Wikipedia talk:Manual of Style/Images

Legibility of thumbnails at default size

Moved to Wikipedia talk:Manual of Style § Legibility of thumbnails at default size

Poor-quality image

There is a lively discussion at Talk:January 2025 Southern California wildfires#Poor-quality image regarding removal of a poor-quality image (just my opinion though). Your input would be appreciated! --Magnolia677 (talk) 16:22, 13 January 2025 (UTC)

Proposal to change the text of MOS:CALLIGRAPHY

There is a proposal to change the text of MOS:CALLIGRAPHY at:

Wikipedia talk:Manual of Style/Islam-related articles#Proposal: clarify "user-generated" in MOS:CALLIGRAPHY

Editors are kindly invited to participate in the discussion there. ☿ Apaugasma (talk ☉) 20:06, 6 February 2025 (UTC)

Fake rules

I've seen three personal preferences passed off as rules, usually in the context of a lead/infobox image:

• Photos are always best.
• Color images are always best.
• Images from the time period when the person was most notable are always best.

Especially for people who lived in the 19th century, we may have photos only from the end of the person's life. The "pro-photos" crowd claims that a low-resolution black-and-white photo of a frail elderly man is the best way to represent a man once known for his vigorous activity, because it's a photo! But the "pro-color" crowd rejects this, because an oil painting from his younger years is in color, and color is best! And the "pro-timeliness" folks reject both of those, arguing for a sketch published in the newspaper on the day of a key event in his life, because the time period is the most important!

None of these are actual rules, but enthusiastic proponents sometimes stray into presenting them as such. Is it time to explicitly reject all of these, and tell editors that they have to use their best judgment and common sense? WhatamIdoing (talk) 07:44, 8 February 2025 (UTC)

Is it time to explicitly reject all of these, and tell editors that they have to use their best judgment and common sense?: How you you propose editors be told? It's a timeless problem of people passing off their personal preference as an actual rule. —Bagumba (talk) 10:02, 8 February 2025 (UTC)

I suggest adding a paragraph to this guideline. Maybe something like this?

There is no requirement to prefer photographs over paintings, color images over black-and-white images, or images from the peak of the person's notability over recent photos. Editors should make decisions about which image to place in the lead, which to place elsewhere in the article, and which to exclude, based on their best judgment and common sense. WhatamIdoing (talk) 21:52, 8 February 2025 (UTC)

I agree none of these are or should be rules, although I think the last should be the norm, though I've rarely or never seen seen it suggested quite as starkly as in your example. In many cases they conflict - 19th-century people with oils portraits from their prime and wrinkly photos in old age, or 20th-century people with young b/w or old colour photos. My personal bug-bear is bushy Victorian beards - most later 19th-century French artists had these, and are nearly indistinguishable. Johnbod (talk) 17:42, 8 February 2025 (UTC)

RfC on images containing Islamic honorifics or calligraphy

There is now an RfC at Wikipedia talk:Manual of Style/Islam-related articles#RfC on images containing Islamic honorifics or calligraphy, offering various proposals on the text of MOS:CALLIGRAPHY. Editors are kindly invited to comment there.

☿ Apaugasma (talk ☉) 11:20, 16 February 2025 (UTC)

Portrait direction

I've read once that images of people in infoboxes should have the face look at the text (left) but I can't seem to find it anymore. Could anybody help direct me to it? ―Panamitsu (talk) 01:06, 27 February 2025 (UTC)

We generally have this advice MOS:PORTRAIT.... but I do remember we said something of this nature about the infobox.. if I'm not mistaken it was removed or moved because of odd conflicts like at Christopher Columbus. If I remember correctly it was determined that the most prominent/recognizable image should be used when possible regardless of which way it's facing. Moxy🍁 01:40, 27 February 2025 (UTC)

Dispute over images in some name articles

Can some people please give their input at User talk:Bookworm857158367#Edit warring 2 about whether the second image here and the image here meet the requirements of this MOS and thus are better included or removed from those pages? Thank you! Fram (talk) 17:30, 28 February 2025 (UTC)

The general guideline I have followed regarding inclusion of images in name articles is that they have something to do with either the meaning of the name or are mentioned in the article itself. If a source has mentioned that a particular person or character influenced usage of the name, it seems appropriate to include a picture or image of the individual. Otherwise, an image that relates to the meaning of the name seems appropriate, with a caption explaining how the image relates to the name. The particular disagreement in this instance is over the inclusion of a painting of a girl holding a kitten in the article Minou (given name). One of the meanings of the name is "cat." In the article Khaleesi (given name), Fram has disputed including a picture of a cat in Wikimedia Commons that was labeled "A cat named Khaleesi." The article also mentions that the name has been a popular one for pets, which the image illustrates. There are other articles that also include similar images -- Hazel (given name) has a painting of two girls picking hazel nuts, for instance. None of these images are purely decorative; all relate to the meaning or usage of the name. It should also be pointed out that people are more likely to read articles that have some sort of illustration. That's Communications 101 and it also applies to an online dictionary. Bookworm857158367 (talk) 17:38, 28 February 2025 (UTC)

Images in a "notable people" section

Should a photo in a "notable people" section be a casual photo, or a photo of the person "in costume"? Your input at Talk:List of Virginia Commonwealth University alumni#Inclusion of poor-quality image would be appreciated. Magnolia677 (talk) 17:33, 10 March 2025 (UTC)

Images of gods

I have an unusual dispute. It is not the first time someone adds images of gods or demons or spirits by a modern nonnotable artist to wikipedia articles. Recent example is Slavic gods, such as Dazhbog which has an image "Dazhbog by Andrey Shishkin". Of course, if an article has an image of an ancient idol, then it is an illustration how our ancestors imagined this god. If it is a painting by a famous painter, it is also a valid illustration of a notabkle work of art. But I do not think that an unknown artist has a right to tell Wikipedia users how this god is supposed to look like. Especially in large quantities, which smells like an undue promotion a nonnotable guy via Wikipedia.

Here is my previous discussion with Wojsław Brożyna (talk · contribs) moved from user page:

YOu wrote: " Even if the painter is not recognisable enough to have his own biography on Wikipedia, his painting is still valuable" - In English Wikpedia we do not add information because it is "valuable": we add only information which is verifiable, i.e., coming from reliable sources. There is no evidence that the painter of unknown ethhographic/hiscotrical expertise draw the image of a pagan god the way as our ancestors imagined it. --Altenmann >talk 15:53, 6 March 2025 (UTC)

Why did you assume that it even pretended to represent "the way as our ancestors imagined it"? Do you have confirmation of that for the other, recognisable artist? And how such an argument can be valid in case of fakelore, which wasn't imagined by our ancestors at all? This is not a proper argument against those works. --Wojsław Brożyna (talk) 18:41, 10 March 2025 (UTC)

I do not have to have any confirmation because I do not add information. And in the case of fakelore it makes no sene to add any images a all, because how you draw image of something that did not really exist, nor imagined by our ancestors. Once again: per our wikipedia rules, all information must come from reliable sources. An unknown painter is not a reliable source. --Altenmann >talk 21:23, 10 March 2025 (UTC)

Please voice your opinion. --Altenmann >talk 21:37, 10 March 2025 (UTC)

At first, you have misunderstood en.wiki encyclopedity with recognisability. This artist haven't meet the requirements to have his own biography on the en.wiki, but it doesn't mean that he is unknown. Second question is that no artist has a right to tell Wikipedia users how this god is supposed to look like - it is every time only an artist imagination. There were famous painters, whose works are used on Wikipedia articles, who have literally zero historical/mythological knowledge - a countless examples of great minds from the era of romanticism, who just unleashed their creativity and/or based on early works on ancient religions, that aren't treated as reliable anymore. Any artist expression of mythological motifs which is technically good (and works of Andrey Shishkin are like that) can serve as an illustration in encyclopedia, because it is supposed to tell Wikipedia users how those motifs are imagined in art - that's all. And especially in case where we haven't any other paints of a deity, erasing images of a painter just because he doesn't have a bio on en.wiki is not constructive. Wojsław Brożyna (talk) 18:55, 11 March 2025 (UTC)

Sorry, disagreed. In your words, I can draw some random picture and upload it as an image of an "imagined god" - which would be plain ridiculous. it doesn't mean that he is unknown" I did write that artist is unknown, but later I clarified: of unknown ethhographic/historical expertise. Therefore we cannot verify that this artist's drawing faithfully represent the description of the god in question. --Altenmann >talk 20:38, 11 March 2025 (UTC)

So how did you know the ethnographic/historical expertise of authors of images included in, for example, Radegast (god) article? Or in Leshy? Those are only examples related to the Slavic mythology; there are plenty of examples of images, that are not accurate from the current state of knowledge. So why we still keep them in the articles? Wojsław Brożyna (talk) 13:31, 14 March 2025 (UTC)

These are pictures made by famous people and have an independent historical value. --Altenmann >talk 16:24, 14 March 2025 (UTC)

So let's be consequent. Either the fame of person matters or their knowledge? Andrey Shishkin is known person. Wojsław Brożyna (talk) 18:11, 14 March 2025 (UTC)

Agreed 100% with Altenmann. For similar discussions with much more extreme examples see Talk:Oshun#Lead image and Special:PermanentLink/1160854150#Your "fashion editorial" and DeviantArt-style AI-generated images of African mythological beings (involves also the topic of AI-generated images too; see c:Commons:Deletion requests/Files uploaded by MiddleOfAfrica) —Alalch E. 17:12, 14 March 2025 (UTC)

(The second discussion includes the following notable quotes: what about these fashion editorial depictions of African deities makes you feel as though they are not good illustrations? and What you did and are doing will be judged in the spirit world)—Alalch E. 17:18, 14 March 2025 (UTC)

• Agreed, and have removed of these specific ones (along with some often worse examples on the same articles). Have done the same with self-made art and AI art on articles about creatures of myth or cryptoids as well in the past. We wouldn't include one man's description of any of these without good sources or without any indication that their view was in some way influential, so why would we include someone's depiction, no matter how well executed? Fram (talk) 17:55, 14 March 2025 (UTC)

  The uploader of these images is clearly not interested in getting consensus and prefers to edit war to include the images... [1], they are busy reinserting all her uploads Fram (talk) 18:10, 14 March 2025 (UTC)

  Are you serious? You have started edit war before we got any consensus here. Actually now I came here to ping you, suprisingly discovering that you have already been there, but you started do erasing those pictures before we agreed to any conclusion. Wojsław Brożyna (talk) 18:12, 14 March 2025 (UTC)

  You have had warnings about this practice since 2018, you had here two other people already disagreeing with you, and at the link I gave, you had already reverted the removal of your uploads on 6 March and on 10 March, so it's hardly correct to claim that I started the edit war. We don't need your agreement. Fram (talk) 18:19, 14 March 2025 (UTC)

  About which practice? The problem of those images emerged in this year. Sorry but you are breaking the rules of discussion here. Wojsław Brożyna (talk) 11:18, 15 March 2025 (UTC)

• Why should we indulge one person's fantasy? What encyclopedic value is conveyed by how that one person imagines this? At least with older, published images, we reproduce images which have already been presented to a wide audience through some publication and indicate how publications imagined the entity. But we reproduce such published things, we aren't the place to start them. Fram (talk) 18:21, 14 March 2025 (UTC)

  Most of those illustrations, if not all, were published before. Images of Marek Hapoń, whose work you also deleted from some articles, are even used on Polish governmental education platform ([2]) or at the cover of book written by some scholar ([3]). Paintins by Andrey Shishkin were used too at book's covers ([4]). Works of Dusan Bozic were printed in branch journal about Slavic culture and mythology ([5]). Those are not anonymous artists, but well known creators in their niche. Their works are valuable. Wojsław Brożyna (talk) 18:46, 14 March 2025 (UTC)

  As I wrote, Wikipedia policies do not operate with the concept "valuable", because it is subjective. In the past we already had discussions on a similar subject: "important" vs. "notable". --Altenmann >talk 18:52, 14 March 2025 (UTC)

  So aren't they notable if they are used in such cases? Wojsław Brożyna (talk) 11:19, 15 March 2025 (UTC)

  P.S. As for Andrey Shishkin, you should have known better: you did participate in Wikipedia:Articles for deletion/Andrey Shishkin, do you? --Altenmann >talk 19:31, 14 March 2025 (UTC)

  Known better what? Wojsław Brożyna (talk) 11:20, 15 March 2025 (UTC)

  that wikipedians do not recognize this person as notable. --Altenmann >talk 15:40, 15 March 2025 (UTC)

  But this is not about wikipedians. There are criteria about notability (WP:N). I'm sure that most of wikipedians don't know most of the persons having biography on Wikipedia and it doesn't matter if they meet those criteria. Shishkin doesn't meet on English Wikipedia (but he still have bio's in another language's Wikis), but it doesn't mean that he is unknown in general. It's just niche. Wojsław Brożyna (talk) 16:46, 17 March 2025 (UTC)

• I want to pay special attention to the fakelore gods. If some mythological entity is not even atested, but it's probably or even firmly invented by some scholars, artists or so, then what is a point of erasing images "of unknown historical accuracy"? They are depicting issue that is openly not histically accurate at all. I mean eg. Chislobog and List of Slavic pseudo-deities. --Wojsław Brożyna (talk) 18:19, 14 March 2025 (UTC)

  The point is that all information in wikipedia must be verifiable and relevant. A figurne of Chislobog by a forger is relevant, because it directly illusrtated article text. A painting by Rembrandt is relelevant because it shows how a notable person imagined something and we have a reason to believe that Rembrandt's vision influenced the vision of many people. A painting of a god by an anthropologist is relevant if it was pblished as part of trheir scientific work to illustrate how in their opinion the god was described. A painting of a god by some Random Shiskind is WP:UNDUE: we either do not trust his/her opinion or dont care about it, and do not want Wikipedia to promote Shiskind's vision all over the world by the authority of Wikipedia. --Altenmann >talk 15:40, 15 March 2025 (UTC)

  

  Paintings of "Random Shishkin" are well known and recognisable. This is why I even asked him to let his illustration be included on Wikimedia: because I meet his works many times earlier. I had even took a photo of someone's home altar with Shishkin's painting. Wojsław Brożyna (talk) 16:52, 17 March 2025 (UTC)

  Shishkin, born in 1960 in Moscow, is a Russian self-taught painter of quasi-academic realist kitsch. His works have not been a subject of serious art criticism. He creates colorful, romanticized, eye-candified, depictions of historic subjects, Slavic gods and mythological figures, that appeal to popular aesthetic taste, relying on online sales through a-shishkin.ru and on commissions for calendars, postcards, and Tarot cards, in addition to cover art and decorative illustrations for books and journals (including a Polish edition of National Geographic).

Looks decorative to me... And thought--provoking :-) (You know which kind of thoughts)

His output is decorative by definition. Therefore, use of his paintings as illustration of Wikipedia articles is adding purely decorative images. While it's impossible to formulate a general distinction between decorative and educationally illustrative, images should not be purely decorative, and sometimes it is possible to make this determination when an image is inherently purely decorative. —Alalch E. 17:36, 20 March 2025 (UTC)

And here we are entering the gray area of the usability of decorative images. And here comes in my another criterion: the meld of WP:UNDUE/WP:TRIVIA criteria: is the painter (or a painting in question) of reasonable fame, so that his opinion about the appearance of a mythological being merits inclusion into wikipedia? My daughter can easily draw a Jesus Christ and I can upload it to Commons, no? --Altenmann >talk 19:44, 20 March 2025 (UTC)

zigzag image placements

I believe the zigzag (left and right) placement of images should be discouraged. This practice disrupts the layout and can negatively impact the reading experience. Astropulse (talk) 15:40, 22 March 2025 (UTC)

It used to be recommended (many years ago), but is now less common. I don't agree that it "disrupts the layout and can negatively impact the reading experience" - it is a typical layout used by professionals in contexts such as large-format magazines. Some images need to be placed on the left anyway. Johnbod (talk) 18:23, 22 March 2025 (UTC)

Academic recommendations for accessibility and ease of reading.... That seen text sandwiching is much more of a concern for accessibility here. We'll never get all images on the right.... as many people will always put images of people looking towards text.... as we recommend.... and simply looks better. Moxy🍁 19:25, 22 March 2025 (UTC)

https://accessibility.huit.harvard.edu/images-and-media they are not exploring image placements. there is much more to that just accessibility. Astropulse (talk) 01:13, 23 March 2025 (UTC)

Magazines use pictures to capture users’ attentions. A zigzag layout is good for that. Pictures are a big part of magazines; I personally just look at pictures and read a lot less in magazines.

Wikipedia is a reading-focused site. Users often scan quickly for relevant info. A consistent image alignment (e.g., all images on the right or left) supports a linear reading flow, making it easier for users to absorb content. Astropulse (talk) 01:04, 23 March 2025 (UTC)

@Astropulse: You should close the thread, Talk:India#Current_image_placements, you have recently opened where you have observed, "nothing wrong with placing all images to the right. current zigzag placement kind (sic) sucks. i prefer all images to right - except select ones."(link to the edit) Forumshopping is not a good idea. As for the reading experience, is this something rooted in the psychophysics of the human visual system, a widely-agreed upon aesthetic, or your personal preference? Fowler&fowler«Talk» 19:02, 22 March 2025 (UTC)

Sorry, disagreed about closing/shopping. This is not forum shopping: this is placing discussion into the proper, because the raised issue is not for a single page India. --Altenmann >talk 19:50, 22 March 2025 (UTC)

Your defense of Astropulse here belongs only to the letter of the law, but not its spirit. They should have either closed that discussion and/or told other participants they are simultaneously pursuing the topic elsewhere in greater generality. In my book what Astropulse did was less than ideally transparent. I’ve seen editors do this with sources. When their discussion about the reliability of a source doesn’t proceed favorably on a page, they open a slightly more general discussion at RS/N. Fowler&fowler«Talk» 22:34, 22 March 2025 (UTC)

there is no bad faith here. that discussion is separate based on existing policies and guidelines. one editor has said i may have an argument and no one has seriously disagreed yet.

I raised this at a broader venue because I felt the issue extended beyond a single article (like India) and related more generally to layout practices across Wikipedia. I wasn’t trying to forum shop — I genuinely want to hear broader input and understand community consensus.

The suggestion that I opened this discussion here simply because I didn’t get my way in another thread is incorrect. Astropulse (talk) 00:05, 23 March 2025 (UTC)

Opening another thread before you have come to a resolution in the current one, may or may not be bad faith, but not informing the other participants that you are, very much is. It is an egregious example of the genre. Fowler&fowler«Talk» 01:11, 23 March 2025 (UTC)

No see this https://www.nngroup.com/articles/zigzag-page-layout/

1) Decorative images used in an alternating list (zigzag) layout caused users to stumble when scanning the page in an eyetracking study. Users scanned efficiently on pages where text and imagery were vertically aligned.

2) aligning similar items (like in the aligned layout) makes them easier to scan.

3) when images were more than just decoration, both layouts worked equally well

4) The zigzag layout is less predictable, making it difficult to scan around image obstacles.

5) “Image placement causes residual fixations” means that even when an image isn’t important or helpful (like a decorative one), users’ eyes still tend to pause or briefly fixate on it — just because of where it’s placed.

6) Zigzag layouts made it more difficult to ignore decorative imagery and caused users to stumble over these unhelpful images and immediately redirect their fixations. ( this may be true for less helpful images as well )

there probably other research about this as well. Astropulse (talk) 00:13, 23 March 2025 (UTC)

@Johnbod what do you think about this. its by Nielsen Norman Group Astropulse (talk) 00:43, 23 March 2025 (UTC)

I like the word "discouraged", which is not the same as "forbidden". There are real problems with "sinistral" images: they screw up section headers and especially lists. Therefore at least an advise must be given to check the appearance of the layout and possibly recommend alternatives:

• Galleries (especially for uniform things, such as paintings of Venus by different artists)
• Smaller images
• Longer texts :-)
• Separators

Other ideas? --Altenmann >talk 20:04, 22 March 2025 (UTC)

As Vector2022 imposes a standard width on pages, there isn't a need to be concerned about potential problems with sinistral images. If they do cause problems, obviously they should be removed, but this will be obvious and doesn't require a separate check. (Under the previous Vector2010 you are more likely to get problems, so even editors using that won't miss it.) CMD (talk) 02:08, 23 March 2025 (UTC)

Does MOS:NOETHNICGALLERIES also exclude a single image used to represent broad groups of people?

Hello, I looked at the MOS:NOETHNICGALLERIES redirect and the past RfCs/discussion and would like some clarification? While I understand that photomontages/galleries representing broad groupings of people have had an RfC/discussion against it, thus infoboxes for pages of ethnic groups do not include them, is this guideline against having a single image in an infobox to represent broad groupings of people?

Although I know that there are technically no set-in-stone rules in Wikipedia and users on talk page discussions may reach a consensus that makes an exception to these various guidelines, I see quite a few pages for ethnic groups that include a single image in the infobox without any talk page discussion, and such image maybe contentious (i.e: why choose a single image in the infobox representing one child of the ethnic group with no men, women, elderly people, etc in the picture).Clear Looking Glass (talk) 09:24, 25 March 2025 (UTC)

NOETHNICGALLERIES was a guideline to address the following situation. Obviously, a single image cannot represent an ethnic group in its diversity. Therefore people started creating collages from notable people of a particular ethnicity, which is clearly original research (why cherry-pick some?). Other than that, if there is no guideline about infobox for ethniocities (Template:Infobox ethnic group), ethnic images must be decided in a case-by case basis. For example, the top image for Kayan people (Myanmar) shows an extremely distinct feature, so it makes sense. On the other hand, the Khoisan image makes some sense for a European person, to show at least some ethnic distinctness, but some may object a cherry-pickd image: for a European all khoisan may look the same, but I am sure there is a wide diversity among them recognized by the natives, and the infobox with a map such as in Bantu peoples would make more sense. --Altenmann >talk 18:57, 25 March 2025 (UTC)

Dispute over population data image

There is a content dispute regarding the relevance of an image at Talk:Durant, Iowa#Population plot. Your input is welcome. --Magnolia677 (talk) 18:18, 1 April 2025 (UTC)

Lead image size

Wikipedia:Manual of Style/Images#Size says:

"As a general rule, images should not be set to a larger fixed width than 250px (the initial base width), and if an exception to this general rule is warranted, the resulting image should usually be no more than 400px wide (300px for lead images) and 500px tall, for comfortable display on the smallest devices "in common use" (though this may still cause viewing difficulties on some unusual displays)."

Does anyone know how old this advice is? WhatamIdoing (talk) 19:24, 28 April 2025 (UTC)

I feel it's potentially sound, right? Not sure we want to give users a much wider thumb gamut to fuss with, but maybe? I do think it should be "normalized" or expressly allowed that lead images can be frameless or wide too, like I fashioned a bit ago at Köppen climate classification. Remsense ‥ 论 19:39, 28 April 2025 (UTC)

The default image width is supposed to be changed to 250px soon, so that probably shouldn't be the largest ordinary option.

The lead image is usually supposed to be larger than the default, as well as the largest image in the article (except for panoramas or similar special circumstances). This usually means 300px or 400px for a landscape image and default width for a portrait (because we'd normally scale a portrait image down).

The image at Landscape painting is (on my screen, with 100%/default zoom, etc.) about 400px wide, and that seems reasonable to me. Landscape photography, on the other hand, uses default image sizes, and it feels much too small, especially for the lead image. WhatamIdoing (talk) 19:54, 28 April 2025 (UTC)

Köppen climate classification looks quite broken for me. With the image appeating under the TOC, which is itself squashed to oneside. The advice at MOS:SECTIONLOC is not to put images at the end of sections like that. That's a separate issue to image sizes though.
I don't see a reason why lead images should be smaller than other images, unless they are in an infobox (as that will add its own width to the width of the image). But I don't think images should be larger than 400px, as that tends to end up with squashed text. -- LCU ActivelyDisinterested «@» °∆t° 21:15, 28 April 2025 (UTC)

Thanks very much, I use Vector 2022 so I had no idea. Do you have any idea of how such images could be well-presented while not limited to a thumbnail? Remsense ‥ 论 21:23, 28 April 2025 (UTC)

Why did you (or someone else?) choose {{TOC right}} originally? WhatamIdoing (talk) 21:46, 28 April 2025 (UTC)

I don't think it was me, but if it was I couldn't tell you. Remsense ‥ 论 21:50, 28 April 2025 (UTC)

So maybe start by removing that. And then think about what having a {{clear}} template above the TOC might do. WhatamIdoing (talk) 21:59, 28 April 2025 (UTC)

Sandwiching and Vector 2022

Given that sand­wich­ing is much less common in the now default WP:Vector 2022 (which the vast majority of readers will use) than in previous versions due to its different layout, does this mean that we only need to prevent sand­wich­ing in the standard settings of Vector 2022, rather than trying to prevent sandwiching in every previously standard Wikipedia skin/layout? Hemiauchenia (talk) 20:34, 28 April 2025 (UTC)

Do you have any examples of articles were sandwiching appears different between skins? -- LCU ActivelyDisinterested «@» °∆t° 21:19, 28 April 2025 (UTC)

For example Theropoda#Biology. In legacy Vector 2010 there's obvious sandwiching, but in Vector 2022 standard settings there is no sand­wich­ing (though switching to small text, or more noticeably to wide margins will cause sand­wich­ing). Hemiauchenia (talk) 21:25, 28 April 2025 (UTC)

That's odd, I use 2010 and see no sandwiching on that article. -- LCU ActivelyDisinterested «@» °∆t° 23:39, 28 April 2025 (UTC)

Whether or not there is sandwiching is heavily dependent on external factors like monitor aspect ratio. This is what makes "sandwhiching" such a difficult issue, as it largely on the users end. Hemiauchenia (talk) 23:42, 28 April 2025 (UTC)

I think you can only do so much. The most important part is not having images directly opposite one another, after that try to minimise image overlap but know it won't be possible in ever case. -- LCU ActivelyDisinterested «@» °∆t° 23:46, 28 April 2025 (UTC)

I think that we should only care about the default view for desktop readers: 100% zoom, default skin, not using Vector 2022's wide width mode. And I think we should only care about it a little bit, because every screen/font/zoom/etc. combination is going to produce a different result.

Because of your edit to Theropoda, I see no sandwiching in that section in Vector 2022. Before your edit, I saw sandwiching in Vector 2022 (affecting about 13 lines of text) and in Vector 2010 (affecting about 2.5 lines of text). WhatamIdoing (talk) 21:57, 28 April 2025 (UTC)

I mean, Tyrannosaurus is another example where there is very litte sandwiching on Vector 2022 and a lot on legacy Vector 2011, and I haven't edited that at all recently. Hemiauchenia (talk) 23:21, 28 April 2025 (UTC)

Simplest way to avoid sandwiching is to follow the first sentence at MOS:IMAGELOC. The vast majority of our readers (64.9% in January 2025) use the mobile skin..... thus this should be the main layout concern for accessibility for our readers. Moxy🍁 23:58, 28 April 2025 (UTC)

Have you ever read Wikipedia on a phone, which is what the vast majority of Minerva Neue readers are using? If you did, you would know that sandwiching basically doesn't exist, because the images are always displayed centrally between paragraphs, which works quite well. Sandwiching is primary an issue on desktop and perhaps tablets, but as far as I can tell there is little modern data (post 2011, anyway) of how many people regularly read Wikipedia on tablets, but I'd have to venture it's quite a bit less than mobile. Hemiauchenia (talk) 00:23, 29 April 2025 (UTC)

Fallow MOS:IMAGELOC and secondly apart from the first recommendation, be more concerned about mobile view layout. Moxy🍁 00:31, 29 April 2025 (UTC)

I think reducing it to desktop readers using 2022 is a bit reductive. It doesn't cover the majority of readers and only opens the quest of what window size on desktop? If the answer to that is full screen the percentage of readers using that is going to be quite small. -- LCU ActivelyDisinterested «@» °∆t° 23:57, 28 April 2025 (UTC)

As I have said above, sandwiching doesn't effect mobile phone readers, though it probably does affect tablets, which I would guess are a smaller fraction than both desktop and mobile, though I have no hard data for that hunch, other than a very outdated survey from 2011 [6] Hemiauchenia (talk) 00:27, 29 April 2025 (UTC)

our latest stats Moxy🍁 00:30, 29 April 2025 (UTC)

I didn't comment on mobile phone usage, was this reply meant to be to me? -- LCU ActivelyDisinterested «@» °∆t° 14:59, 29 April 2025 (UTC)

Thinking of CMD's points below maybe the section should make clear that editors should be targeting how the page is displayed in VECTOR2022 but without the other overly specific details. The main issue that SANDWICH should try to counter is when images are directly opposite each other (as in the example). Beyond that the latest skin is the most important, but the further you specify the less important it is.
Hopefully the WMF will at some point design a mobile view that doesn't look twenty years out of date, but I think it's a smaller concern at the moment as the mobile view displays images differently. But still I would hesitate to put into MOS that mobile view issues, if they do happen, can be ignored. -- LCU ActivelyDisinterested «@» °∆t° 16:34, 29 April 2025 (UTC)

Since en.wiki was moved to VECTOR2022 (and even Commons is now on Vector2022, so it has become a consistent standard) I have considered sandwiching considerations to apply only to VECTOR2022 (even though I mostly still use VECTOR2010). Further, as Moxy notes above, our recommendation is no longer to alternate images, but to place most on the right, meaning sandwiching occurs even less than just with the 2022 switch. The more consistent width also affects other MOS items, such as considerations of paragraph length, how much can fit on a screen, and similar, all of which should be mostly concerned with 2022 than any other skins. (And with mobile, although in many cases that's a different discussion.) CMD (talk) 15:38, 29 April 2025 (UTC)

• I confess to puzzlement at the decades-long hysteria about "sandwiching", like it's a fatal disease or something. If an article is light on images, and if for some reason it's desirable to have some of them on the left, then it's usually not hard to keep those on the left separated from those on the right. But if an article is heavy on images, and galleries aren't appropriate, then I've always reasoned as follows. The guideline provides that upright=1.6 should usually be the largest value for images floated beside text, so I figure if two images oppose each other (fully or partially), then as long as the sum of their "uprights" is <= 1.6, then the width available to text is the same as what the guideline contemplates. EEng

what does the policy have to say on video screenshots?

I feel that even when a screen dump of a paused video isn't so blurry it falls foul of MOS:IMAGEQUALITY, it still feels like a slippery slope.

Should we include screenshots in what we consider "pictures"?

Please note I am not talking about, say, a logo, or television title. That's effectively a still picture repeated in many frames.

I'm talking moving images, like illustrating an article about an actor with a video screen grab from, say, Youtube. We can clearly see the actor's likeness, but the picture is still relatively shitty. (Again, please assume the still isn't pure garbage, because then we could simply apply IMAGEQUALITY. And for the sake of argument, there are no issues with fair use, licensing or other things)

I feel it would be best to simply consider moving video as a substandard source of what we consider "images". Video is one thing, images are another thing, and video isn't just a collection of images.

I'm not asking for an overall ban. I'm thinking editors should be expected to find images elsewhere, and only as a last resort, and with local consensus, screenshotting video to obtain images. And if you can provide a regular image, you shouldn't need to justify switching out a screen grab. The mere fact you are improving Wikipedia by making it less reliant on screenshots is enough.

In other words, screenshots might be tolerable, but never desirable. And if an article is illustrated by a screen grab, we should never consider that article properly illustrated. The hunt for a proper image should go on. Screenshots are "temporary images": they're better than (can be better than) nothing at all, but should be replaced wherever possible.

Currently, our policy doesn't even mention screenshots here. I think it should. CapnZapp (talk) 18:55, 8 June 2025 (UTC)

I think you could apply IMAGEQUALITY to already accomplish what you're proposing: if a better image is available, that other image should be preferred. It doesn't matter why it's better, technically. And similarly, if the only image available is of very poor quality for whatever reason, finding a better one is to be encouraged. Nikkimaria (talk) 20:41, 8 June 2025 (UTC)

What I'm aiming at is to buck the thrust where good-faith editors see articles with no pictures, find a video, take a screenshot and then go "nice, now the article is illustrated, job complete".

I would like us to consider screenshots a stopgap measure at best. I'm not saying we should prohibit them (that is screenshots posing as actual pictures), I'm saying we should be clear they're a second tier solution that should never be considered a permanent fix.

At the very least, shouldn't our policy discuss or at least mention the category of images that are harvested from video, i.e. screengrabs? Currently there is absolutely nothing that makes a good-faith editor hesitate to populate BIO articles with still taken from video. I think there should be. Something feels cheap and off when we present frames from moving pictures as still pictures. CapnZapp (talk) 09:58, 9 June 2025 (UTC)

Keeping in mind NFC and living person's, there may be cases that using a screenshot from a video where the person was most famous for may make sense, and there are otherwise no free ages. We have a free image of Ed Sullivan, for example, but if we lacked any such images and there were no high quality non free images, the use of a shot of him hosting his show would absolutely be reasonable. Masem (t) 22:31, 12 June 2025 (UTC)

As I said, I am not advocating for a ban on video still images. I can definitely see a corner case where a screengrab is the only solution, and that the need to have any picture at all trumps our desire to not flood Wikipedia with lousy video grabs. So in that sense, I think your argument is reasonable. But I want to discuss the possibility of you still being vague enough for someone to interpret it as you discussing the general case, not just exceptional cases, Masem:

Therefore: let us have our policy be clear; videos should not be a standard go-to source for pictures of people. Using them should, in my opinion, always require active consensus. Compare to how PRODs are added. If anyone removes them for any reason, there needs to be a discussion to develop consensus. The onus is on the person wanting the inclusion. In that case, whether to delete or keep an article (a deletion discussion), in this case whether to add or not add a screengrab as a picture. If an editor years later wants to find out how consensus for adding a shitty screen grab developed, there is something in Talk archives. There's not just the edit that added it followed by... no reaction.

And maybe I don't need to say this but just to be clear, I'm not gunning for any higher standard of active consensus than usual: if I start a talk discussion arguing for the inclusion of a screengrab and I get no replies, then that counts as consensus just fine. At least I have clearly flagged a screen grab was added. The distinction is instead that just passive consensus is not enough: just adding a screen grab as if it was any old picture, and expecting dissenting editors to make the effort to start the discussion. Let's not make objecting to screen grabs the unreasonable position. Let's make adding screen grabs the exception.

As I see it, this cannot be read from the policy as it stands today. That is, I'm arguing our policy needs an edit. CapnZapp (talk) 11:13, 14 June 2025 (UTC)

I'll give it a few days and if there's still no objections I'll come up with a tweak to the policy so video stills are at least mentioned. CapnZapp (talk) 22:02, 12 June 2025 (UTC)

There's a humorous overview on WP's challenges for quality basketball-related images by The Ringer.—Bagumba (talk) 07:21, 22 June 2025 (UTC)

How inform reader of relevant nearby image?

I'm working on an article which has a relevant image adjacent to some text. It would be a shame if readers were not directed to the image, since it really adds insight to the text. The image already has a good caption. What sort of wording is recommended?

The MOS says

• "image placement varies with platform and screen size, especially mobile platforms, and is meaningless to screen readers. As such, article text should not refer to image positions, especially with terms such as left, right, above, or below. Instead, use captions to identify images."

Which makes perfect sense.

Is it common to use words like adjacent image or nearby image? Perhaps a parenthetical comment that says "... blah, blah (see adjacent image) ..."?

I understand that the text must stand on its own, and cannot rely on the presence of a nearby image; but does the MOS permit the text to inform the reader of a nearby image, if the image would be especially enlightening?

Noleander (talk) 13:01, 4 August 2025 (UTC)

Could you use an internal page anchor on the same line as the image in wiki code, and then internal link to point readers to image without specifying the actual direction from prose? Masem (t) 13:24, 4 August 2025 (UTC)

Yeah, that could work. What sort of wording would you suggest in that situation (for the link to the image anchor)? Noleander (talk) 13:50, 4 August 2025 (UTC)

"See image", image here being the internal link) unless you can be very brief in what the image describes. Any other term like map, diagram, graph, etc. could work too. Masem (t) 14:00, 4 August 2025 (UTC)

Alright thanks, I'll do that. Just to confirm: WP does not have a pattern, convention, or guideline that has been established for this situation, correct? Noleander (talk) 14:17, 4 August 2025 (UTC)

I found the instructions at MOS:SEEIMAGE confusing. I was expecting instructions on how to refer to an image from the text. It currently states:

References from article text: Image placement varies with platform and screen size, especially mobile platforms, and is meaningless to screen readers. As such, article text should not refer to image positions, especially with terms such as left, right, above, or below. Instead, use captions to identify images.

This guidance says what not to do, but does not explain how an editor can insert a "see image" reference. Would the following be more helpful to editors?

References from article text: To refer to an image from article text, a link to an anchor point near the image may be used. The article text should not refer to an image with terms such as left, right, above, or below (because image placement varies with platform and screen size, especially mobile platforms, and is meaningless to screen readers). The description of an image should be in its caption, not in article text.

This would explain to editors one way to implement a "see image" reference to images, and also would explain what not to do. Thoughts? Noleander (talk) 17:12, 4 August 2025 (UTC)

It would be nice if we could add anchors to images directly, within the WP:EIS syntax. Something like this:

[[File:Bananavarieties.jpg|thumb|Four varieties of banana|alt=Some bananas of various sizes and colours. Two are large and green; three are small and red; three are small and yellow with dark spots; three are large and yellow.|id=Bananas-image]]
Most fruits are found in several varieties, or ''cultivars''. Four of the many varieties of banana are shown in [[#Bananas-image|the illustration]].

but unfortunately, EIS doesn't recognise |id= and treats it as a caption; see sandbox demo 1.

If you place the anchor inside the syntax, thus:

[[File:Bananavarieties.jpg|thumb|{{anchor|Bananas-image}}Four varieties of banana|alt=Some bananas of various sizes and colours. Two are large and green; three are small and red; three are small and yellow with dark spots; three are large and yellow.]]
Most fruits are found in several varieties, or ''cultivars''. Four of the many varieties of banana are shown in [[#Bananas-image|the illustration]].

this links to the image caption, but the image itself will probably be off the top of the screen; see sandbox demo 2. It's also bad for accessibility, since the caption will be read out but the |alt= text won't.

Placing the anchor before the image will work:

{{anchor|Bananas-image}}
[[File:Bananavarieties.jpg|thumb|Four varieties of banana|alt=Some bananas of various sizes and colours. Two are large and green; three are small and red; three are small and yellow with dark spots; three are large and yellow.]]
Most fruits are found in several varieties, or ''cultivars''. Four of the many varieties of banana are shown in [[#Bananas-image|the illustration]].

see sandbox demo 3; but there is a chance that a subsequent editor will not realise that they go as a pair, and insert something else between them.

We can use a <div>...</div> element to associate the anchor with the entire image/caption object:

<div class=floatright id=Bananas-image>[[File:Bananavarieties.jpg|thumb|Four varieties of banana|alt=Some bananas of various sizes and colours. Two are large and green; three are small and red; three are small and yellow with dark spots; three are large and yellow.]]</div>
Most fruits are found in several varieties, or ''cultivars''. Four of the many varieties of banana are shown in [[#Bananas-image|the illustration]].

and again this works just as we want it to, see sandbox demo 4: the link goes to a point at the top of the image border, so that the alt text and caption are both processed. --Redrose64 🌹 (talk) 07:49, 5 August 2025 (UTC)

Would {{wikicite}} work for application?

{{wikicite|ref=Bananas-image|reference=[[File:Bananavarieties.jpg|thumb|Four varieties of banana|alt=Some bananas of various sizes and colours. Two are large and green; three are small and red; three are small and yellow with dark spots; three are large and yellow.]]}}

Mechanically it works and the link correctly highlights the image, but I wouldn't now if the anchor point is in the correct position for the |alt= text to be read out. -- LCU ActivelyDisinterested «@» °∆t° 08:17, 5 August 2025 (UTC)

@Redrose64 - Thanks for those suggestions on how to implement an anchor. Before you posted that, I had tried "sandbox demo 2" and - as you say - the image is off the top of the screen when you click on the "see image" link. So I implemented "sandbox demo 3" (again, before I saw your post here) and it worked better, but I was aware that future editors may inadvertently separate the anchor from the image (as you say). So, I'm happy to see your "sandbox demo 4" approach ... I implemented it, and you can observe it at James_Cook#JournalPage. If you all settle on a more robust solution - e.g. @ActivelyDisinterested:'s Template:wikicite, I can change it to use that approach. Noleander (talk) 15:18, 5 August 2025 (UTC)

Here is a proposal for new content of MOS:SEEIMAGE (very tentative, pending outcome of this discussion):

References from article text: The article text should not refer to an image with terms such as left, right, above, or below (because image placement varies with platform and screen size, especially mobile platforms, and is meaningless to screen readers). The description of an image should be in its caption, not in article text. To refer to an image from article text, a link to an anchor point may be used. To keep the anchor point attached to the image even if image is moved, this technique may be used: <div class=floatright id=Bananas-image>[[File:Bananavarieties.jpg|thumb|Bananas]].</div> That image would be linked from article text as: Bananas can be many colors, as seen in [[#Bananas-image|this illustration]].

Noleander (talk) 15:45, 5 August 2025 (UTC)

@ActivelyDisinterested: I would not advise the use of {{wikicite}}, because it's using a template for something other than its design purpose. People viewing the wikicode may be puzzled; they may remove it entirely. As regards the q. "I wouldn't now if the anchor point is in the correct position for the |alt= text to be read out" - yes it is, because the alt text is set inside the [[File:...]] and wikicite cannot alter what's in there. The way to check is to save your example to a sandbox somewhere, view the sandbox HTML source and see where the id=Bananas-image attribute lies in relation to the <img /> tag. --Redrose64 🌹 (talk) 19:53, 5 August 2025 (UTC)

I'm on mobile so the html source code isn't easily viewed. I take your point about wikicite, it could be seen as misuse. -- LCU ActivelyDisinterested «@» °∆t° 19:59, 5 August 2025 (UTC)

It should not be too hard to make an equivalent template, like "anchoredimage" to do the same. Masem (t) 20:03, 5 August 2025 (UTC)

The footnote modules would have to be updated to recognise the template, the same way they currently do with wikicite, to avoid false positive no target errors. -- LCU ActivelyDisinterested «@» °∆t° 20:24, 5 August 2025 (UTC)

Which footnote modules, and why? It's not a footnote, it's a means of linking from running text to a point immediately before a relevant image. See for example James Cook, look for the prenthesis "(see the manuscript page)", and follow the link therein. --Redrose64 🌹 (talk) 07:28, 6 August 2025 (UTC)

Sorry I was thinking of the link as using harv template, but this doesn't matter if non-template anchor links are used. -- LCU ActivelyDisinterested «@» °∆t° 09:27, 6 August 2025 (UTC)

Seeing no objection to the proposed text for MOS:SEEIMAGE above (the green text) I updated the MOS guidance at MOS:SEEIMAGE accordingly. Noleander (talk) 16:41, 14 August 2025 (UTC)

WikiWorld cartoons

WikiWorld cartoon currently illustrating the Fake Shemp article

Was there ever a consensus for or against using Wikipedia:WikiWorld cartoons in article space?

It was a comic strip that ran from 2006 to 2008 in the Wikipedia:Wikipedia Signpost, where the artist quoted parts of a Wikipedia article as it was at the time, adding cartoons. I don't know if they were ever intended to be used in mainspace.

Obviously, the text that they're quoting may not be present in the article two decades later. I took the Hammerspace cartoon out of that article last year, because the article text no longer speculated about "the nature of Hammerspace", or mentioned trees and tent poles. (The actual drawing of the monkey with the hammer is perfectly illustrative in isolation, though, and I kept it in the article as a cropped version.)

But even when the text of these cartoons is up to date, a full panel cartoon doesn't really seem like the type of image used for similar purposes in high-quality reference works, and the text is (or should be) redundant to the lead section. Belbury (talk) 16:26, 14 August 2025 (UTC)

Expand AI upscaling definition of "historical images"

The MOS is currently that:

Original historical images should always be used in place of AI upscaled versions.

This occasionally gets pushback on the grounds that somebody doesn't consider a 2024 photo of a celebrity to be "historical" per se. Would there be any objection to expanding it to:

Original historical images, and original photographs of people, should always be used in place of AI upscaled versions.

Belbury (talk) 16:57, 19 August 2025 (UTC)

I would support that change. I'd prefer just dropping "historical" from the status quo. Firefangledfeathers (talk / contribs) 17:00, 19 August 2025 (UTC)

Self-promotion with drawings

I come across the article Sam Kinison where a user replaced a public domain image in the infobox with a drawing he made: [7]

I reverted it, because it seems a bit like tooting your own horn to replace a photo of the subject in an infobox with a drawing. I saw a number of other pages this user put their drawings in them, some got reverted: [8] [9]

Shouldn't adding a drawing you made yourself require consensus? Wikipedia photos become very prominent quickly, if you add a photo to a page, it usually automatically becomes a top 5 Google Image result. Not to mention the brief search I did on here suggests it is highly controversial to use a drawing in a BLP at all. --Quiz shows 18:18, 25 August 2025 (UTC)

Second time this week this happens..... that is me seeing something and seeing that someone else has already posted about it. Are these AI generated images..... best we remove them. Editor seems unwilling to engage when reverted... they just re-implement the image. Lots of cleanup to do here.Moxy🍁 00:04, 26 August 2025 (UTC)

@Quiz shows, if you haven't seen Commons:Deletion requests/Drawn Portraits Uploaded by JoeBugMan it might interest you. Gråbergs Gråa Sång (talk) 07:36, 4 September 2025 (UTC)

A specific sub-set of leadimages

Well, mostly leadimages. If you have an opinion, please join Wikipedia_talk:Biographies_of_living_persons#WP:BLPIMAGE,_again. Gråbergs Gråa Sång (talk) 07:34, 4 September 2025 (UTC)

Image Quality subsection needs to address Dutch Angle photography

Dutch angle photography is not specifically addressed when discussing image quality and it should be. Dutch angle images, especially extreme ones, usually are not encyclopedic. In film, the technique is often used to create tension and uneasiness, which definitely is not encyclopedic.

Please see here and the fourth discussion at this wikicommons link for examples. Tagging @Caterpillar84:, @Infrogmation:, @Omphalographer:, and @Grand-Duc:, who were part of the latter discussion.

My proposed change would be (added text in bold):

Poor-quality images—dark or blurry; video stills; showing the subject too small, hidden in clutter, at strange angles, or ambiguous; and so on—should not be used unless absolutely necessary.

Gb321 (talk) 05:41, 26 September 2025 (UTC)

• Normally I'd interpose WP:MOSBLOAT at this point, but since this is just a one-word addition I think there's no harm, so I've gone ahead and done it (with a different word, however) [10]. EEng 02:41, 27 September 2025 (UTC)

  thank you Gb321 (talk) 03:45, 27 September 2025 (UTC)

Image Quality subsection: Unnecessarily placing photographer's hand in the image

Are pictures like the ones here encyclopedic? Personally, I think not and that inserting your own hand in the image should be highly discouraged. Again this is something I think should be addressed in the Image Quality Subsection

I also think pictures like this one also should be addressed in the Image Quality section, although this issue (shoes/legs in the image) is less significant in my opinion

Gb321 (talk) 16:41, 26 September 2025 (UTC)

• Here I'm going to cry WP:MOSBLOAT. Is this a widespread problem? Has there been controversy over handsiness (or footsiness) that could have been avoided by having a rule on the subject? As to your first link, I've obviated the problem by deleting all the flagpole material, which is rank trivia. EEng 02:52, 27 September 2025 (UTC)

  The user that posted these pics has posted lots of pictures like this. I have had issues with deleting some of their pics in the past, as they often claims its just a difference of opinion unless there is an WP:MOS to cite. I'm already in the process of reaching out on numerous other MOS related image issues, but as far as I can tell there is no MOS to cite for this one

Another issue I've had is pictures like this one, which is used here. I don't think there should be an MOS for an image like this (per WP:MOSBLOAT) but when I deleted this image citing it as being too advertise-y, the user put it back because my deletion was just my opinion. Maybe I should've gone the third opinion route, but I actually didn't know about that option at the time. Is third opinion the best way to go when there isn't a specific MOS to cite? Gb321 (talk) 03:43, 27 September 2025 (UTC)

Well, as to the BBQ article at least, that problem should be solved by deletion of the article on this charming-looking but apparently nonnotable eatery. EEng 11:32, 27 September 2025 (UTC)

Well there's also this one, in an article about a restaurant chain that definitely meets notability guidelines. To be fair, this pic isn't as bad as the other, but do pics like this even belong on Wikipedia? If not, is there a specific MOS (or something else) to cite? Or do I go the Third Opinion route? Thanks for your help. Gb321 (talk) 20:12, 28 September 2025 (UTC)

The image was pointless so I removed it from the article. EEng 02:13, 29 September 2025 (UTC)

Discussion at Wikipedia:Neutral point of view/Noticeboard § Discussion at far-left politics

 You are invited to join the discussion at Wikipedia:Neutral point of view/Noticeboard § Discussion at far-left politics. —Alalch E. 10:24, 16 October 2025 (UTC)