This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
This section is resolved and can be archived. If you disagree, replace this template with your comment. Prototyperspective (talk) 11:11, 2 October 2024 (UTC)
This section is resolved and can be archived. If you disagree, replace this template with your comment. Prototyperspective (talk) 17:56, 6 October 2024 (UTC)
In September 2024, 1 sysop and 1 bureaucrat were elected; 5 sysops and 1 oversighter were removed. Currently, there are 179 sysops, 7 bureaucrats and 3 oversighters.
The last 15 or more requests on the above page are displayed as being archived, although they are not yet so. Somebody pleae fix it. --トトト (talk) 23:20, 7 October 2024 (UTC)
This section is resolved and can be archived. If you disagree, replace this template with your comment. Prototyperspective (talk) 23:25, 7 October 2024 (UTC)
Cat-a-lot is slow
Very slow, about 1 edit per second. Up until recently and since “forever” it was lightning fast, able to edit all 200 files in a cat page in a couple seconds. To whom or what should we thank for this improvment, and what reasons were given for that thankfulness? -- Tuválkin✉✇00:48, 2 October 2024 (UTC)
I was wondering why it's been slow lately. Totally handicapping it just because the site crashed once seems a little ridiculous though. --Adamant1 (talk) 05:03, 2 October 2024 (UTC)
According to a comment from a WMF worker it brought the site down more than ten times and was causing problems consistently. A comment from a different volunteer pointed out it brought down all Wikis for 10-30 minutes at a time. The current state of Cat-a-lot is the result of an emergency fix, but whatever improved version takes it place will have to abide by the API guidelines that it previously ignored. ReneeWrites (talk) 09:40, 2 October 2024 (UTC)
I stand corrected. I thought I had read that it took the site down once in August. Ten times for that long isn't great. --Adamant1 (talk) 16:19, 2 October 2024 (UTC)
Hi everyone, if we use the template concerned, there are automatically in the Category:Cosplay by event the different years for each event, which is duplicated with the sub-templates. Is it possible to modify the main template to limit the number of categories? Ellicrum{bablute [...]}22:37, 5 October 2024 (UTC)
I've seen a number of images similar to these pop up while attempting to categorize new uploads - they appear to be stylized representations of a name, but I'm not sure what that's called. Is there a standard name and/or category for this type of image? Omphalographer (talk) 20:40, 5 October 2024 (UTC)
I mean the seals in particular. While those categories do contain a number of images of this type, they're general to all Arabic calligraphy, not this specific form. Omphalographer (talk) 02:11, 7 October 2024 (UTC)
Invitation to Participate in Wiki Loves Ramadan Community Engagement Survey
Dear all,
We are excited to announce the upcoming Wiki Loves Ramadan event, a global initiative aimed at celebrating Ramadan by enriching Wikipedia and its sister projects with content related to this significant time of year. As we plan to organize this event globally, your insights and experiences are crucial in shaping the best possible participation experience for the community.
To ensure that Wiki Loves Ramadan is engaging, inclusive, and impactful, we kindly invite you to participate in our community engagement survey. Your feedback will help us understand the needs of the community, set the event's focus, and guide our strategies for organizing this global event.
Probably not. If I had my way there would just be flat lists or metacats of actors and actresses by name. Minus the whole "by birth" thing since it's totally pointless trivia. More so with "by decade of birth." I'd almost argue the same for "by nationality" to BTW. --Adamant1 (talk) 18:17, 2 October 2024 (UTC)
It doesn't make it easier to find someone, it makes it more difficult, unless accompanied by a flat list that contains all the entries. If we made searching in Wikidata easier, we would not need these intersection lists. --RAN (talk) 23:22, 2 October 2024 (UTC)
A commercial databank, wouldn't go to this level. We actually only need the Actors name here, Equity demands that names are unique, as it is.
Other websites do it better. Wikipedia is where people are going to look for this kind of info, not here. I suppose computers search progs will use Wikidata. Wikidata gives us a detailed infox already.
Main cats are flat cats, by definition. In this case the name of the actor. As an aside, I would be in favour of a rule that states we shouldn’t have more than 4 levels to any main cat. In this case the main cat being People. Broichmore (talk) 13:49, 4 October 2024 (UTC)
4 levels seems a little extreme but it's a good suggestion in general. There's some pretty cavernous category structures out there and most of this stuff is better off in infoboxes or otherwise stored on Wikidata's end anyway. With people specifically, most of the time we already know the person's gender from their first name. So it seems kind of pointless to have categories for it. Plus there's a risk of it getting really pointless and obscure depending on the situation. --Adamant1 (talk) 18:48, 4 October 2024 (UTC)
Separate issue, but how do you feel about categories for other biographical metadata like Category:Births by location (e.g. Category:Births in Fairbanks, Alaska) or Category:Deaths by year? These, too, feel like situations where Commons categories are being misused as a sort of Wikidata-lite to describe people, rather than classifying media. For instance, the aforementioned "Births in Fairbanks, Alaska" contains categories of images of people who were born in Fairbanks, not images of people being born in Fairbanks. Omphalographer (talk) 22:59, 4 October 2024 (UTC)
I've always thought it was weird that categories for "births" contain images of and categories for people who have already been born. I assume a lot of them are added through templates though. So I'm not really sure there's anything that can be done to fix the issue. That's assuming there would even be a consensus to deal with it to begin with. But if it were me I'd confine categories for "births" to actual images of the birthing process and/or babies being born. But then who knows where the sub-categories would go in that case. "People by birth location" maybe? Or just completely ditch the whole scheme as meaningless trivia outright. --Adamant1 (talk) 23:13, 4 October 2024 (UTC)
"So fix it" = existing in a vacuum. I wouldn't even mention it if it weren't representative of a much bigger problem. As I hinted in the other discussion, I really hate mentioning examples because it ALWAYS provides an excuse for the other person to dwell on the example and ignore the big picture. RadioKAOS / Talk to me, Billy / Transmissions 02:57, 7 October 2024 (UTC)
Strongly agree with Omphalographer. We should minimize time and effort required as well as deduplicate work. For example, I think there could be a bot that suggests categories for Commons categories if the cats differ between WMC and WP so both are in sync (they don't have to be the same but often cats here miss some valid useful cats). Just saying "So fix it" would indeed be a case of existing in a vacuum, it would be a good point but that's not all the user said and elaboration of what the "much bigger problem" would be is missing. I could make an actual argument there for your case but I don't know if that's what you mean(?): Wikidata items may have inaccurate data and its items are less well maintained in the sense of watched for vandalism or flawed edits. An argument against that is that people here can see this flawed cat in the category page, they don't see the change in their Watchlist, but they can edit the Wikidata item if they notice it's flawed and also inaccurate data in the few (imo too few) fields that the infobox auto-adds to categories aren't that common. The main argument against this point is a potential solution: scripts that check for difference that likely need manual checking such as when a (surname) category set on an item differs from the category's first word, or items that have "Miss (surname)" where the first word differs and so on [a comprehensive ruleset could be developed]. Prototyperspective (talk) 12:06, 7 October 2024 (UTC)
Uploading photos taken in a commercial establishment
I took these photos of an old Heidelberg printing press at a commercial establishment. Removed any identification of the location, but didn't ask for permission. I want to upload them as PD, but unsure of their copyright status.
I'm not an expert on this, but my assumption is that it's fine to post these. The focus of these images is the printing press, which isn't copyrighted, and the commercial establishment it's placed in would fall under de minimis as it's barely visible. I would not post the third picture however, and I would crop out the sign at the top of the first. You can use the information on that sign (by paraphrasing or summarizing) but I assume the text itself is copyrighted and can't be copied or posted in full. ReneeWrites (talk) 09:51, 7 October 2024 (UTC)
Also as a non-expert, maybe you could be concerned that the text in some of these is copyright-able, so you may want to crop that out. Seems like a pretty small concern, but better safe than sorry. Edit conflict-y: I agree with Renee—Justin (koavf)❤T☮C☺M☯09:59, 7 October 2024 (UTC)
Thanks for all the comments. I also noticed the first photo with some possible identification. But, I doubt that the establishment would object to the promotion. My concern is acceptability on the Commons. BTW, I already removed a lot of identifying information. As for location details, what does that mean? The 3rd photo with the 1985 date is meaningless. It's when the company folded.— Ineuw talk12:51, 8 October 2024 (UTC)
This section is resolved and can be archived. If you disagree, replace this template with your comment. Prototyperspective (talk) 15:04, 15 October 2024 (UTC)
Exhibition at the M.H. de Young Memorial Museum, 1932
Hi, I am looking for the catalog of the exhibition at the M.H. de Young Memorial Museum, San Francisco, 1932, by the Group f/64, or at least a list of artworks. Any idea? In newspaper reports? This would help finding copyright status for these notable pictures. Thanks, Yann (talk) 14:03, 8 October 2024 (UTC)
Hi, I just learned that Internet Archive has been attacked, and suffered a security breach: [1]. :((( We rely a lot on IA for various reasons, and that's very sad. Yann (talk) 10:36, 10 October 2024 (UTC)
Yes, earthquakes in San Francisco area, a concern that I have expressed several times here, is only one of the risks that may affect Internet Archive. Fortunately, as Prototyperspective said, it seems that contents weren't affected, but if, as their only publicly available information says, they only have 2 production copies of each file (and no proper backups), some attack could possibly delete information completely (Archive stores so many files that a complete destruction due to an attack seems highly unlikely, though). I hope this event will help them to greatly improve their infrastructure, and to get the money they would need for that. I hope that the recent Archive's partnership with Google also helps with this. MGeog2022 (talk) 20:04, 10 October 2024 (UTC)
By the way, while WMF has much better resources and infrastructure than Internet Archive, absolute security does not exist, so this is a reminder that Commons media dumps are a real need (as well as true offline backups ready to be restored, both for media and for text content in all WMF wikis). MGeog2022 (talk) 20:26, 10 October 2024 (UTC)
Category slideshows
Does anyone understand why the UI to start a category slideshow uses such "mystery meat" navigation? Nothing obvious about the symbol at all. - Jmabel ! talk14:32, 10 October 2024 (UTC)
In the Upload Wizard there are separate fields for "Main subjects visible in this work" and "Category" (which possibly should read "Categories"). I wonder why it is necessary to have both. In what case would (or could) a "main subject visible" NOT be a category? ITookSomePhotos (talk) 17:59, 10 October 2024 (UTC)
Agree, it's duplication of work. I think categories should automatically sync to structured data if SD is to be used at all. However, note that a file may not depict some things that are in a category...however people also add depict statements for such things (e.g. when a video is only about sth but not depicting it) and there's also categories that are not about the contents in terms of what is being depicted such as Category:Videos without audio. Prototyperspective (talk) 18:50, 10 October 2024 (UTC)
Concur with Andy Mabbett. Ditto for an event at the building where it is mainly a photo of people, probably for a floor plan, certainly for a building permit, etc. - Jmabel ! talk14:08, 11 October 2024 (UTC)
It makes sense and I don't want to argue against it...I'd just like to leave a note about four things to consider:
people often set SD for depicts: the building (using that example)
when it has some depicts statements set the main thing depicted (door handle using that example) may not be in the SD while it's in the cats if cats are set in ca >95% of cases
the category could (often should) have a dedicated subcategory for pictures like that
the depicted thing can be inferred from the combination of categories set (e.g. close-up photos of door handles + building xy + other cats = pic depicts door handle of that building)
Note that I deliberately asked how a "main subject visible" would or could not be a category, and not the other way around. In the case of the door handle, still, my point holds. We would have to specify "door handle" twice, once as a category, and once as a depicted object, apparently redundantly. ITookSomePhotos (talk) 17:25, 11 October 2024 (UTC)
Though I guess the way things are set up now, if we only specified the categories, "door handle" and "building X", and no separate "subjects visible" information, then we would lose the information that "building X" is NOT visible. If this is important to capture, then I suppose in an ideal world we would specify visible objects once, and then have another field for "other categories". ITookSomePhotos (talk) 17:54, 11 October 2024 (UTC)
Red fruits or flowers?
On a closer look it looks like to be flowers. Google Lens does not help me much. It shows a lot of commercial results with red things and little certainty. I got one Commons result, File:Ilex serrata7.jpg but this is not it.Smiley.toerist (talk) 12:28, 11 October 2024 (UTC)
I suggest you use another plant identification tool instead of asking humans here. These tools could also be useful for suggesting categories and I think several work better for plants than Google Lens but I don't know if there's any that is free software. Prototyperspective (talk) 12:48, 11 October 2024 (UTC)
As it is becoming evident that some submitted files do infringe on sculptural copyrights due to lack of FoP in their countries (I nominated some myself), is it appropriate to create tracking categories of deletion requests concerning WLM-2024-related files?
In this way, organizers may be able to reassess their rules for future competitions and some may even try the stepping stones to reform their copyright laws. We may no longer be able to categorize those of past contests due to the deleted files not being visible to non-sysops like me.
@Adamant1@GPSLeo on a second thought, I think it won't help much. With little to no initiatives of various Wikimedia chapters and local user groups as well as unwillingness of several countries to permit full exposure of their public spaces in commercially-available media, I don't think we can expect improvements in copyright laws very soon. Even the U.S. architectural FoP is not immune to criticism, albeit from a small sector composed of some architects and lawyers who are critical of the 1990 AWCPA as having deprived architects the right to control the U.S. public's images of U.S. buildings. Expect some "unsurprising surprises" on FoP statuses of some countries. Withdrawing my categorization proposal, to reduce some stress on my part. JWilz12345(Talk|Contrib's.)06:24, 12 October 2024 (UTC)
Wikidata infobox forcing Wikidata based DEFAULTSORT
The {{Wikidata Infobox}} template seems to impose an (anglosaxonic?) DEFAULTSORT ordering on categories of people based on Wikidata, without broad discussion with the community and completely messing up a number of categories. Example: Category:Male politicians of São Paulo, try to find Category:Professor Fernando there, even if you know his real full name. It's impossible. Not only he is not know by his full name, and we are expecting an A-Z order of category names, but even if you happen to know his full name, someone added a surname on Wikidata that doesn't even exist in Portuguese. Shouldn't this Wikidata DEFAULTSORT imposition be removed or at least "opt-in" for that template? Pinging @Mike Peel: , who develops the tool.DarwinAhoy!14:16, 6 October 2024 (UTC)
Pinging @Animalparty: who also noticed this problem some years ago. This situation seems to be lingering here for long, I wonder why it hasn't been fixed yet with a simple opt-in policy for that automatic Wikidata based DEFAULTSORT ordering, instead of forcing it everywhere and causing a mess.--DarwinAhoy!14:36, 6 October 2024 (UTC)
You can disable its DEFAULTSORT.
wdib is used millions of times. obviously its practical solution is going by "opt out" instead of "opt in". RoyZuo (talk) 14:50, 6 October 2024 (UTC)
You can just add defaultsort and the bot then de-activates the infobox based sorting. That the default is English comes from category names being in English. ∞∞ Enhancing999 (talk) 14:51, 6 October 2024 (UTC)
Yes, it was incorrectly set up in Wikidata, that's why it was barely findable in that system. Now that I fixed it adding his surname Pieri he is impossible to find there, unless you know that rather obscure information. And please stop picking singular cases among the almost 100 which are there, obviously some of them would be correct, while many others will not. That one you brought now has the sorting key "Pires", so it's not even an appropiate example as it's not using that automated DEFAULTSORT. DEFAULTSORT key is the name the person is generally known for, in that case "Pires do Rio". There's no general rule related to using given names, surnames or pseudonyms for teh defaultsort. DarwinAhoy!15:53, 6 October 2024 (UTC)
If there is no rule, it never should have been implemented by default, instead of bringing this additional burden to the community here. 🙄 DarwinAhoy!17:12, 6 October 2024 (UTC)
Fun fact: the infobox is designed in such a way that it can be replaced by a new system when that becomes necessary - e.g., if infoboxes could be built into MediaWiki, or we migrate away from the category structure, since it's just a way of displaying the relevant information from Wikidata. With this issue, it's been great to see the discussion here, it seems no change is needed for now. Thanks. Mike Peel (talk) 22:15, 12 October 2024 (UTC)
I've had a lot of issues with DEFAULTSORT myself. It really shouldn't be the default, but then that would kind of defeat the purpose of the whole thing. So maybe it should just be axed? There's no reason people can't, or shouldn't, sort categories whatever way they want to without it being imposed on them through a template though. --Adamant1 (talk) 20:47, 6 October 2024 (UTC)
If adding the DEFAULTSORT template manually overrides the Infobox's DEFAULTSORT, I don't really see what the issue is. Most casual or novice editors won't think to add the template themselves (and even experienced editors sometimes forget to, i.e. me). In the vast majority of cases it does its job correctly and the few instances where it didn't it can be fixed manually. The alternative is having to add it manually to all categories. ReneeWrites (talk) 09:28, 7 October 2024 (UTC)
It seems to me entirely natural that the Wikidata rule that, given the right data, is correct well upwards of 95% of the time (98% would not surprise me), is used by default. Just so long as we can override, which we can, that's fine. - Jmabel ! talk09:42, 7 October 2024 (UTC)
Sinigh and I disagree about dashes in category names, especially for dates. I favor the ISO/IEC 646 character known as the hyphen-minus. He favors "–" (which I copy-pasted here, I'm not certain of the code point and I'm headed out the door in about 2 minutes). Our discussion so far can be found at https://commons.wikimedia.org/w/index.php?title=User_talk:Sinigh&oldid=934116270. Probably further discussion should continue here, not there. Other opinions sought, because we each have a decent rationale, but are unlikely to convince the other. - Jmabel ! talk06:00, 7 October 2024 (UTC)
En dash indicates a range, so 1800–1885 would be correct. This is as far as I can tell the norm on Commons for both category and file names. The universal hyphen/minus sign is used in category names that don't indicate a date range, see for example every entry here: Category:Days by day. ReneeWrites (talk) 06:31, 7 October 2024 (UTC)
Category names should use hyphen/minus sign, not en/em dashes. Per Commons:Categories#Category names policy "Basic English characters (ISO/IEC 646) are preferred over national variants or extension character sets (for instance, 'straight' apostrophes over 'curly'), where reasonable." Category names should be something that can normally be typed with a keyboard and should not use extension characters where there is no need. Hyphen is perfectly understandable range sign. MKFI (talk) 06:37, 7 October 2024 (UTC)
The policy says "where reasonable", which affords flexibility, and its use for ranges is not only completely reasonable but also grammatically correct. The en dash serves a specific purpose that is universally recognized and does not introduce ambiguity, which is why most English-language style guides will call for the en dash to be used in this way. That it can't be found on most keyboards seems to be the only real argument against it, and I find it completely unconvincing. ReneeWrites (talk) 09:25, 7 October 2024 (UTC)
I agree, especially since it's very common to use Commons:HotCat for categorizing, so the optimal solution is "Title with en dash" and "Category redirect with hyphen". I appreciate that it is somewhat cumbersome to input en dashes, but they are typographically and semantically correct and while HotCat or similar tools are certainly not mandatory, they can easily resolve the issue for those who really care. —Justin (koavf)❤T☮C☺M☯09:34, 7 October 2024 (UTC)
I just wish to point out that I believe typeability is generally a valid concern, so I don't take issue with the basic premise of the discussion. But I only agree to the extent that I'm certain this could have been a real problem. My main objection, however, is that it seems very unlikely that the range dash (or hyphen) will ever be typed. To clarify, the original examples were Category:Hendrik Koolwijk (1800–1885) and Category:Iraqi Army in the Gulf War (1990–1991), which made me wonder: When is anyone ever going to type full category titles like those? And if they were to, when would they actually have to?
If typeability is indeed relevant here, I don't see why it is only considered important when it comes to the dash, while diacritics and language-specific characters derived from the Latin alphabet are preferred. Don't they, more often than not, cause an arguably worse version of same perceived problem?
"Title with en dash" and "Category redirect with hyphen" sounds like good strategy, and hopefully it's also an acceptable compromise. From that point of view, it's actually quite convenient that categories with range hyphens often already exist.
I hope this will be accepted as a compromise too. I know Commons isn't Wikipedia and we do our own thing here, but I do find it useful to look at how things are done on Wikipedia when an issue like this crops up (one that's not Commons-specific, I mean) on how they've dealt with it, and "Title with en dash, redirect with hyphen" is the compromise they settled on as well. ReneeWrites (talk) 10:12, 7 October 2024 (UTC)
If I understand it correctly the short summary of what this is about is whether the - or – character should be used. If there is some decision either way I think this should be in some guidance page where it's also clarified where each character is to be used as well as a bot/script that automatically moves categories accordingly. Additionally, it would be best if not needed that a technical change is implemented that makes HotCat autocompletes with – show up when using the more common minus - character. This is a broad subject and also affects for example Wikipedia lists which sometimes intermingle both. If there is a code issue or proposal about the autocompletes showing up, please link it here. Prototyperspective (talk) 11:57, 7 October 2024 (UTC)
Agreed on documentation and to be clear, HotCat will correct/autocomplete any category redirect, either those that use standard MediaWiki syntax (i.e. "#redirect[[:Category:Foo]]") or by using {{Catredirect}}. —Justin (koavf)❤T☮C☺M☯12:07, 7 October 2024 (UTC)
To clarify: I meant that it should show the autocomplete without requiring there to be a redirect with the same name but a different hyphen character. E.g. always whenever there is a – between numbers show this cat in the autocompletes even if the user entered a minus - despite of there not being a redirect. Maybe it's not really needed because that character usually comes only near the end of a cat title, this would be needed if sometimes this is near the start of a cat title. Prototyperspective (talk) 12:13, 7 October 2024 (UTC)
That would still be a brilliant addition, though. Would it even be possible to automatically redirect searches whenever someone (understandably) types a hyphen where the dash goes? I mean, sort of like the way searches aren't case sensitive, could they also not be hyphen/dash sensitive? Only when needed, of course, i.e. if there actually is a category with a dash where the person types a hyphen. Sinigh (talk) 13:34, 7 October 2024 (UTC)
It is possible, the question is how difficult it would be and what the best way to implement it would be. In any case, it probably needs a code issue for HotCat. Prototyperspective (talk) 14:11, 7 October 2024 (UTC)
Thats the best solution, the hyphen. Juts because Wikipedia has a preference for en dash) does not need to apply to us. I keep on saying we're a databank, names of cats should be as simple as possible and contain easily available on the keyboard. No need for affectations. Broichmore (talk) 12:40, 12 October 2024 (UTC)
Wikipedia's preference is not and has never been the main argument here; ReneeWrites only told us how they solved it over there and pointed out that "we do our own thing here." As for the keyboard layout, everyone already knows that the hyphen is easier to type. That argument has been addressed in several ways, yet noone seems to have anything to say in response to those objections. Simply reiterating the original argument obviously isn't going to convince anyone. Sinigh (talk) 13:06, 12 October 2024 (UTC)
When using the deepcategory search operator it should detect self-categorization so it doesn't break. Until that is implemented, fixing self-categorization is especially needed (e.g. to enable it to always show results phab:T376440) but it's also a general problem if categories contain themselves (in these cases directly).
The report was empty / out of date for a long time but now MZMcBride (talk·contribs) updated it – thanks for that!
I worked on it until Ludwig_August_Fallon (default sorting not alphabetic sorting) except for year_in_India cats. Fixing these problems is usually quite simple – removing that category or replacing a template that sets the category with the categories. Sometimes one also needs to add other categories such as some taxonomic category. I requested on the talk page that it (a separate page / report) also shows categories that include themselves somewhere in their subcategories.
If this is done and with the two other reports announced above there would then be reports for nearly all issues with categories. After these have been solved to some extent what would remain are the issue of categories with just 1 file (debatable whether that's preferable over just categorizing the file like the cat), empty categories (see Quarry:query/7200 and many/most of them should be deleted), and mismatches between WMC and WP (mainly ENWP) categories which often means some cats of the WMC cat are missing and just need to be synced with WP.
--Prototyperspective (talk) 14:59, 10 October 2024 (UTC)
That would be great. It would only fail for categories that have themselves set via templates. If you neither create such a bot nor request one, I'll sooner or later propose one at Commons:Bots/Work requests. Some of these cats may benefit from human checking (because they lack cats) but that's not worth the time required and may get addressed via the mentioned category-mismatch detection that could implemented at some point. Prototyperspective (talk) 16:36, 10 October 2024 (UTC)
Thanks! I don't know what the query for it is and hope MZMcBride updates it again at some point. Categories that contain themselves in some subcategory are just as much of a problem (e.g. cause deepcategory to fail and are logically contradictory) but aren't included there. If somebody is interested in these or can adjust the query (which is it?) to also scan several layers of subcategories, please see the report's talk page. When it comes to templates I suggest people who added the template are pinged or it's asked about at the template's talk page – often the best solution is to comment out or remove the template and replace it with the categories that it automatically sets except of the self-categorizing cat. Prototyperspective (talk) 09:27, 11 October 2024 (UTC)
Not sure, maybe the templates can be fixed to do it more selectively.
Not sure if loops can be avoided entirely, but we could try to identify the categories where the parent category is also a subcategory. ∞∞ Enhancing999 (talk) 12:15, 11 October 2024 (UTC)
Thanks for the link, I've seen it before but don't know where I found the link and forgot about it – however, that's just some code without any info on how to run it. Can it be run in the Quarry tool? Loops can be avoided by putting all the scanned categories in an array and the checking whether the category currently checked is in that array. If so, it would not look into subcategories of that category and show it as one of the categories that contain themselves. This way it would scan layer by layer and the number of layers to scan could be increased over time. Prototyperspective (talk) 12:20, 11 October 2024 (UTC)
All direct inclusions should be fixed now (if my regex missed some, please tell me).
This means that the remaining cases would be template based (151 categories), Some template talk pages have notes about the problem: [2]. ∞∞ Enhancing999 (talk) 07:44, 12 October 2024 (UTC)
FastCCI's cat path exampleGreat! So this is already solved? Could you please put that in a separate query instead of just in some revision of a query about something else and then put that in some report?
Putting it in a report on Commons is also useful for people here to find that query, not just for accessing the results and working on fixing the issues. When it comes to fixing the issues what is still needed is showing a categorization path per each item showing why the second category is located somewhere in the former like in the attached image. Do you think it would be possible to add a column for that in the query? It would also be relevant to this request for a new tool/gadget and this issue for petscan. Such is one of the most needed functionalities to improve categorizations and petscan often shows files that do not belong into categories without a way to find out why (examples in the two links). Prototyperspective (talk) 13:17, 12 October 2024 (UTC)
If you won't forget about it and keep track of all the relevant info and the query then that may be a good idea. I don't think there's an issue with putting it into a report if nobody finds it anyway. A categorization path is also needed for e.g. the two other things I linked and maybe that could be used there. Didn't you say the number of self-categorizations is way down now and the remaining ones are due to templates? In that case it seems like the problem is already solved: making template creators/writers aware of the problem on the template talk pages so the problem gets fixed at the template and/or simply commenting out the template on the respective pages to replace them with the other cats. If you mean the 7500 cats that contain themselves somewhere, I don't think a bot could correct [many of] these self-categorizations – it does need quick human checking and that is best done by enabling users to only need very little time per correction (quick succession editing) including displaying the necessary info of the categorization path. The optimal solution would be displaying the categorization path with each cat having an x button to remove the cat that caused the self-categorization but requiring the user to click the cat and remove it with a second click is also good. Prototyperspective (talk) 17:21, 12 October 2024 (UTC)
They are all pairs as the sample above (Category:(26)_Proserpina_symbols + Category:Persephone_symbols). Longer chains are not include (e.g. grandparent = child).
Oh too bad, I thought I checked some and it was not a direct self-categorization and also I don't understand then why the query shows so many results while the report only has a few hundred or so or 151 remaining ones. In that case maybe it does need a report with two columns with a button to directly remove the problematic cat on either cat directly, maybe this can can be done with some specific URL and the URL could be attached next to each wikilink. No other idea, e.g. I think even for your example case where it's quite obvious to a human it would be difficult to a bot (one that has some chance of getting developed). Prototyperspective (talk) 17:50, 12 October 2024 (UTC)
The 150 are just the ones that are self-categorizations (1 category involved, as your report above).
i believe we have enough images that represents what QR codes looks like. and also there is many more redundant files. right now, 358 files. i believe we should do something about this. Category:Raster quick response codes. we should make DR'd these redundant files. and, i didnt check, also, i believe there is plenty of advertisements out there. modern_primatඞඞඞ ----TALK17:58, 11 October 2024 (UTC)
Support. I'll go through these and open a batch DR later today. FWIW, you can decode most of these by pasting the image URL into Zxing - much easier than scanning them off the computer screen. Omphalographer (talk) 19:03, 11 October 2024 (UTC)
I uploaded a logo and it is stored locally on the English Wikipedia (en:File:Linpra logo.png).
I'd like to upload it to Commons with the same rationale as logos from Volvo, Ikea, Saab and others.
How do I upload it to Commons?
I don't see any documentation of the threshold of originality for Lithuania. Does someone have something? For the U.S. it should be OK. - Jmabel ! talk16:20, 12 October 2024 (UTC)
From a technical point of view, of course you could, but that doesn't make the copyright status in Lithuania any clearer. - Jmabel ! talk20:11, 12 October 2024 (UTC)
Changing the name of Category:Ivory Coast into Category:Côte d'Ivoire
This section is resolved and can be archived. If you disagree, replace this template with your comment. ReneeWrites (talk) 11:29, 18 October 2024 (UTC)
First of all, I apologize for changing the name of a few categories, I didn't realize there were hundreds of them.
Since April 1986, the government declared that Côte d'Ivoire would be its formal name and has since officially refused to recognize any translations from French to other languages in its international dealings. So we have to change the name for all categories that include the name "Ivory Coast". Like we don't call Birmania anymore, or unlike Costa Rica which has never been translated into "Rich Coast", Côte d'Ivoire has an incredible number of different names. Ivory Coast (English), Costa de Marfil (Spanish), Elfenbeinküste (German), Boli Kosta (Basque), Obala Slonovače (Croatian), Ranná Bóga (Irish), Kotu di Bivora (Galician), Bregu i Elefantit (Albanian) and so on. -- Zorion eko 22:29, 17 October 2024 (UTC)
There is no Freedom of Panorama for sounds Does somebody have info on that? Videos of live music would be useful often and also there are several/many videos of such which if they aren't free would need to be deleted (example). I am trying to find wich music band was playing Use the Wayback Machine when it's functional – if they don't have the relevant page archived and other archival websites like archive.today and the Google Cache don't have neither and Web search engines don't show anything for relevant searches then you're probably out of luck but I don't see any problem with that since it's unlikely to be a very notable band where a photo is much needed and I think most such bands already have some photos on WMC. Prototyperspective (talk) 12:22, 12 October 2024 (UTC)
A band playing not permanent and therefore definitely not covered by FOP in most regions as FOP requires works to be permanent in public space. For cases like bell installations it could be a difficult question. GPSLeo (talk) 11:22, 13 October 2024 (UTC)
Thanks for the info. I wonder about the music but I think it can't be copyrighted if it's played in a public space. Nevertheless, I don't know if a concert etc is a public space in regards to whether videos of the live music can be uploaded under CCBY. Prototyperspective (talk) 11:45, 13 October 2024 (UTC)
The video ends with the arrival in a small train stop. A have been checking the stations in the en:Kyūdai Main Line article, but not finding it. Nearly all the bigger stations have a footbridge and this one does not. There is a Japanese text, but I am not certain that this the station name. Is there a handy way the extract an image from a video? I can do a print screen, but I am not certain this is the best way. (I am not used to working with videos). Is there an Japanese/English speaking forum, where I can ask the question? Unfortunatly OpenRailmap does not have an translate function. Smiley.toerist (talk) 08:29, 13 October 2024 (UTC)
Is there a handy way the extract an image from a video? In Firefox you can simply do what is described in Category:Still images from videos by Terra X. Alternatively, you can open the video locally and press shift+s (or ctrl+s?). A tool to directly upload stills from videos on WMC would be handy (especially for the linked cat). Prototyperspective (talk) 10:11, 13 October 2024 (UTC)
Note: It will be lower resolution than the video if one does not click Original file first as described in the cat. This may be irrelevant in many cases but when uploading stills to WMC the stills shouldn't be lower-quality than the video. Prototyperspective (talk) 11:10, 13 October 2024 (UTC)
If you like specifying train stations, this video (among several others) shows many of them. Maybe some of the places and villages shown could also be identified but I wonder about which use-cases that has, at least in the current form. Prototyperspective (talk) 20:27, 13 October 2024 (UTC)
I used the last modified time parameter of the original MP4 file, to add the correct time. I have added the start location. Is there a format to indicate the start and end location of a video? I have experimented with timetext in Dutch. Can I add the English text or is a completely different timetext necessary?Smiley.toerist (talk) 11:42, 14 October 2024 (UTC)
It does not work for the start position. You have to use a predifined location (item), not dump location coordinates. The destination point use gives a warning. Smiley.toerist (talk) 12:37, 14 October 2024 (UTC)
With this usable photo, the description is pure advertising. This is something I come across more often. What to do? Leave it as it is or reduce the text in this case to “Machu Picchu”? Wouter (talk) 17:43, 13 October 2024 (UTC)
This is common in YouTube video descriptions which are abused for all kinds of spam and junk as well. Ranked:
A meaningful description "Tourists posing on steps at Machu Picchu in 2023"
I often do photo editing on low-quality images and upload better versions (corrected exposure, etc). For example, File:Admiralty Centre Tower 2 (adjusted).jpg Is there an easy way to do this? Its a pain to have to manually go through the upload wizard, copy over the original author, license, description, generate a new file name with "(adjusted)" stuffed into the original name, etc. It seems like some bit of javascript could do all of this with one click. Does such a thing exist? — Preceding unsigned comment added by RoySmith (talk • contribs) 19:52, 13 October 2024 (UTC)
Others might be able to point you to a better tool, but the Basic Upload Form in Special:Upload might be useful for you. You can copy the entire description from one file and paste it here. This should almost definitely save you some clicks. Rubýñ(Scold)20:25, 13 October 2024 (UTC)
While these candidates have been ranked through the vote, they still need to be appointed to the Board of Trustees. They need to pass a successful background check and meet the qualifications outlined in the Bylaws. New trustees will be appointed at the next Board meeting in December 2024.
Copyright question regarding document from "Memory of the People" (pamyat-naroda.ru)
Hi all,
I've uploaded a PDF document to Wikimedia Commons that I downloaded from the Russian "Memory of the People" website (pamyat-naroda.ru). This document details the World War II service record of my grandfather, Georgiy Pavlovich Osipov, and includes information about his unit, awards, and activities during the Siege of Leningrad.
The "Memory of the People" website states the following copyright information:
Правообладателем ИС «Память народа», включая входящие в ее состав программы для ЭВМ, электронные банки (базы) данных, включающие электронные копии архивных документов, является Российская Федерация в лице Министерства обороны Российской Федерации (далее – Правообладатель).
Собственниками архивных документов, электронные копии которых размещены в ИС «Память народа», является Российская Федерация (в части документов, предоставленных российскими государственными архивами, государственными органами и организациями), а также иностранные государственные архивы и иные организации, предоставившие электронные копии этих документов.
Правообладателями аудиовизуальных произведений, предоставленных Российским государственным архивом кинофотодокументов и размещенных на сайтах ИС «Память народа», являются авторы этих произведений.
Оператором сайтов ИС «Память народа» (далее – Оператор) является Министерство обороны Российской Федерации, либо уполномоченная им организация.
Все сервисы, а также информационный контент, предоставляемые сайтами ИС «Память народа», являются государственными информационными услугами, оказываемыми бесплатно.
Обязательным условием использования документов с сайта в изданиях, связанных с историей Великой Отечественной войны, является ссылка на сайт.
English Translation:
Legal Information
The copyright holder of the "Memory of the People" information system, including the computer programs, electronic data banks (databases) containing electronic copies of archival documents, is the Russian Federation, represented by the Ministry of Defense of the Russian Federation (hereinafter referred to as the Copyright Holder).
The owners of the archival documents, electronic copies of which are posted in the "Memory of the People" information system, are the Russian Federation (with respect to documents provided by Russian state archives, state bodies, and organizations), as well as foreign state archives and other organizations that provided electronic copies of these documents.
The copyright holders of audiovisual works provided by the Russian State Archive of Film and Photo Documents and posted on the websites of the "Memory of the People" information system are the authors of these works.
The operator of the websites of the "Memory of the People" information system (hereinafter referred to as the Operator) is the Ministry of Defense of the Russian Federation or an organization authorized by it.
All services, as well as information content provided by the websites of the "Memory of the People" information system, are state information services provided free of charge.
A mandatory condition for the use of documents from the website in publications related to the history of the Great Patriotic War is a reference to the website.
My understanding is that the website considers its services and content to be state-provided information freely available, but requires attribution.
Could someone please advise on the most appropriate license to use? At this point of time, I gave it CC-BY-4.0. I want to ensure full compliance with copyright law and Wikimedia Commons guidelines. Any guidance would be greatly appreciated.
These documents were created during Great Patriotic War (1941-1945), so Creative Commons licenses did not exist at that time. Since Russia is heir of Soviet Union, general rules for government-created documents are applicable. --EugeneZelenko (talk) 14:34, 14 October 2024 (UTC)
Change in editing behaviour for keyboard shortcuts?
I've probably missed an announcement, but now if I try to copy something with command-C I get ''Italic text'', and if I paste something with command-V I get <sup>Superscript text</sup> - any pointers to where this changed, and how to disable it? Thanks. Mike Peel (talk) 22:10, 12 October 2024 (UTC)
With more investigation, this seems to be related to my keyboard overlay - if I switch to querty, it works fine, but with dvorak, it isn't. Weird. (I becomes C and > becomes V with dvorak, so the keyboard shortcuts then make sense...) Thanks. Mike Peel (talk) 16:44, 13 October 2024 (UTC)
Is there any Help/Commons page about empty categories? I'd like to ask if these are noindexed and if not propose that they are made noindexed. Moreover, at some point it may be good to discuss what to do about empty categories created over e.g. 2 years ago of which there seem to be many. --Prototyperspective (talk) 11:14, 13 October 2024 (UTC)
The web inspector should be able to tell you if a page has the noindex command in the head of the page. I personally think noindex is overused and breaks the way google updates their indexes to a degree that makes if very difficult to change the page back to being indexed, and as such should not be used on pages that frequently change and where people have an expectation that the page should show up in indexes quickly once the no index command is removed. —TheDJ (talk • contribs) 12:26, 14 October 2024 (UTC)
Okay but then there is a problem with the many empty categories. I doubt that is a reason why Google (and DuckDuckGo) barely index category pages (and media) of WMC but I think we should leave them as little plausible reason as possible. Prototyperspective (talk) 12:34, 14 October 2024 (UTC)
I just found many categories with only infobox categories are either:
What is best done there? Should anything be done about the former? If not I wonder why there are only so few of these – do they maybe miss a cat for disambiguation pages? Is something needed or existing that adds an image automatically to a category if it's added to the corresponding Wikidata item if it's not in any subcat of it? I don't know how these images were added to the items, if people add them directly to the items via upload, maybe that tool needs a change so it's also categorized. Prototyperspective (talk) 15:10, 15 October 2024 (UTC)
Disambiguation categories are fine. An infobox may be a bit excessive there, but the interwiki link from Wikidata is important, so it is linked to parallel disambiguations in other WMF wikis.
For the ones with an image in the Infobox, my first thought is, "So why the hell isn't this image (and possibly others) in the category?" I'd probably see whether the category can be filled in with some content. And given appropriate parent categories. - Jmabel ! talk19:18, 15 October 2024 (UTC)
Some of the empty cats with images in infobox have the image in some equivalent category, for example in the same cat that is not misspelled. Other than that, it could be a good subject for discussion whether it would be better to have the image added to create a 1-image-category (over-categorization?) or just leave the image in the cats it's in and delete the empty cat for the time being (or if it varies when to do what). Here a separate report may also be best but one could also leave the main report unchanged and check the remaining bluelinked empty categories so people can see what's best case by case. Prototyperspective (talk) 20:25, 15 October 2024 (UTC)
Moving from one data item to another in SDC
I created in Wikidata a new item d:Q130530281 as a subclass of d:Q117075694. I wil use the SDC script to move Commons files to the new item. However there are files such as File:Tramway de Gand - Enmarchement de la rame PCC 34 (6034).JPG where d:Q117075694 was used. Can I later check for files with both tram and train items?Smiley.toerist (talk) 12:57, 14 October 2024 (UTC)
I am creating an open-source app for browsing OpenStreetMap – the OsmAPP. You can access it at osmapp.org. One of main features of OsmAPP is showing a side panel with information on any clicked map feature or POI. This includes showing photos from many available sources, which is mainly from connected Wikipedia articles, Wikidata entries and directly linked Wikimedia Commons images (see eg. this page).
This connection naturally led me to think about making upload possible as well, as this could benefit both projects – supply Commons with real world images under correct license, and add photos to various map features in OpenStreetMap.
After a year of effort (the wiki api is quite challenging 😃), I have the upload script ready. I asked for the permission to make a test run, which succeeded (OsmappBot contributions). Now I would like to ask broader public about your opinions, and ideas how to make the most of it for Wikimedia Commons.
Regarding Freedom of panorama – the Upload dialog would warn users if the specifc country forbids public photos of buildings etc. I can't add the specific {{FoP-country}} templates programatically (that would need a AI object recongnition), but I will review uploaded images manually and add it if needed. OsmAPP doesn't have many users, I expect only handful of uploads per month. Mind that, this is not an import bot, it rather uploads images on behalf of logged in OSM users, any change I make to the bot will only affect future uploads.
Standard upload dialog
FoP in effect when uploading in countries from this list
Interesting. I suggest you create a video demonstrating its features and put it on YouTube and also upload it here. Please enable users to add one or several categories with autocomplete similar to the Upload Wizard on this site (maybe the HotCat script could be used for that). Prototyperspective (talk) 16:21, 9 October 2024 (UTC)
I would recommend rewriting Public photography is forbidden in this country. to Public photography is restricted in this country. as all but five countries with freedom of panorama provisions grant permission for some works but not others. For example, the United States has freedom of panorama for architecture but not sculpture. JohnCWiesenthal (talk) 19:20, 9 October 2024 (UTC)
Looks very interesting and promising! Thank you for the work! It would be a nice addition if login and upload with Wikimedia account would be possible. That would also relieve you from the responsibility at least for uploads done through that way. --Marsupium (talk) 06:42, 11 October 2024 (UTC)
@Prototyperspective – Thanks for the feedback. I think I could design a TextField with Commons category search. It will take some time though. Also good idea with the video. I will do it once I have the redesigned dialog ready :-)
@Marsupium – Thanks for the kind words. I think for Wikimedia users, I can provide a link to the official upload tool, there is already an option in OsmAPP to add wikimedia_commons photo in the Edit section of any feature, eg. here: https://osmapp.org/way/7645354/edit .. the upload dialog makes it only easier for OSM users to contribute to Commons. Zbytovsky (talk) 08:15, 16 October 2024 (UTC)
Need Help Add Image to Page
This page requested cropped images, which I created from the source image. However, the system penalizes if I try to add the new images to the page.
Below are the source image and cropped images:
Love American Style cast 1973.JPG - Wikimedia Commons
Love American Style cast 1973.middle.jpg
Love American Style cast 1973.1.jpg
For the last image, I uploaded several versions which you can see at my uploads.
@Starlighsky: You don't say what page (or even what website) you want these added to. The latter two are not very good photos, they are tiny, I would certainly hesitate to add them to (for example) a Wikipedia article. And "the system penalizes" is unclear. Is this something you were asked not to do? If so, you certainly should not encourage someone else to do it on your behalf. - Jmabel ! talk05:21, 16 October 2024 (UTC)
Here is the site: File:Love American Style cast 1973.JPG - Wikimedia Commons
There is a proposal to test if requiring anon users to confirm their edits by clicking the edit button again and showing a warning to avoid accidental edits can reduce the amount of bad edits. There is currently a consensus to test this but with only three people involved Commons:Village pump/Proposals#Simple edit confirmation. There was also the question on numbers of bad edits. I created a tool that shows how many edits are marked as reverted also compared to the patrolled edits that shows the potential high numbers of unseen bad edits [3]. GPSLeo (talk) 16:55, 16 October 2024 (UTC)
Seeking volunteers to join several of the movement’s committees
Each year, typically from October through December, several of the movement’s committees seek new volunteers.
Read more about the committees on their Meta-wiki pages:
Applications for the committees open on 16 October 2024. Applications for the Affiliations Committee close on 18 November 2024, and applications for the Ombuds commission and the Case Review Committee close on 2 December 2024. Learn how to apply by visiting the appointment page on Meta-wiki. Post to the talk page or email cst@wikimedia.org with any questions you may have.
We use reverse image searches pretty often, mainly for determining if a new upload exists elsewhere on the web, but also for looking for 3rd party usage of the files we host. It looks like Google is deprecating its reverse image search in favor of Lens. What are people using as a substitute? I never got particularly good results with TinEye... — Rhododendritestalk | 14:08, 16 October 2024 (UTC)
I get good results with TinEye and use a browser extension so that one only needs to right click the image to reverse search it. I think it would be better if bots/scripts did the reverse searches. Don't know what you mean with the Google issue – it can still reverse search, maybe click on "Find image source" at the top. I don't think there are other options – they would be on https://imgops.com/. Prototyperspective (talk) 15:40, 16 October 2024 (UTC)
If I am performing a due diligence search to see if anyone is claiming to be the creator, or if a for profit archive like Getty is claiming an active copyright on an image I use both. I like that Tineye tells how many images it searched. --RAN (talk) 14:39, 17 October 2024 (UTC)
Automatic detection of potentially problematic domains in UploadWizard
Mockup of a potential message for problematic external source
You are encouraged to take a look at the project page, and to evaluate the current workflow.
We have also a request for feedback for you: we are considering adding a user alert during the upload process if a problematic external source is detected in the file. On the right, we are sharing some potential designs for the alert. We welcome community feedback about how it is phrased and whether it may be useful.
This section is resolved and can be archived. If you disagree, replace this template with your comment. Prototyperspective (talk) 11:11, 2 October 2024 (UTC) 
Support It's to bad the same thing can't be done for previous competitions. --Adamant1 (talk) 00:56, 12 October 2024 (UTC).png/220px-FastCCI_Commons_gadget_showing_why_a_file_is_erroneously_categorized_into_a_category_(category-tree).png)
.png)
Done - Commons:Deletion requests/Files in Category:Raster quick response codes. I've included the decoded content of most of the images. Omphalographer (talk) 03:17, 12 October 2024 (UTC)
Thank you.! modern_primat ඞඞඞ ----TALK 05:42, 12 October 2024 (UTC)
.svg){kind=link}
.png){kind=small}
22:29, 17 October 2024 (UTC)
.png/12px-Kaixo_(Irulegiko_eskua).png)
10:05, 18 October 2024 (UTC)
.jpg){kind=link}
_-_OsmAPP_(4).JPG){kind=link}
link to a category