Commons:Village pump/Archive/2018/09

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search

Why is Category:Балет -- which is a redirect to Category:Ballet -- automatically a child of Category:Ballet? This causes it to be listed as a sub-category at Category:Ballet, which seems obviously wrong. I can't think of a reason why a category redirect should be a child of whatever it redirects to, and doing so clutters the parent category with totally useless links. Is this normal behavior for Template:Category redirect? --Lambtron (talk) 22:03, 5 September 2018 (UTC)

Per this discussion of the template it seems that it is intentional, but should only happen for non-empty categories. Looking at the code, it is indeed so. So the only issue is caching of the number of pages in Category:Балет. The issue will resolve itself in some time. Compare, for example Category:Professions. —⁠andrybak (talk) 06:16, 6 September 2018 (UTC)
You can remove it by editing and saving (you don't have to make any changes) the redirected category after all files/categories/pages have been moved elsewhere. Otherwise, it will self-resolve in time, though not sure how long. Personally I don't like the extra detritus left behind so once I've moved all of the contents to their proper homes, I do the edit & save which also serves to close the loop on not having any leftover files stranded. I go back to the target category and ouala! it is clean as a whistle. Josh (talk) 07:58, 6 September 2018 (UTC)
A null edit won't work in this case, because the automatic categorization by Template:Category redirect relies on number of pages in category. —⁠andrybak (talk) 08:36, 6 September 2018 (UTC)
Well you have me puzzled. The category is empty so a null edit should cure it. It has always worked for me without issue once there were no files left. Hope it cures itself in time? Josh (talk) 11:00, 6 September 2018 (UTC)
The null edit works for the case of page still appearing in the category via a plain [[Category:Name]] inclusion. Template:Category redirect does something more tricky than just include a categorization entry. It checks the number of pages still in category. This number itself is cached. And I don't think that any kind of null-edits will help with that.
Here's the current number: "0" — using code {{PAGESINCAT:Балет}} —⁠andrybak (talk) 16:09, 6 September 2018 (UTC)
(edit conflict) The category is now empty (except for the cat redirect) but for some reason it still is automatically categorized as "Non-topical/index: Non-empty category redirects". BTW, I first noticed this problem when the category was already empty, before adding the discussion template. --Lambtron (talk) 16:17, 6 September 2018 (UTC)

Hopefully the problem will fix itself soon (since there doesn't seem to be a way to fix it manually). As an aside, it seems illogical to classify any category redirect -- empty or otherwise -- as a child of its redirect target. Instead of automatically assigning a bogus classification to the category redirect, the system should prohibit assignment of redirect categories to files/pages/categories. --Lambtron (talk) 20:53, 6 September 2018 (UTC)

I agree, it seems an imperfect solution. I think the idea was to provide a convenient bridge for moving files over to the new page with the idea that the bridge would disappear once the file move was complete. I know Cat-a-lot is quicker to use when there is a category link existing there. I've moved a lot of categories and this is the first time I've not seen one heal up as soon as the files are cleared. Do you know of any more instances? Josh (talk) 21:39, 6 September 2018 (UTC)
This is the only instance I've seen. I just discovered that the problem has been fixed by another editor, who deleted and then recreated Category:Балет. --Lambtron (talk) 23:45, 6 September 2018 (UTC)
Looks ok now. —⁠andrybak (talk) 08:03, 7 September 2018 (UTC)
This section was archived on a request by: —⁠andrybak (talk) 08:03, 7 September 2018 (UTC)

Please rename

Please rename this File:भामहालङ्कारः.pdf. Thank you. NehalDaveND (talk) 17:01, 1 September 2018 (UTC)

@NehalDaveND: ✓ Done but there’s no need to post here after tagging a file for renaming; I don’t think we have a shortage of file movers. And please provide whatever information you have about the source, even if incomplete: anything at all would be better than “unknown”.—Odysseus1479 (talk) 17:33, 1 September 2018 (UTC)

September 02

Mil.ru? images

If these 2 images are mil.ru images, please feel free to review them...as I cannot read Cyrillic.

Thank You, --Leoboudv (talk) 08:57, 8 September 2018 (UTC)

This section was archived on a request by: —⁠andrybak (talk) 21:31, 8 September 2018 (UTC)

Created on date at Special:ListUsers

Hi all. What does the "Created On" date refer to on this page: https://commons.wikimedia.org/w/index.php?title=Special:ListUsers&group=sysop? Is that the date that someone became an admin, or when they created their account? And, what does it mean if that data is absent for someone? Thanks! Glammmur (talk) 10:02, 31 August 2018 (UTC)

The "Created on" at Special:ListUsers is the date the account was created. Accounts created before 22 December 2005 don't show a date. See Commons:List of administrators by date if you want a list of administrators in order of date they were appointed. —RP88 (talk) 10:18, 31 August 2018 (UTC)
Thank you, RP88. Is there anywhere that shows how the actual date they were appointed? Glammmur (talk) 13:26, 31 August 2018 (UTC)
Glammmur, you can see a list broken down by year at Commons:Administrators/Archive/Successful requests for adminship (although this list includes former admins). Visiting the specific page in the archive for an admin will show the date that admin requested the administrator right. If you want to instead determine the actual date of appointment for administrators you can examine the user rights log at Special:Log/rights. For example the changes to my user rights are shown at [1]. You can also use Special:UserRights to examine a user's current rights and see an excerpt of the user rights log. For example, see Special:UserRights/RP88 for my current user rights as well as log of the changes made to my user rights. —RP88 (talk) 22:55, 31 August 2018 (UTC)
@Glammmur: In some cases, for users promoted before 10:20, 13 December 2004 (UTC), the user rights log doesn't go back far enough. For instance, this log doesn't show when Yann, our longest-serving Administrator, was promoted, so you have to look at Commons:Requests and votes/Yann, which seems to indicate Yann was promoted shortly after 20:00, 17 September 2004 (UTC).   — Jeff G. ツ please ping or talk to me 00:51, 1 September 2018 (UTC)
This is so interesting! I'm trying to figure out how many admins there are, and how long they've "survived", and how often new ones are appointed (as you may have guessed!). If I add up the people listed at Special:ListUsers, there are 225... if I add up the people listed at Commons:Administrators/Archive/Successful requests for adminship and remove those noted as status removed/resigned etc, there are about 472 (from a total of 596 ever appointed). This opens the question about what Admin means... Is there a list somewhere that shows all possible user rights, like autopatroller or file mover? Is anyone with any of those rights considered an admin of some sort? Thanks again! Glammmur (talk) 10:27, 3 September 2018 (UTC)
@Glammmur: Successful requests for de-adminship here are categorized here in Category:Successful requests for de-adminship. Actual de-adminship here is reflected in the rights log on Meta (and from 10 December 2004 or earlier m:Meta:Bureaucrat log; search both for "@commonswiki" without quotes). COM:A details what Admin means here. Commons:User access levels details user rights / access levels here, some of which were unbundled from adminship.   — Jeff G. ツ please ping or talk to me 11:41, 3 September 2018 (UTC)
Thanks, Jeff G. ツ. I'm afraid you lost me there. I can't figure out how to operate the pages you directed me to, sadly. I don't expect you to - rather, I'll proceed with these two different snapshots of current admins: the one that's listed on Special:ListUsers (225), and the one I've generated rather manually, using Commons:Administrators/Archive/Successful requests for adminship and removing those noted as decommissioned there (472 active, 124 decommissioned, total 596), unless someone can easily suggest why those two admin counts are that different. Thank you, though! Glammmur (talk) 15:23, 3 September 2018 (UTC)
@Glammmur: I think many of the de-adminships were voluntarily initiated at Meta, with no corresponding decommissioning notes filed here. You are welcome to file such notes as you trawl through the logs on Meta, or file {{Editprotected}} requests if the pages are protected.   — Jeff G. ツ please ping or talk to me 02:53, 4 September 2018 (UTC)

Wikidata does not allow "-" character in pictures

Or at least this is what it says when you try to add a picture that has one in the name.

So for every new person that adds a picture on Commons without knowing this you get a "renaming request", that consists in a direct request to file movers or a discussion in the Bar. Have I got it? I don't like at all this whole procedure; on the contrary it seems to me that giving file "owner" rights to rename its file would be way more practical. Has anyone already proposed this?

Anyway, can someone remove the "-" from my uploads?

Ogoorcs (talk) 19:07, 31 August 2018 (UTC)

Seems to work just fine for me. I'm not sure what the issue is. WD supports a pretty wide range of characters, including non-Latin characters and even emojis. GMGtalk 19:15, 31 August 2018 (UTC)
@Ogoorcs: Were you using Upload Wizard or some other upload tool? (If some other upload tool, what tool?) Is it possible that something else was actually wrong with the filename (e.g. two consecutive spaces, which isn't allowed)? - Jmabel ! talk 22:39, 31 August 2018 (UTC)
I used Firefox for android at first and Chromium for linux later. A popup reporting that files containing ":", "-" and other special characters could not be added appeared. Now it works. This is strange. Sorry for being a silly. Ogoorcs (talk) 01:56, 1 September 2018 (UTC)
Maybe you just left the "File:" prefix at the start of the name. That always gives an error: File names are not allowed to contain characters like colons or slashes. Only paste the file name after "File:", please. I'm not sure why the Wikidata interface doesn't just discard the prefix. --ghouston (talk) 05:03, 1 September 2018 (UTC)
yeah, wikidata can be a little wikicode-ish. but they are very friendly at wikidata:Wikidata:Project chat. Slowking4 § Sander.v.Ginkel's revenge 12:57, 3 September 2018 (UTC)

September 01

Is this a free license?

Just curious what other people think about these photos uploaded by User:Tyw7, for example this one and others in their Gallery. It doesn't strike me as a free license. Sionk (talk) 12:18, 1 September 2018 (UTC)

It seems OK to me except for this part: "If you are a (commercial) publisher and want to relicense my pictures please email me first". There should be no other conditions. 4nn1l2 (talk) 12:28, 1 September 2018 (UTC)
Well the license tag was run pass a few admins first at the time of creation. They said it's OK since that line meant that if they are relicensing ie wanting the picture under a different license, they should contact me first. --Tyw7  (🗣️ Talk to me • ✍️ Contributions) 12:41, 1 September 2018 (UTC)
(Edit conflict) How's the rewrite? PS if you do reply to me, please ping me as I'm not actively on this Wiki. --Tyw7  (🗣️ Talk to me • ✍️ Contributions) 12:46, 1 September 2018 (UTC)
I see nothing wrong with stating if a commercial publisher wants to relicense to email the photographer/user, since a commercial publisher may want to use a photograph in such away it would in fact be in violation of the license. Bidgee (talk) 12:45, 1 September 2018 (UTC)
@Tyw7: I think the new wording [2] is more clear. The word "relicense" was somewhat vague in my opinion, at least to my non-native understanding. 4nn1l2 (talk) 12:59, 1 September 2018 (UTC)
vastly free-er than User:Fir0002/credits. if you want to change consensus about custom user license reuse messages, go for it. Slowking4 § Sander.v.Ginkel's revenge 12:48, 3 September 2018 (UTC)

16:47, 3 September 2018 (UTC)

September 04

Photo challenge July Results

Concert photography: EntriesVotesScores
Rank 1 2 3
image
Title Singer - Tambours du Bronx.jpg Zapato 3 - La Ultima Cruzada Tour 2012. Merida, Venezuela. The Rolling Stones
Author Roumpf Davevzla Les Zg
Score 24 12 11
Telephones: EntriesVotesScores
Rank 1 2 3
image
Title Phone of George Clemenceau during World War I Telephone booth in the land. Alarm phone on the_platform of_a
suburban train station, France
Author Ibex73 Roumpf Ibex73
Score 19 18 15

Congratulations to Roumpf, Ibex73, Davevzla and Les Zg. -- Jarekt (talk) 03:37, 4 September 2018 (UTC)

Please help to remove profile pics from tweets

User:BevinKacon is doling out (speedy) deletion tags for PD-text tweets faster than I can fix them. I would consider these profile pics to be DM anyway (nobody cares about them), but BK keeps trying to get them deleted.

Please help to blur/crop them. I have done many but just can't keep up like this. - Alexis Jazz ping plz 21:47, 27 August 2018 (UTC)

  •  Question: Photos and other images can be easily blurred, but isn’t the text copyrighted, too? -- Tuválkin 22:40, 28 August 2018 (UTC)
    Yes. Seeing as you wouldn't be able to put one of those tweets on a t-shirt and sell it without compensating the author I don't see how blurring the photos is really going to do anything. --Majora (talk) 22:42, 28 August 2018 (UTC)
    Interesting read: Brock Shinen: Twitterlogical: The Misunderstandings of Ownership, including these lines: "The question is not: Are Tweets Copyrightable. The question is: Is This Tweet Copyrightable. The copyrightability of Tweets is not dependent on the fact that they are Tweets. Rather, it’s dependent on the analysis of the Tweet in question. The all-encompassing response that all Tweets are either protected or not protected is misguided. The real response is that it depends. However, when you analyze most Tweets, they would never individually pass copyright muster." Vysotsky (talk) 23:19, 28 August 2018 (UTC)
    I fully understand that. My mistake for using a black-and-white statement like that. Such things are rarely so clean cut. It would certainly depend on the ability to type something that would meet the threshold of originality in some way. Is that possible in 280 characters? Of course it is. But as there are no court cases that I know of that have dealt with this there actually isn't a clear answer (one lawyer's opinion notwithstanding). Without a clear answer we can only make analogies. For example, taking a screenshot of a tweet and taking a photograph of a few lines of a book are the same idea in my mind. The latter is clearly protected and can't be hosted here, without the underlying work being under a free license. Logically the former should follow the same ideas provided it is unique enough to merit TOO. Even {{PD-text}} has a disclaimer in small text at the bottom regarding this ambiguity. --Majora (talk) 23:36, 28 August 2018 (UTC)
    Well, two lawyers. Well, three. Vysotsky (talk) 00:12, 29 August 2018 (UTC)
    Most experts agree the response should not be an all-or-nothing answer, but rather "it depends.". Directly from one of your sources. Like I said, my original black-and-white answer shouldn't have been said. My mistake. Nor should it be black-and-white the other way. --Majora (talk) 00:15, 29 August 2018 (UTC)
    @Tuvalkin: thanks! @Majora: quite some tweets that get uploaded here come from US federal government employees and are work related. The list above happens not to have many of those (except the DR) because I had already blurred/cropped those myself. Besides that, most tweets are not eligible for copyright protection. If a notable person tweets "I'm sorry for your loss @username" that may be relevant and within our scope (depending on who said it to who), but it can never be copyrightable. Jokes also can't be copyrighted. If I say "SILENCE! I KILL YOU!" you probably know what I'm referring to. Alas, Dunham can't sue me (and win..) if I put those four words on a t-shirt. The wording chosen for a joke can be eligible, but in that case the joke must be original (many are recycled) and the wording sufficiently original to be eligible. In 280 characters, this won't be too common. For old tweets with 140 characters, it's nearly impossible. The difference between books and tweets is that a photo of a few lines from a book would be a part of a larger work. In the case of tweets, the tweet usually is the work. Also, if I take a photo of the page of a book and the only readable text is "I'm sorry for your loss", I don't think the author will be able to come after me. I agree it is complicated, but in the end I agree with "However, when you analyze most Tweets, they would never individually pass copyright muster". I suppose the follow-up question might be "so I can import all the tweets some person has ever written and it's all PD? I don't know, it depends on the case, but that's very hypothetical. If any user actually attempted such a thing I would generally discourage it. If the tweets can be combined to become bigger than the sum of their parts, copyright may be possible. But we won't delete any photo that shows a single word from a book either. Unless someone uploads photos of every single word in a book. But that doesn't mean the author can copyright individual words. - Alexis Jazz ping plz 04:27, 29 August 2018 (UTC)
    I agree with all that, but most tweets are out of scope. They are not educational in themselves, and not needed to illustrate anything. So the question is moot at best. Regards, Yann (talk) 09:34, 29 August 2018 (UTC)

Please also note this category: Category:Soroush messenger comments on Google play. Some files have been used in a FAWP article just to "discredit" the messenger by user comments: سروش (پیام‌رسان). I think they are out of scope. What do you think? 4nn1l2 (talk) 09:58, 29 August 2018 (UTC)

  • I note that the list includes some examples of fraudulent claims of "own work" and false claims of copyright ownership. (One does not become the copyright owner of what someone else wrote by taking a screenshot of it!) Any tweet images uploaded by anyone other than the original twitter author would need to have an explanation for why it is permissible to upload as free licensed - for example if it is the work of a US Federal Gov't employee written as part of their Gov't job, or the text is public domain for one reason or another, etc. Any screenshots of tweets taken by someone other than the author without explanation for why the original tweet is free licensed or PD should be deleted just like any other derivative work copyright violation inappropriately uploaded to Commons. -- Infrogmation of New Orleans (talk) 13:41, 4 September 2018 (UTC)

August 28

Cat-a-Lot performance degraded

Since when did Cat-a-Lot start asking for edit summaries? I've just given up on a batch of work, because the pop-up dialogue was slowing me down so much. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 09:08, 30 August 2018 (UTC)

I haven't seen that specific problem, but a day or two ago I was trying to move 200-300 files, and it kept hanging after 50 or so. --Auntof6 (talk) 09:14, 30 August 2018 (UTC)
I have seen the long pause after file 43, but the tool resumed after a while. --Jarekt (talk) 11:14, 30 August 2018 (UTC)
I have not seen this problem either. Please take screenshots, it may be worth raising a Phab ticket to capture the bug. -- (talk) 09:17, 30 August 2018 (UTC)

: I've had Cat-a-Lot ask me for edit summary. I'm not sure, but I think it happened for me when the pages weren't fully loaded. —⁠andrybak (talk) 17:09, 30 August 2018 (UTC) My bad, I confused HotCat and Cat-a-Lot. —⁠andrybak (talk) 13:39, 4 September 2018 (UTC)

I have seen this problem too: if you do more than 100. It keeps hanging around 46. If you do around 60 it also stops a moment at 46 but proceeds after a while. a Rudolphous (talk) 15:55, 31 August 2018 (UTC)
I have had exactly this problem as well. Smaller sets seem to work through okay, but larger groups (select all routinely means ~200 selected items) hang at ~40 to 52 (screen shot added to discussion). Josh (talk) 11:41, 4 September 2018 (UTC)

No source/no permission templates

I see the templates {{No source since}}, {{No permission since}}, and {{No license since}} often added to files that clearly have this information, but where the person including the template does not seem to believe this claim. I always was under the impression that the templates in question were for cases where the information is missing completely, and for other cases {{Speedy}}, {{Copyvio}}, or regular DRs are appropriate.

I don't like this practice and have started to convert such requests to regular DRs. As an admin I often have to divine why the tagger thinks a file does not have the appropriate information, especially since the former templates do not allow to include additional comments. This also has the unfortunate effect that often files that should have a proper DR get automatically deleted after a week without much scrutiny by the community. On the other hand it increases workload and time-to-delete when those files get converted to a regular DR after a week.

What's the current consensus on that? Sebari – aka Srittau (talk) 10:57, 2 September 2018 (UTC)

@Srittau: The templates you mentioned represent should represent a 100% lack of a particular Information field. What if we had {{No confidence since}} and {{Ncd}} templates to express a lack of confidence in textual format, with a 2 week timeline? Would that work for you?   — Jeff G. ツ please ping or talk to me 12:45, 2 September 2018 (UTC)
@Jeff G.: I think the existing templates are sufficient, but should be applied correctly. In case of non-confidence, I'd prefer a speedy deletion request (in obvious cases, like googled images) or a regular DR (in other cases), since the latter are much better suited for community discussion than files that simmer in another maintenance category until they pop up in the admin backlog. Sebari – aka Srittau (talk) 12:51, 2 September 2018 (UTC)
@Srittau: Okay, I promise to apply them correctly.   — Jeff G. ツ please ping or talk to me 13:12, 2 September 2018 (UTC)
sorry, what is the standard of practice for proper use? what is the warning if improperly used? where is the discussion board? where is the "license disputed" maintenance category, and workflow. without these changed processes, the "no source" tags will remain discredited from widespread misuse. it is used as a prod delete with a sword of Damocles, not much collaboration there. Slowking4 § Sander.v.Ginkel's revenge 21:00, 2 September 2018 (UTC)
The main disadvantage of {{No source since}} and {{No permission since}} is that these tags somehow move away the responsibility to explain what's wrong from nominator to processing admin. I think if we would get rid of these two tags completely and demand a regular DR for such cases, we would have a lower workload for the admins and a higher quality of their decisions. The {{No license since}} works fine IMHO, I hardly see files with a license template having this tag. Jcb (talk) 13:05, 2 September 2018 (UTC)
@Jcb: I think removing those tags would be premature.   — Jeff G. ツ please ping or talk to me 13:20, 2 September 2018 (UTC)
Yes, of course some steps are needed between one user voicing an opinion on these two tags and the actual deletion of these tags. Only if a discussion would lead to a community consensus that these two processes are obsolete, we can move forward to dismantle them. Jcb (talk) 13:26, 2 September 2018 (UTC)
Some ideas, mostly brainstorming:
  • Allow {{No source since}} and {{No permission since}} only for autopatrolled users (or possibly some form of Extended confirmed, but Commons afaik doesn't have that group currently)
  • The same for {{Copyvio}} when the file is not claimed as "own work".
  • Copyvio must include a good rationale (like "see watermark" in some obvious cases, "Disney characters" or "movie screenshot", what is a good rationale varies) or a valid link. A Google image search link does not qualify, the link has to show a higher quality version (beware of upscales) or prove the image was online before it was on Commons. Maybe the gadget could be made into some sort of multiple choice thing.
  • Make some users completely immune to speedy deletions (including no source/permission/license). This would primarily include users who are known to create their own work and understand copyright as well as deceased Wikimedians who can't possibly defend their files.
  • Quite a while ago I saw a photographer start a DR as an IP-user. The report was hazy, didn't include the name of the photographer and without intervention may not have resulted in deletion. We may consider a guided form so copyright holders, even without an account, can create sufficiently informative DRs.
Some of these may be possible with an abuse filter, others would require some more substantial changes. These are just ideas, comments are welcome. - Alexis Jazz ping plz 13:58, 2 September 2018 (UTC)
I share the concerns from Srittau by the way. If some clown tags one of my original photos with "no permission" or "no source" while I'm on vacation, wikibreak or retired I wouldn't be surprised if they end up in the bin. - Alexis Jazz ping plz 14:56, 2 September 2018 (UTC)

I think the status quo works, but we need to add some documentation to {{Npd}}, {{Nsd}} and {{Nld}} templates and explain when to use each tag. Here are my suggetions:

  • {{Npd}} can be used when the file needs to go through OTRS including 1) the uploader has not claimed own-work and their username differs from the name given in the author field; 2) the uploader has claimed own-work, but metadata suggests otherwise; 3) the uploader has claimed own-work, but the work has been published before somewhere else (whether obviously implied or explicitly expressed in the description itself). If there are other things to consider (such as size, resolution, username, etc) a normal DR should be instigated instead.
  • {{Nsd}} can be used when 1) no source is given at all; 2) the given source is utterly vague such as "Google"; 3) the image cannot be found in the given URL.
  • {{Nld}} can be used when 1) the uploader has claimed own-work but has not indicated a license at all; 2) the uploader has not claimed own-work and has not chosen a suitable copyright tag and it is not obvious which tag should be used to fix it yourself.

4nn1l2 (talk) 14:11, 2 September 2018 (UTC)

@4nn1l2: I also think {{Npd}} should not be used when there is no realistic chance we will get OTRS permission for the file.
As for previously published "explicitly expressed in the description itself", https://en.wikipedia.org/w/index.php?title=Wikipedia:File_Upload_Wizard&withJS=MediaWiki:FileUploadWizard.js ("This is a free work.", "This file is entirely my own work.") is not smart:

Publication: (text field) Please indicate here if you have previously published this item elsewhere, e.g. on your own website, your Flickr or Facebook account, etc., providing a link.

It might as well read "if you want your file to be deleted, enter something in this field". - Alexis Jazz ping plz 14:56, 2 September 2018 (UTC)
@Alexis Jazz: I don't understand the enwiki link (I hardly edit there). File:Erythrina Mulungu.jpg is an example of the situation where the uploader themself has expressed that the file has been previously published elsewhere. I think adding another field to ask the uploader about the status of previous publication is a good idea. I have seen many users who have provided this information themselves manually in the description field.
I think if the evidence of permission cannot be provided even using OTRS system, then the file should not have been uploaded in the first place. However, I agree that {{Npd}} should not be misused to get rid of unwanted/problematic images. 4nn1l2 (talk) 16:14, 2 September 2018 (UTC)
@4nn1l2: I disagree with many of your examples. This is exactly the admin guessing game I was alluding to. {{Npd}} is completely insufficient when the file needs to go through OTRS. How will the uploader know about this option? How will the admin know the user has been informed? How will the admin know that this is what the tagger meant? Same with all the other examples. If the metadata does not match the author, say so in a DR to give the uploader a chance to explain themselves and to keep the admin from guessing. If the file has been previously published, either do the same or tag as copyvio, if it is obvious the file was taken from the web. When a source like "google" is given, tag the file as a copyvio, don't let it simmer for 7 days in a maintenance category nobody looks at. If the image can't be found, file a DR, since this is not an obvious copyvio and how should the admin know that this is what you mean by tagging it with {{Nsd}}? Sebari – aka Srittau (talk) 16:34, 2 September 2018 (UTC)
@Srittau: I don't disagree with your proposal, but I think that extra DRs will decrease the overall speed of maintenance and will create heavier backlogs as few users monitor new uploads and fewer admins close DRs. Writing per se is demanding especially for non-native speakers of English. We need to reach a proper balance between speed and diligence. However, I agree that one should always festina lente. So, again, I don't disagree with your proposal, but I'm inclined to suggest whatever decision is made, it should be clearly reflected and explained in the documentation pages of the respective templates. I was hungry for such information when I started tagging activity on Commons, and I know I am not the only one. 4nn1l2 (talk) 17:27, 2 September 2018 (UTC)
I was under the impression the templates {{No source since}}, {{No permission since}}, and {{No license since}} were for cases where the information is partially or completely lacking leading to the files ability to be hosted on commons to be called into question. Not necessarily entirely missing. As was said, {{Speedy}}, {{Copyvio}}, or regular DR as appropriate any other files in question.
A user has to enable Quick Delete under gadgets to use the three previous templates in question so unless misuse by new uses is readily documented, in which case by all means, I don’t feel restrictions are necessary. Though little harm would result from having them in place.
I see Jcb’s point. When a user tags a file with {{No source since}} or {{No permission since}} I think in many cases it’s simply a file the tagging user feels isn’t allowed to be hosted on commons and rarely on files that I come across does the issue look like a source or permission issue. There is a shift of responsibility being moved to the closing admin in that case. It directly related to how this conversation began and tagged files being converted to proper DR’s.
If more files tagged were nominated with proper {{Speedy}} or DR’s with legitimate stated reasons by the nominating user I would imagine we would higher quality of decisions on the files and it would be easier for the closing admin. -- Sixflashphoto (talk) 15:22, 2 September 2018 (UTC)
exactly, "no source" is used to bypass DR to get a speedy after 7 days. cannot rely on a consensus process to delete files, which is the point of the exercise. why don't we test the thesis, by having a "no license" holiday, and requiring a DR or speedy from the edtors. it would require more collaboration, (such as it is). would the world come to an end? Slowking4 § Sander.v.Ginkel's revenge 21:07, 2 September 2018 (UTC)
@Slowking4: At least en:WP:PROD has a freeform text reason.   — Jeff G. ツ please ping or talk to me 21:12, 2 September 2018 (UTC)
yeah, good point, you do have to guess what the dispute is, since there is no requirement nor inclination to say just what you want the uploader to do. it is the limbo method of quality improvement. just keep iterating until the deletionist is satisfied. Slowking4 § Sander.v.Ginkel's revenge 12:42, 3 September 2018 (UTC)
 Comment I sometimes use {{Npd}} as it is much faster than speedy deletion (one click only). I usually use it when the work is obviously not own work, but there is a remote chance to be fixed. Regards, Yann (talk) 17:09, 3 September 2018 (UTC)
@Yann: I have multiple two-click speedies among the AjaxDeleteExtraButtons in User:Jeff G./common.js.   — Jeff G. ツ please ping or talk to me 02:59, 4 September 2018 (UTC)
Interesting. At least the COM:NETCOPYRIGHT option should be included in the gadget. Regards, Yann (talk) 03:19, 4 September 2018 (UTC)

I find 'no source' sometimes problematic. In the clarification text 'postcards' should be added as a posibility. As much detail should then be included. (wich editor etc) 'personal collection' should never used as source. Sometimes among the old postcards I find lose pictures with no clue as to the origin. This could taken removed from an old family album or cut out of some periodical. I see no reason that this 'anonymous' should be treated differently as old postcards.Smiley.toerist (talk) 11:03, 4 September 2018 (UTC)

September 03

Could someone help me to describe this picture and help me to put it into the correct Categories? I do not really know what that is exactly named nor what this is. But I am very sure, that some people will see it. I am sorry, normally I speak german, my English is not very well. Kann jemand mir helfen, dieses Bild korrekt zu beschreiben und den korrekten Kategorien zuzuordnen? Ich kenn mich mit Militaria nicht aus. Jedenfalls erscheint es mir interessant für viele Menschen. Zabia (talk) 04:48, 4 September 2018 (UTC)

✓ Done I have added a description in German and English. De728631 (talk) 11:48, 4 September 2018 (UTC)

Duplicate categories

Hi, Here are duplicate categories:

What's the proper name? Thanks, Yann (talk) 09:24, 4 September 2018 (UTC)

Assuming the intent is to include things that are along the lines of museum wall texts, I'd lean toward Category:Signs at Agra Fort, since we don't use the term "wall texts". Also:
  • some of the images categorized here probably don't belong in these categories, e.g. File:Agra fort- Diwan i aam.jpg.
  • Why are these wall texts OK in copyright terms?
- Jmabel ! talk 15:09, 4 September 2018 (UTC)
The Agra Fort is managed by the Archaeological Survey of India, so the texts are under {{GODL-India}}. Thanks for the answers. Regards, Yann (talk) 16:29, 4 September 2018 (UTC)
This is good news then. Thank you, Yann, for clarifying this. As to the categories, I would also tend to "Signs at Agra Fort" with a redirect from "Agra Fort description". De728631 (talk) 16:42, 4 September 2018 (UTC)
✓ Done Yann (talk) 16:45, 4 September 2018 (UTC)

Build a MAMCS bot?

Hello there. The artist Alfred Jungbluth is in the public domain (he died in 1914) and the Strasbourg Museum of Modern and Contemporary Art (MAMCS) holds 665 works by them, which are stored in that illustrated database: [13]. Nobody, including me, would want to transfer all these files manually, but if somebody could build a bot to do it, that would be very useful! The work of that MAMCS-bot could then be extended to the works of Gustave Doré ([14]), and of Lothar von Seebach ([15]). And that could only be the beginning, since the database of the MAMCS is just one among the many of the Videomuseum network, which also comprises museums from Grenoble, Lyon, Nantes, Nice, Paris (of course), Toulouse etc. In fact, there is a treasure trove of well over 5,000 works by artists in the public domain to be explored there! Will somebody make a start? --Edelseider (talk) 19:59, 4 September 2018 (UTC)

Transparent clock (1 min interval set) is an example of an image set. There should be wide consensus that its images should not simply be moved to a parent category, where they would be mixed with those in similar image sets.

After moving many image sets from broad subject categories to categories of their own, I have created this page about the topic. Currently it is an essay, but I think some of it is common sense and should become a guideline. Feel free to leave your thoughts on the discussion page.
TL;DR: Scroll to the examples on the bottom of the page, and ask yourself if the images in these sets should be mixed with other images in a parent category. (Example: The images in this and this set have originally all been directly in Star polygons.) Greetings, Watchduck (quack) 12:38, 1 September 2018 (UTC)

  • (You need to read about the difference between categories and galleries.) Multiple categories can be created to account for all aspects of each media file. F.i., there could/should be a separate category for only and all of those clockfaces, regardless of other categorization, such as categorization by time. -- Tuválkin 13:56, 1 September 2018 (UTC)
(Galleries have nothing to do with this. This is about categorisation and nothing else.)
My point is that multiple categories should be created to account for all relevant aspects of each file, and that being part of the same image set is one of these aspects.
I am suggesting that bundling image sets in dedicated categories should become a guideline, because throwing all such images directly in broad categories creates chaos and destroys information. Watchduck (quack) 14:17, 1 September 2018 (UTC)
  • (Fair enough. I would gladly get rid of the gallery namepace altogether, once the very few and far in between useful content is moved to the relevant category.) Okay, understood. But this is what’s already in COM:CAT — if not expressed at least implied. Then again many of us, including some admins, do have weird ideas about categorization, and often the kind of destructive dissimination or even uncategorization you mention does occur — and in that case, go ahead. -- Tuválkin 15:56, 1 September 2018 (UTC)
Yes, I would just like to make it explicit. (More importantly I would like to make bundling the default for mass uploads.) Is there an official way to propose a guideline? (PS: Your indentation is weird and unsemantic.) Watchduck (quack) 11:49, 2 September 2018 (UTC)
(Good to have something to disagree about. Feel free to bring it up on the Commons:Talk page guidelines talk page.)
Here and here (cat for discussion) and originally here (user talk) is a real life example of why I would like explicit guidelines on image sets. For some reason Arthur Baelde is convinced that his images should be directly in subject categories. (He does not tell me if he wants to empty the other sets as well.) Watchduck (quack) 12:53, 5 September 2018 (UTC)

Index categories

There appear to be a couple of different approaches to index categories (Topic by Sort criteria) used on Commons and I cannot find any consensus on which is the right way to go forward. One is rather strictly just to include categories which intersect the Topic with a named Criteria that is then sorted and grouped by said Criteria. The other is to add on top of this an entire lattice-work of sub-indices representing sub-categories of the Topic. These two seem to exemplify the different approaches:

  1. Category:Products by manufacturer - very limited, contains:
    1. Categories which combine the topic of "products" with a named "manufacturer" (e.g. Category:Kodak products or Category:Products of Bayer), sorted by "manufacturer" name ("Kodak" or "Bayer") and hence grouped by first letter of the manufacturer name ("K" or "B")
  2. Category:Objects by color - has a three different types of categories listed:
    1. Categories which combine the topic of "objects" with a named "color" (e.g. Category:Blue objects), sorted by "color" name ("Blue") and hence grouped by first letter of the color name ("B")
    2. Index categories for sub-categories of the topic "objects" in the format "Type of object by Color" (e.g. Category:Carpets by color), sorted with a space key + topic so they appear as an un-grouped list ahead of the "Color objects" categories, in order of topic.
    3. Double-intersection index categories (e.g. Category:Objects by color by location), sorted with a space key so they appear in the un-grouped list at the very top of the page.

It seems rather natural to sort something like Category:Balloons by color under Category:Objects by color, but looking at the result which is sort of a double list, especially one that obscures the primary items of the list (the "Color objects" categories), it maybe is not the greatest idea. Also, while it seems natural to sort it that way, it is not a use case that makes sense. If I am looking at Category:Fire hydrants and want to browse them by color, Category:Fire hydrants by color is listed there, so I do not need it listed in Category:Objects by color to find it. However, if I was for some reason to look for them in Category:Objects by color, I still would not find them because I would look under "F" for "Fire hydrants" and would come up empty. Either way, it seems that having this be done hap-hazard category to category is definitely not optimal, so I am looking for any discussion, guidelines, or consensus that can be looked to for help, or if anyone has any input on which method is the better way to do these types of categories. Thanks! Josh (talk) 11:34, 4 September 2018 (UTC)

After twelve years of editing and uploading here, I've never seen any guidelines on this subject, so I doubt you'll find any. As you note, when such a category has different classes of subcategories, it's normal to sort them separately (so you don't have "Red objects" sitting between "Rats by color" and "Roses by color"), but that's the only common practice that I've seen. Seeing that this isn't a problem unless multiple classes of subcategories exist, I don't think we ought to introduce any guidelines, but we'd do well to create more meta-subcategories, e.g. the "Category:Objects by color by location" that you note. We could have "Objects by color by type" and "Colored objects by color", for example. Nyttend (talk) 22:43, 4 September 2018 (UTC)
@Nyttend: Thanks for the input. I share your experience as far as seeing them gathered like that when you have both types of sub-cats...I certainly wouldn't want to see Red objects in between Rats by color and Roses by color, that would seem silly. I guess meta-metas might be the way to go if there gets to be too much clutter? I'm certainly not looking to get into a policy-setting effort, but I've seen both methods made out to be the one and only right way to do it, but I guess that is just some folks opinions. I wanted to make sure I'm not missing some neon direction sign somewhere. Thanks! Josh (talk) 07:45, 6 September 2018 (UTC)

500px geocoords

So far, every single set of geocoords from 500px that I've checked (over a dozen so far) have been incorrect, typically by at least 500 metres, often more, in one case by half a continent. When I can work out the correct geocoords, I'm replacing them, but when (for example) a closeup of a flower has geocoords in the middle of a large body of water, it's helpless, and I'm just deleting that location info. Are others having a similar experience, or have I just hit a particularly bad bunch? (I've been dealing with the ones that were placed in Category:Seattle). - Jmabel ! talk 17:51, 4 September 2018 (UTC)

  • I’ve only seen a few, but they do often seem suspect. Offhand I’ve noticed:
– (approximate) city centre for an image from somewhere else in or around the city,
– (approximate) object location instead of camera location for a distant landscape, and
– location of a building (that might be the photographer‘s home or office) for an image evidently taken in a park elsewhere in the city.
I left them alone at the time, but if it’s agreed that geodata from this site tend to be unreliable, I’ll be more proactive in correcting (if possible) or removing them. BTW I’ve also seen quite a lot of Unsplash coördinates of the first kind: precise city-centre locations for subjects from anywhere around the city.—Odysseus1479 (talk) 18:04, 5 September 2018 (UTC)

Museu Nacional, Rio de Janeiro

The Portuguese Wikipedia has placed a sitenotice advertising pt:Wikipédia:Comunicado Museu Nacional, which is basically a plea for anyone and everyone with pre-fire images of the museum's collections to upload those to Commons as a small-scale method of preservation. We're an international project. What if we did the same thing? I'm imagining a bilingual banner (Portuguese and Special:Mylanguage, if possible) reading something like "Upload your images of artifacts from the Museu Nacional, Rio de Janeiro, and preserve them for the world". Nyttend (talk) 22:35, 4 September 2018 (UTC)

I've placed a note in Spanish Café (https://es.wikipedia.org/wiki/Wikipedia:Caf%C3%A9/Archivo/Ayuda/Actual#Museo_Nacional,_R%C3%ADo_de_Janeiro). It cannot harm! B25es (talk) 08:01, 5 September 2018 (UTC)

September 05

This 2014 image was uploaded by a reviewer but the permission link appears to be non-functional. If someone can access the permission and verify the license, please feel free to review this image...or tag it as 'no permission'. Thank You, --Leoboudv (talk) 08:48, 5 September 2018 (UTC)

I think the fact that it had been sitting at Wikipedia for one year before uploaded here is enough evidence to let it pass without additional review. If De728631 (talk) 15:02, 5 September 2018 (UTC)

Identifying two sailing ships

In Aarhus there where three sailing ships. One I could identify, not the other two. One has a logo 'J/L' and the other one has no name.

the nameles one, also File:Aarhus harbour 04.jpg, File:Aarhus harbour 05.jpg, File:Aarhus harbour 06.jpg
The 'J/L' one, also File:Aarhus harbour 02.jpg

Smiley.toerist (talk) 10:08, 5 September 2018 (UTC)

Can see IMO number in first image 5128417 GEORG STAGE https://www.marinetraffic.com/en/ais/details/ships/shipid:157020/mmsi:219417000/imo:5128417/vessel:GEORG_STAGE Oxyman (talk) 10:23, 5 September 2018 (UTC)
For 'J/L' ship: Could be found on the port's marine traffic page. —⁠andrybak (talk) 11:02, 5 September 2018 (UTC)
"J/L" is the logo of J. Lauritzen A/S, a Danish shipping company. I would guess that it's the "Lilla Dan".[16] Lupo 14:30, 5 September 2018 (UTC)
Yes, this is Lilla Dan, compare File:Lilla Dan1.jpg. --De728631 (talk) 14:51, 5 September 2018 (UTC)

Ship in Grenaa

What kind of ship is this?

Smiley.toerist (talk) 20:53, 5 September 2018 (UTC)

This goes into Category:Jack-up crane ships. I'll create a category for the ship name though. De728631 (talk) 21:00, 5 September 2018 (UTC)

September 06

Danish house

What is going on?

Smiley.toerist (talk) 20:51, 5 September 2018 (UTC)

Looks like the resident is an avid collector of maritime stuff. De728631 (talk) 21:15, 5 September 2018 (UTC)

@Smiley.toerist: I've found a Danish newspaper article from 2010 about it. Two years before the man in the house began collecting things to get out of a depression. The things has to be a generation old as well as good or fun. The exhibition in the garden is there to make people curious and stop up for a talk. The things are not for sale however. --Dannebrog Spy (talk) 14:19, 6 September 2018 (UTC)
@Smiley.toerist: Yes, there is a lot of maritime stuff, but the roof looks like it is in need of repair.   — Jeff G. ツ please ping or talk to me 14:27, 6 September 2018 (UTC)
I can't tell where the collection ends and the mural on the next building begins :) Josh (talk) 18:56, 6 September 2018 (UTC)

My common.js showing up in undesired categories

I use my User:Joshbaumgartner/common.js file to add useful snippets to my editor, but in doing so, now it shows up in undesired categories. e.g. I have a snippet that quickly adds Category:Aircraft by registration with one click. This is very useful and efficient. However, my commons.js now shows up in that category as well, which is not useful or in any way good. It is also somehow recognizing some of the templates that are in the snippets even though they are separate pieces of the code! I use the "includeonly" tag for templates to prevent this, but it does not work on the script page. Does anyone know an elegant way of keeping my script page out of this category while not reducing the utility of the snippets? Josh (talk) 08:05, 6 September 2018 (UTC)

Try to add some js comments into the middle of the category names like: [[Cat//Comment//egory:Aircraft by registration]]. I hope this will disable the categorization of you JS-script. --JuTa 11:19, 6 September 2018 (UTC)
@Joshbaumgartner and JuTa: I have been having a similar problem with en:User:Jeff G./JWB-settings.js being categorized, as documented at en:User talk:Jeff G./Archives/2018/August#Userpage in category redirect (and before that at en:User talk:Jeff G./Archives/2018/August#categorisation of JWB-settings.js further up the page) since 13 August.   — Jeff G. ツ please ping or talk to me 13:26, 6 September 2018 (UTC)
Perhaps [//<!-- -->//[Category:Aircraft by registration]//<!-- -->//] could do the job to disable the double brackets for Mediawiki. --JuTa 13:55, 6 September 2018 (UTC)
@JuTa and Jeff G.: Figured it out! Add nowiki tags to the start and end of the JS and it appears to eliminate all of the problems of unintentional template transclusion and categorization while not impeding function:
//<nowiki>
//
//...your JS here...
//
//</nowiki>
Thanks for the help and we all keep learning! Josh (talk) 18:49, 6 September 2018 (UTC)

Read-only mode for up to an hour on 12 September and 10 October

13:33, 6 September 2018 (UTC)

Structured Data on Commons - Structured statements

Mockups of structured licensing and copyright statements on file pages are posted. These potential file page design changes are specific to Commons only, file pages on other wikis will not change. Please have a look over the examples and leave your feedback on the talk page. Keegan (WMF) (talk) 15:54, 6 September 2018 (UTC)

September 07

How to upload a new version of the file that is larger than 100MiB?

Here's the problem, let's look at File:World Strategy Conference - 13.04.2015 Opening Ceremony (eng).ogv, currently it has borders, I would like to remove the borders and reupload that file, but I also think that we should keep the history, so "Just nominate it for deletion and then after it is deleted reupload" is not the option. I also think it will be silly to upload under the different name just because of this issue. If I try to upload new versions of files, I can get to about 20 or maybe 30 MiB until the connection times out. And even if that were not the case, there is a hard limit of 100 MiB on the basic uploader that is used to put new versions of files. This is something to do with the way server is configured. File uploader and every took I checked refuses to overwrite the file if it already exists. Am I missing something simple? Is there a checkmark somewhere saying "If you cannot find this checkmark while trying to overwrite, you are an idiot"? How would I go about doing this? ℺ Gone Postal ( ) 07:39, 7 September 2018 (UTC)

@Gone Postal: Have you tried using Help:Chunked upload#Chunked uploads with script bigChunkedUpload?   — Jeff G. ツ please ping or talk to me 11:45, 7 September 2018 (UTC)
Thanks, I will give it a try. ℺ Gone Postal ( ) 12:01, 7 September 2018 (UTC)
@Gone Postal: You're welcome. Chunk size should be no more than your "about 20 ... MiB", and no more than 1% of filesize.   — Jeff G. ツ please ping or talk to me 12:33, 7 September 2018 (UTC)
It worked like a charm on a smaller, 40MiB file. I will give it a go on 100+MiB a little later. This should really be linked to from somewhere on the "upload new version" page. ℺ Gone Postal ( ) 14:44, 7 September 2018 (UTC)

I do not wish to edit war. I have reverted removing of categories once, but can somebody else please look at this and see what is happening. It is a featured image. ℺ Gone Postal ( ) 22:14, 7 September 2018 (UTC)

I have reverted my revert for now, since at this moment it appears to be a case of poorly communicated recategorisation attempt. ℺ Gone Postal ( ) 22:24, 7 September 2018 (UTC)
Agreed on the first point. I would prefer macaque to monkey, though (maybe even change Sulawesi monkey to Celebes macaque?), and the present title should probably be kept as a redirect because of its popularity in the media (and that WMF conference).—Odysseus1479 (talk) 03:46, 8 September 2018 (UTC)

September 08

Accidental categorization by template

It looks like some (Lua-based?) template is incorrectly putting a ton of categories in Category:Institution template home categories. - Jmabel ! talk 15:24, 8 September 2018 (UTC)

It looks like Module:Institution. @Jarekt: BMacZero (talk) 17:16, 8 September 2018 (UTC)
Jmabel and BMacZero, the logic of how Category:Institution template home categories is added did not changed for years. Is this a new issue or something that was wrong for a long time? --Jarekt (talk) 02:39, 9 September 2018 (UTC)
Fixed, I found the issue. It was broken since Lua version of Institution was written. Thanks. --Jarekt (talk) 02:54, 9 September 2018 (UTC)

Dating London postcard

I suspect this could even be before World War I.Smiley.toerist (talk) 11:06, 4 September 2018 (UTC)

According to Alamy, this shot was taken in 1910, so it seems you're right. De728631 (talk) 11:19, 4 September 2018 (UTC)
The paddle steamer Pepys operated on the Thames between 1905 and 1912. Broichmore (talk) 10:22, 9 September 2018 (UTC)

How to deal with factually incorrect title from a GLAM

File:Bonney-Watson Building, funeral parlor, 1702 E Broadway at Olive Way, Seattle (CURTIS 1071).jpeg Bonney-Watson was at 1702 Broadway, not 1702 E Broadway (which is about a mile away) and at the corner of Broadway and Olive Street, not Olive Way (which is 2 blocks away). If this were from a random contributor I'd just move the file and make the relevant edits to the content, but do we need to somehow preserve (even wrong) information from a GLAM in a case like this? Should/ can we somehow pass back to them that there is misinformation in their archive?

(As it happens, we have a previously uploaded retouched version of this same photo, File:Seattle - Bonney Watson Funeral Home 1916.jpg, which I uploaded about a decade ago.) - Jmabel ! talk 18:17, 8 September 2018 (UTC)

If you are correct than the bad info come from this page, which has "contact us" on the bottom of the page. I would check with them first. --Jarekt (talk) 02:58, 9 September 2018 (UTC)
Thanks, I'll follow that up.
Meanwhile, another problem from the same source: File:Crowd assembled in Occidental Square in front the Occidental Hotel for memorial service for President James A Garfield, Seattle (CURTIS 283).jpeg and File:Crowd in Occidental Square in front of Occidental Hotel for the visit of President Hayes, Seattle, 1878 (CURTIS 319).jpeg are clearly variant prints of the same photo, but claim to be for entirely different occasions! For the former occasion, we also have File:Occidental Square in front of Occidental Hotel during memorial services for President James A Garfield, Seattle, September 27 (PEISER 141).jpeg which is certainly from a different date, given the stable next to the hotel. - Jmabel ! talk 20:31, 9 September 2018 (UTC)
same photo, but claim to be for entirely different occasions — Could have President Hayes visited Seattle for the memorial service for President James A Garfield? —⁠andrybak (talk) 20:36, 9 September 2018 (UTC)
@Andrybak: No, seven years apart, two presidents in between, and as I point out there is a stable next to the hotel in File:Occidental Square in front of Occidental Hotel during memorial services for President James A Garfield, Seattle, September 27 (PEISER 141).jpeg that is not present in these other photos. - Jmabel ! talk 22:06, 9 September 2018 (UTC)

September 09

Page creator vs. uploader

Please take a look at the history of File:Interior Of Transperth's Mercedes-Benz O405NH (Volgren CR225L).jpg. File page has been created by one user, but the actual photograph has been uploaded by another user. How is this even possible? I had never seen a similar case. 4nn1l2 (talk) 08:18, 9 September 2018 (UTC)

It's probably T179884 on Phabricator. --ghouston (talk) 10:27, 9 September 2018 (UTC)
Thanks. It's sad to see that we are facing this problem since at least 9 August 2017, but the issue has not been solved yet (even not triaged). Any file with this problem has to be tagged with {{Nld}} and some may get deleted eventually, and the blame is on the software, not the uploader. 4nn1l2 (talk) 11:46, 9 September 2018 (UTC)
Don't bother anymore, I think יניב הורון might be able to help. Lotje (talk) 12:35, 9 September 2018 (UTC)

Hebrew language users

Hi, anybody hanging around who is able to point me to a an active user speaking Hebrew? I am looking for a possible translation of for example this text. Thank you for your time. Lotje (talk) 11:53, 9 September 2018 (UTC)

@Lotje: Dorian Gray Wild, Geagea and Muhandes came to my mind. De728631 (talk)
@De728631: thank you very much, I asked Dorian Gray Wild and see what he says. Lotje (talk) 03:32, 10 September 2018 (UTC)

Do you have small tasks for new contributors? It's Google Code-in time again

Hi everybody! Google Code-in (GCI) will soon take place again - a seven week long contest for 13-17 year old students to contribute to free software projects. Tasks should take an experienced contributed about two-three hours and can be of the categories Code, Documentation/Training, Outreach/Research, Quality Assurance, and User Interface/Design. Do you have an idea for a task and could you imagine mentoring that task? For example, do you have something on mind that needs documentation, research, some gadget or template issues on your "To do" list but you never had the time, and can imagine enjoying mentoring such a task to help a new contributor? If yes, please check out mw:Google Code-in/2018 and become a mentor! Thanks in advance! --AKlapper (WMF) (talk) 13:50, 9 September 2018 (UTC)

Administrator doesn't answer

How can I discuss with administrator his wrongful edits (see here, here and other related edits) if he doesn't answer my request? His edits aren't based on consensual decision (see his UserYalk page). --VAP+VYK (talk) 17:02, 8 September 2018 (UTC)

@VAP+VYK: Stop edit warring about the picture of the day. If you disagree with the picture, discuss it on an appropriate forum (such as this page), but do not just change the image. (@Jdx: for reference) Sebari – aka Srittau (talk) 17:42, 8 September 2018 (UTC)
Excuse me, where is those forum? --VAP+VYK (talk) 17:45, 8 September 2018 (UTC)
  • He finally ansered on my UserTalk page (somehow not on his one, where I'd asked him), but I didn't see him to understand the situation. His answer can't be satisfactory. --VAP+VYK (talk) 17:45, 8 September 2018 (UTC)
  • I don't think that the problem is over. The administrator protected files so other users can not edit them. Thus he consolidated the situation in his favor without discussion and consensus. --VAP+VYK (talk) 16:59, 9 September 2018 (UTC)
—⁠andrybak (talk) 18:10, 9 September 2018 (UTC)
  • «You have been edit warring at this file since at least June.» — I have not any edit warring since June because nobody reverted my edits (excepting Jdx). «The image has a feature status and was selected as picture of the day for 27 September 2018,» — how it was decided to do just this one POTD? Who had given authority this user to do it? And why other users (me, for example) can not change POTD? «The consensus is COM:NOTCENSORED, as described on the talk page by jdx.» — I have alredy answered him, please, don't repeat disproved argument. The point is not about "protest" or something else. It's about procedure of appointment POTD and why I can't change the image if there aren't any objections from other users? Jdx didn't object, he didn't discuss it with me, didn't ask me why I had done it. He autocratically reverted my changes. --VAP+VYK (talk) 18:26, 9 September 2018 (UTC)
    • This edit reverted yours, but you kept removing the POTD template – even one for the Bengali Wikipedia that may have had local consensus there. The procedure for POTD on Commons is simple: Anyone can select an image from the featured pictures and "most images which have been promoted to featured status are all equally worthy for inclusion on POTD once promoted." I.e. there is no requirement for any consensus to add a POTD candidate. However, changing the POTD in the queue is generally discouraged and there are only a few general reasons where it can be done without consensus. That means any objections should have been made at Commons talk:Picture of the day, but that was not the case for this image. De728631 (talk) 18:46, 9 September 2018 (UTC)
      • I don't know what had wanted the anonimous user just in one Wikipedia, but then he/she didn't object more. Thus it can't be considered as warring (one revert during few monthes). «That means any objections should have been made at Commons talk:Picture of the day» — OK.
        PS It doesn't mean that the point of flagrant autocratical behaviour of the administrator is closed. --VAP+VYK (talk) 15:48, 10 September 2018 (UTC)

Dutch Wikipedia question

Here's what striked me yesterday: a Dutch wikipedian reverted this reference commenting: geen valse informatie geven Lotje, Google Books is niet je vriend, meaning: Do not give false information, Lotje, Google Books is not your friend.
If that is a suggestion, please feel free to comment. Lotje (talk) 03:43, 10 September 2018 (UTC)
Lotje, did you mean to respond to the "Google Code-in" section? it seems unrelated to the questions AKlapper is asking. —⁠andrybak (talk) 05:13, 10 September 2018 (UTC)
@Andrybak: no, I did not mean to repsond to the "Google Code-in", it was just something which I wanted to bring to the English speaking daylight. But thank you for asking. Lotje (talk) 05:21, 10 September 2018 (UTC)
@Lotje: Commons just handles media, you could start a discussion on English Wikipedia but they may also be puzzled why nlwiki comes there. https://meta.wikimedia.org is for cross-wiki issues, I don't go there often but there is probably a section for this. - Alexis Jazz ping plz 08:38, 10 September 2018 (UTC)
Thank you for your suggestion, Alexis Jazz, that seems like a good idea. Lotje (talk) 09:21, 10 September 2018 (UTC)
  • @Lotje: this user is f.o.s.: GoogleBooks may not be your friend (Google itself certainly is not, «don’t be evil» not withstanding), but linking to a digitized/uploaded version of a hardcopy book available from GoogleBooks is abosolutely not a case of spreading «false information». If (and it’s a big if!) the case can be made that the uploaded version of this book in GoogleBooks is somehow modified and not the same as it is/was available elsewhere, then the matter must be carefully and clearly raised in the talk page. This reversion and its summary is meanspirited and unhelpful. Any nl.wp admin should give this user a rebuke. -- Tuválkin 12:37, 10 September 2018 (UTC)
@Tuvalkin: , thank you for your time. In fact, that is wat this user pretends very often, that I spread false information. I'll remember your wise words. Lotje (talk) 12:45, 10 September 2018 (UTC)
Just to be clear. The problem isn't using google books but Lotje who links to google books to support a claim insterted by Lotje whilst this claim doesn't match the source given by Lotje. Natuur12 (talk) 15:17, 10 September 2018 (UTC)

Varbergs station

I could not find the category Varberg station so I created it. Later I discovered that there is the category:Varbergs station and moved my uploaded files to Varbergs station. Should all files be moved to category:Varberg station? I put all the correct upper categories in 'Varbergs'.Smiley.toerist (talk) 13:24, 10 September 2018 (UTC)

If Varberg station is the correct name, then yes, moving all media to the correctly named category would be best. Afterwards, you can create a redirect from Varbergs station to Varberg station. See: Commons:Redirect --oSeveno (talk) 15:02, 10 September 2018 (UTC)
✓ Done. —⁠andrybak (talk) 19:13, 10 September 2018 (UTC)

22:35, 10 September 2018 (UTC)

September 11

Date at sea categories

We have many year categories by country, region, city etc. But for pictures taken in international waters far from shore there are no such categories. In this case I could use '2018 at Kattegat' and the head category: 2018 at sea.Smiley.toerist (talk) 11:32, 7 September 2018 (UTC)

Maybe, however if the image is relevant to climate change, seasonal algae bloom variations, marine life depletion, being able to sort them out by date would be useful. -- (talk) 15:15, 7 September 2018 (UTC)
"presumably" (Laconic phrase). ℺ Gone Postal ( ) 15:18, 7 September 2018 (UTC)
However, personally... even though I classify files uploaded by me in all available categories, I do think that it would be better to just keep top-most and then work on making category intersections an easier task. So you should have date categories, that are inside month categories, which are inside year categories; then add a category for the body of water, and for the person who was presumed not to exist (i.e. the person who need this body of water in a specific year) can just intersect year category with the body of water and look at results. ℺ Gone Postal ( ) 15:34, 7 September 2018 (UTC)
^ +1 --El Grafo (talk) 16:26, 7 September 2018 (UTC)
Most images are taken in coastal waters so the year can be attributed to a country. However I do have examples of algae blooms: File:Zeealgen in de Baltische zee 1.jpg, File:Zeealgen in de Baltische zee 2.jpg. There are also whales, dolfins and other sea 'subjects'. However 'nature' pictures on land are not often dated.Smiley.toerist (talk) 21:04, 7 September 2018 (UTC)
@: I can see getting down to century or even decade for that, and I can even see getting to [year] at sea, but by year in a particular body of water would mostly function as a way of getting relatively useless microcategories. - Jmabel ! talk 21:16, 7 September 2018 (UTC)
Taking images in a category and filing them away into many date by year categories is a way of hiding files and making them useless to users. It makes it impossible to compare photographs and choose anything for illustrations, since the user would need to click over dozens of categories. It's called over-diffusion. See the discussion at https://commons.wikimedia.org/wiki/Commons:Village_pump/Archive/2018/08#Overdiffused_categories. In short its unwitting vandalism. Sadly it's spreading through the project like an uncontrolled virus. -Broichmore (talk) 09:19, 9 September 2018 (UTC)
Every time an image is classified by one aspect, it becomes somehow hidden for people searching by another aspect. This is a problem even for the most useful aspects and the main solution is the "all images" button. Since we have a working solution for some categories, I don't see how the problem worsens by adding more categories since somebody is willing to maintain them.--Pere prlpz (talk) 10:21, 9 September 2018 (UTC)
Basically I agree. Though the "all images" button is not impressive; using Deepcat is far superior. The problem is not adding categories, the problem is taking ALL the images in an existing category, say History of ..., and filing them away by date (year). Broichmore (talk) 12:04, 11 September 2018 (UTC)
I dont see a problem with adding a category. It does not diminish the other search possibilities. In general there should always be time dimension category and a location category. There is no optimum whereby everybody is satisfied. Subdivide to large categories, otherwise you get swamped. With searches with File names, categories and the text descriptions I usually manage to find what I look for. The worst cases are files with no categories and very summary descriptions, These are not 'visible'. We all keep up the good of adding useful categories and descriptions and doing research on de images. These activities are as important as uploading a lot off images. If you cant find them there are of no use.Smiley.toerist (talk) 10:51, 9 September 2018 (UTC)
Again I effectively agree. Adding not deleting. However filing by year, is not usually maintained, once set up; it's too complex and time demanding to do so. Also bear in mind that the most common error in Commons, is the date field. Before the digital camera, dates are prone to error or inaccuracy. Illustrations from books are often the year before first publication, sometimes even as much as 5. Splitting by date is the least helpful way of categorising images. If you want a specific date (and your eye is not enough) it's best to use petscan or type in a specific year as a search keyword to narrow the results. Broichmore (talk) 12:04, 11 September 2018 (UTC)

Does anybody have observations about errors by sysops caused by poor customization of MediaWiki? One thing I worry about is confusion induced by redirect pages.

Case:

  1. One Hassanamin994 uploaded File:Black-brow-cat-in-grass.jpg (histlogsabuse log), a downsample of the File:Chaton2.jpg (histlogsabuse log).
  2. A mass delreq was created on several Hassanamin994‎‎’s uploads, without thorough examination of individual files.
  3. I, Incnis Mrsi, noticed that Black-brow-cat-in-grass.jpg is eligible under Commons:CSD #F8 and made respective tagging.
  4. A sysop JarrahTree deleted the downsample and replaced File:Black-brow-cat-in-grass.jpg with a redirect.
  5. When closing the delreq, Hystrix slipped through the redirect to Chaton2.jpg and deleted the latter, “per nomination”.

Solution: make the mw-redirect class (links to a redirect) easily distinguishable from common internal links. It can be done with a trivial tweak placed into the personal CSS, or with a gadget. It will alert the operator before doing any action with the linked page.
Incnis Mrsi (talk) 16:22, 9 September 2018 (UTC)

September 10

Template "move" doesn't display correct text

Since the last week(s), template {{Move}} doesn't display it's correct text. Is somebody able to detect and resolve the problem? — Preceding unsigned comment added by ŠJů (talk • contribs)

ŠJů, could you please clarify what is incorrect in the current text? —⁠andrybak (talk) 07:04, 11 September 2018 (UTC)
Andrybak That is incorrect that the text doesn't display. The template display {{{2}}} instead the text. See Category:Junkernstrasse-Bau-GmbH building for an example. --ŠJů (talk) 11:01, 11 September 2018 (UTC)
ŠJů, I found at least one issue with {{Move/layout}}: Special:Diff/319688285. Template scheme: {{Move}} uses {{Move/i18n}}, which in turn uses {{Move/layout}}. Trying to figure out why its showing {{{2}}} on Category:Junkernstrasse-Bau-GmbH building. —⁠andrybak (talk) 11:17, 11 September 2018 (UTC)
Does not work on Category:Lochs of Scotland, as well, for example. —⁠andrybak (talk) 11:20, 11 September 2018 (UTC)
ŠJů, this fixed the issue: Special:Diff/319689175. —⁠andrybak (talk) 11:27, 11 September 2018 (UTC)
It seems that it was broken since at least 2013: Special:Diff/101584983. I'm not sure why the FooBar from example code stayed so long in the template... —⁠andrybak (talk) 12:01, 11 September 2018 (UTC)
@Andrybak: Thank you very much, the template works correctly! --ŠJů (talk) 15:26, 11 September 2018 (UTC)
Btw, until this summer, the template worked OK. --ŠJů (talk) 15:28, 11 September 2018 (UTC)

Uploading monument pictures.

I am a new contributor with a wealth of photos to contribute to the request for monument photos. I note that many monuments of note are not available to upload photos for. How can I propose new options? Thank you. — Preceding unsigned comment added by Goreymurphy (talk • contribs) 20:09, 11 September 2018 (UTC)

@Goreymurphy: the WLM organizers in each country work from the official lists of registered or protected historic buildings that are maintained by governmental or cultural agencies, so sites that aren’t listed as of the starting date don’t qualify as subjects for the contest. However, we welcome all freely reusable photographs of interesting or important sites, whether eligible for a competition or not, and there are existing categories for all manner of locations. Can you give some examples of sites you’re thinking of?—Odysseus1479 (talk) 21:03, 11 September 2018 (UTC)
@Odysseus1479 Thank you for your reply. I wasn't thinking in terms of competition as my photos would be more about the subject than about photographic merit. I'm new to this resource, so your guidance has been most helpful. Goreymurphy.

September 12

Hayley Atwell Picture

An IP user has nominated File:Hayley Atwell by Gage Skidmore.jpg for deletion, claiming they represent her and that she "doesn't feel happy with it being displayed especially without her permission" and that she "would very much prefer another more flattering photo." I told them that we are unlikely to delete the image just because they don't like it, but if they make an account and prove their identity as representing her, we could probably work on using another image instead. I can't remember if we have a way to prove identity in a case like this, but if we do, what is that way? Also, anyone else up to helping out with this? --Elisfkc (talk) 18:36, 11 September 2018 (UTC)

@Elisfkc: they could prove their identity with a message to COM:OTRS from an email address associated with Atwell or her manager/agent/etc. That would also serve as evidence of permission should they provide a replacement photo. clpo13(talk) 18:48, 11 September 2018 (UTC)
Hardly an unflattering photo! What is it with public figures who think they can completely control what pictures of them are published? - Jmabel ! talk 20:00, 11 September 2018 (UTC)
Her current enwiki pic makes her look like the female Joker from DC Comics so a win win i say..--Stemoc 00:59, 12 September 2018 (UTC)
@Clpo13: That's what I thought, I just wanted to make sure. --Elisfkc (talk) 21:45, 12 September 2018 (UTC)
i normally honor subject deletion requests, since personality rights is so inadequate. needs verification though, and should suggest looking at the category to pick one or suggest uploading a better one. Slowking4 § Sander.v.Ginkel's revenge 02:32, 13 September 2018 (UTC)

I received a request at OTRS to update a logo, something I've done dozens of times.

In most of the requests, the updated logo is subject to copyright, intended to be used only in the INFOBOX in an article in the English Wikipedia, so I upload as a fair use image and fill out the required rationale.

In fact, that was my first step in this case, and the relevant article contains the updated logo as a fair use image en:Juki.

After doing so, I realized that the image was also used in three other wiki's, two different languages and wikidata. I contemplated replacing each of those but I now think there is a better approach. I see that the prior image was not a fair use image but deemed to be a public domain image because it did not meet the threshold of originality. It was created by converting an EPS file to an SVG file and can be found here: File:Juki_company_logo.svg

I decided the best approach was to update that image with the newer version of the file. I requested that the company sent me an SVG version of the file. They did so but when I tried to update it, I received an error about incompatibility. I've worked enough with SVG files in the past to know that they can be tricky, so I wasn't totally surprised that the upload failed.

In short, I'm looking for help.

Three options occurred to me:

  1. If someone willing to help is an OTRS agent, The file is attached to the latest email in the OTRS ticket: ticket:2018091110004658 It has the less than optimal name: Untitled-1.svg
  2. If someone willing to help is not an OTRS agent, I can email the file to you
  3. The prior version of this file was created by going to Brands of the World and I believe that currently holds the most recent version so if someone could convert this EPS to SVG the update could be completed--Sphilbrick (talk) 14:43, 12 September 2018 (UTC)
Sphilbrick, done? It looks like maybe they didn't actually make an svg, but just made an svg file name? I think I've fixed it now. GMGtalk 15:17, 12 September 2018 (UTC)
@GreenMeansGo: Thanks!!! Can you tell me what you did?--Sphilbrick (talk) 15:57, 12 September 2018 (UTC)
No problem Sphilbrick Honestly? I searched around online for the error message until I found this thread, so I took the original jpg, and used on online site to convert that into a new svg. Then I had to go back in and sample the color from the jpg and add it back into the new svg using wordpad. (Because for some reason it created the new svg in black and white.)
It's possibly the lowest tech solution to the problem imaginable, but it works as far as I can tell. GMGtalk 16:01, 12 September 2018 (UTC)
Well, I was asking because I hope you had a simple one or two click solution that I might use in future situations but you did a fair amount of work. Thanks again.--Sphilbrick (talk) 16:05, 12 September 2018 (UTC)

Question for native English speakers: "Fire brigade" vs. "Fire department"

There is the Category:Fire brigades (with some subcategories) and the Category:Fire departments (with some subcategories). However, I didn't find any semantic difference between these two terms, and English Wikipedia suggests that it's the exactly same thing, one term being British English and the other American English. So my understanding is that the two category trees should be merged, and probably all "Fire brigades" categories renamed to "Fire departments". I opened a CfD page but did not receive a single comment on that in 2 weeks. I would very much appreciate if some native English speakers could comment there. Thanks in advance! --Reinhard Müller (talk) 19:06, 12 September 2018 (UTC)

They are both correct but one is indeed (British) English and the other is American which some people seem to think is also a form of English. As long as you set up appropriate redirects I am not sure it matters which you prefer, but of course you open yourself up to charges of linguistic imperialism either way which may be why no one dared to comment where you raised this before. Something about sleeping dogs? Leben und leben lassen? Success Charles01 (talk) 19:23, 12 September 2018 (UTC)
I've never heard Fire Department used in the UK. The people that come to save you are the Fire Brigade. The organisation they come from can be the Fire Brigade, the Fire Service, the Fire and Rescue Service, and maybe other similar terms. I don't know what specialist airport fire services call themselves though, or sea-port fire-ship organisations for that matter. Would "Fire Service" be more neutral than brigade or department? --Northernhenge (talk) 20:58, 12 September 2018 (UTC)
Well...In the US, if you say "fire brigade" I think w:Fires brigade, AKA field artillery. They're basically the opposite of a fire fighter. GMGtalk 21:02, 12 September 2018 (UTC)
yeah - i am not going to comment on category ontology ideology. big waste of time. Slowking4 § Sander.v.Ginkel's revenge 02:43, 13 September 2018 (UTC)

Infobox Wikidata interactions

As I noticed now, a patrol link [Mark this page as patrolled] displays below the {{Infobox Wikidata}} and pushes all content of the category page below, out of the screen. --ŠJů (talk) 20:17, 12 September 2018 (UTC)

See discussion at Template_talk:Wikidata_Infobox#Text_wrap. It seems to be a bug with Monobook css that needs tracking down and fixing. Thanks. Mike Peel (talk) 23:10, 12 September 2018 (UTC)

This youtube clip has no license given. Does anyone know if it is free? Best, --Leoboudv (talk) 21:33, 12 September 2018 (UTC)

September 13

Strange stuff from Wikidata on Category:Frederick II, Holy Roman Emperor

This category is in some categories where it doesn't belong, presumably related to the way it's connected to Wikidata. Specifically:

Can someone look at this and see what's causing it? Extra brownie points if you explain it to the rest of us so we can learn. :) --Auntof6 (talk) 11:55, 10 September 2018 (UTC)

Crap removed. Incnis Mrsi (talk) 12:10, 10 September 2018 (UTC)
Not totally sure why User:Mhmrodrigues added that category on the Wikidata item. Maybe just a mistake? GMGtalk 12:14, 10 September 2018 (UTC)
Hi! I added this category, because I thought it could be more easy to categorize the Emperor's images not by events of hs life, but by sources; and that's why there are miniature, statues, paintings, engravings, etc.. Best hope for your understanding. I've already corrected the wrong link. Mhmrodrigues (talk) 12:32, 10 September 2018 (UTC)
@Mhmrodrigues: please, learn better what does Commons category (P373) mean. It is not for supercategories. Incnis Mrsi (talk) 12:45, 10 September 2018 (UTC)
I don't really have any strong opinion about what the best way to categorize these on Commons might be. But we ought not be linking the Wikidata for a Commons cat on figurines to the item for the individuals birth, because they're just not the same thing. Also because a Commons category on these figurines is unlikely to have corresponding categories on other projects, it will most likely not meet d:Wikidata:Notability. GMGtalk 13:48, 10 September 2018 (UTC)

Thanks, folks. The "in miniature" category is gone, but not the other one. Any ideas there? --Auntof6 (talk) 19:41, 10 September 2018 (UTC)

Hey Auntof6. Category:Frederick II, Holy Roman Emperor exists on seven difference projects. So it would be appropriate for it to have its own Wikidata item containing all those seven pages. Am I misunderstanding the problem? GMGtalk 19:50, 10 September 2018 (UTC)
The problem was that when you look at Category:Frederick II, Holy Roman Emperor here on Commons, it was in one or more categories (it's varying because people are working on it) that it shouldn't be in. When I looked earlier, it was in Category:Frederick II, Holy Roman Emperor (itself). When I looked just now, it was in Category:Death of Frederick II, which redirects to Category:Frederick II, Holy Roman Emperor (again, itself). It shouldn't be in either of those. However, I may have fixed it by removing the death-of category from the Wikidata entry (I figured, if it's a redirected category here, it doesn't need to be attached to a Wikidata item. We'll see if it stays fixed. --Auntof6 (talk) 20:08, 10 September 2018 (UTC)
@Auntof6: Looks like there were two more issues. First was this erroneous P373 value, the second was a bug in the infobox that I've now fixed. Those together were causing the category loop (Category:Frederick II, Holy Roman Emperor was in Category:Frederick II, Holy Roman Emperor). Does that look better now? Thanks. Mike Peel (talk) 19:55, 10 September 2018 (UTC)
Yes, thanks. I wasn't sure if the change I made (described in my reply above) had fixed anything, but it looks OK now.
By the way, I see that you've worked on that template a fair amount. Is there a way to make it default to hidden/collapsed? I don't like the way it forces subcategories to appear in a single column, at least on my tablet. --Auntof6 (talk) 20:08, 10 September 2018 (UTC)
@Auntof6: I *think* you might be able to auto-hide the template by adding $( "#wdinfobox" ).addClass( "mw-collapsed" ); to Special:MyPage/common.js - can you try that and see if it works? Thanks. Mike Peel (talk) 22:10, 10 September 2018 (UTC)
@Mike Peel: It worked, both on my tablet and my laptop. Thanks! --Auntof6 (talk) 08:43, 13 September 2018 (UTC)
Great. :-) It's now in the template documentation in case anyone else wants to do the same in the future. Thanks. Mike Peel (talk) 10:43, 13 September 2018 (UTC)

"Building construction" categories

I find it very odd when a category like Category:Building construction sites in the United States is placed as parent of a category for a building, rather than individual photos or a subcategory specific to the construction of the building. Can someone explain the purpose of this? @Epicgenius and Justin.A.Wilcox: who have done this with Category:111 West 57th Street and Category:400 Park Avenue South, respectively, in hopes that they can tell me what I'm missing. - Jmabel ! talk 03:57, 13 September 2018 (UTC)

@Jmabel: At the moment, these buildings themselves are construction sites. That's why I categorized it this way. I think they should be removed once completed. If this isn't the right category, what would be a more appropriate parent category? epicgenius (talk) 12:42, 13 September 2018 (UTC)
I'd probably create a subcat (e.g. Category:construction of 111 West 57th Street), even though that means right now no pictures will go directly in the main category. After all, that is where these should eventually end up once the building is completed. Why not just do it now? Or the construction category can be directly on the photo if there are only a few of them. Any other solutions require future maintenance, so they are much more prone to future problems. - Jmabel ! talk 15:55, 13 September 2018 (UTC)

Own talk page

Excuse me. it is allowed to delete all the discussions on the own talk page or it is not allowed? I know it's not allowed, but a user questions this thing. What is the rule? Thank you. --DenghiùComm (talk) 07:43, 13 September 2018 (UTC)

@DenghiùComm: Archiving is far better, please see COM:ARCHIVE and {{Dont remove warnings}}. In the case of removal of text which is not unaddressed legitimate warnings, that is allowed.   — Jeff G. ツ please ping or talk to me 08:10, 13 September 2018 (UTC)
Thank you very much. Cheers, --DenghiùComm (talk) 10:02, 13 September 2018 (UTC)
@DenghiùComm: You're welcome.   — Jeff G. ツ please ping or talk to me 11:20, 13 September 2018 (UTC)

FOP in Romania

Can anyone explain to me the decisions here [23] [24]? This is not a certain building, it is the "Youth House", built in 1972, chief architect being Haralambie Cocheci (1924–2007), collaborator Ivan Stern (b. 1945, living). For the building, the Union Architects Prize was awarded. On ro.wp it is known that in Romania the FOP is applied as such. What's the difference that kept these photos while dozens of other photos of buildings in Timișoara were deleted? I ask because as an OTRS volunteer I need to know all the subtleties in order to be able to exploit them in other deletion discussions. --Turbojet (talk) 09:28, 13 September 2018 (UTC)

This diff Special:Diff/169960111 changed Romania in 2015. Vandal Edit? - the only edit by the user... Ronhjones  (Talk) 17:54, 13 September 2018 (UTC)
The text from Commons:Freedom_of_panorama#Romania was correct. Understanding the text was inappropriate.
Examples for "if the work is not the main subject of the image": File:Ansamblul urban interbelic „Corso”.jpg (depict 5 buildings & an monument, has no main subject a particular building or the monument), File:Catedrala din Timisoara.jpg (depict 9 buildings, has no main subject a particular building). But in an image like this: File:Cazinoul Delvideki (2).jpg it is clear that the building is the main subject (to this building the architect's copyrights have expired). An example at the limit is: File:Palatul Johann Hochstrasser, Timisoara.jpg (depict 2 main buildings — but this image does not raise problems, as architects' copyrights have also expired for all buildings.) --Turbojet (talk) 20:55, 13 September 2018 (UTC)

Probably a stupid question

How come when I'm viewing a diff on Commons I don't have a restore option via rollback? Seems to be working just fine in en.wiki and WD. I've got the option when looking at an editor's contribution history. And I get kindof that I don't have it in a page's history (that's on WD but not on WP), but it seems very odd that there should be no option when viewing an individual diff. Have I screwed up my preferences somehow? GMGtalk 15:52, 13 September 2018 (UTC)

@GreenMeansGo: not sure exactly, but I believe the restore function you describe is provided by Twinkle (along with the “GF” and “vandal” rollback variants), not a built-in feature. I don’t know about WD: if they don’t have TW there it might be some other Gadget instead.—Odysseus1479 (talk) 18:39, 13 September 2018 (UTC)
There is no Twinkle for WD as far as I know, and I don't have any fancy gadgets enabled there other than the one that lets you nominate for deletion. Not being able to rollback from the page history looses like half the functionality of the tool though. GMGtalk 18:52, 13 September 2018 (UTC)

Delete a file

I have uploaded this file today : https://commons.wikimedia.org/wiki/File:Tirdad_kiae.jpg The file is my own work but I would like to delete it. Please help me, thanks .Khobi1396 (talk) 18:37, 13 September 2018 (UTC)

@Khobi1396: to delete a file you’ve recently uploaded (less than a week ago) you can just tag it with {{SD|G7}}. See COM:CSD#G7.—Odysseus1479 (talk) 18:45, 13 September 2018 (UTC)
✓ Done This was uploaded today and was not in use, so I deleted it per Khobi1396's request. De728631 (talk) 19:15, 13 September 2018 (UTC)

September 14

How to see all available languages?

I've been adding descriptions to media files, in a couple of languages. Is there a way to get all available languages to display at the same time? I'd like to be sure the edits turned out correctly, without having to set my language preference to each one in turn. Frenezulo (talk) 16:55, 8 September 2018 (UTC)

The above is specifically about the LangSwitch template; so far it's the only template I've found with this issue. Frenezulo (talk) 23:46, 14 September 2018 (UTC)

{{Extracted from}} in "Source" vs "Other versions"

It seems that {{Extracted from}} is mostly used in "Other versions" field of {{Information}}. I'm wondering, if instead of using {{Derived from}} in field "Source", {{Extracted from}} could be used instead, in some cases removing duplication of links to the same file. —⁠andrybak (talk) 08:35, 14 September 2018 (UTC)

@Andrybak: IMO it can and should be used where applicable. Although note that {{Derived from}} is more general – it can be applied to works/files which are derived from more than one work/file while {{Extracted from}} cannot. --jdx Re: 05:22, 16 September 2018 (UTC)

Category tree tools: +1

I made a small JavaScript tool (User:Fractaler/taxonomy, my first experience) for navigating the category tree, copying the categories chain in wiki-format (why). This is my first experience, so I do not know if it's enough to just inform here? --Fractaler (talk) 10:15, 14 September 2018 (UTC)

Commons:User_scripts may be interesting for you. Ruslik (talk) 20:37, 15 September 2018 (UTC)
Thank you! --Fractaler (talk) 06:54, 16 September 2018 (UTC)

Problem uploading large file

phab:T204408

Hello! I'm unable to upload this large book: https://archive.org/details/greekenglishlex00liddrich/ (~200 MB)

It says: The upload was completed successfully, but the server hasn't returned any information about the file. — Preceding unsigned comment added by Hermolaou (talk • contribs) 12:03, 15 September 2018 (UTC)

Trying upload by URL, I got Our servers are currently under maintenance or experiencing a technical problem. I will try again later. Yann (talk) 12:11, 15 September 2018 (UTC)

@Hermolaou: , @Yann: the file is uploaded here: file:A_Greek-English_lexicon,_1883.pdf - Father of Lies (talk) 15:46, 15 September 2018 (UTC)

September 16

Licence review of File:NASA 60th- The Leading Edge of Flight.ogv (today's MOTD)

I can't see a licence name on the YT page, but is review really needed? There is primary source specified in the description – link to a NASA's web page where the video is available. --jdx Re: 05:07, 16 September 2018 (UTC)

I've gone ahead and reviewed it. Standard NASA work, it's public domain. Huntster (t @ c) 06:27, 16 September 2018 (UTC)

Does anyone know if this image is free. I will sign off for tonight but hopefully someone knows the answer. Best, --Leoboudv (talk) 10:10, 16 September 2018 (UTC)

What is the difference between these two categories? —⁠andrybak (talk) 10:30, 16 September 2018 (UTC)

The latter category is the root category for categories filled in by templates while the former is a broader category. The latter is also a member of the former. Although this distinction is not always obeyed. Ruslik (talk) 19:47, 16 September 2018 (UTC)

SVG rendering problem

Red edge only in second solid

Commons does not correctly render this image (while Inkscape does). See file talk. All the yellow faces have red edges, but Commons shows them only for the second solid. Watchduck (quack) 13:28, 16 September 2018 (UTC)

I seem to remember a similar issue with badly-rendered SVG thumbnails, but I can't recall what it was. De728631 (talk) 20:13, 16 September 2018 (UTC)

September 17

Can someone take a look at this thread and advise? Thanks. —Justin (koavf)TCM 22:28, 14 September 2018 (UTC)

The original author (user indianastatearchives) describes their claim to authorship pretty well. Nomitated for speedy deletion per COM:COPYVIO. —⁠andrybak (talk) 01:29, 15 September 2018 (UTC)
andrybak: Do not create such request. This is clearly public domain due to age, whatever claim exists. BTW indianastatearchives is not the author of the map, so the claim is a nonsense anyway. Regards, Yann (talk) 03:40, 15 September 2018 (UTC)
Yann, I didn't know such aspect of copyright. Thanks. —⁠andrybak (talk) 10:01, 15 September 2018 (UTC)
@Andrybak: Why did you do this? —Justin (koavf)TCM 03:51, 15 September 2018 (UTC)
please do not engage GLAM institutions in this way. telling people they are doing it wrong is counter-productive. and where is the source? no link to reddit, no link to archival site? wow. why don't you query the glam-wiki group or wikimedia library to engage. Slowking4 § Sander.v.Ginkel's revenge 02:02, 16 September 2018 (UTC)
@Slowking4: You are free to post to the Reddit thread and I've invited someone from the institution here. —Justin (koavf)TCM 03:33, 17 September 2018 (UTC)
i provided an imgur source. please reread Commons:Essential information. you realize that newbies have had similar files deleted as source missing? Slowking4 § Sander.v.Ginkel's revenge 10:27, 17 September 2018 (UTC)
@Slowking4: I can't understand your point. —Justin (koavf)TCM 18:08, 17 September 2018 (UTC)

September 15

21:58, 17 September 2018 (UTC)

Thumbnails not animated

The GIF on the left has an animated thumbnail, that on the right has not.

For the animation on the right no animated thumbnail is created. It also says so on the page: Note: Due to technical limitations, thumbnails of high resolution GIF images such as this one will not be animated. Fair enough. But that message would be much more helpful if it told me the limit. I would rather not make test uploads until one is small enough. Watchduck (quack) 00:31, 13 September 2018 (UTC)

I vaguely recall this was discussed in Phabricator perhaps 3 years ago, it would be worth searching the archives there. I hit a limit at 12MB (warning: my human memory about this may be faulty) but the specific limit is fuzzy due to the nature of WM rendering. Whether the limits have changed as the underpinning software has changed, or whether the limit is by arbitrary development choice, I have no idea. -- (talk) 05:09, 13 September 2018 (UTC)
File:Polyhedron 12 big.gif is only 5.54 MB, and it has the same problem. Maybe it actually is about resolution rather than file size. Anyway, I will stick to linking animations instead of showing them. Watchduck (quack) 07:46, 13 September 2018 (UTC)
@Fae and Watchduck: the limit is in decoded pixels of the original. So it is W x H x frame count that should be smaller than the limit. The first image is 256*256*120 == 7.864.320 bytes (7.8MP), whereas the second image is 600*600*360 == 129.600.000 (130MP). The public config shows that the limit for animated images is 100MP. This quickly exploding size is also why GIFs are problematic. —TheDJ (talkcontribs) 11:20, 18 September 2018 (UTC)
Oh, while the limit is in theory 100MP, it should be noted that stress on the servers can cause a timeout to occur with images under that limit. The limit protects the server (as does the timeout), it is not a promise to deliver results at all costs. —TheDJ (talkcontribs) 11:26, 18 September 2018 (UTC)

Is this image appropriate to review. The claim of cc by is not correct as the uploader did not take this image in the early 20th century. If it is passed, the image license has to be changed to EU-Anonymous or something like that. I don't really know if an author took it or if it is safe to review? Best, --Leoboudv (talk) 19:20, 17 September 2018 (UTC)

Is this logo too complex to be copyright free? What do you think? If it is please give a reason to file a DR. Complex design? Thank You, --Leoboudv (talk) 05:53, 18 September 2018 (UTC)

September 19

Request to blur an image

A reader (ticket:2018092310005116) requested that the registration plate numbers in this image be blurred:

File:Boston Fire Department, Engine 5 Firehouse (1).jpg

At a high resolution I can confirm the numbers are readable.

If you ping me when completed, I will do a revision deletion of the prior versions. --Sphilbrick (talk) 20:22, 24 September 2018 (UTC)

@Sphilbrick: ✓ done. —⁠andrybak (talk) 20:45, 24 September 2018 (UTC)
@Andrybak: Thanks for taking care of this so quickly.--Sphilbrick (talk) 21:02, 24 September 2018 (UTC)
This section was archived on a request by: —⁠andrybak (talk) 21:23, 25 September 2018 (UTC)

Portrait right in Japan and South Korea.

On Chinese Wikipedia Village Pumps, someone mentioned that a certain users uploaded 100+ photo with various subjects in Japan and South Korea which included face of bystanders in those photo, and the one who raise the question think that those face must be blurred before uploading onto commons in order to comply with local law in those countries. Is this really necessary for the domestic regulation in those countries to be followed when uploading photos taken from those countries onto commons? An example of image being mentioned would be File:Car Free Ginza 201806.jpg. C933103 (talk) 02:44, 14 September 2018 (UTC)

editor needs to point to legislation. no mention of Japanese personality rights, ([29]; [30]) Chinese personality rights is draft [31] "commercial use" requires consent. see also w:Personality rights, and Template:Personality rights -- Slowking4 § Sander.v.Ginkel's revenge 03:42, 14 September 2018 (UTC)
It seems like the editor pointed to a law reported by media that doesn't actually exists in those countries. C933103 (talk) 04:25, 14 September 2018 (UTC)
I've heard that there were some civil cases where photographers lost the suit and were required to reimburse those photographed in public areas, ie. Subway station. — regards, Revi 13:13, 17 September 2018 (UTC)

Personality rights aren't copyright. If personality rights are a problem for some uses, it should be enough to put {{Personality}} to warn reusers.--Pere prlpz (talk) 16:30, 19 September 2018 (UTC)

I'm seeking something I can't easily search for as I don't know the proper terms or they are too common or vague to Google for wiki help. I know I've seen something like what I'm after before, perhaps only on Wikipedia.

I'm uploading images that are related. I'd like to add thumbnails to related images in a set. "Set" and "album" and "gallery" are vague words. I don't think a gallery is what I'm after. I'm not making a photo album. I'm simply trying to offer options of similar shots taken of the same thing on the same day. Please feel free to just point me to an appropriate help page.

For example, I'd like to cross reference these:

And, as I said, I have other sets I'd like to "join". — Preceding unsigned comment was added by 98.143.77.163 (talk) 03:13, 17 September 2018 (UTC)

Also, I'm uploading Flickr select images from a prolific talented person on Flickr who's given CC2.0. It's a little tedious and I'd appreciated advice on automating or batch uploads. I'm not going to do his/her whole gallery, but a dozen or two very select choices I intend to use on WP and WV articles.

Thanks in advance, ~ JasonCarswell (talk) 01:14, 17 September 2018 (UTC)

I think a use of gallery was actually what I was after, but now I far more prefer that Parkour example. Thanks a bunch. That Living Computer Museum trick is nifty. Any advice on Flikr to Commons? ~ 98.143.77.163 08:25, 17 September 2018 (UTC)
https://tools.wmflabs.org/flickr2commons/#/: some aspects of the UI are a little tricky at first, but it works well. Feel free to get hold of me if you have trouble using it. Expect to do some editing on the pages it generates, it does a lot of the work, but not everything. - Jmabel ! talk 16:19, 17 September 2018 (UTC)


Thanks! ~ JasonCarswell (talk) 02:08, 20 September 2018 (UTC)

The source has a cc by sa 3.0 license at the bottom of the source but I am not sure it refers to the photo since it is next to wikipedia's article on the subject in the photo. Secondly, the image resolution is very small and this is the uploader's first image on Commons. Is this image acceptable for Commons or a copy vio? Best, --Leoboudv (talk) 08:25, 19 September 2018 (UTC)

Need help categorizing an image

Hi, en-wiki regular, first time using commons. This image https://commons.wikimedia.org/wiki/File:Victoria_Station_geograph-4047117-by-Ben-Brooksbank.jpg seems to be in the wrong categories, they're all 1998 instead of 1988 .. I can fix easily but it's strange to me that there are so many incorrect cats. Would like a second opinion. --LaserLegs (talk) 11:28, 19 September 2018 (UTC)

Hello there. There are some typos in category names, just go ahead and change them. Rodhullandemu (talk) 11:49, 19 September 2018 (UTC)

For possibly more objective (or more) results (or if you do not want to read 50 KiB plain text, or if you want relieve the Wikimedia servers) there is now also for Commons this as gadget (universal version, analogous [32],[33]). PS: Comments can also be submitted here. -- User: Perhelion 15:02, 19 September 2018 (UTC)

September 20

Structured Data - IRC Office hours in October and November

There's a lot happening in for Structured Data: there's an ongoing look at structured statements, a talk page discussion about starting the property creation process on Wikidata, coming next week is a search prototype, coming next month is the first feature release (multilingual captions, you'll be hearing more about that next week or so). As such, I've scheduled two IRC office hours in the next six weeks for the project.

  • Thursday, 4 October from 17:00-18:00 UTC in #wikimedia-office
  • Thursday, 1 November from 17:00-18:00 UTC in #wikimedia-office

Some more information, including a link to join and date/time conversion, is on Meta. I'll post a notice and send out an email to the Commons mailing list as we get closer to the dates. Keegan (WMF) (talk) 20:55, 20 September 2018 (UTC)

September 21

Hermann Lukowski

Hi, Some German and/or Polish speaker(s) are needed to find out about this person: Category:Hermann Lukowski. Thanks, Yann (talk) 06:45, 21 September 2018 (UTC)

I need 1000 deleted selfies, to develop selfie detection in the Commons Android app

As a developer of the Commons Android app, I am programming selfie detection (when the user tries to upload a selfie the app will warn them that selfies are off-topic, and ask them whether the person in the picture is really notable).

For machine learning, I need a lot of real-world selfies. I could use Category:Cell phone selfies but it contains only 39 pictures (I need at least 1000), and the selfies in that category were kept for some reason, whereas I prefer selfies that have actually been deleted, as they are more representative of what we don't want on Commons.

Problem: Only administrators have access to deleted images.

Would an administrator be kind enough to send me at least a thousand deleted selfies? Technically, you could maybe search for deleted files that have the word "selfie" in their close reason? Not sure how deleted files are stored. If it is too complex, just send me the latest 10000 deleted files and I will sort them by myself (if possible, only images uploaded from the app: Category:Uploaded with Mobile/Android or https://quarry.wmflabs.org/query/10587).

Please ping me or see my email address at https://github.com/nicolas-raoul thanks a lot! :-) Syced (talk) 03:55, 13 September 2018 (UTC)

@Syced: Unfortunately, searching deleted files and pages is not easy. I would take a look at Category:Personal files, which might have some of what you're looking for. The only real distinction between selfies we keep and selfies we delete is whether the person is a notable person or Wikimedia contributor. Machine learning probably can't tell the difference. I would use this logic: "Warn if the photo is taken with the front-facing camera AND depicts one or two people and little else AND user has less than total 100 edits across Wikimedia projects." Guanaco (talk) 04:32, 13 September 2018 (UTC)
Wouldn't it be possible to search for the deletion request pages, and then get the file names of the associated files? Front facing camera (should be in the metadata?) sounds like an excellent suggestion btw. Effeietsanders (talk) 05:05, 13 September 2018 (UTC)
@Guanaco: 99% of uploaded selfies are of non-notable people, I know this because I review new uploads a lot, and because out of thousands of uploaded selfies only very few are left on Commons. As you say the app will only show a warning, so machine learning not being able to tell whether a person is notable or not is expected, and it is not a problem. We will use the logic you describe. Thanks for the personal files link, I should be able to select a hundred mobile-taken selfies using its intersection with the Category:Uploaded with Mobile/Android category. I still need 10 times more samples, though. Syced (talk) 05:21, 13 September 2018 (UTC)
Hi, Someone will an account on tools labs should be able to create a list of deleted files. Regards, Yann (talk) 06:51, 13 September 2018 (UTC)
just do not delete the 1% such as File:Dorothea Lasky.JPG. the fact that notable people do not want to upload self-portraits is a bug not a feature. Slowking4 § Sander.v.Ginkel's revenge 04:07, 14 September 2018 (UTC)
@Syced: List of 1931 deleted selfies. A few more selfies are also found in Category:Selfies and 30 in a category that I created a while ago, Category:Selfies with arm visibly stretched out. But not all are taken with a cell phone and some are simply not yet further categorized. - Alexis Jazz ping plz 12:14, 17 September 2018 (UTC)
@Alexis Jazz: Wonderful! Unfortunately these files are red for me... probably because I am not an administrator. Would any administrator be kind enough to download them, put them on a network drive, and send a link to my email address (link to it in my first message above). Thanks so much! :-)
@Syced: I would have thought "front camera => reject" would be the simplest and just use EXIF data. Don't see the need to augment that with any other attributes such as identifying people or counting user contributions as Guanaco suggests. It is a poor quality camera on most phones and often linked with filters such as skin softening. Requiring that contributors use the rear camera to take photos for Wiki would not IMO be any hardship and result in higher quality images and no selfies. -- Colin (talk) 12:23, 17 September 2018 (UTC)
Unfortunately most phones do not put this information in the EXIF. Syced (talk) 02:50, 18 September 2018 (UTC)
@Colin and Syced: If the user has been a contributor, selfies are welcome for userspace. Also, on occasion there are photos of notable people or natural phenomena taken with the front camera. Some people also have a broken rear camera. I think a warning can be useful, but outright rejecting images could turn editors off from the app and/or Commons. Guanaco (talk) 17:41, 18 September 2018 (UTC)
Well I think the balance between warning, strong warning, and just banning has to be determined by admins who will end up having to delete the photos that are either multiple selfies or too poor quality. The point is that we don't need fancy AI to virtually eliminate a whole class of image. There is a stronger danger I think that false hits will end up with valid images rejected, and that could also put users off. I think it will be hard to tell the difference between a selfie and a photo of a celebrity taken close up, or even a photo of a painting portrait in an art gallery. Whereas "is the camera facing the photographer" test is much simpler. -- Colin (talk) 20:54, 18 September 2018 (UTC)
While we don't want selfies of non-notable people unless they are active Wikimedians who want them for user pages, etc., they're pretty harmless. I'd put this at the low end of Commons' concerns. - Jmabel ! talk 00:10, 19 September 2018 (UTC)
Once again, nobody has ever talked about rejecting selfies. The final goal of this is to display a gentle reminder of what kinds of pictures are totally off-topic, from time to time, to users who AI guess are uploading mostly selfies. I am glad to see you guys are not too extreme in your rejection of selfies: A few years ago we had a person who really wanted the whole app to be shut down just because a few % of the uploads were selfies. Cheers! Syced (talk) 01:58, 19 September 2018 (UTC)
If the info in in the EXIF and is in a relatively standard format, it should be possible to query to find which images on Commons were taken by the front camera of a mobile phone. Then we could get some stats on what number of those are useful. I think that would be a valuable exercise, though biased because so many selfies and other unwanted images will already have been deleted. If that was compared to the rear camera, we might get a feel for whether enabling the front camera is really providing any significant advantage. The concern above that the rear camera is broken and the user only has a front camera, must apply to such a tiny number of people as to be insignificant. Remember that it is perfectly possible to get a portrait of oneself by asking someone else to take the picture with your camera, or another camera. So there is not really any significant use-case for requiring Commons to accept front-camera photos, if rejecting them leads to the elimination of a lot of admin hassle.
All things are a balance and many times people forget just how few admins there are on Commons or editors on given Wikipedia articles. I remember Wikipedia had some dreadful experience with Canadian Psychology 101 students in classes of 1000+ dumping plagiarised and ignorant sentences into medical articles. The prof thought that Wikipedia had millions of editors with nothing better to do that fix these, and even thought they could use the "was it fixed/removed" as a free way to get the student homework marked. Mobile phones are a particular problem as the numbers are massive (literally billions of photos taken a year) and the userbase likely less educated about Commons scope. If you start introducing AI into the equation, there is a risk people will have fun trying to defeat the algorithm for fun, or who mock us when the algorithm misfires on another subject. -- Colin (talk) 09:20, 19 September 2018 (UTC)
This thread is not about whether enabling such AI feature. The thread is about where to find images to produce that kind of feature. Once developed, we can see if it's useful.
Furthermore, it makes an interesting point that if we had keep those images, they would be useful to somebody - there are lots of developers interested in training AI to recongize some kinds of images. The images we delete as personal or redundant and therefore useless would be useful to a lot of people.--Pere prlpz (talk) 15:51, 19 September 2018 (UTC)
"If you start introducing AI into the equation, there is a risk people will have fun trying to defeat the algorithm" = heavens, we would not want to automate the soul crushing backlog, since there are so few admins to do it. better admin personal control of a broken and bankrupt system, than introducing some tools, not entirely within their control. if we introduce AI, there is a risk admins might be shown to be redundant, or counter-productive. Slowking4 § Sander.v.Ginkel's revenge 03:30, 21 September 2018 (UTC)
Slowking4 above: just do not delete the 1% such as File:Dorothea Lasky.JPG. the fact that notable people do not want to upload self-portraits is a bug not a feature. Colin above: Requiring that contributors use the rear camera to take photos for Wiki would not IMO be any hardship and result in higher quality images and no selfies. Whatever its value or the notability of its subject, File:Dorothea Lasky.JPG used the rear camera and yet is a selfie. (But change "no selfies" to "a lot fewer selfies", and I'd agree.) -- Hoary (talk) 04:23, 21 September 2018 (UTC)
I never claimed that selfies can't be taken (or unwelcome portraits of oneself taken by others) with a rear camera. Just that I think the front camera selection, if the EXIF is available, is an easy win and likely have very few false positives. Pere prlpz, no, Commons Scope is not "We will host any image that someone might want to train an AI algorithm on". That's not our purpose. I actually find it baffling that anyone might think that Commons would be a suitable source of selfies to train an AI algorithm. There are just one or two other websites with probably several billion selfies that one could train on. -- Colin (talk) 08:36, 21 September 2018 (UTC)
I agree that storing images to train AI algorithms is not our stated purpose, and that embracing that purpose would need a discussion that doesn't fit this thread. However, Commons has a great advantage for those purpose: we have millions of labelled images. Other sites host large amounts of images but they are not labelled. In fact, the most important modification to become a perfect repository for AI would be stop deleting images with the (otherwise reasonable) rationale "we have plenty of better images about that" (like selfies, family photos or blurred birds).--Pere prlpz (talk) 11:23, 21 September 2018 (UTC)
i find it baffling, that people can tl;dr, why commons cannot do this or that task. scope is the new "i don't like it"; commons is not paper, and a few thousand selfies, do not harm the other images. meanwhile, we have thousands of crotch-shots, maybe they should train on those. "very few false positives" = do not do creative work here, go anywhere else. it is a failure of imagination. Slowking4 § Sander.v.Ginkel's revenge 12:21, 21 September 2018 (UTC)

lower case a...z in Categories

At the moment, in Categories of all wiki projects the links are indexed under A...Z upper case letters (latin or other). BUT: they look so similar (A = a? α? B = b? β? and so on). Has there ever been a thought to change them into nice big lower case letters? Thank you. Sarri.greek (talk) 15:39, 17 September 2018 (UTC)

  • Since categories are supposed to be in English whenever that is reasonable, that shouldn't be an issue. Do you have a specific problem in mind (e.g. a case where this is unclear)? - Jmabel ! talk 16:21, 17 September 2018 (UTC)
    • The names of files can be in any language, @Jmabel: , especially for pronunciation-audios (Thank you for your help with those). Each .ogg file is indexed/alphabetized manually. e.g. words under K (k) and under Κ (κ) in this Category. Some Cyrillic letters (Russian), and Greek, may look very similar to latin letters.
      I am suggesting, that at an international platform, it would be very helpful to have lower case letters as headings, because they are more precise than the capitals. Of course, greek is a small exception. Here are some greek letters which could be perceived as latin: Α Β Ε Η Ι Κ Μ Ν Ρ Τ Υ Χ Ζ and Ο = α β ε η ι κ μ ν ρ τ υ χ ζ and ο (this is identical to latin o, but coded differently). Thank you. Sarri.greek (talk) 08:42, 18 September 2018 (UTC)
    • I see, you are talking about how content of categories are indexed, not how the list of categories is indexed. - Jmabel ! talk 15:14, 18 September 2018 (UTC)
  • @Sarri.greek: Subcats and other category content will be sorted alphabetically by name (incl. filename, for files) unless a sort key is added in the categorization, introduced by a pipe character. When you add [[:Category:Greek letters by letter]] to its subcat Category:Psi (letter) is will sort under "P" in Category:Greek letters by letter, but if you add instead [[:Category:Greek letters by letter|Ψ]], it will sort under "Ψ". (Adding such sort keys to files is also possible although not a good idea in most cases.) See also this. -- Tuválkin 00:17, 21 September 2018 (UTC)
    • Thank you @Tuvalkin: Tuválkin, I did not explain myself well (not an anglophone). It is not the sorting, that I meant. It is the use of upper case letters as these little headings e.g. at this Cat the latin and greek symbols are sometimes identical. Suppose large lower case were used: latin: a instead of A, greek: α instead of Α and so on. But, never mind, thanks. Sarri.greek (talk) 00:37, 21 September 2018 (UTC)
  • The headings are always capitalized regardless of how you case the sortkey. E.g., File:Ell-Hlios.ogg will always be under the heading "Η" (upper case eta) regardless of it being categorized as [[:Category:Greek pronunciation|ηλιοσ]] or as [[:Category:Greek pronunciation|Ηλιοσ]]. Simply, the first character is taken, capitalized if its Unicode properties allow it, and used as the heading in the parent category. The only exception are characters like "ß" or "ʔ", with unusual casing rules (cf. "ẞ" and "ɂ"<"Ɂ", resp.). -- Tuválkin 01:27, 21 September 2018 (UTC)
  • Hmm, new idea: You can use "𝛈" or "𝝶", from the Unicode block of en:Mathematical Alphanumeric Symbols. It’s phishing, but it’s for a good cause, I guess. -- Tuválkin 01:32, 21 September 2018 (UTC)

The GFDL license on Commons

18:11, 20 September 2018 (UTC)

Wait one cotton pickin minute...Commons allows local uploads? News to me. GMGtalk 18:18, 20 September 2018 (UTC)
@GreenMeansGo: Really? How have you been uploading? - Jmabel ! talk
I mean. it's kindof a silly artifact of "post this message to all projects that allow local uploads". Also sarcasm. GMGtalk 01:09, 21 September 2018 (UTC)
@GreenMeansGo: the Local upload configuration info (few more details and working links) actually doesn't list Commons, so I had to make an exception to make sure Commons wouldn't be excluded. But maybe it really was a mistake? Otherwise how can we explain "Cross-wiki upload from commons.wikimedia.org"? If it's a cross-wiki upload from Commons, where did it get uploaded? - Alexis Jazz ping plz 01:34, 21 September 2018 (UTC)
The sound of one wiki clapping... - Jmabel ! talk 04:36, 21 September 2018 (UTC)
Oh. Okay then. I was just being silly. GMGtalk 10:07, 21 September 2018 (UTC)

Some help with agricultural things

A colleague has uploaded File:John Deere 4720.jpg and we have a category for the tractor. But we would like to know is if there is a category for the additional machinery on the tractor. Thanks. B25es (talk) 16:20, 21 September 2018 (UTC)

Village pumps

Here in Pakistan there is different types of water pumps that drawn water from the ground mechanically. The people of Pakistan were first time introduced by a newly formed of mechanical pump after the flood 2010.The government of Saudi Arabia installed many pumps in Southern Punjab province. — Preceding unsigned comment added by Vet Hassnain Khan (talk • contribs) 17:22, 21 September 2018 (UTC)

  • Signing your posts on talk pages is required and it is a Commons guideline to sign your posts on deletion requests, undeletion requests, and noticeboards. To do so, simply add four tildes (~~~~) at the end of your comments. Your user name or IP address (if you are not logged in) and a timestamp will then automatically be added when you save your comment. Signing your comments helps people to find out who said something and provides them with a link to your user/talk page (for further discussion). Thank you.   — Jeff G. ツ please ping or talk to me 12:00, 22 September 2018 (UTC)
@Vet Hassnain Khan: This is our virtual village pump, where we discuss topics related to Wikimedia Commons. Your photos and other media depicting water pumps and other subjects are welcome, as long as they comply with our policies.   — Jeff G. ツ please ping or talk to me 12:00, 22 September 2018 (UTC)

September 22

Template:Latvian coins

The template {{Latvian coins}} is outdated and has dead links. It is titled coins but I think it only applies to bank notes. Euro latvian coins cannot use this license. Triplecaña (talk) 16:11, 23 September 2018 (UTC)

Hi, A workshop on Copyright and free licenses will be conducted on 19 evening-21 October 2018 at Bangalore (or slightly around). It is aimed for users actively involved and recently active on Wikimedia projects (especially Wikimedia Commons). Wikimedians from South Indian states of Andhra Pradesh, Karnataka, Kerala, Tamil Nadu, Telangana can participate in the workshop. I will be the primary trainer.

The workshop is sponsored by CIS-A2K. Please see m:CIS-A2K/Events/Copyright workshop: South India for details. As a measure to bridge the gender gap, female Wikimedians are encouraged to apply. If you are interested to join in the workshop as a participant, please fill this form. The last date for submitting the form is 2 October 2018. Regards, Yann (talk) 05:24, 24 September 2018 (UTC)

@Yann: I wish you luck with this. Will languages other than English (like Hindi) be used or supported?   — Jeff G. ツ please ping or talk to me 05:46, 24 September 2018 (UTC)
We had a workshop in Delhi 3 weeks ago in 3 languages: English, Hindi, and Penjabi. But in South India, many people don't know Hindi, so it will be mostly in English, with eventually translations in Kannada, Tamil, etc. Regards, Yann (talk) 06:14, 24 September 2018 (UTC)

Public domain of Madagascar postage stamps

I found these postage stamp from Madagascar, but I did not find any information about the expiration of copyright of postal stamps from this territory in Wikimedia Commons. Without knowing, I will not put on this site. Mário NET (talk) 10:05, 22 September 2018 (UTC)

When was that stamp issued? Ruslik (talk) 20:55, 22 September 2018 (UTC)
Here on this page I found the information that belongs to 1963. Mário NET (talk) 21:34, 22 September 2018 (UTC)
The copyright term in Madagascar is 70 years after the author death and can not find any exceptions for postal stamps. Ruslik (talk) 20:54, 23 September 2018 (UTC)
Thank you for give-me these information, Ruslik. Mário NET (talk) 03:33, 25 September 2018 (UTC)

September 23

Picture of the Day caption, 24 Sept

The English caption for the Picture of the Day is bizarre. It identifies the woodpecker in the picture, but not the (palm?) on which it is perching. Then there is a digression for two sentences about the Republic of Trinidad and Tobago, and the first meeting of its Parliament.

There is no explanation offered as to a connection between the woodpecker and Trinidad and Tobago. Are the captions in other languages equally bizarre? --EncycloPetey (talk) 00:23, 24 September 2018 (UTC)

Also, there is a Speedy deletion notice on the Main Page, though I cannot track the source of the problem. --EncycloPetey (talk) 00:25, 24 September 2018 (UTC)

@EncycloPetey: This is being discussed at COM:AN#Main page issues.   — Jeff G. ツ please ping or talk to me 01:20, 24 September 2018 (UTC)
Thanks, I had e-mailed Billinghurst, as I knew he was around and an admin, so the functional problems seem attended to now.
However, there is still the odd photo caption:
"A female red-crowned woodpecker. . . Trinidad and Tobago became a Republic. . . September 24 is Republic Day. . ."
I mean. . . what??? Three seemingly unrelated statements, only one of which seems to connect to the Photo. --EncycloPetey (talk) 02:24, 24 September 2018 (UTC)
@EncycloPetey: Look at the file description page – the photo had been taken on the island Tobago, part of Trinidad and Tobago. And today is 24th September, the anniversary of T&T became a republic. This is the common style Charlesjsharp describes his photos that have been chosen POTD. --jdx Re: 06:02, 24 September 2018 (UTC)
As I say, the photo caption makes none of this clear. --EncycloPetey (talk) 13:29, 24 September 2018 (UTC)
Indeed, the caption in English is confusing. I've created a protected edit request to clarify that over at Template talk:Potd/2018-09-24, but only noticed the Village pump discussion now. —⁠andrybak (talk) 20:28, 24 September 2018 (UTC)

Can we standardize category names for regions of Peru?

I wanted to put template {{Regions of Peru}} on various "of Peru by region" categories and their subcategories. However, the subcats aren't set up consistently, so the template won't work on all of them.

There are two main issues:

Here are some ways this could be resolved (listed in my order of preference):

  1. Standardize both the capitalization of "region" and use of "the" across all subcats of "of Peru by region" categories so that the template works the same in all of them. This would require renaming categories. If we choose lower-case for "region", the template would also need to be changed.
  2. Standardize the capitalization and use of "the" individually in subcats on a case-by-case basis in individual by-region categories so that the template could be made to work, although the coding would be different in different places. This would also require renaming some categories, but probably fewer than with option 1. The template would need to be changed to require specifying the word "region" in the suffix parameter so that the needed capitalization could be specified.
  3. Leave everything as it, but create selected category redirects to make the template work in individual areas.
  4. Leave everything as it is, and just accept that the template won't work everywhere.

Is there any interest in or support for standardizing these category names? If there's a better place to discuss this, please let me know. --Auntof6 (talk) 07:40, 24 September 2018 (UTC)

I agree that in such cases, the category names should be standardized. --Reinhard Müller (talk) 08:48, 24 September 2018 (UTC)

Access GeoJSON via Commons API

I would like to download the GeoJSON of all Amsterdam districts via the Commons API. But it only seems to be working for pictures. Does anyone know a workaround? Thanks in advance! --Renek78 (talk) 13:08, 24 September 2018 (UTC)

15:23, 24 September 2018 (UTC)

Automated Wikidata galleries

Hi all. In case it's of wider interest, I've set up {{Wikidata Gallery}} that takes inputs of a property number and a QID and uses those to create a gallery (using Listeria) with an infobox. It has to be substituted. A live example is at Dytaster. Thanks to @Christian Ferrer for suggesting this, and to @Magnus Manske, RexxS, and Lockal: for the tools this is built on. Note that more complex galleries can be created using different sparql queries, e.g. see User:Mike Peel/Radio telescopes by country. Feedback welcome! Thanks. Mike Peel (talk) 21:29, 24 September 2018 (UTC)

@Mike Peel: I'm assuming a bot will then populate the gallery? Assuming I did the example correctly. Bidgee (talk) 21:39, 24 September 2018 (UTC)
@Bidgee: Yes. You can manually tell the bot to update it by clicking on the 'Manually update list' link at the top-right - I just clicked on that on your page and the bot's updated the page. Thanks. Mike Peel (talk) 21:53, 24 September 2018 (UTC)
Ok awesome. Been working on adding data to WikiData and this would be a good way of having a gallery that you don't have to painstakingly update. Bidgee (talk) 00:41, 25 September 2018 (UTC)

Geology rock outcrops

Here is an typical example of what is often seen in Sweden. Rocks sticking out on an otherwise sedimental plateau. This was gletser territory. Is there a name for there kind of formations?Smiley.toerist (talk) 12:06, 24 September 2018 (UTC)

That doesn't match my understanding of glacial erratics. I checked the Wikipedia article: Glacial erratic to see if my recollection was flawed, but I think I'm on solid ground (so to speak). The key point is that they have been moved by glaciers from their original location, and are almost always completely almost the exposed. That doesn't apply to this photo.--Sphilbrick (talk) 20:38, 24 September 2018 (UTC)
I agree the example appears to show something bigger than erratics, which are boulders that have been picked up and carried from elsewhere by glaciers, floods, &c. If “sediment” in the OP means soil or other recent deposits, then I’d just call them outcrops. But if these are igneous rocks appearing where the surrounding bedrock is sedimentary (and, being softer, eroding faster), they may be intrusions. Given what little I know about Scandinavian geology, my first guess would be granite shield bedrock (bared by glacial action) outcropping from recent cover.—Odysseus1479 (talk) 06:07, 25 September 2018 (UTC)
That would have been my guess as well, a geological map of the area could probably help clearing things up. In any case, clearly there has been quite a bit of en:Glacial polishing going on. --El Grafo (talk) 07:48, 25 September 2018 (UTC)
Oh, there's a bedrock map online at [35] (found here). Looks like there's only igneous rock (granite, pegmatite) around that area, so we're very likely dealing with outcrops rather than intrusions. --El Grafo (talk) 08:03, 25 September 2018 (UTC)

Czechoslovakia by year categories

Category:Rail transport in the Czech Republic by year has the years 1973 to 1992 where the Czech Republic does not exist, it was part of Czechoslovakia. (From 1918, earlier Austria-Hungary and before that the Bohemia and Moravia kingdoms). Category:Rail transport in Slovakia by year starts correctly from 1993 on. I suggest the years 1973 to 1992 be moved to a new series Category:Rail transport in Czechoslovakia by year. There is only one tram category involved: Category:1992 in tram transport in the Czech Republic. I did a quick check but nearly al images come from the Czech part of the Czechoslovakian republic. But images from Slovakian side could always come. Any other ideas?Smiley.toerist (talk) 10:39, 25 September 2018 (UTC)

  • I don't if that can help but in France, we had new regions in 2016, and someone had the idea to create categories with "in present-day" (note that I have no specific opinion if it is the good way or not)
Category:2015 in Languedoc-Roussillon always exist as a useful category however it is also now part of Category:2015 in present-day Occitanie, itself part of Category:Occitanie by year. We also added infos in some templates or category, example in Category:Seasons 2017 in Occitanie, you can read some infos at top. But is seems that it is quite more difficult for you case as it is the opposite, I mean a big entity "Czechoslovakia" was divided into several smaller entity, while in France some regions have been merged into bigger new regions. Sadly, I can not help you more than to give you that example. Sorry, Christian Ferrer (talk) 16:53, 25 September 2018 (UTC)
You could use Category:Czechia by year, which covers the whole history of the region.Joostik (talk) 17:10, 25 September 2018 (UTC)
I am sorry, but you get a confused mess: Category:Built in the Czech Republic in 1258? There is certainly no republic in 1258, and there is no political Chech entity before 1918. Only Moravia and Bohemia where political regional entities. Czech nationalism started in the 19th century. If 'Czech Republic' is replaced by 'Czech lands' it would be clearer. Smiley.toerist (talk) 22:52, 25 September 2018 (UTC)
Just a note that this discussion is also underway at Commons:Categories for discussion/2016/12/Category:History of Czechia. - Themightyquill (talk) 12:43, 26 September 2018 (UTC)
The reverse French example would be: '1990 in Czech lands of historic Czechoslovakia' and '1910 in Czech lands of historic Austria-Hungaria'. It is simpler to just use '1990 in Czech lands'. Even better is to use a more regional '1910 in Moravia' or '1990 in Bohemia'. '1910 in Austria-Hungary' is correct but much to broad. Austria-Hungaria is an constitutional settlement from 1867 for de the Habsburg domiminions. The farther you go back in time the less defined the states are. Before the French revolution large parts of Belgium where Habsburg territories.Smiley.toerist (talk) 12:52, 26 September 2018 (UTC)
As an aside if your going to catalogue by year then the files should also appear in a main category; for example "Rail transport in the Czech Republic". Otherwise what you have here is a good example of over-diffusion. See https://commons.wikimedia.org/wiki/Commons:Village_pump/Archive/2018/08#Overdiffused_categories. - Broichmore (talk) 13:06, 26 September 2018 (UTC)
@Smiley.toerist: It's true of almost all countries: Category:1532 in the United States, Category:804 in Italy, etc. It might be anachronistic/teleological, but it's also a logical way for people to find the (pre-)history of a certain country. I'm not sure it's always such a problem. - Themightyquill (talk) 14:31, 26 September 2018 (UTC)

Partial blocks

m:Wikimedia Forum#Does your wiki want to be an early test wiki for the new partial block feature?

Maybe interesting for Commons? It's not fully clear to me if namespace blocks are already possible, which is what would be the most interesting thing for Commons. - Alexis Jazz ping plz 20:37, 25 September 2018 (UTC)

Less interesting on Commons than Wikipedias. Restricting one account from editing, say, a couple of articles about Israel (it has to be fewer than 10), is quite different from restricting them from all photographs of Israel. -- (talk) 20:39, 25 September 2018 (UTC)
I was more thinking about the options for namespace blocks. This makes upload-only accounts as proposed by Donald Trung technically possible. - Alexis Jazz ping plz 06:06, 26 September 2018 (UTC)
@: Wouldn't it be possible to restrict an account from making any edits to "master Category:Israel and all of its subcategories"? But yeah, this would make that more possible, and also perhaps partial blocks against automation, or against users who systematically insert bad categories into existing uploads, but it'll probably take years before this feature is correctly implemented because the current generation of administrators will probably still go for full-sitebans first before these more direct solutions on preventing disruption will be taken into account, though I think that will be more habitual than anything else. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 06:18, 26 September 2018 (UTC)
Lots of things are technically possible, but not necessarily desirable or especially usable.
Think about how Commons works in practice. Administrators are mostly inactive, with a minority doing most maintenance stuff, including blocks. A "good" compliant block following COM:BP would include good faith warnings to the user. Unfortunately unlike en.wp, Commons Admins do not follow the principle of escalating warnings, so blocks tend to be abrupt and without especially friendly warnings, in many cases with no warnings or a block notice because the user has been thought of as a vandal or sockpuppet, even if the evidence for this may not withstand scrutiny.
In these circumstances, complex options for blocks are unlikely to help, an observation made more concrete by considering how many topic bans or interaction bans are in place currently on Commons (can you name any?). What would be far more helpful would be to see more administrators using a process of agreeing behavioural changes, with the authority of a block behind them. This becomes a social contract rather than a technical enforcement, so say, "Your behaviour is unacceptable [quotes evidence], if you repeat the behaviour on [topic or locus] in the next [period] this will result in a block of your account for [period]. If you have any questions please raise them here..." is a way of using BP without resorting to the ultimate step of using sysop tools and yet effectively does the same thing for us as the "refinement" of blocks. Nothing stops administrators doing this now, though only a minority are seen to do this, possibly because it's more hassle. -- (talk) 10:17, 26 September 2018 (UTC)

A possible implementation of partial blocks on Wikimedia Commons Guanaco could be happy with is when deleting a file 📁 that the option "prevent this account from re-creating this page" which could could prevent the same page name from being used by that particular account, although this should probably only be restricted to non-Autopatrolled users and obvious copyright © violations (thus excluding OTRS-pending). But people can brainstorm after the feature is already implemented. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 06:54, 26 September 2018 (UTC)

September 26

Tolerence on the wrong naming of categories?

Categories about Hong Kong (and Shenzhen) have issues (of the naming of the categories, categorization of files) emerged in the next few years, the features of the issues include:

I highly suspect those users are somehow related but I do not have any evidence for this accusation (which might be very severe). Here are my questions:

  • Are they really related?
  • Are they vandalizing?
  • What should the community do to deal with this situation?
  • What should the community do to prevent this situation?

Hope someone can share their ideas and answer my queries. MNXANL (talk) 06:34, 26 September 2018 (UTC)

  • @MNXANL: have you tried discussing it with the user first? Generally speaking categories don't always have to have English names, but this is preferred. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 08:46, 26 September 2018 (UTC)
  • @MNXANL: The project policy is that for Category names should generally be in English, excepting some of proper names, biological taxa and terms which don't have an exact English equivalent. Generally speaking I would stick to latin script and English, if only to increase the visibility and useability of images worldwide. Regarding file names I don't have a problem with combining English and Chinese in a title. The question to ask is "Can I find this image if I type in a search using English". This would mean that there should be an English translation of the images "description" within the file itself as an absolute minimum. Broichmore (talk) 11:33, 26 September 2018 (UTC)
  • I think one of MNXANL's concerns is sockpuppetry. In this case however, the presumed sockpuppets look more like a user who keeps forgetting his/her login credentials. It is not very classy, but as long as he/she does not use multiple accounts for illegitimate purposes (like influencing a vote or evading a block), the user is not doing anything against the rules. --HyperGaruda (talk) 19:09, 27 September 2018 (UTC)

Why doesn’t the Wikimedia Foundation help out with non-controversial maintenance tasks?

Why doesn’t the Wikimedia Foundation help out with non-controversial maintenance tasks and take over the development of certain tasks that help the volunteer community? Specifically in cases where a volunteer editor with a highly active bot that does certain maintenance tasks or even just the maintenance of certain applications like Flickr2Commons? In fact actually using the talent of its own engineers seems like a better way to spend money than to give it to grants which rarely if ever are (1) granted and (2) produce anything. Having a separate Commons:Bot requests/Wikimedia Foundation for fulfilling either indefinite, extremely tedious, and/or backlog-related tasks could greatly benefit the project. This is not a proposal, just wondering why the Wikimedia Foundation takes such a hands-off approach with fighting backlog. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 08:44, 26 September 2018 (UTC)

This has a long history. Years ago the WMF positively refused to give any support to the development of GWT, instead a number of chapters put money in the pot based on a project proposal put together by a couple of us committed (unpaid) Commons volunteers and eventually we paid out a couple of hundred thousand dollars of (non-WMF) development effort making it work.
Bit of an aside here ... The WMF has shifted its stance, and currently sees the promotion of GLAM as within its remit. This remains controversial as in theory the WMF is supposed to be there for operational support of our open projects, rather than promoting any specific types of content, or for controlling content and contributors. As the WMF has shifted to dabbling in aspects of content and user moderation, their legal exposure has increased. Back in 2010, the WMF could have claimed that content was not under their control, so a legal case resulting from a user doing something criminal on the websites would have had thin grounds to raise a claim against the WMF. However with a history of office bans and investigation, along with actively promoting GLAM/academic content and increasing marketing statements that the WMF is taking action to control behaviour of users, it would be fairly straightforward to assert the WMF's liability if they were turning a blind eye to website misuse.
With the perspective of an old timer, no I do not think you can expect the WMF to pick up F2C. Honestly it does not fit their organizational scope, even though you can probably point to Wikipedia tools which the WMF has invested in, not always successfully.
Any volunteers(s) can put a proposal together for funding and the WMF may or may not chose to help, or one of the Affiliates may feel they can justify helping with funding. However big-ish proposals like creating a F2C replacement or improving it take a long time and probably work best if driven from within a large Chapter or Thorg that has access to staff support. From my experience of doing these things, you can expect to burn out good volunteers along the way, and I would recommend any technically capable volunteer take some time out to tot up how much a minimal fee for their development time should be, and consider being paid for that effort within a proposal. At least that way, if the rug is pulled out from under the feet of the project at the last minute, say by unplannable operational changes by WMF Dev, they will not feel so hard done by, and instead can just put in a "Proposal phase 2" and get paid for a second tranche, or find another developer to do it instead. -- (talk) 10:02, 26 September 2018 (UTC)
@: "Thorg"? Does that mean "thematic organization" (offhand guess) or something else? - Jmabel ! talk 16:05, 26 September 2018 (UTC)
m:Thorg -- (talk) 16:20, 26 September 2018 (UTC)
I take it this has some currency, or there wouldn't be a redirect, but do note that the term thorg does not appear on that page. - Jmabel ! talk 23:32, 26 September 2018 (UTC)
yeah. it is a long history of neglect. the WMF would do well to consider a tool development, and life cycle process improvement, where useful tools are on boarded and supported. rather we have seen tools sun-setting, when their volunteer maintainers burnout, with loss of functionality. there is a lot of "not invented here" and lack of flexibility. we now have wishlist and phabricator to communicate with devs, but there is a fundamental lack of consensus on a design and tool process. and yeah, mass upload tools are hacked together by volunteers and chapters, while dev time is spent on upload wizard (which i will not use). community feedback tends to be frustrated and unproductive. it is bad tactics and strategy. need many more WMF ambassadors (such as wikilibrary) to facilitate and coordinate, there are ways to thread the needle, but WMF seems reactive to problems rather than problem avoidance. Slowking4 § Sander.v.Ginkel's revenge 04:06, 27 September 2018 (UTC)
Unfortunately, my impression is that WMF's upper management is more concerned about empire building and promoting "lighthouse projects" that further their own career than about improving existing projects. That situation has improved a bit in the last few years, but there is still a lot the WMF could do better. Sebari – aka Srittau (talk) 06:20, 27 September 2018 (UTC)

About Commons:Flickr2Commons. WMF paid for alternative tool Special:UploadWizard, which I am a big fan of and it does a great job working with flickr images, and can upload then in batches up to 500 at a time. There are security concerns that prevent it from being open to all users, but maybe we should revisit it, patch the security issues and make it available to all. I wonder how other tools like Flickr2Commons deals with the same security issues. In the past I also in the past, I heard concerns about not making flickr upload too easy, because with wider use there would be more abuse (flickr-washing, etc.) which might outpace our policing capabilities. --Jarekt (talk) 12:07, 27 September 2018 (UTC)

September 27

Misidentified species, bird

The image File:Stagonopleura oculata.jpg is misidentified I thought, and I found an annotation on the image that agrees and suggests Stagonopleura bella. I removed the image from teh english wikipedia and wikidata, but not sure of the best way to have it renamed and removed at other languages. cygnis insignis 15:45, 27 September 2018 (UTC)

Done, a filemover should perform the rename soon. Since the original photographer already changed the description over at Flickr, I don't expect anyone to contest that. Suggest to wait with removing it from other language versions of Wikipedia until the rename has been done and CommonsDelinker has replaced the old file name with the new one on the other wikis. That way, it should be clear why you are removing the figure even across language barriers. --El Grafo (talk) 16:03, 27 September 2018 (UTC)
Delinker! That was what I recalled seeing in action, and now I know how to invoke it. Cheers, cygnis insignis 16:55, 27 September 2018 (UTC)

check photos uploaded by students

Hello... I have my 2 ninth-grade group uploading photographs in Commons. You can see their work here Category:Photos uploaded by American School of Durango students. I have already caught a couple that were taken from the Internet basically because they are just too good. None are photographers. Can anyone who knows how to do image searches check any others they have concerns about? If you find anything, please feel free to message me or send me a mail at thelmadatter at gmail dot com.... Students who do this will get a 0 for the assignment. Thank you! Thelmadatter (talk) 16:13, 27 September 2018 (UTC)

✓ Done About half in one group were taken from online. None in the other appear to be. GMGtalk 16:30, 27 September 2018 (UTC)
Thanks for the fast work! But I need to know which users uploaded one or more copyvios so I can give them a zero and a discipline report.Thelmadatter (talk) 16:40, 27 September 2018 (UTC)
If you look at my recent contributions you can see where the accounts have been notified. GMGtalk 16:52, 27 September 2018 (UTC)
Thanks, Im on their assesThelmadatter (talk) 16:58, 27 September 2018 (UTC)
If you look on my talk page, I provided the list of photos deleted, and those nominated for deletion. Thanks for keeping up with the little darlings. Cheers! Ellin Beltz (talk) 17:06, 27 September 2018 (UTC)
@Thelmadatter: an image-based search is quite easy nowadays. If you'd like to try it yourself: have a tab or window open with Google's Image Search, and the suspect online file in another tab/window. Then just drag the image to the Google Image search bar and it will come up with identical and similar images. There are shortcuts if you know how to, but this is the most basic way and should work with most desktop environments. --HyperGaruda (talk) 18:51, 27 September 2018 (UTC)
There are also Gadgets to put a reverse search through Google Images or TinEye in a tab-menu on file pages, passing the preview as a matching target to the respective service automatically.—Odysseus1479 (talk) 19:32, 27 September 2018 (UTC)

Future changes to how old file page revisions are viewed

Hi all,

The Structured Data team is looking at an important change to how old file page revisions are served back to users when looking at them. This change will only affect the File namespace on Commons, it will not affect any other namespace here or the File or other namespaces on any other wiki. You can read about the change on the Structured Data hub talk page. Keegan (WMF) (talk) 19:41, 27 September 2018 (UTC)

September 28

Banana trees in Milan?

Are these banana trees in Milan? What are the other plants?

Smiley.toerist (talk) 22:57, 17 September 2018 (UTC)

Others, look like Hardy palms, Botanical name Various: Chamaerops, Jubaea, Trachycarpus. Broichmore (talk) 09:06, 27 September 2018 (UTC)

September 18

Title format of file

Hi there, I come here about a dispute over a title format in an image file. The dispute affects two similar files, this and this, although there is a whole series uploaded with a similar... description. Note that I added a description, the one in place now, replacing the former one (now added by the editor in question to the title).

I hold the title format to be conspicuously wrong, and that it does not adhere to regular conventions. However, I was reverted. The other editor refuses to discuss it further. I call him here to make his point. I can settle with a regular, conventional title format for these two files, not as it stays now. Iñaki LL (talk) 21:39, 24 September 2018 (UTC)

I have discussed it a lot. I've simply stated that (a) I uploaded the file (in fact it was more than 9,000 files from this valuable source, which, I guess, deserves some credit); (b) I stated that the sourced information about what the original title in the source was must be maintained (stating that it was the original title in the source); (c) I accepted with no problem your additions (even if they're a primary source) and to exchange the fields in the templage to provide a better location for your additions; (d) I noticed that there is no "regular" convention, but your mere opinion about your proposal being "better" than the initial content of the template, which I provided. Therefore, I end up noticing that, even if your proposal is valid, the current content is valid as well and therefore, there's no need to change a correct content just because you prefer another "correct" content. That's a very basic consensus in wikipedia: respecting the work of good faith uploaders; being ready to reach a consensus, even if that's not the exact result you want; and don't removing sourced information. That's the reason why I refused to go on the discussion. I'm not obsessed with these files (see your contributions in the last days, all about the same) and have a lot of work to do. Feel free to spend your time in another constructive tasks. Best regards --Discasto talk 21:48, 24 September 2018 (UTC)
I will not add more noise here, our histories are there for everyone to see (see also discussions in the files), for which I am happy. Just saying again, that is not a title format, and that is not the original title, the title is the file name, as it is stated in the very source, clear. O, I uploaded a good number of these files as well, but not automatically. Iñaki LL (talk) 21:57, 24 September 2018 (UTC)
Sure (I can use bold typography too). Compare your regular convention here with this. Even the date is wrong. Did you do it in that way with the "good number of these files"? Hopefull not. --Discasto talk 22:37, 24 September 2018 (UTC)
If it is not right, you can fix it. That is called a constructive attitude, not accusatory style. Can someone intervene, strictly on Title conventions and formats?

This is how it stays now, defended by the editor above:

Title: "Español:
Título original: Republicanos cargando un cañón del 21 (2/2)
Localización: Guipúzcoa"

My proposal, as I defend:

Title: "Republicanos cargando un cañón del 21 (2/2)"

Iñaki LL (talk) 09:15, 25 September 2018 (UTC)

To sum up, you want to remove the language tag and avoid to state that, per source, Republicanos cargando un cañón del 21 (2/2) is actually the original title, as stated by the source. Cool.
On the other hand, just to note:
In spite of this, you keep on finding imaginary mistakes in other guys' uploads. Seriously, why don't you focus on your own faulty work, fix your mistakes and stop annoying your fellow wikipedians? I have better ways to waste my time. What you do with yours is up to you. Yo me bajo aquí. --Discasto talk 10:00, 25 September 2018 (UTC)
Sigh... Calm down, stick to content and do not derive into personal. Any improvement made to a faulty contribution of mine is welcome and appreciated, angry and reactive edits are not. So my question remains, is that a regular title that cannot be put into a regular format? I have never seen any such a title format so far. Iñaki LL (talk) 13:00, 25 September 2018 (UTC)
There is no problem with the title as it stands. - Jmabel ! talk 16:54, 25 September 2018 (UTC)
O no? So Title: "Español: /Título original: Republicanos cargando un cañón del 21 (2/2) / Localización: Guipúzcoa" is a very normal title and it does not deserve to clean it up? That is not what the source states [36], first of all, nor does the Kutxateka source web page state it, since "Título original" is the label and the title content is "Republicanos cargando un cañón del 21 (2/2)". I think that is the rational way to put it, or no, am I wrong? So "Localización:" is also part of the title and that is fine... Iñaki LL (talk) 19:32, 25 September 2018 (UTC)
What precisely are you saying is inaccurate? I was not commenting on accuracy, just that it was a perfectly normal way to title a photo here. File names here do not have to be 1-1 to any external source, though they should not be actively misleading. Does it show Republicans loading a 21cm cannon? Does it come from Marín-Kutxa Fototeka? If so, then the title is fine. If not, then it needs to be corrected. Please don't tell me I need to read 3 or four other pages to work out what you are asserting is wrong, spell out what you are saying is wrong. - Jmabel ! talk 00:00, 26 September 2018 (UTC)
Did I say 'inaccurate'? On 1-1, I completely agree with you, but that is precisely what the editor above claims, not me, but even if it were so, it is very-very simple: 1. That is not the title in the source, just the application's rough importation of all the data in a bundle. 2. If someone is going to quote the title, this is what s/he would get: "Español: /Título original: Republicanos cargando un cañón del 21 (2/2) / Localización: Guipúzcoa". Not about aesthetics, but I cannot think of a more clumsy and inconvenient title to quote anywhere, and hardly descriptive of the picture itself, they are metadata that partially includes picture description. Iñaki LL (talk) 06:32, 26 September 2018 (UTC)

I didn't want to waste my time again but just to illustrate this artificial controversy (I have to say that, in the more of the ten years I've been working in Commons this is one of the most moronic discussion I've been brought to). Does it come from Marín-Kutxa Fototeka?. Yes, for sure. The links are pretty clear in the file description (see here). It's indeed a problem that any original title provided by Fototeka is not accurate. For sure. However, we're not here to invent the title provided by the source, but to show the information provided by it (regardless of our personal likings). Considering the lack of judgement shown by the creator of this artificial controversy (see above about invented claims on a person being buried more than one hundred km from the alleged location) it's a real pity that the very same person that has created an edit war on his exclusive likings, made me waste a lot of time that could have been used in more productive ways, cried about imaginary personal attacks, bothered anyone not agreeing with his claims... comes now here to make the community wastes their valuable time. --Discasto talk 09:04, 26 September 2018 (UTC)

That is what you said ("Yo me bajo aquí"), but here you are back again with your signature erratic, ad hominem style. I have been 10 years in the EU and EN WP also, so I have seen you, others, and more... As I said, my point is content-based and cannot be more clear. Iñaki LL (talk) 17:42, 26 September 2018 (UTC)
  • FYI: "localización" has its own field in the photograph information template, namely |depicted place=. There is no need to put Guipúzcoa in the title parameter, unless it were really part of the image title. --HyperGaruda (talk) 19:31, 27 September 2018 (UTC)
And you can put "título original" behind |original description=. Here is the result in one of you example pictures. I am just not sure if the language tag should be left out, as it makes it easy for computer programmes to identify the language of a certain text. --HyperGaruda (talk) 19:48, 27 September 2018 (UTC)
Good suggestions!!! I can try to update the whole set of pictures (about 9,000) --Discasto talk 20:23, 27 September 2018 (UTC)
@HyperGaruda: At last! Thanks for that, that is something. There is no point in adding the language tag, I never have seen a caption starting with the language tag. (But at least you brought some rationality to the title, or caption) Iñaki LL (talk) 10:55, 28 September 2018 (UTC)

September 25

NRHP template -- help needed to bring it more up to date

Please see Template talk:NRHP#NRHP id number format has changed for a discussion about a possible change that would help this template. --Auntof6 (talk) 05:43, 25 September 2018 (UTC)

The two (count 'em: two!) of us who participated in the discussion agree that an update is in order. Would someone with better template coding skills than User:Jmabel or me please look at the discussion and help change the template? I've summed up the discussion on the talk page. If you have questions about exactly what is needed, please ask there. Thanks! --Auntof6 (talk) 15:12, 28 September 2018 (UTC)

Additional eyes on deletion request United States Senate testimony of Christine Blasey Ford

Additional eyes on deletion request United States Senate testimony of Christine Blasey Ford :

Thank you !

Sagecandor (talk) 09:24, 28 September 2018 (UTC)

Are direct flickr uploads broken for anyone else?

On the upload wizard when I hit "get from Flickr" it just grays out the button and does nothing. GMGtalk 13:04, 28 September 2018 (UTC)

@GreenMeansGo: I just tried and it successfully loaded the image from Flickr. It might have been a timeout in the request on Flickr's end. I'd recommend clearing your browser cache and trying again. clpo13(talk) 17:03, 28 September 2018 (UTC)

September 29

Category:Ships by name

User Themightyquill moved the complete Category:Ships by name without consensus to Category:Ships by name (flat list). It is now the only category (flat list) in Category:Categories by name (flat list). I think this move was not correct. I don't correct it myself, as it has a strong impact. What is the opinion of the community? The discussion page is even closed. --Stunteltje (talk) 08:12, 22 September 2018 (UTC)

Perhaps you could explain why you think it was an incorrect move? It certainly solves a problem, and I don't see what problem it creates. I might mention that I did not move the entirety of Category:Ships by name to Category:Ships by name (flat list), as two other categories remain in the former. That was, in fact, the point of the move. - Themightyquill (talk) 08:30, 22 September 2018 (UTC)
See the discussion page. The problem raised had simple been solved bij recategorisation. Not by renaming, as you did. --Stunteltje (talk) 12:02, 22 September 2018 (UTC)

IMHO, the Category:Ships by name by type should simply be removed, it serves no purpose. The following structure would be logical and completely sufficient:

etc. Thanks --Reinhard Müller (talk) 21:09, 23 September 2018 (UTC)

Unless I'm mistaken, both of your are proposing a de facto flat category of ships by name. Stunteltje proposes leaving it parallel with Category:Ships by name by type, which they must see as useful. Reinhard Müller proposes, instead to delete Category:Ships by name by type, which he doesn't see as useful. But you both want Category:Ships by name by type not to be sub-categorized by other qualities. I've made that desire explicit in the name, and still don't see the problem. - Themightyquill (talk) 07:39, 24 September 2018 (UTC)
Not a very big problem. Just that it is the only (flat list) category in Category:Categories by name (flat list). All other categories in that category are just "by name" only. Why? Are all these other categories nominated to be renamed as flat lists too? --Stunteltje (talk) 19:11, 24 September 2018 (UTC)
But all the other categories in Category:Categories by name (flat list) are not flat list categories. Take a look in any of the "Streets in X by name" categories (Category:Urban squares in Bulgaria by name for example), and you'll see that they are subcategorized. I'm clicking at random here and seeing that none of these are flat list categories: Category:International airports by name‎, Category:Photographers by country by name‎, Category:Airbus aircraft by name‎, Category:Churches in Germany by name‎. Some of the categories in Category:Categories by name (flat list) might be de facto flat lists at the moment, but there's nothing stopping someone from sub-categorizing them if they want to. - Themightyquill (talk) 06:52, 25 September 2018 (UTC)
  • I agree with Reinhard Müller's view. Furthermore moving Category:Ships by name to Category:Ships by name (flat list) achieves nothing. It's self evident that the former is a flat list already. We should keep it simple, rather than lengthening and overcomplicating titles. Furthermore this discussion and those like it should be automatically flagged at WikiProject Ships, so we can take into account the views of our largest and most influential client. -- Broichmore (talk) 10:46, 26 September 2018 (UTC)
I can't understand how you can say it's self-evident. Evidentally it's not because it was sub-categorized by more than one commons user. But that's an excellent point about WikiProject Ships. Had I known it existed, I would have added a notification there during the CFD for more input. - Themightyquill (talk) 12:41, 26 September 2018 (UTC)
I do sympathise it's a mess. There's too much nesting. Your right, spring cleaning is required. We need to do away with two cats at least 1). "Ships names" and 2). "Categories by ship name"; they are better served elsewhere. All the main cats in "Ships names" should be in "Ships" The problem you have highlighted is that if a cat exists people feel compelled to fill it whether it needs it or not. Take "Categories by ship name" it's sitting inside "Ships by type"; duh! What about "Ships by function"! or "Ships by name by type‎". Four cats doing the same job. There was a discussion recently about this very subject; "over-diffusion" of cats and images. See https://commons.wikimedia.org/wiki/Commons:Village_pump/Archive/2018/08#Overdiffused_categories. In this case catting by date. There can be nothing more destructive than that, and it's rampant elsewhere on the project. Almost any file from before the invention of the digital camera doesn't need to be filed by date - Broichmore (talk) 15:20, 26 September 2018 (UTC)
Thanks, Broichmore. I agree that overdiffusion is a problem, and worst of all if there is no consensus that a line has been crosed, we have no clear overarching rule to say "this amount of diffusion is okay but no more" and restrict further diffusion. That's precisely why I created this flat list category. If some people want to diffuse, they can do it without disrupting those who want a category for all named ships. - Themightyquill (talk) 12:59, 27 September 2018 (UTC)
If there's further doubt, take a look at Category:People by name which has a category description stating it's a flat list, and yet, it keeps getting sub-categorized. Category:Women by name might also be a flat list in some people's minds since it has over 84000 sub-categories, and yet, it's sub-categorized by various titles, by country, etc. Category:Men by name and even Category:Men of the United States by name are the same. - Themightyquill (talk) 18:27, 28 September 2018 (UTC)
The same can happen to any main category, it's no reason to change it. All categories require curation. I have already sorted out cat:Ships by name of its spurious sub cats. Please move all the "ship by name" cats back to "Category:Ships by name". I had already sorted out cat:Ships by name of its spurious sub cats. Please move all the ship by name cats back to "Category:Ships by name". Thank you. Broichmore (talk) 17:16, 29 September 2018 (UTC)

Editing data imported from GLAMs

jmabel and I had a short discussion about whether it's advisable to change File page contents that were imported from a GLAM and wanted a broader opinion. jmabel suggests that the data should be preserved unchanged, while I think the page contents should be open to be improved on. Is there any policy about this, and what are your opinions? BMacZero (talk) 16:48, 23 September 2018 (UTC)

The edit that raised this discussion is here, though I'm more interested in general best practice than the merits of this specific edit. BMacZero (talk) 16:50, 23 September 2018 (UTC)
For what it's worth, I often add to what came from the GLAM, but am very hesitant to remove anything. In the case of outright errors, I often contact the GLAM, try to get them to make the correction on their site, then bring over the correction. (I've had to do that with the University of Washington Library over a dozen times in the last 2 weeks!) - Jmabel ! talk 23:28, 23 September 2018 (UTC)
yeah, should not "correct" the metadata from GLAMs, such as here [37]; [38]; [39] because we need to maintain the chain of attribution, and reflect what the reliable sources say. GLAMs are a reliable source, we are not. error correcting at the GLAM is the right way, although less summary. you could use a category, which is a commons interpretation. Slowking4 § Sander.v.Ginkel's revenge 23:45, 23 September 2018 (UTC)
@Slowking4: In that case, what should be done with data that's factually correct but not in the best format for the wiki? (In this case, it's a "Depicted Place" of "New York; New York County; New York", which {{City}} isn't able to automatically parse, or as another example I've encountered, replacing the plain name of an author with their {{Creator}} template). BMacZero (talk) 15:50, 25 September 2018 (UTC)
New York County exists in wikidata https://www.wikidata.org/wiki/Q500416 -- Slowking4 § Sander.v.Ginkel's revenge 16:10, 25 September 2018 (UTC)
That's true, but the {{City}} template, which backs the "depicted place" field of {{Information}}, isn't smart enough to associate the text "New York; New York County; New York" with that entry. This causes "New York; New York County; New York" to show up as a redlink in Special:WantedPages. I replaced it with "New York City" to get rid of this error, and I think that's a reasonable change to make. It doesn't lose or change any information about the file. BMacZero (talk) 02:37, 27 September 2018 (UTC)
@BMacZero: Technically, that should be "Manhattan; New York County; New York" or "Manhattan; New York County; New York City; New York". New York City is the only city I know of that contains counties (five of them).   — Jeff G. please ping or talk to me 09:30, 27 September 2018 (UTC)
@Jeff G.: Interesting, I didn't know that about New York. Are you saying that the original metadata is incorrect? Either way, I'm a bit confused and not sure what action is needed at this point. BMacZero (talk) 17:09, 28 September 2018 (UTC)
yeah, i guess location is more complicated than city template allows. and could indicate metadata from the GLAM and supplement with brackets if you must, why you are wasting time visualfilechanging places in metadata is an open question. and the fact you are "simplifying" new york city data, and questioning it is incorrect. do you have evidence about the data, before you changed it? why not create all the categories you want, and link to wikidata? keep in mind place name jurisdictions change, and there are reasons to keep the historical name, i.e. Danzig and Gdansk. Slowking4 § Sander.v.Ginkel's revenge 01:38, 30 September 2018 (UTC)

September 24

Is a newspaper clipping from 1956, with no copyright status indicated anywhere within the paper, now in the public domain? TenPoundHammer (talk) 06:24, 29 September 2018 (UTC)

What country is the said newspaper published in? Always assume copyrighted (with or without copyrighted status in the paper itself), and then depends on the time period that the copyright applies. Bidgee (talk) 06:49, 29 September 2018 (UTC)
United States, 1956. TenPoundHammer (talk) 07:16, 29 September 2018 (UTC)
No U.S. paper outside of New York renewed its copyrights after they expired, as they would have in that era to retain copyright after 28 years, so if it's not a New York paper & it's not their reuse of someone else's material, {{PD-US-not renewed}} should apply. - Jmabel ! talk 07:57, 29 September 2018 (UTC)
I am afraid your blanket statement about the lack of renewals by newspapers outside of New York is definitely incorrect. Some non-New York based U.S. newspapers did renew their copyright registrations, including the Washington Post, Chicago Tribune, Los Angeles Times, etc. In addition, some contributors to newspapers and other periodicals filed renewals separately (particularly syndicated content as you mention), so you can't always rely on the non-renewal of the issue itself. —RP88 (talk) 11:36, 29 September 2018 (UTC)

Leave a redirect behind when moving a file using script

When I want to rename a file using move script, the option "Leave a redirect behind" has three statuses: 1) unticked; 2) ticked; 3) ticked and frozen (locked). The status is selected by the system automatically. My understanding is that when the file is new, the box is unticked, and when the file is old, the box is ticked and frozen. However, I am not sure how new or how old a file should be to tick/untick the box. Do you have any idea? Thanks 4nn1l2 (talk) 11:53, 30 September 2018 (UTC)

What script do you mean? Ruslik (talk) 19:09, 30 September 2018 (UTC)
Unless the old name was specifically problematic (personal attacks, intentionally misleading name, etc) or there's a housekeeping reason to do so (the exiting name is needed for a different image for technical reasons), file redirects should generally always be kept. Pi.1415926535 (talk) 19:34, 30 September 2018 (UTC)
Ruslik, I think MediaWiki:AjaxQuickDelete-dev.js is responsible for that. Ctrl+F: "moveFile" or "Leave a redirect".
Pi.1415926535, In this case, I think the box should always be ticked by default. I sometimes forget to tick the box. 4nn1l2 (talk) 19:42, 30 September 2018 (UTC)

October 01

Defaultsort of churches

Four years ago, there was this discussion:

It's my belief that categories are there to assist users in finding media we host. So it should be a given that users should find media where they expect to. However, we seems to have inherited a system from en:WP of default sorting churches and pubs by location first, then name. So "St Bernard's church, Sometown" is defsorted so that it appears under "S" rather than "B" (obviously sorting under "St" is unhelpful because most would then sort under "S", which would be overwhelmed and thus useless). Similarly, "The King's Arms, Anytown" appears under "A" rather than where a user would expect to see it, i.e. under "K" (Again, "The" is redundant). There is one user who persists in perpetuating this error, as I see it, claiming that "it's the way it's always been done" is a taxonomically valid reason for the current system. So my proposal is this:

Churches and public houses in the United Kingdom should be sorted by name first, then location, except where the name is part of the category, in which case sorting by location is used..

I realise it may take some time to correct this error, but I don't see that as a reason for not doing it, as I'd prefer to get things right than confound our users, and I invite all to participate in a straw poll to determine consensus. Rodhullandemu (talk) 08:32, 10 September 2018 (UTC)

@Rodhullandemu: Who is the "one user"?   — Jeff G. ツ please ping or talk to me 11:49, 10 September 2018 (UTC)
@Jeff G.: I don't want to embarrass him by naming him, but I confirm he has been made aware of this proposal. Rodhullandemu (talk) 17:10, 10 September 2018 (UTC)
This section was archived on a request by: ReaperDawn 12:08, 12 October 2018 (UTC)
Support by four people, in 2018

Convenience link to original discussion: Commons:Village_pump/Proposals/Archive/2018/09# for UK churches and pubs --HyperGaruda (talk) 07:09, 27 March 2022 (UTC)

Plea

  • Pubs and churches are two very different types of facilities.
  • Many churches are protected culural heritage.
  • Even on pubs, the preference of the name is doubtful, as the same pub may have had variuos names in its history.
  • On churches:
    • To understand the landscape of architecture,it is necessary to know the location of each single church.
    • In historic cities and towns, the churches form an ensemble. Some places have a cathedral, some churches were built for parishs, some for convents. It is useful to see in the list of a county, whiich place has how many historically important churches.
    • In some regions of England, most churches have one of oy six or seven requently used names.
    • Though, in some cases, the patron saint indicates something on the sponsor or on the economy of rthe parish, these relations are no strict rules.

Therefore, to use the sorting as a vehicle of scientific information, the location ought to have the preference. --Ulamm (talk) 12:39, 26 March 2022 (UTC) Fortunately, in a lot of counties, the defaultsorts mainly use the preference of the location, actually.--Ulamm (talk) 18:45, 26 March 2022 (UTC)

Discussion

  • If you are proposing to overturn the consensus from the original discussion to defaultsort by name, in favour of defaultsorting by location: I oppose this and support the original outcome. As is mentioned in the original discussion, grouping by location is done by using a super-category for that location (i.e. Category:Churches in [location]). Sorting by place rather than by the category's name goes against all indexing conventions we usually find in books and other works. --HyperGaruda (talk) 07:09, 27 March 2022 (UTC)
  • You are wrong: Typically, scientific books on architectural history have an index of places. For each place, the buildings, there, are listed by their names.--Ulamm (talk) 20:06, 27 March 2022 (UTC)
  • See the structure of the volume on Gelderland in the series Monumenten in Nederland:
Voorwoord – 7
Ten geleide – 8
Historie – 12
Stijl en verschijningsvorm –20
Materiaal en constructie – 33
Regio's – 40
Steden, dorpen, monumenten – 60
Beknopte literatuuropgave – 374
Verklaring van enige termen – 380
Topografisch register – 388
Register van personen – 394
Verantwoording van de afbeeldingen – 411
--Ulamm (talk) 20:26, 27 March 2022 (UTC)
The list you provided is the table of contents found at the beginning of a book, which is sorted according to page number. If you meant for use to look at the Topografisch register you emphasized, I have some bad news for you: places and buildings are all sorted alphabetically. For example: the building Aardhuis, Het (Hoog-Soeren) is found among entries starting with the letter "A", not under its location Hoog-Soeren further down the list. --HyperGaruda (talk) 10:50, 28 March 2022 (UTC)
  • You neglected the content of the list, Ive copied:
    • The main text is "Steden, dorpen, monumenten", page 80 –373.
      • In the text, for each place, buildings are grouped by types, and in each type, they are grouped aphabetically.
      • In the rating of a Commons category by location it is similar: Buildings are primarily rated by place, only buildings in the same place are rated by their names.
    • After the text, on page 388, there starts the topografisch register, the list of places.
Monumenten in Nederland is avalable as PDF, and therefore easy to copy.
  • Günther Binding, "MASSWERK", the most common German handbook on tracery, has after the text the "Bautenverzeichnis", that is index of buildings. Its first column is:
"Albi, Kath (i. e. Cathedral): 82, 84; Abb. 73, 99
Alençon, Notre-Dame: 104
Atlenberg, Zisterzienserkirche: 24, 169, 216, …
Amiens, Kath.: 23, 27, 44ff, …
Amiens, Saint.Germain: 101; Abb. 94
Annweiler, Pfarrkirche: 255; Abb.94
Auxerre, Kath.: 25, 29
Avignon, Papstgrab, 27, 97
Bacharach, Wernerkapelle: 264; 296
Bampton: 142
Barholme: 101
Barnack, St. John the Baptist: 144; Abb. 152
Bar-sur-Seine, Kirche: 103
Basel, Münster: 204, 354
Basel, Barfüßerkirche: 299; Abb.338
Basel, Predigerkirche: 217; Abb. 251
Basel, St. Peter: Abb. 394
Basel, St. Theodor in Kleinbasel: 287; Abb. 323
Bautzen, Dom: 349; Abb. 323
Bautzen, Nikolaikirche: 349."
Scanning it would cause a problem of copyright, I'm afraid.
--Ulamm (talk)+Ulamm (talk) 14:26, 28 March 2022 (UTC)
Then I think we have reached an impasse. The Topografisch register supports sorting by name, the Bautenverzeichnis by location. Ignore the main text; categorisation is a navigational tool and the best "real life" examples to look at would be other navigational tools such as indexes. In any case, I strongly advise you to set up this proposal at the correct venue (Commons:Village pump/Proposals) and to also take into consideration Commons:Village_pump/Archive/2020/05#Defaultsort for UK buildings RFC, another inconclusive discussion on this subject. --HyperGaruda (talk) 07:16, 29 March 2022 (UTC)
The topografisch register supports sorting by location, too, such as the main section of the volume, "Steden, dorpen, monumenten" does. Within each place (that is the lower level!), the monuments are sorted by kinds of building, and within each kind of building by name.
In large cities, the location is more specified, by sub-chapters for suburbs.
The volume could provide a register of churches by denominations (patron saints or the typical "Grote Kerk", "Broerenkerk", "Oude Kerk", "Nieuwe Kerk" that were chosen to overcome the totemist attitude of Catholic parishs to their patron saint.), but it does not.
The presently more common sorting within Commons categories, with preference of the location above the name, ha the same effect: Normally, onroerend objects are sorted by the places. If there is only one object by place, this provides a sure identification. Where there is more than one object (here church) in a place, they are sorted by their names.
The sorting preferred by that gang creates groups of sometimes more than 20 objects with the same name, that have to be sorted by their places. Therefore, you have almost as much sorting by place, as if you use this preferrently.
The difference is: The location of each building is scientifically relevant, and the better identificator, whereas the patron saint is nothing but a not very good identificator.--Ulamm (talk) 13:04, 29 March 2022 (UTC)
∑:
  • For users who like groups of churches of the same denomination, the "Curches by patron saint"-categories ar eprovided.
  • In "Listed churches"-, "Curches by century"-, "Churches by architectural style"- and "Churches by shape"-categories, the sorting by denomination creates large groups of homonymous churches that on the lower level are sorted by location.
  • A sorting by location, in these kinds categories, creates only one sequence for mot villages churches. Where in one location there is more than one church in this category, this local group provides the information, that the churches in this place have some common properties, this information is provided without a complicated system of sub-categories.
  • The obvious priority of location (as upper level of sorting) above denomination in scientific printed books is denied by @HyperGaruda:
--Ulamm (talk) 20:54, 31 March 2022 (UTC)
  • @Ulamm: May we assume that by "denomination" you mean "name", not religious denomination? Or are you suggesting yet another sorting criterion? - Jmabel ! talk 00:38, 1 April 2022 (UTC)
    In UK, most categories of church buildings are formed like "St Peter's Church, Littleton". This way, the denomination is the name of the patron saint.
    But, as I've mentioned, in the lemmata of most such categories, the location suffix is more specific (less homonymous) than the patron saint.
    And in real life, people say "the church", if they talk of the church of their own parish, and "Littleton church", if they talk of the church of a neighbouring village. In a city, they call the church of a neighbouring parish by its patron saint, of course. Ulamm (talk) 17:14, 1 April 2022 (UTC)
2nd ∑: Actually, the churches of much more counties are sorted by location than by denomination. This has been done by active users, probably the majority of them. Therefore, the voting of four people in 2018 can't be called the decision of a representative majority. It has to be called the destructive initiative of a little gang.--Ulamm (talk) 08:39, 3 April 2022 (UTC)

I had a long conversation with Rod shortly before he died. One of his regrets was that he was unable to finish the work he'd started to apply the 2018 consensus mentioned above. I think he said he'd reached "D" in an alphabetic process. That is probably why "the churches of much more counties are sorted by location", and as he said in the quoted original text, "it may take some time, but thats no reason not to do it." So it's no surprise that the chaos in categories he hadn't yet reached from "E" onwards actually precede his work. But the consensus of those who took part in the 2018 discussion was never intended to be global, nor can it have been, and it was left for two weeks during which nobody added any opinion either way, and then closed. So, unless overturned, it remains consensus. It's alsovery unhelpful to describe a consensus as a "gang". If Rod had lived longer, perhaps he'd have tackled, but without much enthusiasm, the damage caused by Motacilla in sorting categories alphabetically rather than by their function. RIP, mate. Sadly missed 94.174.111.230 21:37, 5 April 2022 (UTC)

The majority has sorted by location, since long. That is a kind of consense.
Four people are a non-siginificant minority.
Most "users" of the Commons did not notice the "polling" of October 2018.
Therefore the practice of the great majotity has to be considered the real vote.
Some very few people had begun to create chaos by beginning to sort by denomination, before 2018. They are the problem.--Ulamm (talk) 13:55, 6 April 2022 (UTC)
The alleged consensus of the longevity of the majority is irrelevant. Throughout history, the consensus of the majority has been shown to be wrong. Take, for example the German Election of 1932. Take the election of Donald Trump in 2016.
Four people, with nobody objecting, is de facto and de jure enough to constitute a valid consensus according to COM:CONSENSUS
All users of Commons could have contributed if they were interested enough. They weren't.
The "real vote" consisted of doing something here that was on Wikipedia. Commons categories are mostly different from those on Wikipedia.
The problem is not those who sought to get it right, which the consensus reflects, but those who unthinkingly followed the wrong thing.
I don't see this discussion going anywhere, and it should be shut down, for Rod's sake if nothing else. He would be turning in his grave if the coroner had not embargoed his body until the outcome of the inquest. 94.174.111.230 22:34, 7 April 2022 (UTC)
The decision tof hose four people was presented to me dug out of the archive. That shows that except of a handful of persons nobody knew of it. That situation is quite different from general acceptance.
Many, probably most users visit the village pump very seldom. I was active in WM Commons also in autumn 2018, uploading own photographs, searching, categorizing, but I was too busy to visit the village pump.
I have objected against that decision on sorting pubs and churches in UK, as soon as I got informed of this gang decision. The problem is, how to cause publicity for this matter.
I have shown that the sorting of immobile objects by denomination rather then location contradicts scientific practice, and that in the case of churches (as well as pubs), it creates groups that (beside the denomination) have nothing in common that distinguishes them from other objects (of the same class). Sorting by location, in contrast, provides an addtional information and creates groups with effective common properties.The location registers usual in scietific books on architectural heritage have a good reason.
These books have registers of artists, of scientists and, sorted by location, of the buildings, but no registers of denominations.--Ulamm (talk) 22:59, 8 April 2022 (UTC)+--Ulamm (talk) 23:26, 8 April 2022 (UTC)
The problem is, how to cause publicity for this matter. - As mentioned earlier, you should make an official proposal at COM:VPP. If accepted by the rest of the community, some sort of rule should probably be added to COM:Categories or even an entirely new policy/guideline page, lest you get lost in endless discussions. --HyperGaruda (talk) 08:16, 10 April 2022 (UTC)