This page is used for discussions of the operations, technical issues, and policies of Wikimedia Commons. Recent sections with no replies for 7 days and sections tagged with {{Section resolved|1=--~~~~}} may be archived; for old discussions, see the archives; the latest archive is Commons:Village pump/Archive/2025/03.
Please note:
If you want to ask why unfree/non-commercial material is not allowed at Wikimedia Commons or if you want to suggest that allowing it would be a good thing, please do not comment here. It is probably pointless. One of Wikimedia Commons’ core principles is: "Only free content is allowed." This is a basic rule of the place, as inherent as the NPOV requirement on all Wikipedias.
Any answers you receive here are not legal advice and the responder cannot be held liable for them. If you have legal questions, we can try to help but our answers cannot replace those of a qualified professional (i.e. a lawyer).
Your question will be answered here; please check back regularly. Please do not leave your email address or other contact information, as this page is widely visible across the internet and you are liable to receive spam.
Purposes which do not meet the scope of this page:
SpBotarchives all sections tagged with {{Section resolved|1=~~~~}} after 1 day and sections whose most recent comment is older than 7 days.
March 13
March 2025 update from WMF Legal on "Vogue Taiwan and possible Copyright Washing" discussion
Hi, I’m LRGoncalves-WMF, from Wikimedia Foundation’s legal department, and I just wanted to provide an update to the Vogue Taiwan situation discussed here: /wiki/Commons:Village_pump/Copyright/Archive/2025/02#Vogue_Taiwan_and_possible_copyright_washing. We reached out to Condé Nast to give them a heads-up about the CC license in their Vogue Taiwan videos and specifically asked them if the content posted on their YouTube Channel is in fact CC-licensed. A couple of days ago they replied confirming that all videos on their Vogue Taiwan youtube channel were not available for reuse. In their words: “All copyrights are owned by the Condé Nast global network. The CC license was applied due to an unknown error. We have immediately fixed it and updated all videos and settings on the Vogue YouTube channel back to the "Standard YouTube License.”
Based on their answer, the Legal Department can’t confirm that the stills of Vogue Taiwan videos uploaded to Wikimedia Commons are openly licensed. As Condé Nast’s counsel and some commentators above pointed out, the attribution of the CC-license was made in error, and not a deliberate choice to freely license these videos. LRGoncalves-WMF (talk) 12:09, 13 March 2025 (GMT-3)
Given that they had mistakenly included CC license in their videos for many years, it isn’t far fetch for them to also missed removing the license for some videos. Maybe we can include a note somewhere for uploaders in the future, so they don’t mistakenly upload the copyrighted images here? Tvpuppy (talk) 16:56, 13 March 2025 (UTC)[reply]
IMO the opinion of WMF Legal here is quite equivalent of a DMCA by the copyright holder. It means to me that WMF Legal would accept such a request if ever they would send one. Unless we have a similar legal opinion about other cases, I don't see any reason to change our decision. However I very much like to know the answer to PHShanghai's question below. Yann (talk) 21:28, 13 March 2025 (UTC)[reply]
As would I. Are Creative Commons licenses non-revocable or are they revocable when the licensor is a giant corporation with a team of expensive lawyers? I agree with Yann that WMF Legal seems to be saying they'd agree to a DMCA and thus I believe Yann's actions to be correct in terms of protecting our site and protecting our reusers (although we should also alert our reusers to this situation with Conde Nast). Abzeronow (talk) 19:10, 15 March 2025 (UTC)[reply]
I second this. Not protecting the copyright on Night of the Living Dead was also a mistake (and a far easier one to make than positively choosing a CC-BY license!) and we don't just all agree "Whoopsies: we'll just take this out of the public domain for you". It's disappointing to see us taking free use media down that is clearly valuable for our mission. —Justin (koavf)❤T☮C☺M☯22:52, 13 March 2025 (UTC)[reply]
@Koavf: IMO there is a big difference between Night of the Living Dead or similar cases, and Vogue Taiwan. There should have been a copyright notice for the film, as it was the distributor's duty to add one. While as Vogue Taiwan is not the copyright holder of these videos, the license was never valid. These licenses were not more valid than the ones added by license washing people we often see on Commons. I would not accept Condé Nast argument (We made a mistake.) if the free licenses were added by them. The whole point is to determine who is the copyright holder. Yann (talk) 19:53, 15 March 2025 (UTC)[reply]
That's a valid concern and I'm not suggesting that you did the wrong thing as such. I respect your decision-making, in case that was not clear. —Justin (koavf)❤T☮C☺M☯05:52, 16 March 2025 (UTC)[reply]
I don't really buy this "mistake". Unfortunate mistake, true, but these were under a CC license and if we are now deleting these it means CC licenses are revokable, which would set a dangerous precedent. On the basis of good faith I would support prohibiting uploading new Vogue Taiwan files from now on (even if CC licensed on YouTube), but those already uploaded on Commons or still CC licensed on YouTube as of 13 March should unmistakably be kept. Bedivere (talk) 22:26, 13 March 2025 (UTC)[reply]
The question to me is whether Vogue Taiwan actually had the legal authority to release the videos under a CC license. If the copyrights are owned by some other part of Condé Nast, and there was never internal authorization for Vogue Taiwan to release the files under a CC license, then the license wouldn't even be valid in the first place. To me, it's equivalent to the situation where a PR employee of a company uploads the company's logo to Commons without their company's legal department authorizing them to do so. Pi.1415926535 (talk) 23:22, 13 March 2025 (UTC)[reply]
There's always the possibility that someone went rogue on his last day of work and slapped "we license everything CC-BY" on a bunch of media, but if the holding company that owns them is just so big or mismanaged that the left hand has no idea what the right hand is doing, that's not something anyone else should have to adjudicate. —Justin (koavf)❤T☮C☺M☯23:34, 13 March 2025 (UTC)[reply]
@Koavf It shouldn't have anything to do with anyone going rogue. If Condé Nast owns the copyright and does not share ownership rights with VT, then VT putting a CC license on it may be invalid, period. VT could even have a good faith belief that they fully own the video--doesn't matter. If that belief isn't actually true, then any licenses they issue are likely invalid. I do not believe that this is a case of joint-ownership, but look at this family of cases for an idea of how a court would treat a situation like this, unless VT clearly had sole ownership rights over the video. You said above "It's disappointing to see us taking free use media down", but I don't think this was ever free use media. Alyo(chat·edits)21:52, 14 March 2025 (UTC)[reply]
What are those assumptions? I see no one in this thread putting forth legally reasonable assumptions that result in a world in which VT had the authority to publish those videos under a CC license. Is is theoretically possible? Sure. But it makes no sense legally, it makes no sense given the IP policies on all CN sites, it makes no sense from a business structure standpoint, and we have a direct statement from the copyright owner saying otherwise. And for what, screenshots from youtube videos? I don't think this is a fight worth fighting. Alyo(chat·edits)23:16, 14 March 2025 (UTC)[reply]
But these are big companies, not a single individual, who left videos with a CC license for years. It's technically possible Condé Nast doesn't even own footage from their photographers (just example), but we assume big companies know what they are doing when they create, publish and license content. REAL💬⬆05:40, 14 March 2025 (UTC)[reply]
Agree with Bedivere that this is setting a dangerous precedent. I believe the WMF legal department's communication with Condé Nast has helped clarify the situation. We can be confident that, moving forward, even if their videos are mistakenly CC-licensed again, these will be considered errors and should not be used. However, CC licenses are supposedly non-revocable, meaning that videos licensed as such within that timeframe should remain free to use, regardless of the reasons Condé Nast published them this way. 👑PRINCE of EREBOR📜09:38, 14 March 2025 (UTC)[reply]
@Bedivere and Prince of Erebor: It doesn't matter if the CC licenses are revokable or not. The files should be deleted so that we aren't endangering our reusers with lawsuits (meritless or not). This is the same reason we shouldn't be hosting images by copyleft trolls. "Welcome to Commons! The repository of technically free-license images you'll probably get sued for trying to use!" Is it any wonder that people prefer paying $40 for public domain images on Alamy rather than getting them here? Nosferattus (talk) 01:11, 15 March 2025 (UTC)[reply]
Thank you for the update. For the record; just want to clarify how the non-revocable part of Creative Commons licenses work in this case? What is the official statement of WMF Legal regarding that? Thanks. @LRGoncalves-WMF: PHShanghai (talk) 20:43, 13 March 2025 (UTC)[reply]
I will reply to @Yann's DMCA comment down here. Looking at the original post WMF Legal doesn't say to delete all photos, just that an error has been made. Rewarding companies that have hypothetical DMCA capabilities and disadvantaging organisations (and regular people) that don't is weird to me. Commander Keane (talk) 22:59, 13 March 2025 (UTC)[reply]
I'd +1 to the PHShanghai's question in a personal capacity, although I do wonder if the Foundation's Legal folks would even be able to offer an opiniongiven that this isn't an attorney/client situation. Associated with that question, I'd opine that Yann's administrative decision to delete all of these images without additional discussion of this new viewpoint was made too hastily. As I've done previously, I'd encourage Yann to be much more careful when taking unilateral administrative actions—especially in a case like this, where deletion means that images have been removed from probably dozens of Wikimedia projects and cannot easily be restored. Ed[talk][OMT]01:44, 14 March 2025 (UTC)[reply]
Dozens if not hundreds of articles are now imageless because the Vogue Taiwan images were hastily deleted, with A list articles such as Adele and Billie Eilish. May I remind you that this decision to delete hundreds of images was done without any consensus of the community and was just a broad action applied. I also don't like how Yann is going about this, they templated me on my talk page for uploading some of those Vogue Taiwan images despite knowing WHY those pages were deleted. Quite rude. PHShanghai (talk) 04:57, 14 March 2025 (UTC)[reply]
Yes, I still agree with the argument Yann made previously; what court or judge will accept "Sorry, but the license which was published by our subsidiary company there for years is wrong."? REAL💬⬆05:34, 14 March 2025 (UTC)[reply]
If the license is actually invalid/VT didn't have the authority to publish it? All of them. Good faith belief/reliance isn't a defense to copyright infringement. Copyright infringement is a strict liability tort. You either did it or you didn't. Alyo (talk) 21:18, 14 March 2025 (UTC)[reply]
Come on, these videos were up with a free license, by a company with apparent authority to do so, for years during which anyone could find them on YouTube and use without knowing about whatever internal situation Condé Nast had. No judge is going to tolerate a claim of copyright infringement against users for using the media under the terms of the license. REAL💬⬆15:49, 16 March 2025 (UTC)[reply]
@999real, I have limited experience in copyright law, but do not agree with this assessment of the situation. More importantly, even if I did agree with your assessment, I have no idea how that supports keeping the files, rather than just being thankful that users wouldn't be on the hook for damages accrued before the WMF/CN statement.
@999real: I think most judges would have no problem following Restatement (Second) of Agency to find there is no apparent authority. Agents have a scope of employment, and VT's scope does not include licensing the content of other CN entities. If you had a reasonable and honest belief that VT could grant a free license (e.g., you did not know other CN websites did not use free licenses) a few months ago, you might escape liability. Given the current discussion about VT and CN's actions, your and Common's apparent authority defense has evaporated. Glrx (talk) 01:50, 18 March 2025 (UTC)[reply]
Comment Opinions seem versatile. Previously, a number of people argue for deletion of these files. Now that we have an opinion from WMF Legal department, people want to keep them... Yann (talk) 09:24, 14 March 2025 (UTC)[reply]
There was obviously a mixed consensus on the actual previous discussion. It was not a broad consensus to delete the files unless we heard from CN themselves. PHShanghai (talk) 12:58, 14 March 2025 (UTC)[reply]
I agree with Yann. And I say this as someone who uploaded 46 images for probably as many articles that were deleted due to this, and that makes me very sad. Yes, technically we could argue that the license was there on the files for many years, that we had every right to take the license as good, that many different articles are visibly the worse for losing these images, all that. But we're here to do a good deed, to make the single largest source of knowledge in human history, and not to be copyright trolls. If there is a reasonable chance that a good faith error was made - and that is what Conde Nast is asserting to WMF - then I can understand us forgiving the error, and letting the images go. So it goes. --GRuban (talk) 19:00, 15 March 2025 (UTC)[reply]
Don't forget about downstream reuse, either. Displaying these images on Commons with a CC license is an assertion that "yes, you can use these images freely given these conditions" - if we have reason to believe that the images may not actually be CC licensed, and that reusing them may actually pose risks, we shouldn't offer them. Omphalographer (talk) 19:17, 15 March 2025 (UTC)[reply]
Info I restored this from the archive. Comment I think this is the case of "license laundering" on Vigue Taiwan's part. As per Alyo's comment below, "licenses are only "irrevocable" if they are properly granted. In this case, those of us arguing for deletion believe no valid license was actually granted. If we are correct, then this situation never reached the question of irrevocability." If this is the case, Vogue Taiwan committed COM:License laundering, because they had no authority to grant license in the first place. That authority comes from their parent firm or their "superiors": Condé Nast itself. I endorse deletion of all Vogue Taiwan files. JWilz12345(Talk|Contributions)04:21, 2 April 2025 (UTC)[reply]
DMCA suggestion
@LRGoncalves-WMF: hello. Can you also inquire Condé Nast if the revocation of CC license applies to Vogue Taiwan content hosted here before March 13, 2025? If so, can Wikimedia suggest them to file a single take down notice vs. all of the said files through COM:DMCA? Without a DMCA notice, there is uncertainty whether those files should be deleted too or not. JWilz12345(Talk|Contributions)16:38, 16 March 2025 (UTC)[reply]
I believe WMF Legal messaged it this way so we would understand that they would agree to a DMCA request if it was ever issued, and they would rather have us deal with it so a DMCA notice would be unnecessary. Abzeronow (talk) 19:58, 16 March 2025 (UTC)[reply]
I think us and everyone who's been through this whole debacle deserve a clear cut explicit answer from WMF Legal instead of doing guesswork. I will be very disappointed if we do not get clarification because all of us have been busting ourb butts off here trying to get a proper consensus on this issue. PHShanghai (talk) 23:43, 16 March 2025 (UTC)[reply]
As I understand it, WMF's lawyers cannot provide legal advice as such to the Commons community. We are not their client. What they can do, and have done, is to communicate a legal position on behalf of WMF. - Jmabel ! talk01:00, 17 March 2025 (UTC)[reply]
We are not asking for *legal advice*, we are asking for a comment regarding how Creative Commons Licenses worked in this context. We can't actually do anything anymore but I feel the community is at least owned a little more in depth explanation, *especially* regarding irrevocability of Creative Commons licenses and how it would work in this context. PHShanghai (talk) 04:41, 20 March 2025 (UTC)[reply]
@PHShanghai The application of legal principles/rules to a specific fact pattern is legal advice. CN is going to say "the CC license was a mistake, we retain all rights to the videos and any derivative stills"--repeating their previous statement. What you are asking for is for the WMF's opinion on whether or not that argument would hold up in court in a situation where we keep the files and CN issues a DMCA, which is legal advice. Again, licenses are only "irrevocable" if they are properly granted. In this case, those of us arguing for deletion believe no valid license was actually granted. If we are correct, then this situation never reached the question of irrevocability.
Hi, I've made several dark mode fixes to VFC, if you guys could try adding importScript('User:Matrix/vfc.js') to your common.js, then changing to dark mode, which is on the sidebar if you are using Vector 2022, that would be very useful. Please report any bugs here or on my user talk page. —Matrix(!)ping onewhen replying {user - talk? - uselesscontributions} 20:16, 18 March 2025 (UTC)[reply]
I did the first two, though the problem with the third one is that Template:Rfcua is substituted so I would have to manually fix it for all checkuser cases before 2025. Whilst this is possible, I don't really think it's a good use of my time. —Matrix(!)ping onewhen replying {user - talk? - uselesscontributions} 11:46, 29 March 2025 (UTC)[reply]
User Allforous just moved these there. So the other category is now a better example. Note that in the former cat there is this video File:Harm Aversion video explanation.webm which is not study-specific short clip. In other categories there are more videos like it, here there seems to be one exception. Below are some examples.
IMO, many of the videos imported by the Open Access bot - especially the ones used as experimental stimuli in neuroscience papers - are out of scope for Commons, as they have essentially no meaning outside the context of the papers they accompanied. I appreciate the intent of that import job, but it was perhaps overly broad. Omphalographer (talk) 03:51, 21 March 2025 (UTC)[reply]
Agree. There are over 10,000 of these clips and the problems are:
that they flood categories where it's then hard to see the actually useful videos / media and
(another example cat is Category:Videos of science which before I subcategorized a bit had 10 k files directly within it and even a category only containing subcategories is better than it being flooded with these clips)
that they are a time sink for contributors who categorize these or deal with files that drown under these clips and
that They can also substantially degrade the quality of search results, making this site significantly less useful
I also appreciate the intent and agree that they "have essentially no meaning outside the context of the papers they accompanied". I think something should be done:
Either deleting all of the clips from studies uploaded by the bot except for the very few clips that are in use (and any clip with more than e.g. 50 views per month if one can query for that) or
removing all their categories except for an Open Access Media Importer Bot-specific one like Category:Videos from studies uploaded with Open Access Media Importer which is removed from cats not specific to the bot since it screws up deepcategory search results (this doesn't address the search results but one could maybe add a deboost template) or
moving all of these to some separate Wikimedia project / related site like https://mdwiki.org/ or
something quite similar to any of the above to the same effect
I think option 1 would likely be the best (easiest to implement, most effective) option. If this isn't the right place to discuss this (and I think it is), maybe somebody or I should make a request for comment. Prototyperspective (talk) 13:55, 21 March 2025 (UTC)[reply]
I don't think file deletion discussions for singular files are the right approach – there's over 10 k of these, and that's the reason they're a problem to begin with as they make so many categories unusable.
People spend so much time deleting single files. Yet here's 10 k of which just a few are useful outside the context of a study whose media is not useful. How can a video of a one second click be useful for example? There's so many of these and they make many categories unusable and drown the high-quality useful content in them such as the one on the right (moved the files by now). Moving files into subcategories also doesn't do anything about the search results. Prototyperspective (talk) 12:23, 31 March 2025 (UTC)[reply]
March 22
Category for technology / software in public administration?
How do we handle imaginary flags that have been uploaded as real?
Tfjbhugdw has uploaded a number of flags that purport to being flags for localities, mainly in Hawaii. However,
A cursory Google search for flags for these localities turned up no matching images, except for the flags for Bristol, Rhode Island & Burrillville, Rhode Island.
In Hawaii, localities are unincorporated as incorporation is only at the county level. Therefore, I know of no means by which these localities could choose such flags.
Tfjbhugdw has cited no sources. Each images source is listed as Own work.
None of the Hawaiian locality flags have any Hawaiian themes.
I do not know if there is a contingency for deleting fictional images that masquerade as something real. They certainly must be at least renamed, as I believe they give a false impression that they are actual flags for these localities. Peaceray (talk) 05:37, 24 March 2025 (UTC)[reply]
Just want to remark: the issue here should be flags that are truly imaginary, not flags that are unofficial. For example, the Cascadia flag is unofficial but widely used, and certainly belongs on Commons, while File:Potentional flag of Cascadia.png appears to be made up by a user, and I have no idea why we are hosting it.
I could not find the flags that Tfjbhugdw posted for Hawaii through a Google search. For instance, Tfjbhugdw uploaded File:Flag of Waimānalo, Hawaii.svg, & doing a Google source on (waimanalo OR Waimānalo) flag simply fails to retrieve the image from anywhere else. Its not even that these files are unofficial. Nobody, aside from Tfjbhugdw, seems to use these images.
Tfjbhugdw added the images for the fictional flags to Wikipedia articles. I have deleted these images in enwiki except for the flags for Bristol, Rhode Island & Burrillville, Rhode Island, which do seem to be versions of real flags. Peaceray (talk) 17:16, 24 March 2025 (UTC)[reply]
I continually find people having added low-quality structured data, mainly depicts (P180), to content I've uploaded. It's not exactly wrong, it's just beside the point. For example, File:Solvej Schou, Michelle Threadgould, & Lucretia Tye Jasmine.jpg is a photo of three reasonably notable writer-musicians, but it is described as depicting blond hair (Q202466), name (Q82799) (presumably because of name tags), microphone (Q46384) (two microphones are visible in the picture, but not in a way that would make this useful as a picture of a microphone), and laptop (Q3962) (twice; there is a barely visible laptop near the lower right corner of the photo).
Is this actually considered desirable, and I'm just out of step? If not, would it be appropriate for me to remove this cruft? - Jmabel ! talk16:44, 25 March 2025 (UTC)[reply]
After a quick glance in the history: these additions look like a perverted usage of Commons:ISA Tool. It may be kind of a game here, with a high score going towards those who put in the most SDC data, even when the sense of using a particular claim is marginal at best. Remove it, I'say, it does not advance our projects. Regards, Grand-Duc (talk) 17:14, 25 March 2025 (UTC)[reply]
Lots of just plain wrong structured data too: e.g. some ignorant user uploads a public domain photograph from 1875, listing themselves as the author, and the "date" as upload date (May 5, 2021, ha!), and faithfully stupid bots dutifully transcribe such incorrect metadata into the structured data. And unfortunately it takes several more steps to correct or remove false metadata than to create it. --Animalparty (talk) 00:48, 26 March 2025 (UTC)[reply]
Again, is it appropriate for me to remove this cruft when it shows up on content I uploaded? Does it have to be outright false (rather than merely useless) for me to remove it (e.g. laptop (Q3962) in the above example)? - Jmabel ! talk01:31, 26 March 2025 (UTC)[reply]
I'd remove it. On Wikidata's end they usually stick to having one best value for something like that. I don't see why Commons shouldn't take the same approach. Otherwise structured data is kind of useless. --Adamant1 (talk) 02:09, 26 March 2025 (UTC)[reply]
I would strongly advocate to remove it, too. Knowingly or negligently using not suitable structured datasets is against Commons' aims, as far as I understand them, as it degrades the value of the repositories, making them somewhat less usable. It's comparable to introducing typographic errors in Wikipedia articles; both are behaviours that should be avoided to the best of one's abilities. Regards, Grand-Duc (talk) 03:11, 26 March 2025 (UTC)[reply]
Jmabel removed the depicts which I agree is best. There is Commons:Depicts and depicts modeling documentation but they aren't really fleshed out and explicit; the project is in documentation and development limbo. I hope things like Commons:ISA Tool are no longer being used like this today.
I was going to say something about that myself. Structured data shouldn't be added by inexperienced users and/or through semi-automated tools like this. --Adamant1 (talk) 06:38, 26 March 2025 (UTC)[reply]
@Donia (WIA) organises campaign377 that involved Jmabel's file. There are gift card prizes associated. I will leave a talk page message and hopefully they can drop by and comment.
@Trade: True. Are you saying that it is useful to add the particular values that were added to this particular photo, or are you just saying that, like categories, something can be worth adding without it being a large portion of the photo? If someone added Category:Microphones or Category:Laptops to this photo, that would also be cruft, no? - Jmabel ! talk00:42, 27 March 2025 (UTC)[reply]
I think a category like "People sitting in front of microphones" would be appropriate if it exists but none of those two. This isn't a very rare incident, lots of files have SD like that and this case just shows how SD is not the panacea it's sometimes thought of by some and has quite some issues even if it was used widely – Trade makes a valid note, those depicts statements aren't even wrong which just shows how many items miss how many depicts statements OR how the ones set aren't necessarily helpful. If one was looking for a representative or good image of a microphone one would be better helped by going to the category. One could make a good case for keeping these depicts statements. I think categories are best for file contents and SD is mostly useful for keeping things out of the categories that aren't really helpful there. With that I'd think of metadata like captured-with:NikonD5000 or "Videos without audio" which could eventually be used by filters in the search, but one could also argue SD could be used to describe each and every thing in files. Prototyperspective (talk) 01:04, 27 March 2025 (UTC)[reply]
I use "depicts postcard" for scans of postcards. It's essentially worthless for anything outside of that though. I certainly wouldn't add something like "depicts ink" or "depicts cardboard" to an image of a postcard. There's a difference between something being "wrong" and being totally pointless. Like adding "depicts dirt" to every photograph taken outside. Your seriously missing the point of structured data if your saying people should do that because it's technically correct. Not Prototyperspective of course, I mean it more generally. --Adamant1 (talk) 07:22, 27 March 2025 (UTC)[reply]
@Jmabel: Yes please remove it. You get a light warning message if you upload something via the Upload Wizard and add more than three items to the structured data, which signals to me that it's really not supposed to be inclusive of background details, but meant to focus on the subject matter of the image itself. ReneeWrites (talk) 09:18, 27 March 2025 (UTC)[reply]
To me, the big question is: why on earth is there a contest encouraging something that we nearly all seem to agree is counterproductive? - Jmabel ! talk16:29, 27 March 2025 (UTC)[reply]
@Jmabel: The top contributor to it User:Blessingedi76 was asked on their talk page to not use the depicts statement that way 10 days ago and it seems like they didn't get the message going by the edits to File:May's medical report.png from today. Perhaps you could give them a more stern warning now that it's clear no one thinks what their doing is a good idea. As I suspect it would deal with a lot of this if they got the message. --Adamant1 (talk) 21:17, 27 March 2025 (UTC)[reply]
"Name the different elements that you can see in this image". That could be changed to:
"Name the most important element(s) that you can see in this image".
@Eugene233 is active in ISA development, maybe they can investigate this change and if it is desirable.
The drop down selector has woman/human/sitting/writing/woman writing. I assume the contest organiser sets those. For future contests I would put more specific examples such as Âmiran Kurtkan Bilgiseven (Q29572828).
women reading (Q30672195) is an artistic theme and I don't know enough about Wikidata to comment on the appropriateness of that one.
I'm not personally all that interested in warning each individual user who follows the terribly conceived instructions of an apparently official contest. If someone else wants to do that, fine. What I'm interested in is stopping having contests that basically invite vandalism. - Jmabel ! talk14:46, 29 March 2025 (UTC)[reply]
Just as a practical idea/solution: There are multiple sources for tagging data which don't match on the wanted abstraction level. For example, most archives and museum databases have a different abstraction level for their keywords than what is wanted in Wikimedia Commons for depicts (P180) values. Also, machine vision generally will create values which are at different abstraction level, and crowdsourcing will do the same. I think that a proper solution would be to have a different property than depicts (P180) for imported, generated, or tool-assisted crowdsourced values. New property should require information about source and method as qualifiers. As the abstraction levels are source specific if values aren't good to be usable alone, but would be usable when combined with other data (for example, if we want to confirm things to be correct or not using multiple methods or we need filters in SPARQL queries). --Zache (talk) 18:56, 29 March 2025 (UTC)[reply]
This whole SD depicts thing just requires ever more efforts and complexity when what's depicted can mostly be read from the categories. Also, machine vision generally will create values which are at different abstraction level That depends on its configuration and its technical design quality. Machine intelligence could for example read the categories and the with machine vision look at the image to see which things are depicted prominently. If SD was to be actually used (and widely set), then I think in a way that's similar to that which also cut or reduce the vandalism / polluted data (as arguably in the example) and maintenance needs. Prototyperspective (talk) 13:07, 30 March 2025 (UTC)[reply]
Hello @Commander Keane and others.
First of all, forgive me for my late reply but these days we have Eid al-fitr so things are getting too busy. As you mentioned, I am the creator of ISA 377 campaign, moreover, I am the facilitator of the tool if anyone face difficulty or challenges, or if we have an issue like what you provide here in this discussion. Actually, many useful campaigns were organised using ISA in several years, and in each campaign, we share in our social media posts and on commons pagethis link of guidelines and a published video on youtube to explain how to edit successfully in ISA campaigns. As I started organising ISA campaign last year, this is the first time to face this issue but I'd appreciate your suggestions. I read one suggestion not to have it as a contest, we may try not to offer prizes the next few campaigns to make sure that participants in the the upcoming campaigns not chase quantity but quality. I may also add the link of the guidelines inside the campaign page on ISA. I also read a suggestion in one discussion (not sure here or another place) suggest editing the sentence on ISA main page to make it explain that the one should depict the most prominant items. I'd appreciate all your suggestions as our aim is the same which is to improve images of wikimedia commons. For the current campaign, I see you already blocked some of the users, and the campaign has ended so no need to furtherly block them, I may discuss to prepare an online meeting and invite all the participants to explain the negative points and how they can participate successfully in the future. Best. Donia (WIA) (talk) 12:31, 2 April 2025 (UTC)[reply]
March 26
VOA likely to be shut down, help copying information
With the very likely shutdown of the Voice of America, I've begun copying what I can to Wikimedia Commons, before they start deleting their various websites. I'd encourage anyone who has a few moments to go to the VOA websites and copy anything that is copyright-free for use here. Especially if you speak another language, there's a trove of good information there that will likely be of interest in the future. I doubt much is getting archived by the current US administration either, making this even more important. Oaktree b (talk) 15:26, 26 March 2025 (UTC)[reply]
I was just going to find people together for this, too.
Does anybody know of any coordinated effort anywhere? reddit?
It seems to me that Archive Team is only backing up the Voice of America videos (https://wiki.archiveteam.org/index.php/Voice_of_America) leaving out the photos of their photojournalists. I found at least 2 places where they publish them:
https://www.flickr.com/photos/voiceofamerica/ - photos from Voice of America but without a free license, is it possible to backup from here under a free license since they are US government employees?
https://www.insidevoa.com/ - it looks like there are only pictures of Voice of America. It would probably be the ideal place to back up the files to Commons.
I doubt the man at left was this cross-eyed, and no human beard ever looked the way his is depicted. I could go on, but I think that's enough. - Jmabel ! talk16:31, 27 March 2025 (UTC)[reply]
That's quite horrible. Any reason not to list it for deletion discussion? If it's loosely based off some original real free licensed photo, the original might be in scope, but this AI nightmare is not. -- Infrogmation of New Orleans (talk) 21:14, 27 March 2025 (UTC)[reply]
If you're putting it up for a DR, then you'll get lots of comments like "Keep, COM:INUSE"... That said, I also have the opinion that AI upscaling mostly never does any good. Regards, Grand-Duc (talk) 23:41, 27 March 2025 (UTC)[reply]
Does anyone have the original image? If it's freely licensed, it could be uploaded to Commons and used instead. If it's not freely licensed, it's a derivative of a non-free image and should be deleted. I see no reason to keep this. ReneeWrites (talk) 10:51, 28 March 2025 (UTC)[reply]
Could categorization requests please be added to Commons:Community portal? There users can add or take up categorization tasks. For people looking for things to do on Commons, that seems like a perfect way to get started and it would help get things done. Prototyperspective (talk) 14:51, 27 March 2025 (UTC)[reply]
'Statistics by field' and 'Category:Data graphics'
Category:Statistics by field is about domains where statistics are applied and does not contain only statistical graphics. It contains lots of subcats and files.
Category:Data graphics is meant to only include data graphics. Is missing many or most subcats / files.
Could somebody help populating the latter category using the former category?
Moreover, there are separate Category:Charts and Category:Statistical charts which could make it more difficult and less organized to find things. I think this kind of content is important (e.g. heavily used in Wikipedia and of public interest) and it needs some thought and work to make things clear and well-organized.
usually meant for buildings but construction work has not begun, or been abandoned, or such land as results of demolition of buildings originally occupying it? RoyZuo (talk) 19:29, 27 March 2025 (UTC)[reply]
@Wouterhagens's suggestion building site (Q11448974) currently has the english description as "place where a building is located, formerly was located, is under construction, or will be constructed"
what i'm thinking of is exactly "place where a building formerly was located, or will be constructed"
they are not natural lands (not yet developed for any purposes), agricultural lands (developed for agriculture) or construction sites (construction has begun).
According to this, it's just land that is not being used so that would be much broader than buildings but construction work has not begun, or been abandoned, or such land as results of demolition of buildings originally occupying it. I'd suggest to simply create a new category about the subject and maybe add the two cats I mentioned to it. If there is in fact some cat that you don't know of that's about the same somebody would eventually notice and merge them. Prototyperspective (talk) 12:46, 30 March 2025 (UTC)[reply]
How to avoid addition of categories bij Location info?
The File:Eizer J church.jpg has 3 categories (Eizer, Churches in Eizer and Cemeteries in Belgium) that can only be removed by deleting {{Location|50|47|57.9|N|4|32|54.6|E|scale:6000_heading:SE}}.
How can I remove these categories without removing the location info? Wouter (talk) 09:29, 28 March 2025 (UTC)[reply]
Somehow the categories managed to end up in the image title EXIF metadata. It has been 17 years since you uploaded the file, so I don't think you remember if you changed the EXIF title before uploading, do you? --HyperGaruda (talk) 10:38, 28 March 2025 (UTC)[reply]
Hello @Wouterhagens, as mentioned above, the categories are there because the code for them was included in the EXIF caption (in the row “Image title”). The caption is currently this: Sint-Maria-Magdalena church in Eizer (Overijse), Belgium. {{Location|50|47|57.9|N|4|32|54.6|E|scale:6000_heading:SE}} [[Category:Eizer]] [[Category:Churches in Eizer]] [[Category:Cemeteries in Belgium|Eizer]]. Tvpuppy (talk) 10:45, 28 March 2025 (UTC)[reply]
Thanks for the info. I don't think I changed the EXIF before uploading, but it's been too long ago. I could download the image, remove the EXIF info and re-upload, but since it turns out the image doesn't appear in any of the 3 categories I'm leaving it as is. Wouter (talk) 15:37, 28 March 2025 (UTC)[reply]
Is there a page where user not very experienced with editing templates can request edits to templates in general, including nonprotected ones? Maybe there are users experienced with editing templates that are looking for things to contribute and/or are simply far more skilled/experienced/efficient with editing templates that could implement the requests.
In such cases I'd usually ask at the template talk page or on the talk pages of the users who created the template or have worked on it. If those users are high-profile template contributors, their talk pages might be watched by other users who might chim in to help.
Yes, but usually these aren't watched by many active users. VP/T is about anything technical and may not be watched by those people or not checked because there's too many posts. For example, if somebody experienced with editing template was looking for tasks to do, then there seems to be no page and VP/T may not have any anymore due to them gotten archived. Prototyperspective (talk) 13:37, 29 March 2025 (UTC)[reply]
I wanted to use a Geograph image in en.wiki, by uploading it to Commons (https://www.geograph.org.uk/reuse.php?id=4552154), and searched on "This search should find it if it has been [uploaded]". It found nothing, so I tried to upload it and was told that it already existed, as it does: File:North Bar Within, Beverley, Yorkshire - geograph.org.uk - 4552154.jpg. So it looks as if the search built in to Geograph isn't finding these files, as the string "geograph 44552154" isn't present, although the two words, separated, are in the file name.
Perhaps I should be complaining to Geograph rather than here - or is there a formal channel for communication between the two systems, through which this could be fixed? (Or am I misunderstanding or overlooking something!) PamD (talk) 21:41, 28 March 2025 (UTC)[reply]
I just saw Phabricator task T390302: Native iOS App upload flow to Commons.
It seems to be a way to upload files directly to Commons from the Wikipedia iOS app, a bit like the cross-wiki upload feature on desktop (note current meta RfC to restrict that feature).
I brought up the issue of iOS and the HEIF format on Phabricator.
I imagine the barrier to entry for Android users of downloading a specific Commons app for uploading means we don't deal with a bigger influx of dickpics and selfies.
The task quotes Wishlist and Village Pump requests for a dedicated iOS app. This is not a dedicted app, and it is unclear to me what advantages this flow would have over the regular mobile upload process.
I would really like to have this. I would not use it as uploading from mobile is not part of my workflow, but I think it is very good to get new people. Especially for the Wiki Loves contests people already see the banner in the app if they could then directly upload their photos to Commons could lead to many more participants. If that is combined with the map the app already offers it is a really great way fill photo gaps for locations. GPSLeo (talk) 05:21, 30 March 2025 (UTC)[reply]
Hi @GPSLeo thanks for responding I agree that it would be powerful to connect actions like uploading images to campaigns. This is something we've thought about, so it is helpful to know that it would be a welcomed feature. Just so you know, folks who read Wikipedia using the Apps do not currently see Central Notice banners like those for Wiki Loves contests. Central Notice is not optimized for display in the app, and most of the links that are included in banner lead to Web pages which we've learned is a confusing experience for App users. We have thought about investing in the creation an announcements system for all types of announcements, events and campaigns in the Apps that would lead to a user friendly experience. While we continue to think about an announcement system, what are your thoughts about sharing about events through other methods like the recent Community Updates module, or the Collaboration list?
Something I would definitely oppose is having an upload feature for media files in the Wikipedia app for iOS but not Android.
I think an iOS Commons app next to the Android app would be good but not overly important and only as long as it doesn't take too much resources (note the Android app is so far developed just by volunteers).
Adding an upload-only feature for iOS and Android to the Wikipedia is something I'm unsure about. It's not something I'd consider important and the downsides may outweigh the positives – people may upload lots of copyvios and mundane useless low-quality pics. Generally, smartphone photos are less likely to be useful than other kinds of photos. Nevertheless, it could make it much easier to upload media and bring new Commons contributors and smartphone cameras have gotten much better. Not unlikely it could be quite a good thing but I'd suggest to first implement proposal m:Community Wishlist/Wishes/Bots detecting copyvios on Commons (image reverse search etc) and maybe also some work in regards to (semi-) automatic categorization but this the importance of such in this context may depend on how many files will actually get uploaded that way. There's thousands of copyvios that lingered on Commons for years undetected. Prototyperspective (talk) 13:01, 30 March 2025 (UTC)[reply]
@Prototyperspective I agree there are tradeoffs between making it easier and more discoverable to upload from Mobile for folks who genuinely want to contribute, and the risk of making it too easy for folks to upload low-quality images, or copyright violations that cause extra burden for moderators. I’m aware of the history of quality issues from mobile uploads, and we would not move forward with something like this before coming to an agreement with the Commons community on a threshold for access, and how files coming from this source can be identified. I am imagining the flow could have similar steps and checks to the Upload Wizard, which has been improved recently to reduce copyright violations. Do you think that the user flow for upload wizard mobile version is a good place to start? HNordeen (WMF) (talk) 17:18, 2 April 2025 (UTC)[reply]
It may be best to reuse (or adapt) the code used for the upload wizard. The functionalities it has such as autocompletion of categories are useful and important. Prototyperspective (talk) 19:04, 2 April 2025 (UTC)[reply]
@Prototyperspective I beg to somehow disagree on the aspect of photos taken from phones, as a user who currently takes photos using a Samsung Galaxy A20 (see File:Pulilan Church 20230410-2jwilz.jpg, for example). However, I do agree with the possibility of several people using the app to upload copyvio images. We have been battling copyvios every day (Category:Copyright violations never gets empty for at least 24 hours), and new WEBP uploads are being uploaded at almost every hour. Some may be decent, like faithful PD 2D works or from freely-licensed Indonesian government sites, but the majority are questionable uploads.
Wasn't saying there aren't many good-quality / useful photos uploaded taken by smartphone cameras (I don't think we disagree in principle – I just would implement methods for spotting copyvios first if possible). This photo of yours of the object would be a better example btw. more effective, proactive mechanism to discourage or at least slow down the uploads of questionable files agree and most of these could be spotted via bots, mainly using image reverse search but possibly also using other methods to identify likely problematic files – these bots could also set speedy deletion templates or make it quick and easy to solve DRs. There could be more methods, yes – like not allowing webp uploads from mobile. Prototyperspective (talk) 08:00, 3 April 2025 (UTC)[reply]
@GPSLeo: as far as I can see, in the Wikipedia app you can't visit Commons without opening a browser tab. I'm not sure what happens when you click the Wiki Loves banner to find out about that project. If it makes your browser visit Commons, and you decide to upload a photo why would you need to upload in the Wikipedia app? Or you mean they see the banner and later when using the app they upload. The dynamic between mobile web and mobile app is interesting; you can't browse Commons categories for example to see if a location is already photographed. Having said all that, the Wiki Loves contests generate lots of new interest and anything to aid that is important. Commander Keane (talk) 12:07, 31 March 2025 (UTC)[reply]
Yes if you click on a link to Commons inside the Wikipedia app it opens a browser inside the app and it is possible to upload files using this browser. But a native app is always much smother than a web browser. The most important feature would be to have a map like the Android app has. GPSLeo (talk) 13:33, 31 March 2025 (UTC)[reply]
I would use it for confirming location and direction where photo has been taken. Another use case could be that if we are adding depicts then use map for selecting depictions and for filtering categories. Zache (talk) 09:19, 1 April 2025 (UTC)[reply]
@GPSLeo, @Prototyperspective I'm curious what you think about a similar idea, but that utilizes the existing map on the iOS Wikipedia App: The “Places” map on iOS currently shows a placeholder “W” icon if an article doesn't have an image. I can see a potential flow being: users click on an article without an image on the map, they're prompted to add an existing image from Commons to the article (similar to our current media insertion flow). They could use Commons search, or see suggestions from the image recommendations algorithm. What do you think about that idea? Another possibility is that if there are no image suggestions already on Commons, only then do we surface the upload flow to users (something like this would be possible if we've invested in the upload flow already). HNordeen (WMF) (talk) 17:22, 2 April 2025 (UTC)[reply]
Could be a good idea and may get more people to edit Wikipedia, it may depend on how difficult it would be to implement this: I think it would need to smoothly show the images in the category (but also consider subcategories) so the user can select one of these or if finding no suitable one select upload new photo. Currently, adding photos to articles is something basically too complex than what I'd do on a smartphone because it requires some browsing and searching around but if the app makes it smooth and easy it could be good. However, I don't know how many and why would use the map to do something like that; maybe few use it with the intent of making sure all nearby articles have a photo but I don't know how many that would be. Media suggestions more broadly nevertheless would be good and I think currently missing is considering which media is set on the Wikidata item which usually is a fairly representative or useful one. Also: do they enable some 'add missing images' mode in the map or do they tap on the placeholder image to add images? I think more users at least of the Wikipedia app who tap on an article item would like to use the map for discovering places and getting info about places etc. Prototyperspective (talk) 19:14, 2 April 2025 (UTC)[reply]
I think the more important feature would be to have an upload link for all objects on the map also if the are already linked to a photo, many photos are bad or outdated. Taking photos has to be done outside where people are around with the app. The management for existing photos is something to be done at home at the desktop computer. I also see participation using the app only as an entry step for people to join our project. Later I want them to move towards taking photos with real cameras and editing them in a more professional way. GPSLeo (talk) 06:35, 3 April 2025 (UTC)[reply]
Hi @Commander Keane, thanks for tagging me in the discussion here! I’m a product manager on the Mobile Apps team, working a lot on the Wikipedia iOS App. The task to Prototype Native iOS App upload flow to Commons is a proposed project for the 2025 Hackathon, so it’s just an idea at this point, and we’ll have to see if there are enough folks interested in working on it at the Hackathon. I’m happy for it to continue evolving as we hear feedback in this discussion & on the task.
The reason we are suggesting iOS first is because a few members of the iOS team are going to Hackathon, and because we were recently talking with volunteers about their ideas around an iOS Commons app. It focuses on the upload flow, because my understanding is that the primary pain point for iOS users is the lack of an easy way to upload media from their iOS device to Commons (let me know if you disagree).
I want to stress that if something is developed at the hackathon, it will just be a prototype, and we would not move forward with something like this before coming to an agreement with the Commons community on a threshold for access, and how files coming from this source can be identified.
This would not preclude a separate Commons app for iOS, but rather would be a simple way to give some iOS editors easier access to upload to Commons from mobile devices. It could also make it easier for image uploaders to add their images to Wikipedia articles. HNordeen (WMF) (talk) 17:17, 2 April 2025 (UTC)[reply]
March 30
API failures
None of my uploads are working: I gett error messages like An unknown error occurred in storage backend "local-swift-eqiad" and Could not store upload in the stash (UploadStashFileException): "An unknown error occurred in storage backend "local-swift-eqiad".". Can anyone help? mr.choppers (talk)-en-02:14, 30 March 2025 (UTC)[reply]
Oh good, it's not just me. Was trying to move+replace a file, and got
"Error while moving the page. A detailed description of the error is shown below: API request failed (backend-fail-internal): An unknown error occurred in storage backend "local-swift-eqiad". at Sun, 30 Mar 2025 02:14:39 GMTserved by mw-api-ext.eqiad.main-8674c8fcdf-rwhwf"
Trying move+replace with a different one similar-but-different error, and some images are failing to load until one (or several) refreshes. I think this is an API error and it clears up by itself in awhile, based on past experience with a similar error while trying to upload in the past. - The Bushranger (talk) 02:18, 30 March 2025 (UTC)[reply]
Probably because of this subcat: Category:Uploaded with VicuñaUploader. It is 3 subcat down of Category:Camels, which will be included in the query you linked.
Checking the history, it was added to the category “Things named after camels” 5 years ago by Tuvalkin. VicuñaUploader is indeed named after the animal Vicuña, but I’m not familiar with the animal so I’m not sure if it is technically a camel or not. Maybe someone else more familiar with it can chime in? Tvpuppy (talk) 04:30, 30 March 2025 (UTC)[reply]
Things named after categories are always a problem in our category system. For example you are looking for photos of a person and you get photos of a different person because the photo was taken in a building named after the person you were looking for. GPSLeo (talk) 04:57, 30 March 2025 (UTC)[reply]
Yes, I have already notice that, but usually we get a few false positive results. In this case we get 2 millions, that is the problem. JotaCartas (talk) 05:05, 30 March 2025 (UTC)[reply]
Yes, I think it should be removed for example because of the problem you just described.
Furthermore, this case raises the subject whether we should delete or somewhat isolate the category "Things named after xyz" because they introduce these issues where entirely unrelated files are shown in such scans (including deepcategory walls of images [here just showing 1 camel image among unrelated ones])
File<->Category pathThanks for looking into this and asking about it here. It's part of a major issue of Commons as is. There are many more cases like that and that's partly exactly why I call again and again for a way to see source of categorization for files so one can fix flawed categorizations causing unrelated images showing up in the results. – see the screenshot and A way to see why files are included in the specified cats (source of categorization)
To Tuvalkin and everyone. In fact the animal "Vicuña" is not a "Camel" it only belongs to the "Camelidae" family. But that's not the problem! In my opinion it should not be in any "Named after any animal" category. I'm waiting for my opinion to be supported by some users before making the removal.--JotaCartas (talk) 06:01, 30 March 2025 (UTC)[reply]
The English word "camel" is an acceptable stand-in for what is more accurately termed a "camelid" (zool. fam. Camelidae) — that makes this categorization correct.
Removal of correct categorization is… ungood.
If search quearies yield incorrect results, then correct the searching algorithm.
If otherwise correct search quearies yield unexpected results, then adjust your expectations or your search queary.
Categories are meant to be browsed, not to be fed as raw text to a brute-force search engine.
Category:Uploaded with VicuñaUploader appears to be a hidden maintenance category that isn't even used directly. At least IMO such categories shouldn't be a part of the normal category structure. As users don't generally browse through or add files to them directly anyway. Nobody searching for an image of a camel is looking for one of a Floral design on the wall of a temple in the village of Maluti though and it follows that if something is in a subcategory of a particular topic then it should relate to it somehow. So the category should be removed. I'd say the same for similar categories. They should really be in their own space that's totally separate from the general category system. --Adamant1 (talk) 18:10, 1 April 2025 (UTC)[reply]
Why does no one ever email permissions-commons@wikimedia.org?
Even if the uploader owns the rights to the photos they upload 95 percent of the time they never bother to email permissions-commons@wikimedia.org with a written permission. Instead they just continue to reupload the deleted images over and over again until they eventually gets banned or takes their frustration out on the community that their images keeps getting deleted
@Trade: could you provide links to the instructions that these users are getting?
Last year's Crosswiki Uploads report recommended for the Upload Wizard:
Surface the VRT process in the upload process (this is #9 at the bottom of the file page).
If that has not been implemented and you can think of a way to achieve it, even in a small way, it may help for future iterations of the Upload Wizard. Commander Keane (talk) 07:50, 30 March 2025 (UTC)[reply]
There is no indication to visit the file page to see the shiny "confirm copyright via email" button (release generator). The VRT release generator is nice to use, but could be enhanced. If the first step included "WAIT! If you place a link to an appropriate webpage with license information on the file page you don't need to bother with this messy email process" then you could simplify the talk page instructions and add a blue "Confirm copyright" button. People like wizards, not wading through documentation.
As far as I can see, the Upload Wizard doesn't include a pathway for providing a link to a webpage if you have already selected "own work", nor is there VRT integration. The UW could display your deletion log at the first step saying "Hey, we noticed you have been struggling, have you followed the instructions on your talk page?".
It's just about portraits. In my personal opinion I find it boring and unimportant. Celebrity pictures can simply be released under CCBY by the celebrities if they wished to. They could release a pic they took themselves or pay/ask a photographer for it for example. It's also not the type of content that is missing much. On the other hand, Wikipedia articles about famous people seem to get read a lot (however many entirely unillustrated articles as well). I wonder why there are essentially no (or nearly no) media coverage about broader issues of missing free media, for example in subjects of science to name one broad domain. Also I think at least one of the photos on the left of that article could be considered better than the new one on the right. Prototyperspective (talk) 12:52, 30 March 2025 (UTC)[reply]
In many cases there are photos in articles taken with compact cameras from the time Wikipedia was founded. There are already many much better photos on Commons but there are not enough people checking if there are better photos available as these articles are not on the missing photos lists. GPSLeo (talk) 13:11, 30 March 2025 (UTC)[reply]
There is something ironic about a news site who cant tell Wikipedia and Commons apart complaining how the site is run Trade (talk) 17:52, 30 March 2025 (UTC)[reply]
It was good coverage. No mention of Wikimedia Commons as Trade eluded to, we need to work better at making obvious links from Wikipedia. I was more concerned that the BBC didn't seem to meet file licensing requirements, unless they got permission from the authors? Commander Keane (talk) 23:31, 30 March 2025 (UTC)[reply]
Hi, For some time now, I see many files tagged with {{Permission pending}}, where they won't obviously be any permission, e.g. File:Sean PNG.png. This last case is typical: no license and author is said as "unknown" and other nonsense information, so no one could send a permission. How brand new users got to know that adding this template might delay the deletion?
I was able to replicate this in a test upload. The template is automatically added in the upload process when when "This work was created by someone else and is free to share" and "I have permission to upload this work from my employer or the creator of this work" are selected. Ciell (talk) 08:36, 31 March 2025 (UTC)[reply]
OK, this should not happen. So now, we have obvious copyright violations pending for one month. Bad, very bad... Who can remove this "feature"? Yann (talk) 08:45, 31 March 2025 (UTC)[reply]
This is a good feature requested from us for cases where people have permission the problem is that some people seem simply lying when clicking the button that they have permission. Maybe there should be a large warning like "When you click this button without having the permission (or if the email is not sent in time) you might be blocked". GPSLeo (talk) 08:56, 31 March 2025 (UTC)[reply]
I tested it and when clicking the "I have the permission" option it is not possible to select a license. This is definitely wrong, choosing this option should always require to set the license. GPSLeo (talk) 20:14, 1 April 2025 (UTC)[reply]
I think simply adding the same license list from the own work licensing to the permission option and requiring the selection of a license from the list would be fine. GPSLeo (talk) 18:55, 2 April 2025 (UTC)[reply]
My inclination is the same as GPSLeo's here, but I'd like to see a little bit of data before making a decision: I'm afraid of flooding the field with junk uploads that look more plausible than before.
Do we have any data on how many uploads per month use "I have permission to upload this work from my employer or the creator of this work"? (I have a few other questions, but they are only relevant if that one has an answer.) - Jmabel ! talk20:16, 2 April 2025 (UTC)[reply]
@GPSLeo Maybe I am wrong here, but if the work is already under a CC license, or any other free license, it would just be better to use another option and to clarify which license the media is released under. So maybe we can rework the prompt here, and make it more clear. Would it work? Sannita (WMF) (talk) 13:59, 3 April 2025 (UTC)[reply]
@Sannita (WMF): If the work is already published under a free license, then they won't be under the "I have permission to upload this work from my employer or the creator of this work" case. This is about the case where they are going to have to use VRT. Commons requires that in addition to marking the file as "permission pending," they also must indicate what license is forthcoming. Right now, the latter requires them to edit again after uploading. Experienced users know to do this, but it is very opaque to relatively new users. - Jmabel ! talk17:50, 3 April 2025 (UTC)[reply]
An appeal to all Commons users: please consider applying to become sysops or nominating users. Departure of multiple sysops exacerbates Commons volunteer staff shortage. Your experience and passion are much needed.
Other news
Thanks to User:Don-vip, Video2commonswas fixed so uploads from Youtube are possible again. This bug impeded Video2commons for approximately two months.
Video2commons is in need of maintainers. Please help keep it running!
It still works. Sometimes one needs to retry (and also refresh page if it just keeps loading). Also it seems like YT (and by extension DuckDuckGo Videos) now also shows nonCCBY videos in the results (for one search I tried at least) even filtering by CCBY but maybe that's just a rare bug. Prototyperspective (talk) 12:52, 1 April 2025 (UTC)[reply]
Maybe it wasn’t working earlier in the day, but it is still working for me now. I just tried and successfully uploaded couple YT videos using v2c. Tvpuppy (talk) 13:01, 1 April 2025 (UTC)[reply]
i see indeed you imported more just now, but i retried mine and it still failed with the bug "Sign in to confirm you\xe2\x80\x99re not a bot..." :/ RoyZuo (talk) 13:22, 1 April 2025 (UTC)[reply]
Wikidata and Sister Projects: an online event
Hello everyone, I’m writing to announce an upcoming event called Wikidata and Sister Projects that will be a mini online conference to highlight the different ways Wikidata can be connected and integrated with the other WM projects.
We are currently looking for session ideas and speakers for our program and wanted to reach out in case there were any editors here that might have a cool idea for a session proposal. More info and templates for sessions can be found on the event discussion page.
As previously mentioned, we would like to showcase the relationship between Wikimedia Commons and Wikidata and how data such as Commons Categories, Structured Data and depicts information help Wikimedians navigate through the projects and find relevant information and articles.
The event is scheduled between May 29 - June 1st, 2025. If you have any questions about the event, would like more information or have a session idea to propose, please feel free to get in touch by replying to this post or writing on the event page or on my talk page. Thanks for reading, - Danny Benjafield (WMDE) (talk) 07:23, 1 April 2025 (UTC)[reply]
Default signature
Default signature looks like this
[[User:Example|Example]] ([[User talk:Example|talk]]) Example (talk)
However, also by default, users initially have no user pages.
So, what's the point of default including a red link to the user page, which often is non existent? An effect of this can be seen on Commons talk:Abuse filter, which is full of new users and hence many red links.
Instead of user page, a link to user talk page and a link to Special:Contributions are way more frequently used and useful.
Most people who lack a user page abandon their accounts shortly after creating one, or participate only very sporadically. For those who do choose to engage long-term, having the user page in their signature provides an easy and natural prompt to create one, and in a way it serves as your "face" on this site - neither the contributions nor talk page serve a purpose similar to this. Removing the user page link from signatures would diminish its visibility, and by extension its role in encouraging community participation. You could make an argument to have the contributions page added to the default signature, but I strongly oppose removing the user page from it. ReneeWrites (talk) 18:16, 3 April 2025 (UTC)[reply]
Red links
Any red links on commons should not open that page in the editor automatically. unlike other wiki projects, commons users dont ususally create pages by writing a text but by uploading a file. as Special:Statistics shows, the majority of pages are in file namespace.
@RoyZuo: what you are saying is true of "File" space on Commons, as on any other project. Of course, we use "File" space a lot more than other projects, but is there any other page space on Commons of which this is true? - Jmabel ! talk23:20, 1 April 2025 (UTC)[reply]
Another use case is red link categories. /wiki/Category:Example without opening the editor automatically, you can start using hotcat, but hotcat doesnt work if you click Category:Example and land on the editor.
If they are redirects or text, e.g. Category:Жизнь (text), then we certainly do accept them. I've also seen categories for books being written in native script, but I don't know whether that's acceptable (though it might be for old books that have never been translated). For subjects that might not have an English name (e.g. a church in a small village), I tend to either translate the name (if possible) or romanize it. Nakonana (talk) 15:04, 2 April 2025 (UTC)[reply]
Diff post, 1 April 2025, How crawlers impact the operations of the Wikimedia projects
I think that this could be interesting here too. WMF just posted a Diff post on how after early 2024 Wikimedia has experienced a 50% increase in bandwidth usage primarily from AI scraper bots collecting training data from their image repositories, with at least 65% of resource-intensive traffic coming from bots. This is not from the WMF posts but i think that situation could be helped if there would be least pre-scaled image repository where to download images instead of hammering the main servers.
I've been getting random loading errors here and there for at least the last couple of months. I wonder if the load from AI bots has something to do with it. Hopefully they come up with a way to keep the load from effecting the user experince regardless. Off-site backups and hosting of popular files would certainly help. --Adamant1 (talk) 19:53, 2 April 2025 (UTC)[reply]
you can add the subtemplates directly to categories for use, so you have many customisation options and can choose what works best for the categories you use. RoyZuo (talk) 19:40, 2 April 2025 (UTC)[reply]
Can I split this into two separate archives (one for 2022 and the other for 2023) for consistency of archived pages, despite 2022 only having one listed DMCA takedown notice?
Clarification on OECD's new licensing policy and compatibility with Commons
Hello everyone,
After a brief discussion] on pt.wikisource, I'm bringing this question here for the Commons community to weigh in. I’d like to clarify the compatibility of OECD publications with Commons’ licensing requirements, particularly for documents published before July 2024.
The OECD recently updated its Terms and Conditions as part of a wider policy change toward Open Access, retroactively applying the following permissions for pre-July 2024 content:
Use, copying, and distribution: Free for commercial and non-commercial purposes, provided the original source is cited: "OECD (year), Title, URL."
Adaptations: Allowed (commercial or non-commercial), but without the OECD logo/visual identity. Must include: "Adapted from OECD work. The views expressed do not represent the official OECD or its member countries' views."
Non-commercial translations: Permitted without prior authorization (no OECD visuals). Requires: "In case of discrepancy, the original text prevails."
Commercial translations: "you must request authorisation to translate OECD written content published before 1 July 2024 for commercial purposes. Please send your request to comrights[at]oecd.org with the title of the OECD publication and the name and address of the entity requesting authorisation. If granted, authorisation will be free of charge. No exclusive rights will be granted."
For post-July 2024 content, the OECD now uses CC BY 4.0 as the standard.
Questions:
Do pre-July 2024 documents (e.g., 2007 Open Science Declaration) meet Commons’ licensing standards, given the non-commercial restriction on translations?
For post-July 2024 CC BY 4.0 content, are there any additional requirements for upload? Should there be a specific license tag for OECD files?
These files are officially more than six days long. However, if you play them, you hear a single word or phrase, then the recording stops playing. Can anyone guess what's going on with them? Based on the descriptions (which match the recording) and the file sizes (7 KB in each case), the length is erroneous; I'm not concerned that we've lost more than six days of sound. Note that all three are uploads over originals that were 0.0 seconds long (but each original sounds the same as the one that overwrote it); all six versions should probably be something like a second or at least half a second. Nyttend (talk) 19:24, 3 April 2025 (UTC)[reply]
This one, on the other hand, may be as long as it says. However, the original upload and the current version both begin with a few words and then go to silence. I don't feel like listening to 163 hours of silence; is there some way to review the file to see if it's just silence the rest of the time? And if it is just silence, is there a way to "crop" it (not sure what the right term is with sound) to just the first few seconds? Note that the file history claims that the original is 0.0 seconds long, but obviously it's not. Nyttend (talk) 19:24, 3 April 2025 (UTC)[reply]
This file wasn't actually 163 hours long; it was merely 15 hours and 40 minutes of silence. I've trimmed that off. However, the contents of the file don't even remotely match with the filename - the user reads the words "rondo, sango, sédio, speco", then makes some comments in Portuguese about "How do you pause? Am I pressing the right button?". It would appear that they didn't. Omphalographer (talk) 20:04, 3 April 2025 (UTC)[reply]
I then put it aside when I couldn’t figure how to overwrite, I was only reminded of it recently when it got tagged as duplicate. This version only has a slight quality improvement, so it will be fine if it was deleted, but still I wanted to ask is it possible to overwrite the current version with the slightly better version? Thanks. Tvpuppy (talk) 21:38, 3 April 2025 (UTC)[reply]
using v2c, you cant overwrite; you cant even upload to a filename that was once used but then deleted.
@Tvpuppy right now with your new intended version already uploaded, the usual approach is let a sysop move the new to the old covering the old, or as you mentioned, "tag as duplicate" and the sysop will do just that. RoyZuo (talk) 22:35, 3 April 2025 (UTC)[reply]
Comment Opinions seem versatile. Previously, a number of people argue for deletion of these files. Now that we have an opinion from WMF Legal department, people want to keep them... Yann (talk) 09:24, 14 March 2025 (UTC)
Info I restored this from the archive. 
Done will upload VFC changes soon. —Matrix(!) ping onewhen replying {user - talk? - _Ordinary_December_day_in_Berlin_17.jpg)
.png)
.png&diff=prev&oldid=1014422511){kind=link}
