with a single image, let alone massive conflicts like World War II. On the other, I agree that generally you see very little in most montages and that many don't feel effectively chosen for their significance versus the "niceness" of the image itself or the decoration they provide (to use the WWII example, the atomic bombing of Japan makes innate sense to me as a major element of the end of the war, or a shot of Stalingrad for its considered role as the "turning point" of the war. Images of tank or aircraft make some sense in terms of the mechanization of war but the choices seem overall random, and the entire collage doesn't do a great job illustrating the global nature of the conflict, its civilian toll, etc. Some of this feels like it wouldn't be as much of an issue if people didn't want an infobox over all else (I've got a collage image as the lead for Art Deco architecture of New York City to demonstrate the different styles the form took in the city across the boroughs it's prominent in, but it's allowed to be more than 30% larger by virtue of not being in an infobox.) Der Wohltemperierte Fuchstalk13:00, 18 March 2024 (UTC)[reply]
In most cases I've seen in these infobox collage discussions where the topic couldn't be covered with a single image, it couldn't really be covered by 3-6 images either. The exact same problem is faced. There are probably some topics out there which are very well summarised with 3-6 images, but perhaps by that very nature they don't lend themselves to extended disputes about image inclusions. CMD (talk) 02:52, 19 April 2024 (UTC)[reply]
I agree. They are awful and should be banned. Otherwise the number of them will inexorably grow, as they are (like over-loaded infoboxes) another thing that editors who can't or won't add text love to do. Johnbod (talk) 13:47, 18 March 2024 (UTC)[reply]
There are good clean examples of info ox collages like in most city articles. But this is generally based on using images that have strong contrast and clear features like city skylines and key buildings or infrastructure, so that as a thumbnail it's still easy to read. Pictures from WWII aren't going to have the same contrast or clarity at small sizes so, a collage doesn't make sense here. Masem (t) 14:08, 18 March 2024 (UTC)[reply]
Collages have long been controversial. There are some editors who wish that our policies and guidelines did more to restrict their use, but to say that the present language of PAG restricts them is a major stretch/overstatement. Please don't conflate an argument with what you'd like them to say with interpretation of what they presently say. Sdkbtalk14:25, 18 March 2024 (UTC)[reply]
There's some policy based arguments to be made against certain collages, but I don't think current policy prohibits them. Certainly poor contrast collages and those that poorly illustrate the subject are bad, and probably a poor choice according to policy. Infoboxes are poor at handling complexity (this isn't just related to images), but that is because they are meant to show simplified information. Dispersing images through out article is a better way to illustrate the subject, but that doesn't mean all collages should go. It certainly seems as if some collages are being added because they are collages, and that doesn't look like careful consideration of the subject or the articles needs. -- LCU ActivelyDisinterested«@» °∆t°19:47, 18 March 2024 (UTC)[reply]
Collages and galleries are an accessibility nightmare on many levels... Be it fragmented images, mini images or scrolling nightmare before reaching pros text (most readers only scroll a few times then go somewhere else)... should really only be used for comparisons in my view. Wikipedia:Manual of Style/Accessibility#ImagesMoxy🍁 02:40, 19 March 2024 (UTC)[reply]
I agree with Cinderella and infobox collages should be strongly discouraged, and certainly not expanded. One good image is better than 6 postage stamps; Actual Studies done by real UI / usability types show that users like Big Images. A collage can be acceptable (if not encouraged) for truly gigantic conflicts where only showing a single image might provoke a nationalist reaction (World War II the canonical example here), which is unfortunate because WW2 is probably the most viewed war article, but it should be the exception, not the rule. That being said, I think that the old hesitance regarding galleries in some policies was misplaced, so I'm fine with the laudable goal of moving more images into articles - just create a gallery section to throw them into, and since the current gallery defaults are still unfortunately postage-stamp size, throw in a custom widths & heights parameters so that they're legibile. Galleries also display nicely on the mobile app, which is how a lot of readers see Wikipedia. On mobile, they can easily be scrolled through, or scrolled past if a reader isn't interested, unlike an infobox collage. SnowFire (talk) 22:53, 21 March 2024 (UTC)[reply]
The question is specifically about "military conflicts", and I think this (i.e., "only showing a single image might provoke a nationalist reaction") is the key difficulty: you can't put up an image from "one side" and claim that inherently biased choice will "illustrate [the] point in an encyclopedic way". Multiple images could, however, be an "important illustrative aid to understanding" – by illustrating that there isn't just one, single important thing to know here. It is perhaps a literal case of the medium is the message: Giving multiple images shows that there are multiple viewpoints.
Strongly concur with Cinderella157, on every point. Infoboxes are for ultra-concisely convenying the key-facts gist, and absoltely do not serve an image-gallery function. WP even discourage (per MOS:GALLERIES) image galleries at all (in the body) unless there are one or more prescribed good reasons for adding one. It's completely inappropriate to make what amounts to one in an infobox, where doing it is also a serious accessiblity problem for anyone without amazing eyesight. — SMcCandlish☏¢ 😼 09:48, 1 May 2024 (UTC)`[reply]
I usually enjoy these collages in articles about cities. With military conflicts, it is different: to give an example, the infoboxes of the Coalition Wars all look essentially the same to me: First, Second, Third, Fourth, Fifth, Sixth, Seventh: some paintings of battles, sometimes involving ships. The twelve infobox images in the Sixth Coalition article might be better placed in the sections discussing these battles. —Kusma (talk) 11:12, 1 May 2024 (UTC)[reply]
You are invited to join the discussion at WP:MCQ § Proactive request for input. -- Marchjuly (talk) 13:31, 23 February 2023 (UTC)
Not done - there is no request. -- Whpq (talk) 18:06, 10 April 2023 (UTC)
.jpeg){kind=link}
.png){kind=link}
![[4]](https://en.wikipedia.org/wiki/Paroxysmal_supraventricular_tachycardia#/media/File:SVT_Lead_II-2.JPG){kind=link}
![[5]](https://commons.wikimedia.org/wiki/File:Supraventricular-extrasystole-leadII.png){kind=link}
.png){kind=link}
