Commons:Village pump/Archive/2026/04

How many people categories is too much?

If we had the image of a historical list with 1,000 people, that we also had wikidata entries for, would we create 1,000 categories for that list? RAN (talk) 04:49, 1 April 2026 (UTC)

Yes, everything that has a Wikidata item should also have a category. GPSLeo (talk) 06:02, 1 April 2026 (UTC)

No. If something has a valid wikidata item, then that's a justification for having it here too, against any questions of 'notability' [sic]. However it's not a requirement to have one, if it's not considered useful to Commons' own goals. Two obvious examples of this might be a highly notable topic where we just don't have any Commons content for it. Another one (which we've encountered previously) was for team photos, where the team had a wikidata item, as did each individual, but the only Commons content was a single photo of the entire team.

We certainly should not bulk auto-create a bunch of empty Commons categories (that are likely to stay empty) from a script run over Wikidata. Andy Dingley (talk) 16:05, 1 April 2026 (UTC)

No, a photo of some list with people's names on it shouldn't be in the categories of all those people but instead a broader category/ies like 'group xy', and/or 'zy lists of people', or 'characteristics zv', etc. Or if you want to categorize lists with merely people's names on it, there's no need to discuss hypotheticals here. There's more than enough challenges and backlogs without discussing hypotheticals and I consider this thread solved. Prototyperspective (talk) 12:02, 1 April 2026 (UTC)

+1. The fact that a name appears in a document is not, in and of itself, a good reason to create a category representing that name. Omphalographer (talk) 18:26, 1 April 2026 (UTC)

• How is it solved with two contradictory replies? Is the assumption that your answer is correct, and the other answer is incorrect? --RAN (talk) 15:12, 1 April 2026 (UTC)

  Maybe I misunderstood your question as to be about which categories to set on an image of a list while you're asking about whether we should create categories for wikidata items. If the latter is the case, then categories still should not be empty. In either way, this seems to be about hypotheticals and there's more than enough nonhypothetical things to discuss. Prototyperspective (talk) 15:48, 1 April 2026 (UTC)

  • And if this is not hypothetical, it would be useful to have a concrete example. But I will venture slightly into hypotheticals. If we had a photo of the entire U.S. Senate of the moment, I would oppose adding a category to that photo for each individual Senator. If we had a picture of a list of 1000 names of individuals, I would certainly not add categories for the all people named in the list, any more than I would add, for a PDF of a book, a category for every place mentioned in the book. - Jmabel ! talk 18:55, 1 April 2026 (UTC)

• Maybe not places, but what about people? In that book pdf, there might be mentioned "Jim Smith" that we have a category already as "James H. Smith". How would someone refind it once the connection has been made? People have synonyms and researchers need a way to aggregate all the information on a person. I can see not needing to index every mention of George Washington in a book, but some people are more obscure. --RAN (talk) 21:12, 1 April 2026 (UTC)
  • If the book has substantial coverage of the person, sure, but (for example) I would not want to tag a history of European art with the categories for 583 artists. If anything, flooding the category with content like this makes it harder to find actually relevant material. - Jmabel ! talk 21:25, 1 April 2026 (UTC)

How would it make info more difficult to find? If I have no interest in the index of a book, I don't look at it. If I need to find someone, I use the index. Text searching has made most indexes redundant, but as pointed out people have synonyms. --RAN (talk) 02:01, 2 April 2026 (UTC)

+1, I think this issue is covered by Commons:Overcat, is it not? My favorite example (not with people, but locations) is that a world map should not be categorized into all hundreds or thousands of location-categories of the places that are shown and/or labelled in that map. Just because one dot in this map is labelled "Lhasa", does not turn that map into a "map of Lhasa", at least in my opinion. The same logic goes for the hypothetical group photos of large-crows, for long name lists, or bound collections of short-bios. There are reasonable exceptions, but I think that most files with way over 10 categories are cases of overcat/miscat. --Enyavar (talk) 22:05, 1 April 2026 (UTC)

Commons:Overcat deals with redundant categories such as adding Category:Albert Einstein and Category:Physicists from Germany to an image of Einstein, not about properly identifying everyone in an image (or I assume list). --RAN (talk) 01:38, 2 April 2026 (UTC)

If that is so, then I think we should add a point to the Categories' policy about not adding hundreds/thousands of categories to files. Even if we could identify each of the people in this painting by name, I argue that we should only do so in respective crop-outs where those are needed... but not in the larger picture. --Enyavar (talk) 11:52, 2 April 2026 (UTC)

@Richard Arthur Norton (1958- ): flooding a category with tangentially relevant files makes genuinely relevant files harder to find. - Jmabel ! talk 20:18, 2 April 2026 (UTC)

Notification of DMCA takedown demand — Brixton riots, 1981 (enwiki)

In compliance with the provisions of the US Digital Millennium Copyright Act (DMCA), and at the instruction of the Wikimedia Foundation's legal counsel, one or more files have been deleted from Commons. Please note that this is an official action of the Wikimedia Foundation office which should not be undone. If you have valid grounds for a counter-claim under the DMCA, please contact me.

The takedown can be read here.

Affected file:

• en:File:Brixton_riots,_1981.jpg

To discuss this DMCA takedown, please go to COM:DMCA#Brixton riots, 1981 (enwiki). Thank you! Joe Sutherland (WMF) (talk) 12:37, 2 April 2026 (UTC)

Notification of DMCA takedown demand — Dragon Bravo Fire Pyrocumulus

In compliance with the provisions of the US Digital Millennium Copyright Act (DMCA), and at the instruction of the Wikimedia Foundation's legal counsel, one or more files have been deleted from Commons. Please note that this is an official action of the Wikimedia Foundation office which should not be undone. If you have valid grounds for a counter-claim under the DMCA, please contact me.

The takedown can be read here.

Affected file:

• File:Dragon Bravo Fire Pyrocumulus 7.28.2025.jpg (edit|talk|history|links|watch|logs)

To discuss this DMCA takedown, please go to COM:DMCA#Dragon Bravo Fire Pyrocumulus. Thank you! Joe Sutherland (WMF) (talk) 12:48, 2 April 2026 (UTC)

Broken Wikimedia image (sizes)

This semester and last the embedded Commons images I've long used have been broken. I was fixing them ad hoc, but today tried to figure out what was going on. It appears that a lot of the sizes that used to be served offsite no longer are?

I have a python script, which might've fixed most of the issues, but though I looked I can not find the discussion of what changed and why.

❯ wikipedia-image-embeds-fix.py talks/
Updated 1 links in talks/180-privacy.md
    1024px to 960px : https://upload.wikimedia.org/wikipedia/commons/thumb/8/8b/CPT-Proxy.svg/960px-CPT-Proxy.svg.png
Updated 1 links in talks/056-brown-learning.md
    1024px to 960px : https://upload.wikimedia.org/wikipedia/commons/thumb/f/fc/Laurentius_de_Voltolina_001.jpg/960px-Laurentius_de_Voltolina_001.jpg
Updated 5 links in talks/075-darknet.md
    512px to 500px : https://upload.wikimedia.org/wikipedia/commons/thumb/8/81/Digital_signature_schema.png/500px-Digital_signature_schema.png
    512px to 500px : https://upload.wikimedia.org/wikipedia/commons/thumb/5/55/Bitcoin_Block_Data.svg/500px-Bitcoin_Block_Data.svg.png
    512px to 500px : https://upload.wikimedia.org/wikipedia/commons/thumb/e/e1/Onion_diagram.svg/500px-Onion_diagram.svg.png
    1024px to 960px : https://upload.wikimedia.org/wikipedia/commons/thumb/0/06/Iceberg_of_Webs.svg/960px-Iceberg_of_Webs.svg.png
    128px to 330px : https://upload.wikimedia.org/wikipedia/commons/thumb/0/06/Iceberg_of_Webs.svg/330px-Iceberg_of_Webs.svg.png

Reagle (talk) 20:22, 2 April 2026 (UTC)

Custom thumbnail sizes got entirely removed because of massive crawler traffic. See mw:Common thumbnail sizes for the available sizes. GPSLeo (talk) 20:38, 2 April 2026 (UTC)

Proposing the removal of the previous version of a file

Shortly after uploading a picture, I noticed a visibile license plate in it and so I uploaded a new version with said license plate blurred. Is there a way to propose the deletion of the precious version?

If not, what is the correct method to solve this sort of issue?

I suppose something like:

1. Ask for deletion first;
2. wait for the file to be removed;
3. Upload the file with license plate blurred.

Thanks in advance,

--Marco (talk) 06:54, 8 April 2026 (UTC)

Yes, see COM:REVDEL.

This section is resolved and can be archived. If you disagree, replace this template with your comment. Prototyperspective (talk) 12:42, 8 April 2026 (UTC)

Prototyperspective (talk) 12:42, 8 April 2026 (UTC)

Action Required: Update templates/modules for electoral maps (Migrating from P1846 to P14226)

Hello everyone,

This is a notice regarding an ongoing data migration on Wikidata that may affect your election-related templates and Lua modules (such as Module:Itemgroup/list).

The Change:
Currently, many templates pull electoral maps from Wikidata using the property distribution map (P1846), combined with the qualifier depicts (P180): electoral result (Q19571328).

We are migrating this data (across roughly 4,000 items) to a newly created, dedicated property: apportionment diagram (P14226).

What You Need To Do:
To ensure your templates and infoboxes do not break or lose their maps, please update your local code to fetch data from apportionment diagram (P14226) instead of the old distribution map (P1846) + depicts (P180) structure. A list of pages was generated using Wikimedia Global Search.

Deadline:
We are temporarily retaining the old data on distribution map (P1846) to allow for a smooth transition. However, to complete the data cleanup on Wikidata, the old P1846 statements will be removed after May 1, 2026. Please update your modules and templates before this date to prevent any disruption to your wiki's election articles.

Let us know if you have any questions or need assistance with the query logic. Thank you for your help! ZI Jony using MediaWiki message delivery (talk) 17:11, 3 April 2026 (UTC)

Wiki Loves Monuments UK upload tool broken

Hi all. https://wlmuk.toolforge.org/ has not been working for several days. It just says "Loading..." and gets no further. I use it frequently (even outside the WLM competition period) to upload photos of historic buildings as it does some useful pre-filling of identifiers etc. Anyone know what the problem is? Dave.Dunford (talk) 12:54, 10 April 2026 (UTC)

@Richard Nevell (WMUK) might be able to help. Ciell (talk) 15:02, 10 April 2026 (UTC)

It now appears to be working again. Thanks, if it was down to manual intervention. Dave.Dunford (talk) 16:05, 10 April 2026 (UTC)

This section is resolved and can be archived. If you disagree, replace this template with your comment. --Prototyperspective (talk) 16:53, 10 April 2026 (UTC)

Idea for tool: "Other versions"

Category:Plain black Pinhead SVG icons

It would be useful to have tool, perhaps a user script, which would allow the user to select (in the manner of selection in Cat-a-lot) two or more images in a category view, like the above, and then add a thumbnail or link to each of them, from the other as |other versions=.

For the images in the above screenshot, it would be equal to these two edits: 1; 2.

It would need to detect and gracefully fail if the link is already present.

Does such a tool exist, and if not could someone please make one? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:55, 5 April 2026 (UTC)

Note: it's not very common that the other versions or all of them are in some same category and even when they are due to many other files being there too and them being named differently, it can be difficult to select files that way and seems rather impractical. Making it easier to add other versions of files could nevertheless be impactful. Prototyperspective (talk) 20:09, 5 April 2026 (UTC)

Shouldn't that typically use {{Other version}} rather than just a thumb? - Jmabel ! talk 21:49, 5 April 2026 (UTC)

Photo challenge February 2026 results

First aid: Entries • Votes • Scores

Rank	1	2	3
image
Title	A labrador retriever of the search and rescue dog unit of the Austrian Red Cross. Vienna 2025, Austria.	Lifting a person in a rescue basket.	REGA rescue team
Author	Aciarium	Julian Herzog	Roy Egloff
Score	20	12	9

Orange: Entries • Votes • Scores

Rank	1	2	3
image
Title	Woman in traditional orange dress called Buhara sun, Buhara, Uzbekistan	Fire burns, but makes you warm and relaxed	People viewing trees across the pond, illuminated by the setting sun.
Author	PetarM	Maryam Yazdanisheldareh	Ka23 13
Score	14	11	9

Congratulations to @Aciarium, @Julian Hendrawan, @Roy Egloff, @PetarM, @Maryam Yazdanisheldareh and @Ka23 13. This is Taiwania Justo speaking (Reception Room) 11:43, 6 April 2026 (UTC)

"Cosplay"

Could someone clarify our working definition of "cosplay"? In particular, is there any wearing of costumes that does not qualify as "cosplay", and if so what? I am increasingly seeing the term applied by others to my photos in contexts where I would never use that word. - Jmabel ! talk 22:13, 4 April 2026 (UTC)

Well, for me, as cosplay (costume and character play) is wearing a costume of a fictional character, like Link, and acting like him. --PantheraLeo1359531 😺 (talk) 07:34, 5 April 2026 (UTC)

So if we have a category of kids doing trick or treat, that would be nested under cosplay? And actors in a theatre play ? —TheDJ (talk • contribs) 14:30, 5 April 2026 (UTC)

Some other examples where it has been applied where I find it dubious:

• File:Elvis dog costume for Halloween 2012 (8106438179).jpg
• File:2015 Fremont Solstice cyclists 232.jpg (NSFW)
• File:2015 Fremont Solstice cyclists 183.jpg
• File:Say No To Trump Demo (43401912921).jpg
• File:Moscow car meet 2025-07-12 NKVD cosplay 07.jpg

For most of these its a matter of people (or a dog!) in costume, but no indication of "play" beyond simply wearing a costume. Jmabel ! talk 21:47, 5 April 2026 (UTC)

To borrow the definition from the English Wikipedia:

A cosplay (a portmanteau of "costume play") is the activity and performance art in which participants called cosplayers wear costumes and fashion accessories to represent a specific character. Cosplayers often interact to create a subculture, and a broader use of the term "cosplay" applies to any costumed role-playing in venues apart from the stage.

(The subculture in question is, more often than not, fandom). This is a pretty broad definition, but also states that stage-play actors are excluded, answering an earlier question. It also gives a definition that doesn't include (role)playing specifically, but puts the focus on the costume to represent a character, which squares with my personal experience having attended cons where people cosplay characters by dressing up as them, but aren't always (or even usually) in-character.

Wikipedia states that "It is generally considered different from Halloween and Mardi Gras costume wear, as the intention is to replicate a specific character, rather than to reflect the culture and symbolism of a holiday event."

Within the context of e.g. a kid dressing up as Batman for Halloween, are they not trying to also replicate Batman? Can it not be both an expression of fandom and something you do to go trick-or-treating? Children don't tend to pick characters at random to dress up as. We could make a separate category for people who dress up as certain characters for Halloween and it would be a pretty unambiguous separation, but the question is also if that's worth the effort. I personally don't see it as that big of a deal to categorize all of these as cosplay, even if "People dressed up as..." might be a better/more accurate way to categorize these photos (except for the dog). The Moscow car meet photo shows a historical reenactor (generally not considered cosplay), and is appropriately categorized as such. ReneeWrites (talk) 23:24, 5 April 2026 (UTC)

But the Moscow one refers to "cosplay" in its title. - Jmabel ! talk 05:59, 6 April 2026 (UTC)

In the file name, yes. Do you want me to change it? (And do the same for others in that set). ReneeWrites (talk) 18:12, 6 April 2026 (UTC)

@ReneeWrites: probably a good idea. And I'd really love for Category:Cosplay to have a much clearer description. The one there now would include theater, reenactors, etc., and the "many other outlets" is hopelessly vague. - Jmabel ! talk 22:19, 6 April 2026 (UTC)

Would this: Special:Diff/1122831921/1192847283, be acceptable? Regards, Grand-Duc (talk) 22:26, 6 April 2026 (UTC)

I think that's a much better description, thank you for adding it. ReneeWrites (talk) 08:35, 7 April 2026 (UTC)

Definitely a step in the right direction. Still unclear on reenactments, and somewhat unclear (at least to me) on whether simply wearing a costume is enough vs. the "play" aspect. - Jmabel ! talk 19:34, 7 April 2026 (UTC)

Is it workable to nominate for deletion most (If not all) files in a category?

Okay, this might be a bit rage-driven but... As per this guideline about derivative works, unfortunately most pictures of toys are not acceptable in Commons. I know, I am talking about deleting thousands of images. It that feasible? Or is it better to do it on a case-by-case basis? --JJ - Schumi4ever (talk) 01:02, 9 April 2026 (UTC)

@Schumi4ever:

• In my experience, mass DRs should take on a group of files that will almost certainly stand or fall together. It's OK if there might turn out to be a few outliers, but (for example) you would not want to mix Raggedy Ann dolls with chess sets, because the facts of the case may be completely different. Similarly, you would not want to mix closeups of toys with an image where the presence of the toy might well be de minimis. And you need to watch out for cases like a well-licensed photo of a U.S. toy from the 1960s that might never have been properly copyrighted. So, while it is reasonable to do these in batches, a mass DR containing hundreds of photos will often be semi-quickly closed as a "procedural keep" for presenting too many different cases.
• For the technical side of how to do this, see Commons:Deletion requests/Mass deletion request. I strongly recommend that you use the method described there using VisualFileChange, not the more manual method. - Jmabel ! talk 02:59, 9 April 2026 (UTC)

Gadgets not working

As of approximately midnight today, the controls of gadgets (HotCat, Cat-a-lot) have stopped appearing on the pages. Can anyone solve this? ŠJů (talk) 04:34, 9 April 2026 (UTC)

@ŠJů, hello! Gadgets are working (at least for me) right now. Deltaspace42 (talk) 10:04, 9 April 2026 (UTC)

@Deltaspace42: Unfortunately, the problem still persists. It makes work very difficult. --ŠJů (talk) 12:59, 9 April 2026 (UTC)

@Deltaspace42: Maybe, it can relate with the last update of Firefox (149.0.2)? In Google Chrome, gadgets work normally for me. --ŠJů (talk) 13:15, 9 April 2026 (UTC) In MS Edge, the HotCat works, and the Cat-a-lot control is not available on pages. --ŠJů (talk) 13:31, 9 April 2026 (UTC)

@ŠJů, I don't think so; both in Chrome and Firefox gadgets work normally for me. Deltaspace42 (talk) 13:22, 9 April 2026 (UTC)

I'm on Firefox (149.0.2) and not having this problem. - Jmabel ! talk 20:43, 9 April 2026 (UTC)

MediaWiki:Gadget-UTCLiveClock.js disappeared for me.

dont notice problems for other gadgets. RoyZuo (talk) 11:05, 10 April 2026 (UTC)

Same, UTCLiveClock disappeared for me as well. Deltaspace42 (talk) 12:06, 10 April 2026 (UTC)

The problem lasted about 24 hours and has now gone away on its own. I have no idea what could have caused it. --ŠJů (talk) 23:21, 10 April 2026 (UTC)

Template:Glamorous

it appears that https://glamtools.toolforge.org/glamorous/ just got updated and the permalinks to queries changed, so the template needs to be updated soon to fix the link. RoyZuo (talk) 21:18, 9 April 2026 (UTC)

Convenience link: Template:Glamorous. - Jmabel ! talk 21:55, 9 April 2026 (UTC)

Fixed it. Also added default parameters. Adjust or remove as adequate, thanks. Prototyperspective (talk) 23:49, 9 April 2026 (UTC)

Lots of other pages also link to glamorous v1 scans. The site of version 1 was down at time of thread creation but it's back up again. I don't know if a version switch is planned or if other pages need to be or would benefit from having their glamorous links changed as well. Version 2 has some major advantages over version 1 but it also has major downsides, one of which is that pagination seems broken (the issue is now visible after migration to codeberg): https://codeberg.org/magnusmanske/glamtools/issues/110 Prototyperspective (talk) 17:06, 10 April 2026 (UTC)

Wikidata Infobox conflicts with marker for unpatrolled pages

see [1], {{Wikidata Infobox}} conflicts with the marker Mark this page as patrolled (it shifts all the content down to the end of the page). No idea where this can be fixed. It is possible that you must have patrol rights to see the effect. User:Mike Peel fyi. best --Herzi Pinki (talk) 21:59, 10 April 2026 (UTC)

The clear: both; in .patrollink forces the break after the infobox, but I am not certain if this should be changed. It makes immediately apparent that the category has not yet been patrolled, is not visible to normal users without patrol right and the issue disappears if you press the button and mark the category as patrolled. Since this is a Mediawiki feature an interface admin is likely needed to change this. MKFI (talk) 08:01, 11 April 2026 (UTC)

I have the right to patrol and thus see the box. Never thought about what the task of patrolling comprises (ok, read now Commons:Patrol). So sometimes I click patrolled, sometimes I don't. But the situation is a bit different, there is this lengthy Infobox and thereafter is the patrol box. Would be better the other way round. Patrol box at the top (so see it immediately, Infobox can be longer than the page), Infobox thereafter and left to the Infobox no more empty white canvas. best --Herzi Pinki (talk) 10:11, 11 April 2026 (UTC)

A. J. Sawyer?

There is discussion on en.Wikipedia of File:Akilagpa Sawyerr.jpg, which is described as "taken by Photographie Leopold Dubois in Poitiers, France [and] marked A. J. Sawyer 1882 on the reverse", but said to depict en:Akilagpa Sawyerr (lawyer), who was born in 1883.

Thoughts welcome, there or here. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:37, 5 April 2026 (UTC)

• Yes, odd to have a photo showing a birth year rather than the year taken. I wonder if his father had the same name, and the image was taken in 1882, which seems to fit the style of clothing. --RAN (talk) 17:18, 8 April 2026 (UTC)

The uploader indicates that he saw the words "A. J. Sawyerr 1882", researched him [sic] and drafted an article. However:

• Akilagpa Sawyerr's initials were not A J Sawyerr. His birth name was Gosford Collins Sawyerr and he later adopted the name Akilagpa Osabambra Sawyerr.
• the person in the photo is I would say at least 20; Sawyerr would have been 20 in 1903
• that doesn't look like a 1903 or later photo
• there's nothing in the research to suggest Sawyerr travelled to the US where the uploader found the photos in his family's photo albums.

Suggest the file be renamed A J Sawyerr, circa 1882. Mmemaigret (talk) 22:29, 13 April 2026 (UTC)

Getting some books from HathiTrust

Hi, There are some books from HathiTrust which I can't get. I have tried several times each, and all times failed. HathiTrust Download Helper doesn't work anymore, and it is a pity. It was very reliable. I use the Internet Archive Downloader, but it is not reliable at all, specially for big books. Any idea? Yann (talk) 20:05, 12 April 2026 (UTC)

The books:

• Annales de chimie et de physique Ser. 1, t. 63-64 (1807) to be uploaded over File:Annales de chimie, tome 63-64, 1807.pdf
• Annales de chimie et de physique Ser. 1, t. 65-66 (1808) to be uploaded over File:Annales de chimie, tome 65-66, 1808.pdf
• Le Mouvement socialiste v.5 (1901) to be uploaded over File:Le Mouvement socialiste, Jan-Jun 1901.pdf (US only)
• Le Mouvement socialiste t.6 (1901) to be uploaded over File:Le Mouvement socialiste, Jul-Dec 1901.pdf (US only)
• Le Mouvement socialiste t.7 (1902) to be uploaded over File:Le Mouvement socialiste, Jan-Jun 1902.pdf (US only)

I can try in about eight hours. —Justin (koavf)❤T☮C☺M☯ 21:20, 12 April 2026 (UTC)

Wow, I not only cannot download them, but they won't render in my browser at all. I'm stumped. —Justin (koavf)❤T☮C☺M☯ 07:44, 13 April 2026 (UTC)

@Yann: If you know, how to set a VPN tunnel, you can use a fake US based IP address, if the files cannot be displayed due to copyright considerations in your home country. For instance, Norton 360 provides a comfortable way to use fake IP addresses via VPN with three mouse clicks. NearEMPTiness (talk) 07:05, 13 April 2026 (UTC)

@NearEMPTiness: Yes, I know. I do have a VPN, and that's not the issue. These are very big books (>700 pages). In the highest resolution, they are more than 500 MB. The issue is that downloading fails repeatedly. And I don't know why, there is no meaningful error message. I have also tried getting them not on the highest resolution, but it doesn't work either. It seems that the issue is rather ralated to the number of pages. Yann (talk) 08:04, 13 April 2026 (UTC)

Try to ask on w:en:Project:WikiProject Resource Exchange/Resource Request, I think more users reading that will know how to handle these downloads. – b_jonas 08:49, 13 April 2026 (UTC)

Well, this request has nothing to do with English Wikipedia. It concerns Commons and Wikisource. Yann (talk) 09:51, 13 April 2026 (UTC)

To that point, you can try s:en:Wikisource:Scan Lab. —Justin (koavf)❤T☮C☺M☯ 09:59, 13 April 2026 (UTC)

I see 2 possible solutions: 1. Getting access to a university account, or 2. using the command line tool hathitrust-downloader, which is probably more reliable than the browser extension, but this gives a 403 error. Yann (talk) 09:21, 14 April 2026 (UTC)

 Done by Koavf. Thanks a lot! Yann (talk) 13:47, 14 April 2026 (UTC)

I found a way around. I discovered that Ayesha can download part of a file. So cutting a large file into smaller chunks, it doesn't fail, and it is actually faster. The whole file has to be reconstructed, but it is more reliable. Yann (talk) 18:10, 14 April 2026 (UTC)

Government images

I want to add a government image to Wikimedia Commons, but don't know if it's allowed to upload. The image is in an official government website that can be seen by everyone who has access to Internet but don't know if I can safely upload it. — Preceding unsigned comment added by LavaSalt402 (talk • contribs) 18:09, 16 April 2026 (UTC)

@LavaSalt402: Hi, and welcome. What is the URL of said image, or how can we get to it?   — 🇺🇦Jeff G. ツ please ping or talk to me🇺🇦 18:31, 16 April 2026 (UTC)

There is no URL, but the photo is the one that looks like this located in this Turkish government website. LavaSalt402 (talk) 18:45, 16 April 2026 (UTC)

@LavaSalt402: I can't reach that website (it takes too long to respond), but I can reach this archive of the source of File:Logo of the Turkish Application to the European Union.svg.   — 🇺🇦Jeff G. ツ please ping or talk to me🇺🇦 19:21, 16 April 2026 (UTC)

The picture is actually the logo of the old Ministry for European Union affairs of Turkey, retrieved from a government website. Do you think would that be a problem? LavaSalt402 (talk) 04:57, 17 April 2026 (UTC)

@LavaSalt402: Yes, that would be a problem unless Turkey has carved out a copyright exemption for their Ministry for European Union affairs and we have documentation to that effect. I have been trying to reach that website from multiple networks in the US for multiple days; perhaps it is geofenced.   — 🇺🇦Jeff G. ツ please ping or talk to me🇺🇦 13:10, 20 April 2026 (UTC)

In my region, I can access the picture but can't find any copyright. LavaSalt402 (talk) 13:50, 20 April 2026 (UTC)

@LavaSalt402: What region are you in? I found that per COM:Turkey Turkey joined the Berne Convention 1 January 1952, so anything created in Turkey after that date was automatically copyrighted from the moment it was fixed in tangible form.   — 🇺🇦Jeff G. ツ please ping or talk to me🇺🇦 14:04, 20 April 2026 (UTC)

@LavaSalt402: What do you mean by "can't find any copyright"? Just "can't find an explicit copyright notice"? Or something else? The default is that everything copyrightable is copyrighted. We would need a specific reason for something copyrightable not to be copyrighted. - Jmabel ! talk 20:43, 20 April 2026 (UTC)

@LavaSalt402: for most countries, the fact that something is a "government image" doesn't change anything about its copyright status, though it may affect ownership of the copyright. The issues are pretty much the same as for any other rights-holder.

It sounds like your specific question has been answered, but if you are working on other government content from Turkey, you will find Commons:Copyright rules by territory/Consolidated list T#Turkey useful. Analagous pages/sections exist for virtually every jurisdiction in the world. - Jmabel ! talk 23:50, 16 April 2026 (UTC)

I couldn't find anything about government images in that page's Turkey part. LavaSalt402 (talk) 04:58, 17 April 2026 (UTC)

I think there are a lot of government logos, so uploading it won't be a problem. LavaSalt402 (talk) 13:46, 17 April 2026 (UTC)

I saw a lot of government logos in Wikimedia Commons, so I think there shouldn't be a reason for me to not upload them. What do you think? LavaSalt402 (talk) 19:59, 17 April 2026 (UTC)

Again: some logos are OK for Commons (usually because they are below the threshold of originality, but also sometimes because they are very old, or because some government entities either license everything they create or place it in the public domain, but that last does not generalize. The lack of anything specific about government-created images in the section on Turkey would mean that they are handled exactly the same way as non-government images. - Jmabel ! talk 20:57, 17 April 2026 (UTC)

Which means I can upload it, right? (It's a former ministry logo) LavaSalt402 (talk) 05:12, 18 April 2026 (UTC)

No, it does not. Unless there is a verifiable reason that a file is in the public domain, it is copyrighted and cannot be uploaded here. Pi.1415926535 (talk) 05:32, 18 April 2026 (UTC)

What is the evidence for the logo not to be protected by copyright? Trade (talk) 00:01, 19 April 2026 (UTC)

I have no idea. LavaSalt402 (talk) 08:38, 19 April 2026 (UTC)

Guess there's the answer to your question then Trade (talk) 11:31, 20 April 2026 (UTC)

And the answer is LavaSalt402 (talk) 13:48, 20 April 2026 (UTC)

@LavaSalt402: Don't upload it; if you do, I promise that I will report you for violating the copyright of the rightsholder.   — 🇺🇦Jeff G. ツ please ping or talk to me🇺🇦 14:00, 20 April 2026 (UTC)

Ok? LavaSalt402 (talk) 16:32, 20 April 2026 (UTC)

Will you still report me if I find out that you can upload it freely? LavaSalt402 (talk) 16:49, 20 April 2026 (UTC)

I would suggest you get the permission before uploading other people's stuff rather than doing it afterwards Trade (talk) 17:04, 20 April 2026 (UTC)

I didn't upload the file yet and who should I get permission from (the Minister of EU affairs is not present) LavaSalt402 (talk) 17:11, 20 April 2026 (UTC)

@LavaSalt402; US Government logos are different because the US Government has excepted the employees of said government from acquiring copyright while performing their duties. Many other governments have not. See, for example, COM:US#US Government agencies, {{PD-USGov}}, and Title 17, Chapter 1, Section 105 of the US Code.   — 🇺🇦Jeff G. ツ please ping or talk to me🇺🇦 12:50, 20 April 2026 (UTC)

And this only generally applies to the US federal government; works created by most state and local governments can still be copyrighted. Omphalographer (talk) 20:45, 20 April 2026 (UTC)

This section is resolved and can be archived. If you disagree, replace this template with your comment. --Prototyperspective (talk) 12:24, 20 April 2026 (UTC)

Telephone box/kiosk/booth categories in the UK

Hi all. The category tree below Category:Telephone booths in the United Kingdom is inconsistent: category names refer interchangeably to these structures as either boxes, booths or kiosks. This is quite unhelpful when trying to assign new or uncategorised photos of telephone boxes to the most specific category. IMO "telephone box" is probably the commonest UK usage, but I don't want to impose my own assumptions and overall, booth seems to be the most frequent, and reflects the international usage higher in the category tree. Anyone object if I tidy up, renaming these categories as booth throughout (leaving category redirects, of course)? I have category renaming rights, so I'm proposing to rename each category manually, but are there any tools that might be useful? Dave.Dunford (talk) 10:49, 15 April 2026 (UTC)

@Dave.Dunford, hello! I think you might find this tool useful: Indic-TechCom/Tools/MassMove. Deltaspace42 (talk) 11:02, 15 April 2026 (UTC)

@Deltaspace42 Thanks. I've got User:Jeff G./massrename.js installed already, which seems to do much the same job but, like MassMove, doesn't automatically move the files in the old category into the new. So it looks like it's going to be a semi-manual process, using either massrename.js or massmove.js to rename the category and then Cat-a-Lot to move the files. Dave.Dunford (talk) 12:52, 15 April 2026 (UTC)

• Oppose Commons:Categories for discussion/2019/07/Category:BT Group telephone boxes Andy Dingley (talk) 12:55, 15 April 2026 (UTC)

  So what do you suggest? The category tree below Category:Telephone booths in the United Kingdom is inconsistent (e.g. Category:Telephone booths in Wiltshire, Category:K6 telephone kiosks in Wiltshire, Category:Grade II listed telephone boxes in Wiltshire). Do we adopt a British convention (either "boxes" or "kiosks") below Category:Telephone booths in the United Kingdom, or go with the international word throughout for the sake of consistency? Dave.Dunford (talk) 14:13, 15 April 2026 (UTC)

  Things like this are generally/usually discussed in CfDs which is why more participants in these are needed since we can't have 3000 or so threads about such here – see thread above. Prototyperspective (talk) 14:35, 15 April 2026 (UTC)

  Thanks – noted. I'll take the discussion back to Commons:Categories for discussion/2019/07/Category:BT Group telephone boxes, though it's wider than that. Dave.Dunford (talk) 15:46, 15 April 2026 (UTC)

• Generally I'd favour boxes, per COMMONNAME, and I'm fine with any push to make these uniform across the UK. Maybe 'kiosk' if the type is qualified as the K2 / K6 etc. (it is after all, what the K stood for). But 'booth' is the one I'd oppose. Andy Dingley (talk) 14:54, 15 April 2026 (UTC)

Photographs of schools

Something I've noticed while looking for photos on WikiShootMe is the lack of photos of primary and secondary schools, at least within England and Wales. Obviously there are good reasons why people wouldn't want to be taking photos of schools while walking around their town, but nonetheless it would be useful within the scope of Commons.

I wonder whether there's some way of encouraging people to upload photos— either school staff and LEAs, or volunteer photographers as usual. (I suspect it looks less suspicious if you do it during the holidays.) Marnanel (talk) 13:16, 10 April 2026 (UTC)

@Marnanel: The extension of this would be use on English Wikipedia, where such schools are not considered to be notable.   — 🇺🇦Jeff G. ツ please ping or talk to me🇺🇦 14:22, 10 April 2026 (UTC)

Yes, but

• that's enwp's policy and not ours
• enwp is not the only place that has an interest (cy.wp, kw.wp)
• a notable event could suddenly happen there and one of those Wikipedias would need an image
• ditto Wikinews
• it doesn't matter what any project but Commons thinks, anyway: per COM:PS, Commons "acts as a common repository for all Wikimedia projects, but the content can be used by anyone, anywhere, for any purpose."

Marnanel (talk) 18:24, 10 April 2026 (UTC)

This is a good point, Commons needs a broader coverage of public buildings in general. I think, the awareness is not that high. I try to cover the municipalities in my home county, and when I look up what's already there, then it is mostly churches and cultural heritage monuments, but rather less town halls, main streets, prominent shops and of course schools, etc... I was allowed to take images of classrooms; something like this would be a super addition, too, but it is of course not so easy to get a permission --PantheraLeo1359531 😺 (talk) 16:24, 10 April 2026 (UTC)

Commons needs a broader coverage of public buildings in general […] I was allowed to take images of classrooms; something like this would be a super addition, too, but it is of course not so easy to get a permission I made this photo challenge competition proposal about/relating to this: 'People in public institutions buildings'.

encouraging people to upload photos— either school staff and LEAs I have some doubt school building photos are of special importance compared to notable buildings more broadly. May be good to focus on that for eg a campaign but then I think other types of buildings would be good to include in later iterations of a series. Images may be missing in WikiShootMe but there is little practical use of that. It's kind of still valuable to have free-licensed images of schools, eg people would want to have images of the places they went to for long times and many schools have Wikipedia articles where some image(s) are missing. One could also scan Wikipedia articles about schools with 0 images in various ways. I have some doubt it's a media gap of high importance though but would support this more than the Nth iteration of Wiki Loves Monuments when I don't think there's many monuments left that don't have photos. In the Wikipedia Nearby Places map one can see school articles on the map and an image would help identify which building it is but it probably can usually be identified without it and I'd like to filter away school articles from the map because these aren't particularly interesting places to discover/explore/visit in cities imo. Prototyperspective (talk) 16:52, 10 April 2026 (UTC)

Yes, something like WikiShootMe is a good idea. Interest will grow with the time. There will be some time, where people are glad that somebody when, of all people, someone actually documented it --PantheraLeo1359531 😺 (talk) 13:39, 11 April 2026 (UTC)

Same as User:PantheraLeo1359531 said, even though I live in a foreign country 2 decades after wikipedia / commons wikimedia have been created, I have uploaded the first photos of many ordinary things in this place. A very recent example: Category:Freie Waldorfschule Berlin-Mitte, in the center of the capital and an international metropolis!

Because I am more curious. Whenever I see something that stands out even just a bit from the rest, I'd take and upload a photo. I believe most buildings and most organisations (among them all the long-established ones) are worth documenting. Schools, hospitals, government agencies, are all important. Maybe not all kindergartens or doctor's offices. When I am on the road taking photos, I usually dont check whether the objects have been photoed and uploaded to commons. I just photo them anyway. Like the example above, I realise only now that I am the first.

Also, I often dont just take a photo of the front view / facade, but also try to document its sides, back, inside, details... Many "photographers" have an eye for only the landmarks and only their facades, especially from afar. RoyZuo (talk) 21:38, 10 April 2026 (UTC)

and because so many things were not documented until i did, i often had to create the commons categories and wikidata items for them, which i had no clue what they were and spent a great deal of time googling, reading and checking the maps to find out. i got so burned out by this chore that now i just upload and leave the chores to other users who have way better local knowledge. RoyZuo (talk) 21:43, 10 April 2026 (UTC)

this legend Category:Photographs by Anonymous Hong Kong Photographer 1 (e.g. 2024 uploads) is an inspiration. they probably have been photoing every building on every road they walk past for over 2 decades now, uploading 10k to 20k photos each year. RoyZuo (talk) 21:54, 10 April 2026 (UTC)

Wow, this is very enthusiastic work --PantheraLeo1359531 😺 (talk) 16:30, 11 April 2026 (UTC)

I'm a keen documenter of everything around me, so yes, I think it would be great if there were greater awareness about documenting schools, especially public ones, which, in a country like mine, can even house important works of art. I usually do it in the afternoons when no one is around, or on weekends to protect the privacy of minors. I would just mention that something I *don't* find particularly useful is that on Wikishootme you're more likely to find a nearby hotel than a school. ProtoplasmaKid (talk) 15:25, 15 April 2026 (UTC)

i think many wd items, which show up on shootme if it has coordinate location (P625) set, were created by importing from certain databases. so for some areas things are very well represented in wd if say a particular city has a database with coords and that database gets imported. for example, every street and every protected structure in some cities (like berlin the current capital of germany) show up on shootme, but some (like bonn the former capital) only have some landmarks. RoyZuo (talk) 09:26, 16 April 2026 (UTC)

unjustified massive requests for deletion

Hi! the partially blocked user Henblo has opened several requests for deletion of files related to symbols of the Colombian government. The user claims that they are not official symbols, without providing any further justification. Even if that were the case, the files were created and uploaded by Commons users; they were not copied from other sites. Government symbols in Colombia are in the public domain under the law according to “Law 23 of 1982 on Copyright, Articles 11, 21-29”, amended by “Law 44 of 1993, Article 2” and “Law 1520 of 2012, Articles 4, 6 and 10.

Is there anyway to dismiss the requests collectively?

The deletion requests are the following:

• Commons:Deletion requests/File:Presidential Seal of Colombia (1922—2002).svg
• Commons:Deletion requests/File:Función Pública.svg
• Commons:Deletion requests/File:Departamento de Planeación (DNP).svg
• Commons:Deletion requests/File:Gobierno de Colombia.svg
• Commons:Deletion requests/File:Coldeportes.svg
• Commons:Deletion requests/File:Colciencias.svg
• Commons:Deletion requests/File:Ministerio de Vivienda de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Tecnologías de la Información y las Comunicaciones de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Transporte de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Cultura de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Ambiente de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Educación de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Minas de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Trabajo de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Agricultura de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Defensa de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Justicia de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Hacienda de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Relaciones Exteriores de Colombia.svg
• Commons:Deletion requests/File:Vicepresidencia de Colombia.svg
• Commons:Deletion requests/File:Ministerio de Salud de Colombia.svg
• Commons:Deletion requests/File:Prosperidad Social.svg

I hope I included all the requests here. Any idea about how to proceed in these cases? - Sahaquiel - Hast du eine Frage? 17:12, 14 April 2026 (UTC)

Offhand, these look to me like reasonable DRs. If these are not official symbols, then for each we should have at least one real-world citation indicating that something at least very like that symbol was actually used. Otherwise, these are out-of-scope fictions. - Jmabel ! talk 20:02, 14 April 2026 (UTC)

And it seems improbable that all of these government ministers would have official insignia which differed only in color, sometimes very slightly - e.g. File:Ministerio de Hacienda de Colombia.svg, File:Ministerio de Trabajo de Colombia.svg, and File:Ministerio de Defensa de Colombia.svg are all very similar shades of blue. Omphalographer (talk) 21:16, 14 April 2026 (UTC)

hola @Jmabel y @Sahaquiel9102. Le dejo los link de los Manuales de identidad visual de los diferentes ministerios y departamentos administrativos del gobierno de Colombia, reglamentado por la Ley 2345 de 2023:

• Gobierno de Colombia: https://www.minagricultura.gov.co/SIG/DocumentosNormograma/MANUAL%20DE%20IDENTIDAD%20VISUAL%20-%20GOBIERNO%20DE%20COLOMBIA%201.pdf#search=manual%20de%20identidad%20visual
• Función Pública: https://www.funcionpublica.gov.co/nuestro-manual-de-identidad-visual
• Ministerio de Ambiente y Desarrollo Sostenible: https://www.minambiente.gov.co/wp-content/uploads/2024/07/MANUAL_IDENTIDAD_VISUAL_AMBIENTE.pdf
• Ministerio de Minas y Energía: https://www.minenergia.gov.co/documents/12012/Manual_de_Identidad_Visual-Minenergia2024.pdf

Henblo (talk) 21:19, 14 April 2026 (UTC)

Probably the most relevant of these is the first link - page 11 ("Usos incorrectos") specifically forbids the use of circular seals resembling the ones in these DRs. Omphalographer (talk) 21:25, 14 April 2026 (UTC)

el primer link recoge todos los del gobierno de Colombia, para claridad. Pero cada uno saca de los ministerios, departamentos administrativos y consejerías presidencial crea un documento independiente para que no se mezcle con otras dependencias gubernamentales.

La Ley 2345 de 2023 también recoge para otras entidades como empresas comerciales del estado, las entidades territoriales (Departamento, Municipios, Distritos), entre otros. Henblo (talk) 21:32, 14 April 2026 (UTC)

Lo que importa (para VP) es que los DRs son razonables. Podrían tener éxito o fracasar, pero plantean dudos razonables. Quizás estuviera mejor combibarlos todos en un DR masivo, pere es dificil combinarlos despues de las discusiones han comenzado. Si alguien hay comentarios que se pueden aplicar a todos, es posible escribir con detalles en un lugar, y vincular de los otros páginas. No creo que hay una ventaja en discutir aquí en el VP. La situación de haber tantos discusiones separadas no me parece ideal, pero no me parece tampoco un gran problema. No se como hubiera docenas de discusiones. - Jmabel ! talk 22:50, 14 April 2026 (UTC)

There is no reason, according to current Commons guidelines, to make these nominations. In any case, it could be clarified in each file that these are independent productions and are no longer in use, and current logos could be prioritized as valid. Our current tools allow for this. ProtoplasmaKid (talk) 15:00, 15 April 2026 (UTC)

@ProtoplasmaKid: that presumes these seals were ever actually used, which last I looked had not been demonstrated. Again, these are perfectly reasonable DRs, which may or may not result in deletions.

Gracias a por las opiniones. Entiendo entonces que la DR es procedente aunque los archivos sigan los alineamientos de Commons. Por lo tanto request deberá decidir el resultado. Mi posición es que las imágenes están formadas por componentes libres y como mucho podrían necesitan un cambio de nombre. -02:45, 16 April 2026 (UTC) — Preceding unsigned comment added by Sahaquiel9102 (talk • contribs)

No es solamente un asunto de derechos de autor. Si son efectivament ficciones, no serían en el alcance de Commons. - Jmabel ! talk 05:51, 16 April 2026 (UTC)

Category for the act of surrendering / piling up weapons

• 
• 
• 
• 
• 

cat for this? RoyZuo (talk) 11:42, 7 April 2026 (UTC)

• Category:Disarming (events)
• Category:Groups of weapons
• Category:Stacked rifles

Nakonana (talk) 11:53, 7 April 2026 (UTC)

disarming could happen to single persons and outside wars (like police disarming someone holding a knife). i was thinking how best to name this kind of massive disarming / laying down arms.--RoyZuo (talk) 21:52, 7 April 2026 (UTC)

Previously you asked for the category, not how to name a category for this. Prototyperspective (talk) 12:38, 8 April 2026 (UTC)

Going by the dewiki article that is linked to Category:Disarming (events), the category is rather meant to be about (military) mass disarmament, and is seemingly supposed to be different from Category:Disarmament. Nakonana (talk) 15:25, 8 April 2026 (UTC)

Disarmament refers to a country reducing its military, rather than disarming actual persons. RoyZuo (talk) 09:45, 12 April 2026 (UTC)

i'm gonna move Category:Disarming (events) to "surrendering weapons". hopefully it's clearer that this refers to the physical act of all soldiers giving up their weapons rather than Category:Disarmament the policy of reducing military. RoyZuo (talk) 14:59, 17 April 2026 (UTC)

An intersection category of Category:Disarming (events) and Category:Groups of weapons doesn't yet exist but could be created. Questions like this one (of type is there a category for XYZ or what's the category closest fitting for zw) could be asked a proposed editor assistant tool which would be especially useful for newcomers.--Prototyperspective (talk) 12:17, 7 April 2026 (UTC)}}

Undelete some of my files

I had uploaded some of my screen recordings for tools, and I thought I had given them a proper license, but apparently not because they were deleted. Can someone undelete them please, I don't have admin rights here.

• File:Relator2 create new item.webm
• File:Relator2 add relation manually.webm
• File:Relator2 set manual relation by clicking on another entry.webm

They will all be {{Cc-by-sa-4.0}}. Thanks! --Magnus Manske (talk) 10:19, 24 April 2026 (UTC)

 Done. --Túrelio (talk) 10:52, 24 April 2026 (UTC)

Please see Commons:Undeletion requests (one can also request undeletion of multiple files at once; maybe the template/page should better consider that). --Prototyperspective (talk) 13:44, 24 April 2026 (UTC)

This section is resolved and can be archived. If you disagree, replace this template with your comment. --Prototyperspective (talk) 13:44, 24 April 2026 (UTC)

Way to find categories with number of files over certain threshold?

Apart from manual patrol, is there a reasonable way to find categories with number of files over certain threshold? I sometimes like to unwind performing different patrol actions in Wikipedia. For example I would like to deep search category Churches in Poland to find all categories that have over 200 files, so that I would clean up main categories of specific churches. I thought PetScan tool might be able to do this, but so far no success. — Preceding unsigned comment added by Tupungato (talk • contribs) 15:02, 2 April 2026 (UTC)

@Tupungato: there is Category:Categories requiring diffusion (200-item threshold), but it requires a category to be manually tagged first with the CatDiffuse or Diffuseat template. --HyperGaruda (talk) 06:50, 6 April 2026 (UTC)

I know about this category (I even have a link on my user page, because I visit it regularly), I also patrolled categories to add this template. I'm looking for ways to up my game in patrolling categories for diffusion.--Tupungato (talk) 08:23, 7 April 2026 (UTC)

Special:WantedCategories Special:MostLinkedCategories. RoyZuo (talk) 17:36, 9 April 2026 (UTC)

@Tupungato You can use Quarry to find these types of things, an example for Churches in Poland would be https://quarry.wmcloud.org/query/104408 . I know the syntax is a bit confusing, but you should be able to make it work for any category you chose by just altering the first line with a different category name. Bawolff (talk) 21:49, 15 April 2026 (UTC)

I'm really confused: why does somthing like Category:Wymysorys pronunciation (Józef Gara's version) show up in those results? (Yes, it's a big category, but what does it have to do with Category:Churches in Poland? - Jmabel ! talk 05:44, 16 April 2026 (UTC)

You can see the deep path going from category to category in the right column. The category depth is set to 12. Setting it to something like 5 helps mitigate the situation.Tupungato (talk) 08:37, 16 April 2026 (UTC)

...Interior of the Poznań Cathedral → Category:Buried in Cathedral of Saints Peter and Paul in Poznań → Boleslaus I of Poland → Wendish Crusade...

as Commons:Village_pump/Archive/2025/12#c-RoyZuo-20251217184100-Category:Burials discussed before, all these relations -- which are not very relevant for most of the files contained, but represented as categories -- lead to these wild cat trees. RoyZuo (talk) 09:15, 16 April 2026 (UTC)

Preventing that sort of semantic drift was one of the motivations for SDC i think. Its one of the things that make the category tree really hard to work with as its super hard to get all the subcats of essentially the same thing without subcats that are totally different creeping in. Bawolff (talk) 15:35, 16 April 2026 (UTC)

When it comes to deepcat views, one could filter off such offtopic cats via this (incl premade filters for a compiled list of cat-types that contain offtopic files such as Burials in… subcats). And more broadly, categories that are not really about the subject such as Burials in… subcats containing cats about people could be converted to other solutions with two ideas being: 1. list/gallery pages (could even be maintained by ListeriaBot) and 2. {{Seealsocat}} which often is better to use in place of categorizing. Prototyperspective (talk) 16:00, 16 April 2026 (UTC)

You can also order the results by depth of category, if that is helpful https://quarry.wmcloud.org/query/104437 Bawolff (talk) 19:39, 16 April 2026 (UTC)

Boleslaus I of Poland → Wendish Crusade Pretty weird to put the crusade under a person, no? We wouldn't put WWII under the category for Hitler.

Preventing that sort of semantic drift was one of the motivations for SDC talk about dropping an atom bomb on an ant! And missing, since SDC still doesn't describe the relation between any two Commons categories, and is way to computationally inefficient to be an at all likely substitute for Commons categories. It would have been infinitely easier to add a third field in category links after the category and sort key, describing the relation, with a set of potential values ("geographic narrowing", "instance of", "subclass of", "chronological narrowing", "notable because of", "gender narrowing", "by geographic areas", "by chronological periods", etc.); that also could be done in SDC, of course, but hasn't been, and would be less computationally efficient. - Jmabel ! talk 23:40, 16 April 2026 (UTC)

I would agree that SDC has largely been a swing and a miss. I think the original idea was that you would have depicts statements that were Q numbers and then you could use wikidata to ascertain the relationship between the different depicts values. I suppose that technically exists now, but actually doing that type of query is hard, and SDC is largely unpopulated so its moot anyways. As a product, i think SDC spent way too much time just blindly bolting wikidata on to commons and not enough time asking how do we actually make this meet user requirements/what do the users actually need. A lot of that could probably be papered over with a better UI imo, but i digress. Bawolff (talk) 00:19, 17 April 2026 (UTC)

I could go on for hours about what they got wrong with SDC, but two things head the list, and they both fall under a focus on theory rather than practice: (1) little or no concern about how data was going to be put into the database by humans and (2) a tendency to compare what the wikibase could do in a "frictionless environment" (no costs of computation, experts doing the modeling and somehow having everyone fall in line, plus a little bit of just plain waving away issues like the difficulty of modeling dates well in a rigid system) vs. what wikitext was achieving in the real world. That plus some arrogance about their ability to make a hierarchy of instances and subclasses that would be ontologically solid, whereas in fact they just ended up generating another folksonomy with only a slight edge on the once that Commons had developed with far less planning. Not that I think Wikidata + wikibase + SDC is useless: I love the resulting infoboxes, and it's done a very good job of managing interwiki connections and getting our various projects to line up their categorization better, and it's pretty good for describing things like artwork, books, etc. But not so good for describing photos, which is the bulk of what we do with metadata here. - Jmabel ! talk 04:47, 17 April 2026 (UTC)

While i think its far from the only issue, for me, the biggest issue is that SDC is hidden behind that pannel you have to click to get to. How can we possibly expect users to add SDC when they don't even see it? How could they possibly figure out the schema when we dont even have the most basic querying built in to see what other media with the same value is like? With categories if you dont know what should be in cat X you can just click on it to find out. Why would the user bother maintaining metadata when the only way to effectively query is a half maintained blazegraph instance off site that basically nobody even knows exists and you have to be an expert to use? Categories can actually be used to find other files by the average user. Bawolff (talk) 13:28, 17 April 2026 (UTC)

Currently SD is visible on mobile Web and in the two Wikipedia apps, but not categories. Both should be visible. A way to easily see files with a given depicts may be useful...but not for setting the depicts because depicts is described on the SD help page and the subject defined on the hyperlinked Wikidata item (with a short desscription directly displayed). Also worth noting is that lots of files have wrong SD so seeing which files have it would be more useful for correcting SD rather than seeing which files should get the SD. Currently, SD is used to uprank and surface files in the MediaSearch search results so that would be the main motivation. Prototyperspective (talk) 13:37, 17 April 2026 (UTC)

Agree with what you said basically and a further use of specifiers that are set quite like sortkeys currently are are timestamps/ranges for videos that are substantially showing/about a subject but the overall video is much longer as proposed here. And I doubt UI improvements would suffice to make SD better fit for that purpose and solve its issues which doesn't mean SD isn't useful. Prototyperspective (talk) 12:44, 17 April 2026 (UTC)

Getting back on topic, "It would have been infinitely easier to add a third field in category links after the category and sort key, describing the relation". I suppose nothing is stopping commons from doing the lazy version of this ourselves and having a convention in the sortkey that something like :: is used to mean relation. So you could have [[Category:Poland|{{PAGENAME}}::Geographic narrowing]] for subcategories with that relation (probably in practise with the details hidden behind a template). Very much a "We have SemanticMediaWiki at home" sort of solution, but it would at least record the relation in a way that is at least accessible to quarry. Bawolff (talk) 13:46, 17 April 2026 (UTC)

i have very little knowledge of graph and algorithms.

one problem i think the "category tree" system has. cat tree is a directed tree, but in reality concepts are often interconnected (so not a tree) and it's difficult to say which direction the relation should be. example: a singer set up a company that organises a music festival that includes the singer.

instead of a tree system, what if categories are like "tags"? and the tags are connected in a graph? (so we can have Cycle_(graph_theory).) then the "deepcategory" search could be instead searching x degrees from a starting vertex. i dont know if this will be too computationally difficult. RoyZuo (talk) 14:25, 17 April 2026 (UTC)

Having a small number of cycles isn't as big a deal as it appears at first glance, because usually you are starting at some point and you keep track of what nodes you have seen before and skip anything you have already seen. Bawolff (talk) 15:27, 17 April 2026 (UTC)

Btw, if you think that example is weird, consider Dogs → Dogs in religion → Dog deities → Dog gods → Ares → Offspring of Ares → Nike (mythology) → Nike (mythology) in art → Nike by medium → Nike in heraldry → World War II Victory Medal (United States) → World War II Victory Medal recipients → John F. Kennedy → Documents related to John F. Kennedy → Cuban Missile Crisis. Bawolff (talk) 04:26, 18 April 2026 (UTC)

There are 2 probable flaws or issues in that chain:

• Nike in heraldry → World War II Victory Medal (United States) → World War II Victory Medal recipients: files in World War II Victory Medal (United States) all display Nike so that doesn't warrant a subcategory for files actually showing the medal; however it could be inappropriate to categorize the recipients under the award since it categorizes the people under Category:Awards established in 1945 and they are neither awards nor were established in 1945 – instead as with the other case, it should probably be a dynamic bot-maintained list/table
• Dog gods → Ares: looking at Ares there is no claim of a dog god or dog in general until In Renaissance and Neoclassical works of art, Ares's symbols are a spear and helmet, his animal is a dog, and his bird is the vulture. so the Dog god should be set on a subcategory about Ares as a dog god [in Renaissance and Neoclassical works of art]

Since such miscategorizations are currently hard to find and correct I can't stress enough the potential impact and need for W397: In Commons category deepcategory view mode (wall of images), allow easily filtering offtopic subcats as well as better tools to detect miscategorizations such as seeing the cat chain you gave under a file showing John F. Kennedy in a deepcat view of Category:Dogs, including especially W393: A way to see why a file is somewhere underneath a specific category (tool to show cat-path). Since cat paths were not visible for such a long time, it will take quite some time until the rate of miscategorization goes down. Prototyperspective (talk) 12:30, 18 April 2026 (UTC)

In general though, it seems like commons likes also using categories to mean "Person (or entity) has property of parent category". I don't know that i think that is even wrong per se, but it is one (among many different) common sources for the tree to become quite wild. Even if you accept those as incorrect, whether Ares is a Dog deity, the real problem seems to be not that, but going from actual dogs, to deities that have a dog aspect, then going from a deity that has a dog aspect to their offspring which presumably doesn't, going from a medal with a specific deity on it to people who have received it, and then going to a category of pictures about a certain person to documents related to them and from there to a world event they are associated with. All of these steps seem like significant meaning shifts. Even if Ares was all dog, the rest would still be problematic. As an example, Artemis has always been associated with dogs (In as much that she has dogs not that she is a dog), and we get the same crazy chain with stuff relating to her like Dogs → Dogs in religion → Dog deities → Dog goddesses → Artemis → Deeds of Artemis → Metamorphoses in Greek mythology → Metamorphoses (Ovid) → Events in the Metamorphoses by book → Ovid Metamorphoses Book 4 → Andromeda → Things named after Andromeda → Andromeda (constellation) → Andromeda Galaxy → Amateur photos of the Andromeda Galaxy.

I think sometimes there is also ambiguity over how much categories are just a see also sort of link. e.g. in Felis silvestris catus → Domestic cats → Domestic cats in art → Cats in art by medium → Cats in decorative and applied arts → Cat costumes → Catsuits → Catsuits by colour → Plug suits by color → Red plug suits → Asuka Langley Soryu → Cosplay of Asuka Langley Soryu → Cosplay of Asuka Langley Soryu by clothing → Cosplay of Asuka Langley Soryu wearing Japanese school uniforms. The most obvious wrong part is Catsuits under Cat Costumes. But maybe the idea is see also Catsuits since the term was inspired by people thinking they looked cat-like. (Going from Red plug suits → Asuka Langley Soryu is also pretty bad, but i also sort of get the logic as that anime character is super famous for red plug suit, so its kind of like, see also the most protypical example). Bawolff (talk) 15:30, 18 April 2026 (UTC)

Well again the two approaches to go about this would improve the situation. Maybe you're saying this also needs addressing at root cause where one changes how categories are thought of and used. If you have concrete ideas regarding that in terms of how to improve things I'd be interested to hear them. However, it needs to be said that often in cases like the one you described there are subcategories of the starting category that are more narrowly about the subject such as about biological dogs – so if one is interested in that specifically (depictions of actual biological dogs) one can use these categories and the way they are surfaced is via sortkeys. Simply go Category:Dogs→Category:Dog types or Category:Views of dogs and the immediate problem may well be solved, especially if you know how to exclude subcats via deepcategory when using the wall-of-images view. An issue there is that lots of files can be missing.

And as described in W397, there could be premade filters one can enable with a click in a category for subcats of common types of relations that introduce offtopics files, here 'Domestic cats in art' (and maybe one could also have the filter find the Dogs in religion → Dog deities subcat to filter out) Prototyperspective (talk) 15:57, 18 April 2026 (UTC)

My suggestion above was to append to the sortkey the relation "type" so that inappropriate relations could be filtered out. If i understand right, your idea is to use category naming conventions to filter out inappropriate category relations from deepcat. Both are kind of the same idea just different methods of figuring out what to filter. The sortkey idea has the con that it would require commons adopting it and updating everything which sounds like a huge amount of work. The naming convention thing has the downside of being less flexible. In principle, both are possible, using naming conventions generally, but using the sortkey hack as an override for things that don't fit the naming conventions. However the more i look at actual category tree, the less sure i am that category links can be so cleanly classified, so now i am less sure. In any case, if you had a list of all the category naming conventions you think should be excluded from deepcat, i'd be interested in hearing as we could at least do some experiments to see how well it works in practise. Bawolff (talk) 07:16, 19 April 2026 (UTC)

I think it's a good idea and could be combined with the setting of the timestamp(s) when categories are set on videos where the category applies to only a part. One could also fill a lot of that data using scripts based on things like the category naming – so for instance xyz in art categories which I suggested to be in a default-available quick-filter could get that relation type specified this way. However, I think quite early on or from the start it would need a more intuitive user-friendly interface instead of setting [[Category:Poland|{{PAGENAME}}::Geographic narrowing]] – this could come in the form of a change to HotCat, eg in the form of a input field for the relation below the category (a tiny + button beneath the cat input field that expands an input field with autocompleting relation types to select from). Additionally, it would need some specification of how that's set on the category. Sortkeys are specified to be [[Category:cat name|sortkey]] and maybe it could be [[Category:cat name|sortkey//relationtype]] for example.

If one is able to see which categories are the top sources of files in a cat with an x button to filter any of these out (if likely introducing many offtopic files) then this would be very dynamic (a downside can be that this quires some user interaction and maybe some minor effort instead of showing the desired results right away to users unfamiliar with the site). As shown in the mockup image, the naming conventions are just used for selectable filters so one can filter out common sources of offtopic files quickly easily – so none are excluded by default but one can easily select one.
I guess one could also have a reverse-filter (whitelisting-like) where instead of filtering things out, one just includes cats with certain relation types for example – so when viewing Category:Dogs one can easily go to a view where it just shows depictions of dogs. However, I'm not sure that's desirable or needed – e.g. I think the ideal results would just all be on-topic and I'd find information graphics like charts and diagrams about parts/aspects of dogs more interesting than just plain photos or at least to make the overall results more diverse and interesting and likely to contain relevant/useful files. I don't have a list of naming conventions of categories commonly introducing offtopic files but the wish names "xyz in fiction", "xyz in art", and Category:Named-after categories. The more cat-paths one can see, the more such offtopic-files-introducing-cats can be listed and until now it was very cumbersome if possible at all to see the cat path and there was little activity on the wish so I haven't compiled a comprehensive list. Another commons source are cats of Category:Letter combinations by letter included and per this discussion it seems like Awards, Burials and, Births categories may also be commonly introducing such. Prototyperspective (talk) 11:34, 19 April 2026 (UTC)

How to categorize 49,000 media needing categories as of 2021?

Are you experienced in categorization? If so, do you want to help, please, to categorize 49,000 media needing categories as of 2021? Do you have a good idea, how this could be made more effectively or even automatically, for instance by sending a message to uploaders of uncategorized files? Some hints are shown on Commons:WikiProject Minimum One Category, but now we look out for more volunteers. --NearEMPTiness (talk) 23:52, 15 April 2026 (UTC)

An idea other than better aid with categorization in the UploadWizard would be to develop some tool that suggests categories for an individual image as well as especially sets of images.

For this it could read the file title, file description, exif metadata, as well as the file contents. It needs to work well and not make the edits itself but instead just display the suggestion to the user who can then select the cat to add or not and unselect images to which it should be added. (A simple example would be to identify portrait photos of people to add Category:Unidentified people but more users checking that Unidentified cat to at least identify the files that are in use would be good.) Prototyperspective (talk) 12:29, 20 April 2026 (UTC)

FWIW, this is now at 43,336 files. Making some progress, but not enough. At this pace, it would take the rest of the year 2026.

I can imagine some sort of bot assist, but we would have to deal with the fact that there is a definite history of users being much too willing to ratify but suggestions without verifying. They get the dopamine hit at a liability to the site.

I would suggest that if a bot does this, rather than work at all directly with adding categories it does something like what we do with certain tools that bring in Flickr tags, and uses "other fields" to make a list that would just be more data for any user who tries to do the categorization. - Jmabel ! talk 20:40, 20 April 2026 (UTC)

Sure, that's good point – the same applies to things like AWB and cat-a-lot though. Things work out because they are revertable and one could make these tools only available to a subset of users or only on request. It wouldn't necessarily involve a bot but that seems like the likeliest way this could work well: a gadget on the user end would probably do too many requests for a cat of tens of thousands of files or just work on small sets of files which would be less beneficial; a bot wouldn't have to add visible flickr tags either – it could also add its suggested categories in the form of comments. That would work as well if one can search invisible html comments with the insource: search operator (haven't tested it). An alternative would be to add it via some template that is collapsed. However, that template or the comments should probably be removed upon categorization just like {{Uncategorized}} is. An idea for that would be to add the suggested categories to that template, e.g. {{Uncategorized|suggested=Unidentified lakes;;People hiking}}. Prototyperspective (talk) 22:08, 20 April 2026 (UTC)

Model release template

Just yesterday I created {{Model release}} as I couldn't find anything like that already existing. (see File:Two police officers on police bicycles in Noord-Brabant, Netherlands.jpg for an example use)
Today while browsing maintenance categories I stumbled upon {{Consent}} which I don't remember ever seeing before, but it's used on 22551 files. It serves a similar purpose, but is quite different. Now I'm unsure what to do. Should they somehow be merged, co-exist, or something else entirely? - Alexis Jazz ^{ping plz} 04:40, 20 April 2026 (UTC)

@Alexis Jazz: With all due respect, {{Consent}} seems more useful, and you might want to think of how your considerations might be added to the possible set of conditions there, or as an additional parameter. - Jmabel ! talk 21:02, 20 April 2026 (UTC)

Jmabel, some of the options of {{Consent}} seem not entirely clear. The difference between "basic" and "full" is the appending of "under a free license". But I'm not sure how that would affect something or why it would be relevant. If the original image is CC-BY and the consent type is basic, does that mean derivatives must use an unfree license? If the consent type is "full", does that mean derivatives must use a free license, is it like a ShareAlike clause? The "hosted" type doesn't actually include what the terms are, so it's susceptible to link rot. The "published" type is probably not really useful? The "public" type, isn't that the default and already covered by {{Personality}}? The "query" type should maybe just be a DR instead. I could be misunderstanding. - Alexis Jazz ^{ping plz} 21:38, 20 April 2026 (UTC)

The {{Consent}} template is mentioned just a few sentences after discussion of model releases at Commons:Photographs of identifiable people, so I'm not sure how you missed it. Do you have any suggestions for how to make it more prominent or discoverable? "Basic" and "full" are different levels of subject consent, but they don't impact the licensing. The licensing is still whatever it says it is. I guess the distinction has more ethical implications than legal. Yes, "public" is basically the default, but it still needs to be an option as it's useful information to know that consent was definitely not given as this is important in some jurisdictions. That's a good point that "hosted" could perhaps be divided into several options (or a separate parameter). Nosferattus (talk) 23:28, 20 April 2026 (UTC)

Category:Logos of political parties in Peru

Hi

Did you know if all logos from Category:Logos of political parties in Peru are under the threshold of originality? Panam2014 (talk) 11:04, 17 April 2026 (UTC)

Of the ones directly in that category, I'd say about a 20-35% could arguably have a problem, though Peru has a high TOO, so even those might be fine. File:Logo de Libertad Popular.jpg, for example, looks to me like a stretch. File:Logo of the Agricultural People's Front of Peru.png looks awfully complex and creative to justify saying it is below TOO. I suspect that with their high TOO, File:BlancoAR.png and File:Frente Independiente Moralizador.jpg are fine. File:Escudo del partido naciona democrata.jpg is probably fine on simply being a minor variant on something quite old (even if not necessarily as the logo of that particular party):

• 
  a stretch for TOO
• 
  a stretch for TOO
• 
  probably fine with high TOO
• 
  probably fine with high TOO
• 
  probably fine with high TOO and historical precedents for the design

On several of the files there, the claims of "own work" and that the creator/uploader is somehow in a position to offer a CC license do not increase my confidence. - Jmabel ! talk 21:10, 17 April 2026 (UTC)

Then again it always seems to be treated as no big deal so i can see why people do it Trade (talk) 13:36, 20 April 2026 (UTC)

In at least one respect it is no big deal: since the SVG is probably copyrightable, the license is appropriate for that aspect. But they seem to have forgotten that the visual aspect also matters. - Jmabel ! talk 20:47, 20 April 2026 (UTC)

@Jmabel and Trade: should PPNPL be verified? Panam2014 (talk) 11:06, 21 April 2026 (UTC)

Do we really care about verifying an account with 4 edits, all to one PD file? I certainly don't. - Jmabel ! talk 16:39, 21 April 2026 (UTC)

Wiki Loves Bangla 2026 has started, Join Now!

Hello,

We are excited to announce that Wiki Loves Bangla 2026 has started! This year’s theme focuses on Bengal festivals, inviting participants to capture and share images and videos of the diverse cultural celebrations across Bengal.

Wiki Loves Bangla is an international photography contest on Wikimedia Commons aimed at documenting Bengali culture and heritage worldwide. It is organised annually as part of the Bangla Culture and Heritage Collation Program, with a dedicated theme each year.

How You Can Participate, it's easy and simple, and every upload contributes to the world's largest free knowledge repository:

Winning image from Wiki Loves Bangla 2025. Attribution: Ashraf747 / CC BY-SA 4.0

• Capture: Take photos or videos of Bengal festivals.
• Upload: Share your files to Wikimedia Commons between 14 April and 15 May 2026.
• Win: A total of USD 1,100 in prizes.

Ready to get started? Click here to upload your media, or visit the main project page for full details.

Your contributions help document and preserve Bengal’s rich cultural heritage for the world.

For any questions, email us or join our Telegram group.

Warm regards,
Wiki Loves Bangla Team.

#WikiLovesBangla

Moheen ^{(keep talking)} 20:56, 21 April 2026 (UTC)

Mathematics → Data visualization images

Jochen Burghardt removed File:Chart of number of works in the field of Mathematics in the OpenAlex database by year.svg from Category:Mathematics with explanation per COM:OVERCAT via Statistics of global scientific publications / Literature statistics / Statistics by field / Statistics / Probability and statistics / Mathematics by topic (see Category:Probability and statistics and Category:Statistics).

• Mathematics defines Mathematics as "a field of study that discovers and organizes methods, theories, and theorems that are developed and proved either in response to the needs of empirical sciences or the needs of mathematics itself. There are many areas of mathematics, including number theory (the study of integers and their properties), algebra (the study of operations and the structures they form), geometry (the study of shapes and spaces that contain them), analysis (the study of approximating continuous changes), and set theory (presently used as a foundation for all mathematics)."
• Statistics defines Statistics as "the discipline that concerns the collection, organization, analysis, interpretation, and presentation of data."

The change to me raised the question whether that category path should be changed and if so how. (And if not it probably needs extensive categorization work of files.) I think random data graphics like charts showing one count number over several years (also numbers) in the form of a bar chart (one bar per year) don't belong into the Mathematics category. The category I think is bloated and there needs to be some work such as differentiation between things involving numbers and the field of Mathematics. Please help develop a good solution.

• Note that many files in the Probability and statistics cat and its Statistics subcat indeed are about the subfield in terms of Mathematics such as about mathematical statistical methods. I thought maybe removing the link between Category:Probability and statistics→Category:Statistics would be a good solution but again lots of subcats and files there do relate to the subfield and if it was removed, the cat would be uncategorized.
• Also note that they have some very broad subcategories that probably don't belong there in the current broad format such as Probability and statistics → Category:Randomness
• Application of and content of COM:OVERCAT as it relates to cases like this or similar to it (eg long chains where subcat has deviated far from original cat) may also be good to discuss at some point or here
• There is an open CfD about whether Category:English-language statistical charts should be merged into Category:English-language charts (regarding that note that which category is chosen for charts so far seems rather random)
• Note that cat Mathematics is in Category:Academia→Category:Academic disciplines

Any input on what is best done? Prototyperspective (talk) 16:42, 15 April 2026 (UTC)

If there is a 7-removes connection to Mathematics because it is a statistical study, but it is about Mathematics, then it is not OVERCAT to also place it directly in Category:Mathematics (though possibly Category:History of mathematics would be more appropriate). To give some examples each at only one remove, but also acceptable in terms of OVERCAT:

• If the category for a person is a subcategory of the category for their more famous spouse, and we have a picture of them together, we would still put both people's categories on that picture.
• If we have a picture that is mainly a picture of a street, but it prominently features a building that has its own category that is a subcategory of the category for the street, it is acceptable to use both categories for the picture.

Jmabel ! talk 20:20, 15 April 2026 (UTC)

Makes sense. The thread isn't much about the categorization of the file though but about the categorization structure of Mathematics→…→Data visualization images. Subcategorizing the file into a subcat like Mathematics publications for example would have been appropriate imo. Regarding your second point, that's how I thought about it when categorizing but now the file is just the context of what got me aware of this and an example for the issue(s) discussed here. It's not a simple problem; sooner or later the categorization path and/or how the files in it are categorized needs to be overhauled. One could start a CfD at Probability and statistics maybe (and tag some related cats with the link to the CfD) but it's a problem difficult enough that it needs more contributors to work out a solution(s) probably and also it's so broad that even this already-existing CfD is kind of tied to it. Maybe it would be good to amend the OVERCAT policy and how long cat chains etc are dealt with in regards to OVERCAT could be discussed here too (eg to have sth more concrete to summarize or draft at the Commons:Categories talk page). Prototyperspective (talk) 21:58, 15 April 2026 (UTC)

It's not a matter of length of the chain. It's a matter of whether the relation of the ancestor category to the child category (or file) corresponds to the relation expressed by the hierarchy. Category inheritance can represent all sorts of thing: geographic narrowing, chronological narrowing, something being named after something else, etc. So in this case: sure, statistics inherits from mathematics, but statistics about mathematics also inherits from mathematics in an entirely different way. - Jmabel ! talk 06:02, 16 April 2026 (UTC)

That's a helpful clarification, making this part/aspect of the discussion much more ready to be discussed at Commons:Categories talk page.

Relating to this, Mathematics→Mathematics by topic→Probability and statistics→Statistics→(all sorts of data graphics that don't really have to do with math as that term is commonly understood or narrowly defined in the quoted definition above) has some flaw in it that makes it not a narrowing with the files relating to the high-level cat. Prototyperspective (talk) 13:05, 16 April 2026 (UTC)

As an aside, I would be in favor of treating data graphics (charts, diagrams, infographics, etc) as a special case and not categorizing them under "generic" categories like Statistics. These types of graphics are common enough, and relate to enough different topics, that it's not helpful to categorize them in relation to their identity as "data graphics". It'd be like categorizing all photographs under Photography. Omphalographer (talk) 18:14, 16 April 2026 (UTC)

Not sure if I understand you correctly. First of all, charts, diagrams and infographics are very different things and of these three usually only charts show data with some exceptions where infographics include charts. The files are not all directly in the Statistics cat. Of relevance to the thread here are only data graphics. Prototyperspective (talk) 21:07, 16 April 2026 (UTC)

The issue is that e.g. Category:Line charts → Category:Frequency (statistics) → Category:Statistics, so all line charts in this category are transitively categorized as "statistics" regardless of their actual topic. Similar issues apply to many other types of charts. Omphalographer (talk) 23:04, 16 April 2026 (UTC)

All of which is related to a comment I just made on another thread here. - Jmabel ! talk 23:44, 16 April 2026 (UTC)

Well hence this thread; it's what it's about. Examples:

• Removing the subcategories like Category:Line charts from that page and instead listing these on the page as hyperlinks, not subcats, with the remaining files contained in the cat being about Frequency in statistics as a subject (Frequency (statistics)) or
• Creating a subcategory for Frequency in statistics that's about files with that as a subject, not instances of line charts, pie charts, etc or
• Removing Category:Statistics from Category:Probability and statistics and creating (and populating!) a new subcat of it that is for files about Probability and statistics as a subject
• (any more good/better ideas?)

Prototyperspective (talk) 12:56, 17 April 2026 (UTC)

The most problematic link in the chain, IMO, is the first one - Category:Line charts can be used to visualize many different relationships; they are not inherently related to the statistical concept of frequency. And in general what I'm trying to suggest is that categories for general types of visualizations like Category:Line charts, Category:Sankey diagrams, etc, should be kept isolated to category trees which describe the form of an image, like Category:Information graphics; they shouldn't be treated as a topical category. Omphalographer (talk) 20:31, 21 April 2026 (UTC)

Good point; one could disentangle the two branches based on whether it's topical or about the form of the image. I'll remove Frequency (statistics) from Line charts but I don't think this solves the thread topic. Prototyperspective (talk) 17:51, 22 April 2026 (UTC)

Android phones strip coords before upload, since April 2026

posting here to raise awareness of this new problem, in case you also usually upload photos through android phone mobile browsers and want to retain the coords (gps / locations / geotags) in exif/metadata. to be more precise, any kind of file transfer thru any app that doesnt have the specific permission required for coords will strip the coords, so it affects not just "upload". see Commons:Village_pump/Technical#c-RoyZuo-20260420223600-Phone_stripping_coords_/_geotags_since_April_2026! for more info.

you can easily check your commons uploads by checking the presence of {{Location}} or scrolling down to the exif at the bottom. if you dont see the coords, then they are most likely stripped. you can also use online tools such as https://jimpl.com/ to check the metadata. RoyZuo (talk) 11:37, 21 April 2026 (UTC)

@RoyZuo: may I presume that transferring files to a PC and uploading from there should work fine? - Jmabel ! talk 16:41, 21 April 2026 (UTC)

yes "copying via usb cables" seems to be the most straightforward method.

"transferring files to a PC" should not go through a phone app though, which may result in coords stripped if that app doesnt have the relevant permission. RoyZuo (talk) 16:54, 21 April 2026 (UTC)

so can't have "coords" without "cords"...? - Jmabel ! talk 17:29, 21 April 2026 (UTC)

It also works fine if one uploads using the Commons app. The app settings permit to choose whether one wants to include or exclude coordinates while uploading. Nakonana (talk) 19:48, 22 April 2026 (UTC)

Category for wheel guides

Is there a category for the "wieldwingers" in Dutch or "wheel guides" in English (if that is the correct translation) as shown in the image. Wouter (talk) 14:16, 26 April 2026 (UTC)

I do not know but I see Category:Wheel stops, may be something around it. Ymblanter (talk) 17:37, 26 April 2026 (UTC)

Category:Roadside barriers of sorts? Nakonana (talk) 18:18, 26 April 2026 (UTC)

Category:Guardrails, very short ones? MKFI (talk) 06:37, 27 April 2026 (UTC)

Thank you. I think 'guardrails' is the most comparable existing category name. Wouter (talk) 08:25, 27 April 2026 (UTC)

This section is resolved and can be archived. If you disagree, replace this template with your comment. Prototyperspective (talk) 22:10, 28 April 2026 (UTC)