Commons:Village pump/Archive/2020/12

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

images located on earth's diagonal

Hi folks, at the moment there are more than 5000 images with geolocation set to lat=lon: https://quarry.wmflabs.org/query/50218, which is rather unlikely. As we depend to a certain amount on coordinates being set correctly, those coords should either be fixed or deleted. --Herzi Pinki (talk) 18:23, 1 December 2020 (UTC)

User:Herzi Pinki - looking at the query's results, is there a way to filter those with lat/lon = 0? File:Mantowoc North Pier.jpg doesn't appear to have any geolocation info, and if that is getting included, I'm surprised that the query only came back with ~5000 results. Chris857 (talk) 00:39, 2 December 2020 (UTC)
The problem with the file you mentioned is that that wrong coordinates are in the exif-data. And not in the file description. This can also happen with coordinates /= (0/0). These cases would need a solution in the software either by removing such coordinates from Exif or by ignoring them. What you can do is to overwrite such wrong data with correct data using {{Object location}}. You can sort the results of the query by column, which means also by lat or lon. A modified query https://quarry.wmflabs.org/query/50240 will filter away those with (0/0). Still > 5000. Don't know how to grep only those files that do have explicit coordinates aside the exif data. --Herzi Pinki (talk) 06:29, 2 December 2020 (UTC)
There is a data inconsistency with redirects, thanks for letting me look into it in more detail. --Herzi Pinki (talk) 13:38, 2 December 2020 (UTC)
Tracked in Phabricator
Task T269228

transfer pictures from :bar: for en:Schuhplattler

one done
24.7.56.99 01:50, 2 December 2020 (UTC)
 Not done, they are not marked with a compatible license.   — Jeff G. ツ please ping or talk to me 01:57, 2 December 2020 (UTC)
I've updated the source information for bar:Datei:Schuahplattla-Weabung.jpg according to information from the Oldenburg University's library system. {{PD-old-assumed}} should apply and it would go into Category:Collectible cards of Liebig's Extract of Meat Company (or better: Category:Liebigbilder_1892,_Serie_221_-_Nationaltänze_III). I'll see if I can convince the FileImporter to accept this … --El Grafo (talk) 13:46, 2 December 2020 (UTC)
Here you go: File:Schuahplattla-Weabung.jpg. --El Grafo (talk) 14:05, 2 December 2020 (UTC)

FAQs are outdated

Hello, I just realized how outdated (parts of) COM:FAQ are. Examples:

I'm sure there's much more outdated information that should be updated. Just in case someone is bored and needs something to do ;-) --El Grafo (talk) 11:47, 2 December 2020 (UTC)

Special:MediaSearch

Hello, folks! At the Community Wishlist Survey 2021, I added a proposal to have an improved image/video search page. I was told that it already exists, Special:MediaSearch, and it looks awesome!!!

Were any of you aware of it? How could we encourage people to use it? --NaBUru38 (talk) 21:52, 2 December 2020 (UTC)

You can also use filters like "filemime:audio" in the normal search. One benefit of the standard search is that it shows you how many total matches there are, which the special pages does not. I may be overly technically minded, but this way feels more flexible for me. --Fæ (talk) 22:30, 2 December 2020 (UTC)
@NaBUru38: glad to see that you like the new MediaSearch. It's still in active development and I've been slowly spreading the word about the new tool as features become available. I think one of the best ways to encourage people to use it right now is to link to it and reference it in conversations here on-wiki. Word-of-mouth is a pretty good way to share new tools in a small way that can contribute to good, active feedback about using the tool while it's still being developed but before it's ready for any sort of big announcement. Keegan (WMF) (talk) 18:40, 3 December 2020 (UTC)

Commons:Deletion requests/Files uploaded by Coagulans

All these "unsourced" files are my own works, as pointed out in the description. Please someone review the reasons for this deletion request, per WP:POV. Thanks. - Coagulans (talk) 07:28, 3 December 2020 (UTC)

The rationale given was "Scope" and the files were deleted after only 2 days. Should these "unused" files have been used at some MediaWiki-powered FanArt-Wiki via InstantComnmons, than they shouldm't have been deleted at all and clearly not after two days. --C.Suthorn (talk) 12:20, 3 December 2020 (UTC)
@C.Suthorn: Sorry to contradict you, but the DR dates from November 17 and the files were deleted on November 24. How do you count the two days? --Pafsanias (talk) 12:32, 3 December 2020 (UTC)
I did misread your vote from 22nd for the signature of the DR author (and your vote is the only diskussion that took place). --C.Suthorn (talk) 14:24, 3 December 2020 (UTC)

I messed up the Structured Data of File:All Your Base -22-10-10).ogg

I added

and then I got an error that English speech synthesis (Q103169476) should only be used as a qualifier. Is this wrong? Shouldn't I be able to specify what type of English the audio file has in the structured data or are categories supposed to cover that like [[Category:Speech synthesis sounds]]?

I also made some clumsy edits after that which I regret. Instead of doing those I should have posted here immediately after getting the first error on my first edit. LotsofTheories (talk) 08:14, 3 December 2020 (UTC)

That is a bug caused be the Warnings on Wikidata they make no sense on commons are still shown here. This should be fixed in the future. Now you can just ignore this. --GPSLeo (talk) 09:44, 3 December 2020 (UTC)

Kindergarten, day care, preschool

See also: Commons:Categories for discussion/2019/07/Category:Kindergarten.

We now have separate category trees at least for Category:Kindergarten, Category:Day care and Category:Preschools. We also have e.g. Category:Child development centers, Category:Child day care, Category:Pre-kindergartens and Category:Nursery schools. Some of these are linked by {{Category see also}}, but only some, and only to a few of the others – and not all are under Category:Early childhood education or "see also"-linked to it. There are differences, some major, some subtle, but not consistent across countries.

I suppose it makes sense to have categories for specific systems like pre-kindergartens, or Montessori preschools, but shouldn't those preschools be in "by country" categories for more general terms? At least we should agree on a main category they all would belong to (or a few "see also"-linked categories).

Not all these categories are linked to Wikipedia articles by the same name, and few have descriptions, so what they are intended to cover remains unclear.

–LPfi (talk) 12:46, 4 December 2020 (UTC)

Surname category remains empty

Recently I created Category:Bettina Gilois. From Wikidata was added the not yet existing Category:Gilois (surname). I created that category, but the Category Bettina Gilois was not visible in that surname category. The null edit + purge did not help including that of Category:Surnames. As the Category:Gilois (surname) remains empty the sorting is according to the Bettina (given name) as can be seen in Category:Female screenwriters from the United States. What I fear is that at a certain moment a bot comes along and speedy deletes all empty categories in Category:Surnames. How to solve these problems? Wouter (talk) 14:10, 4 December 2020 (UTC)

@Wouterhagens: I fixed it with this edit, purges, and a null edit.   — Jeff G. ツ please ping or talk to me 14:46, 4 December 2020 (UTC)
Thanks, is this just a trick or is there an explanation too? In the latter case, there should be a warning that the label in the official language must also be entered in the label table. I made many of these surname categories and will check them as far as I can remember them. Wouter (talk) 15:40, 4 December 2020 (UTC)
The infobox uses the English label to generate these categories, so you have to set that for the category to be populated. It's also good to add the commons sitelink if you can, as the infobox might switch to using those in the future. Thanks. Mike Peel (talk) 15:43, 4 December 2020 (UTC)
@Mike Peel: That happens even if the family name originates from another language. It seems too English-centric because proper names are excluded from the English language requirement for cat names here. I didn't try using a label in another language.   — Jeff G. ツ please ping or talk to me 15:53, 4 December 2020 (UTC)
I tried it with Category:Franck Balandier. The category Category:Balandier (surname) is empty. In the Wikidata Balandier the label in the official language is indicated as French. I also added for other sites the Category:Balandier (surname). The Category:Balandier (surname) remains empty and does not give a link to Wikidata. Wouter (talk) 16:32, 4 December 2020 (UTC)
Yes, again, you need to set the English label, at the top of the Wikidata item. Sorry if it's English-centric, that's one of the reasons we might want to change to the sitelinks in future. Thanks. Mike Peel (talk) 16:54, 4 December 2020 (UTC)
I went ahead and fixed it. — Huntster (t @ c) 17:34, 4 December 2020 (UTC)

Add color

Please. Can someone make the tree green?
File:Menlo Park California Logo.svg
The color you can see here:
File:Menlo Park California Logo.gif
--Vlixes (talk) 04:40, 5 December 2020 (UTC)

Category redirects, redirects, or both?

I feel like I've not seen a good answer to this.

For category redirects, we have {{Category redirect}}, which helps with maintenance work, and #REDIRECT, which actually pushes the user to the target page. I know the template is standard for categories, but I've seen a couple times now people who seem to want to have both. Doesn't this obscure when there are indeed files/pages in the redirect category? I don't see a documented preference not to use both, however -- why? What reasons take precedence? — Rhododendrites talk |  16:39, 5 December 2020 (UTC)

Our current guidelines are probably insufficient. Anything added to a cat redirect should be automatically moved to the target cat by bot, eventually. Both cat redirects and the second option of redirecting the page, may be deleted by some sysops who make a subjective personal judgement as to how useful they are. In this process the normal user no longer sees these as options when they try typing categories they expect into the category entry field on an image page. A good example of an essential redirect is Category:Crows where the plain English/Common usage word would be used by 98%+ of our contributors.
Categories remain messy, and most of our longer term users tend to avoid getting locked into category 'discussions' which go on for years, sometimes long after those interested have retired from this project. --Fæ (talk) 14:13, 6 December 2020 (UTC)

Eliminazione foto

Buona sera. SONO L'AUTORE DELLA FOTO" Basilica di Nostra signora della Pace " a Yamassoukruo in Costa d'Avorio. Volevo precisare che va bene se eliminate la foto. Grazie. Buona sera (Ex Cooperazione) 5/12/2020 21,00

Do you mean "File:Yamoussoukro Costa d'Avorio Basilica di San Pietro.jpg" which is already nominated for deletion? --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 22:52, 6 December 2020 (UTC)

Someone tries to appropiate my images

Some IPnr tries to put an ownershipname on it. I have the original prints from 1979.Smiley.toerist (talk) 11:45, 6 December 2020 (UTC)

I have uploaded 5 1979 rail Greece pictures (see Category:1979 in rail transport) It was first foreign trip on my own.Smiley.toerist (talk) 11:52, 6 December 2020 (UTC)
@Smiley.toerist: This was done by two different Greek IP addresses in the range 178.147.0.0/16. Per special:contribs/178.147.0.0/16, that range has no valid contributions since 1 October 2020.   — Jeff G. ツ please ping or talk to me 12:26, 6 December 2020 (UTC)

sorting files in a category

How could I sort files in Category:Aryamehr Quran? Hanooz 20:02, 5 December 2020 (UTC)

@Hanooz: Are you talking about the order of the files in the category page? See [1]. – BMacZero (🗩) 17:58, 6 December 2020 (UTC)
@BMacZero: Yes. I uploaded all pages of a book but I see page 101 is located before page 11. Hanooz 19:29, 6 December 2020 (UTC)

Is there a Cat-a-lot to add images to my watch list

Is there something as the Cat-a-lot to add a number of images (say 20) to my watch list without the need to open them each individually? And vice versa if possible. Wouter (talk) 21:16, 6 December 2020 (UTC)

You can use AWB which has an option to add all to watchlist. -- Ricky81682 (talk) 22:31, 6 December 2020 (UTC)
If you have a list of filenames, you can simply paste the filenames into the raw edit window of the watchlist. --C.Suthorn (talk) 22:39, 6 December 2020 (UTC)
Thanks, but what is AWB? Wouter (talk) 10:20, 7 December 2020 (UTC)
@Wouterhagens: You can use AWB or JWB (if allowed).   — Jeff G. ツ please ping or talk to me 10:55, 7 December 2020 (UTC)

2020 Coolest Tool Award Ceremony on December 11th

Hello all,

The ceremony of the 2020 Wikimedia Coolest Tool Award will take place virtually on Friday, December 11th, at 17:00 GMT. This award is highlighting tools that have been nominated by contributors to the Wikimedia projects, and the ceremony will be a nice moment to show appreciation to the tools developers and maybe discover new tools!

You will find more information here about the livestream and the discussions channels. Thanks for your attention, Lea Lacroix (WMDE) 10:55, 7 December 2020 (UTC)

Photo challenge October results

Curves and spirals: Entries • Votes • Scores
Rank 1 2 3
image
Title stairs on an oil tank Bienenweide sequence photography
of Rescue helicopter
from REGA Switzerland
Author Tcolburn52 Pawel Kostal Roy Egloff
Score 15 14 14
Long distance views: Entries • Votes • Scores
Rank 1 2 3
image
Title Monument Valley, seen from U.S. Route 163 Mt. Ararat via Khor Virap Reconstruction of the
top station on the summit of
Germany's highest
mountain, the Zugspitze.
Author King of Hearts Tanjagari PtrQs
Score 29 15 15

Congratulations to King of Hearts, Tanjagari, PtrQs, Tcolburn52, Pawel Kostal and Roy Egloff. -- Jarekt (talk) 02:36, 8 December 2020 (UTC)

Books from Internet Archive...

What started as a small project to ensure Commons mirrored some of the scanned public domain works on Internet Archive has grown to contain a lot more books than imagined. Impressive!!

However it would be appreciated if other Commons contributors were able to assist with :

  • identifying authors and ensuring that appropriate {{Creator}} tags get created, and that authors get entered appropriately into Wikidata as well
  • identifying the small number of works which Commons regrettably can't host, due to licensing and copyright reasons.
  • pairing up single volumes of mutli-volume works and 'serials' (into appropriate categories), and identifying if there are additional volumes (not yet uploaded) that could also be on Commons.

More information on the upload project is at : COM:IA books

@Fæ: Anything you want to add? ShakespeareFan00 (talk) 09:43, 8 December 2020 (UTC)

Sample searches and categories are on the above project page.
We will probably hit 1,000,000 uploads at the start of 2021, so that might be a good time for a more detailed update on progress. --Fæ (talk) 10:07, 8 December 2020 (UTC)
By comparison, how big is the Library of Congress current stacks? ShakespeareFan00 (talk) 12:59, 8 December 2020 (UTC)
Catalogue of the Library of Congress "contains 18 million catalog records for books, serials, manuscripts, maps, music, recordings, images, and electronic resources in the Library of Congress collections." Vysotsky (talk) 13:02, 8 December 2020 (UTC)

Elliot Page, related changes on Commons

Page is trans person who recently came out as trans (or non-binary depending on the sources), generating a lot of press attention. There's been a pretty good response across our projects, updating articles and references to their gender. Checking our main category for Page, Category:Elliot_Page, raises some potential further corrective action. I'm raising here for general comment and suggestions, and because this notable example might be a good case study for future policy.

  1. Where Page is named in the filenames, their deadname is being used. In line with best practice by journalists, and the practice on the English Wikipedia, the respectful change would be to rename these to their current name to de-emphasize the deadname. I can technically do this, but it's not currently covered in a clear way by COM:FR, apart from a possible "harmonization" rationale.
  2. Wikidata has been updated, so the category infobox has also updated. Unfortunately Page's deadname is listed in the "Date of birth" field. A respectful change would be to design the category infobox to avoid emphasizing deadnames, and it looks like obvious avoidable emphasis here. Could someone with access to Wikidata and knowledgeable about the {{Wikidata Infobox}} design find a way to improve this?
  3. The category name has been moved and some of the parent categories relating to gender identity corrected, so that seems to have been mostly dealt with, and very promptly. Thanks folks!
  4. The descriptions of photographs have not all been changed. Again though we have no clear guidelines, the respectful approach is to swap the uses of their deadname to Page's current name. As far as I can see, that would be unlikely to cause anyone to be confused.

Feedback on these changes and suggestions for any further fixes would be welcome, especially if expressed in a respectful way. Partly my interest here is via discussions within the Wikimedia LGBT+ User Group. Thanks --Fæ (talk) 22:17, 2 December 2020 (UTC)

WRT (1), the relevant filenames have been moved today on the basis of no objections being raised here. --Fæ (talk) 14:59, 3 December 2020 (UTC)
@Fæ: There is a parallel discussion at Commons:Administrators'_noticeboard#Elliot_Page_-_dead-naming_in_historical_images. – BMacZero (🗩) 17:28, 3 December 2020 (UTC)
@Mike Peel and RexxS: for the infobox. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:04, 3 December 2020 (UTC)
I suggest changing the birth name (P1477) statement on Wikidata to have deprecated rank, or asking at d:Wikidata:Project chat about the best way to handle this in the data. On the infobox side of things, I'm not sure it should change, since it is appropriate e.g., for married people who change their surname. Thanks. Mike Peel (talk) 21:17, 3 December 2020 (UTC)
  • Not sure
    • Yes this is an important issue in general
    • Yes there is a lot of wiki community conversation about this particular case, and it makes a good case study
    • Yes a wiki guideline should govern this issue, and also I think if there were a proposal and conversation we could come to some consensus on this issue including with the Wiki LGBT+ group's participation
    • No, we do not have a guideline to apply right now
    • Yes this course of action seems reasonable seems reasonable for this case and also would apply to many other cases
    • Yes this course of action could be a draft proposal for a guideline and a practice to repeat
    • Yes I support executing all of these actions now, but only for this one case until further discussion
    • No I do not immediately support standardizing the above actions as a guideline for all cases
Problems with the above course of action include the following:
  • Lots of people have strong opinions that a particular course of action is ethical and other courses of action are unethical. The issue is too heated to resolve without organizing public discussion.
  • Expert LGBT+ organizations have differing opinions or the odd absence of opinions on this issue; we need to talk it through.
  • In general, being without a guideline gets less criticism than setting a debatable policy without first having community process. We can delay taking a firm position.
  • This is an issue with no global consensus, as people from different language communities and cultures would handle it in different ways.
  • Being trans is a personal and individual experience, and in addition to a default practice, we should also clearly state when we entertain exceptions. The reason to clearly state conditions for exceptions is that if we do not, then the precedent is that every one of these cases gets treated as exceptional and meriting a huge discussion. While it is not sustainable to debate every individual case without trying to form some general principles, we need to be ready in advance to clearly give conditions under which we will make exceptions.
  • The Wiki Community has already done a lot to try to address this. We organized meta:Queering Wikipedia, a summer 2020 conference to sort this and other issues. COVID was the cause of cancelling that meeting and delaying progress. Anyone who wants faster progress can direct concerns to the Wikimedia Foundation and encourage them to more completely sponsor and support community conversation on challenging issues which cross language and cultural barriers.
Blue Rasberry (talk) 23:31, 4 December 2020 (UTC)
@Fæ and Bluerasberry: and others. Thank you for taking on this task. Any effort to make Commons and other Wikimedia projects to be more friendly, respectful and inclusive to LGBT+ community (and other communities) are important. For me changing filenames, to deemphasize person's deadname seem like a logical step; although I can also see the logic in the argument that the name should reflect gender identity at the time when it was taken. So on file remaining issue I would argue for it should be settled within Wikimedia LGBT+ User Group first (perhaps it already was) and than we should adjust our policies to clearly allow such reason for file rename. As for the issue of deadname showing up in Wikidata, I think that we should have it as "also known as" alias and in birth name (P1477) statement, but other places should be changed. --Jarekt (talk) 18:41, 9 December 2020 (UTC)
@Jarekt: Thanks what you propose is reasonable to do now.
Wiki LGBT has been having discussions and will continue to have discussions. It is not a easy issue to resolve so I do not expect any recommendation from that organization soon. For now the usual wiki community way of addressing individual cases is probably best. When the organization can come up with something it will bring it here. Thanks. Blue Rasberry (talk) 20:40, 9 December 2020 (UTC)
Conversely: may I assume we have consensus that we do not delete images of someone taken before they transitioned simply because they no longer like the fact that such images exist? - Jmabel ! talk 23:52, 9 December 2020 (UTC)

Low-value creator statements in structured data

We have very many thousands of items with a structured data statement creator (P170) = "some value", qualified with author name string (P2093), when there is a Wikidata item about the creator.

Can this be fixed, by a bot, and by obtaining the QID of the creator by either traversing categories, or looking at the use of {{Creator}}? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:55, 9 December 2020 (UTC)

More effective would be to fix the Upload Wizard, so that newbies are not driven to enter 'captions' before even writing a description. Fixing the workflow so that Commons is not driving everything to structured data, but getting the primary entries correct first, would probably eliminate newbie low-value creator statements to a small fraction of what they are now.
BTW, if folks want to spend time writing bot jobs related to structured data, then updating MediaWiki to document these features of the API (if they exist in the API and are functioning reliably) would be the place to start. --Fæ (talk) 12:38, 9 December 2020 (UTC)
Few if any of the statements to whiich I refer have been created by 'newbies'; most are the work of bots. This has nothing to do with captions, so changing that aspect of UW would have no effect whatsoever on the issue. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 21:36, 9 December 2020 (UTC)

Deletion due to lack of permission from photographer

I am Ulf Larsen. I am a 64 year old Norwegian that found Wikipedia in 2004, and I have contributed ever since. In my civil life I am a chief mate on a large Norwegian coastal ferry. People that I work with know I am a professional seaman, that do not take light on safety and security, that I write on Wikipedia, and that I am an exhibitionist and amateur porn model.

Regarding the last, I lately found out that pictures of me in amateur porn could possibly be of interest for Wikimedia Commons as it endeavors to cover all aspects of human activity. Not so, it seemed. The pictures were marked for deletion, as they were pornographic. But that was the whole point of them being uploaded. Then they were added for deletion due to they were seemed as revenge porn. With effort, I managed to have my ex-girlfriend document that she indeed was a willing accomplice in shooting them. Then the last came, what about the photographer, did I have that person's accept to upload them, (if not they would have to be deleted)?

I did not, as at the time it was various persons I paid to do it, mostly using my camera, with a short, five minute course in how to use it. So in short, if this is an absolute, then the pictures must be deleted, all of them. But to be correct, I have probably over the years also uploaded other pictures that in fact was shot by someone else. So if this will stand, I will have to use some hours to look closely through the few hundred uploads I have done.

I do of course respect the general principle that a photographers rights shall be respected. But without being into law, for me it seems that this can be viewed in a narrow and a wide way. The narrow way is as described above - if the photographer can not issue a statement releasing the pictures, then they must be deleted. The wide way is that ordinary people do not look at this as professionals, and why look for trouble - if someone claim the rights - then delete them later on. For me, it seems that the latter is within the range of how we work on Wikipedia - it used to be called be bold. This is what I have to say about this issue, the rest is up to the community. Ulflarsen (talk) 22:33, 9 December 2020 (UTC)

  • @Ulflarsen: I hate have to say it to you but: Commons is very strict in general with reference to copyrights and, for sexual imagery, we are comparably strict about personality rights. I'm sure you understand in general why this policy was adopted. And, really, I'm sure you were operating in good faith, but unfortunately that's not enough. We really need written releases, because of the assurance we need to be able to pass on to our reusers.
  • In case you don't know: sites like Flickr are much less strict about this (as long as you tag your images appropriately in terms of "safety", i.e. how sexually explicit). You can free-license them on such a site, and make them almost as readily available for re-use as here. If you want to do similar Commons work in the future, though: get the same releases you would for commercial work, because it is a matter of principle that anything on Commons should be able to be used that way. - Jmabel ! talk 00:03, 10 December 2020 (UTC)
To talk: You write as follows: "I hate have to say it to you but: Commons is very strict in general with reference to copyrights and, for sexual imagery, we are comparably strict about personality rights." As far as I can see, these are separate questions. If its about sexual imagery, then due to personality rights the one's I have uploaded - beside my ex-girlfriend, that through the OTRS have explicit accepted the uploading - should be deleted. But regarding copyrights? Professionals care about that, ordinary people as me, mostly do not. What harm would it do to Wikimedia Commons if any such hypothetical question was solved, if it ever emerged...?
But regarding personality rights. How can you defend that Wikimedia Commons are " for sexual imagery, we are comparably strict about personality rights"? For me, it seems that either we have to accept that people have accepted being taken pictures of - unless there is reason to believe the opposite - or we have to apply the same strict view on everyone. Below are a number of various pictures I have shot where some people may be recognized, should those pictures also be deleted?

https://commons.wikimedia.org/wiki/File:City_ferry_Fredrikstad,_Norway_06.jpg

https://commons.wikimedia.org/wiki/File:Moss_railway_station,_looking_north.jpg

https://commons.wikimedia.org/wiki/File:Pier_and_building_used_by_ferry_to_check_for_clear_of_pier_02.jpg

https://commons.wikimedia.org/wiki/File:Wikipedia_workshop_03.jpg

https://commons.wikimedia.org/wiki/File:Norwegian_contributors_to_Wikipedia_meeting_at_Litteraturhuset_in_Oslo_03.jpg

I know that in some of the pictures I told those people that pics would be taken, while in others I assumed there was very little possibility that any harm would be done. If I understand you correctly, then I need to have a written agreement for all of them. Its no problem for me, but I assume that an equal rule here would severely restrict the number of pictures available on Wikimedia Commons. Ulflarsen (talk) 00:44, 10 December 2020 (UTC)
How can we defend that "we are comparably strict about personality rights ... for sexual imagery"? First, in many countries, there's a big legal difference in publishing a fully clothed picture of someone in public and publishing nudes taken in private. Second, that's because people tend to get much more annoyed about their nudes getting passed around than a photo of them in public. I'm a fan of public photos being public, but certainly don't want to demand that ex-lovers and paparazzi can publish whatever they want no matter how intrusive.--Prosfilaes (talk) 10:17, 10 December 2020 (UTC)

We have that situation at OTRS every day, and we deal with it consistently: If neither a permission nor a transfer of rights from the photographer can be submitted, then we cannot accept the pictures – regardless of their theme. --Mussklprozz (talk) 08:34, 10 December 2020 (UTC)

I do not know the laws in Norway. But in Germany I can ask a person to take a photo and the person can agree that I can publish this as I have all usage and publishing rights on this photo without writing this down and sign this. The oral agreement is enough. --GPSLeo (talk) 09:38, 10 December 2020 (UTC)
This applies only to photographs where picture details, focus, etc. have been determined by the owner of the camera and the "accidental passer-by" is merely releasing the shutter. In this specific case the photographer obviously did not only determine the image details, but also directed and reacted to the scenery, which is clearly a creative act covered by copyright. --Reinhard Kraasch (talk) 12:16, 10 December 2020 (UTC)
I do not understand you point. One example: If you give me you chocolate and you say it is okay that is eat it. I eat it you can not say that I stole it because I got the permission. But we do not need to write down this contact. Written down contracts are only needed when you do not trust the other one. Like in a shop, where you get a receipt to show that you paid for the chocolate and did not stole it. If you say you never said to me that I can eat the chocolate and we did not write this down, then we have a problem. I am not sure if this was the case here. If there was an oral agreement that this is okay but now there is a written mail that there was not such an agreement we and Ulflarsen are having a problem. --GPSLeo (talk) 12:57, 10 December 2020 (UTC)
We are not talking about chocolate, but about images and copyright issues. Copyright as an immaterial right is different from material rights. The question here is: Who is holding the copyright of the photo, since only the copyright owner can release the image under a free license. Regarding the rights assuming "good faith" is explicitly excluded here in Wikimedia Commons, see Commons:Project scope/Precautionary principle. --Reinhard Kraasch (talk) 13:36, 10 December 2020 (UTC)
It is not about the person owning the copyright is it about who is owning the right for publishing under any license that is (al least in Germany) an important difference. But in both cases the question is: Did the person the uploader stated to have the right to upload the photos of now wrote a mail that the person never agreed to this? --GPSLeo (talk) 14:52, 10 December 2020 (UTC)
The issue is not the "right to upload" (there is no such thing), the issue is: who is able to give the permission to publish the work under a free license. In German law there is indeed (or: may be) a difference between Urheber ("copyright owner") and Nutzungsrechtsinhaber ("holder of the rights to use"), but German law does not apply here, and even if, it would not make a difference, since a proof would be necessary that the rights to use have actually been transferred. It is not necessary to wait for the actual copyright holder to complain (again: see Commons:Project scope/Precautionary principle). --Reinhard Kraasch (talk) 18:14, 10 December 2020 (UTC)
Back to the question of this thread. Fact: Ulflarsen though he had the right to upload the photos taken by an other person. Question: Did this person now wrote a mail to OTRS and said never gave the permission for this? --GPSLeo (talk) 18:48, 10 December 2020 (UTC)
No, this is not the question. EOD. --Reinhard Kraasch (talk) 20:40, 10 December 2020 (UTC)

@Ulflarsen: Two further remarks:

  1. On the copyright front: "Professionals care about that, ordinary people as me, mostly do not." Given that Commons from the outset made the decision that our imagery should be available for reuse in commercial contexts such as newspapers, books, etc. we have to go by the standards of those professionals, not that of "ordinary people". You are participating here in what is, among other roles, the media repository of what may be the most important encyclopedia in the world (with the possible exception of Britannica, but I really do think Wikipedia has passed them for importance). While we have done this in no small part by including the contributions of "ordinary people" it has been on the basis that "ordinary people" are fully capable of coming up to the standards of "professionals", not that the standards normally associated with ordinary people suffice.
  2. On the personality rights and sexuality front: it is very unlikely that someone's life will be damaged or destroyed by a picture of them looking out at the skyline from a ferry. A picture of them engaged in a sex act is a very different matter. To which I will add: there are countries where a picture of two men holding hands and smiling at each other falls in the latter category. You and I are both fortunate enough to live in countries where pictures such as those you uploaded are not usually a terribly big deal, but they would be if the subject of the picture had not clearly consented. You are, by your own description, an "exhibitionist". Fine. As they say, "Your kink is not my kink, but it's OK." I personally would be appalled if an ex-girlfriend of mine posted a picture of me in flagrante on the Internet, especially a picture where I was fully identifiable. I am actually happy that such images exist, as anyone could tell by reading between the lines in my remarks above, but we need to be very clear that all individuals who are the subject of such photos genuinely consented to be in them. - Jmabel ! talk 00:08, 11 December 2020 (UTC)

Standardise PD-country category names

I raised a question about the non-uniform names of the subcats of Category:Public domain by country. Please discuss at Commons:Categories for discussion/2020/08/Category:Public domain by country.--RZuo (talk) 20:23, 10 December 2020 (UTC)

Autopatrolled

Sounds like I have a tracking device attached to my body.

A few days ago I got a message saying I was Autopatrolled. It gave me a link to check it. But, when I checked I wasn't on the list. I forget about the "occasionally the twain shall meet" relationship between Commons and Wikipedia, so I don't know which system it came from, but the message gave a link. I'm not sure this ?honor? will affect my life but somehow these inconsistencies gnaw at me. -- Jim Evans (talk) 23:58, 10 December 2020 (UTC)

  • @Jim Evans: It just means that in the future when you make an edit, it won't be marked for someone more experienced to check ("patrol") it. Auto-patrolled in the sense that you are now considered to have been a consistently good contributor that the fact that you made the edit is sufficient that it is inherently considered to have been "patrolled". - Jmabel ! talk 00:18, 11 December 2020 (UTC)
But, my question is, why don't I appear when someone searches the list of Autopatrolled members? -- Jim Evans (talk) 00:55, 11 December 2020 (UTC)
How do you come to this conclusion? See [2] --Reinhard Kraasch (talk)
I think I did this [3] and assumed E would come before F. I guess it was a case problem -- Jim Evans (talk) 01:39, 11 December 2020 (UTC)

Wikidata connection delays?

Not sure if I'm doing something wrong here, or if there's a technical problem, or what. The uploading tool I'm using (iNat2Commons) requires connections between Commons and Wikidata to work. So I created Category:Symphyotrichum_nahanniense, and made the connection on Wikidata under "other sites" https://www.wikidata.org/wiki/Q15555214. Pretty much immediately, the Wikidata Infobox populates properly on the mobile website (https://commons.m.wikimedia.org/wiki/Category:Symphyotrichum_nahanniense). But now, 18 hours later, the infobox is still not working on the non-mobile website and iNat2Commons can't detect the connection either. What is going on? I don't understand why there would be this discrepancy between the mobile and non-mobile websites. Somatochlora (talk) 14:52, 3 December 2020 (UTC)

It is now working properly. I can't imagine that it was a coincidence that this happened within 5 minutes of me posting it after 18 hours not working, but as no changes have been made at either Wikidata or Commons that I can see, I don't know what has happened. Another page still isn't working: Category:Symphyotrichum_simmondsii Somatochlora (talk) 15:06, 3 December 2020 (UTC)
No queries with Wikidata, or tools that depend on them, are guaranteed to work. Larger queries are almost certain to suffer from multiple drop-outs over a year, and eventually all suffer permanent breakages. The onus is on the unpaid volunteers that create these tools to constantly maintain them, year after year, which is obviously not sustainable. There is no commitment on WMF dev to consider second order breakages when they make whatever changes they feel may be interesting. Though the WMF has enjoyed benefits in promoting Wikidata, including selling special access for the Googles and Amazons in the commercial world, there is far less strategic commitment to keep Wikimedia Commons operational or functional. One need only look at how ghastly and clunky the Upload Wizard interface and basic workflow is today for newbies, to understand how the current strategic priorities actively damage this project and limit our future here. --Fæ (talk) 15:35, 3 December 2020 (UTC)
"selling special access for the Googles and Amazons in the commercial world,": Are you saying that the wmf makes money directly off the efforts of volunteers? Thanks in advance, Ottawahitech (talk) 17:32, 3 December 2020 (UTC)
This is under development, there's been no move to stop it and yes, it is a way to fund and grow the WMF into more directly commercial services. It's not a secret, just most volunteers remain unaware that this is the WMF strategy and implicitly it means that the whole of Wikidata is going to be promoted for commercial use, from biography data through to training AI systems, of precisely the sort that Amazon are leading in. Refer to m:Strategy/Wikimedia movement/2018-20/Recommendations/Iteration 1/Revenue Streams/1 and m:Strategy/Wikimedia_movement/2018-20/Recommendations/Increase_the_Sustainability_of_Our_Movement states:
Building enterprise-level APIs (with high standards of availability, throughput, and usability).
Engage partners in the development wherever appropriate, incorporating the needs of a spectrum of small, non-commercial, and larger commercial reusers.
Explore fees or sustainability models for enterprise-scale commercial reusers, taking care to avoid revenue dependencies or other undue external influence in product design and development.
--Fæ (talk) 18:29, 3 December 2020 (UTC)
The WMF is a non-profit foundation; any income from OKAPI, as the service is to be known, are to be spent on running our projects. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:01, 3 December 2020 (UTC)
Yes sure.
This can include paying a commercial marketing manager a six figure sum, coming out of the Google fees. Of course if Amazon pay an access fee of $1m to the WMF and makes a profit of $100m, precisely none of that goes to volunteers that created the data that is the content that Amazon is selling.
The WMF remains a non-profit, so that's all fine. Completely unrelated, there's been no action recently to give volunteers second-hand laptops the WMF has written off. Waiting is up to 2 years now... it would be nice to feel slightly more cared about as we make this content that fuels all this commercial interest. --Fæ (talk) 20:53, 3 December 2020 (UTC)
Fæ, you've gone way off topic to this thread... Mike Peel (talk) 21:24, 3 December 2020 (UTC)
Mike, you know the village pump is for varied discussion, this is not a proposal or a RFC. The reason you would rather close down pointing out these facts about the promotion of Wikidata by the WMF for commercial sales is not really because this is a meandering discussion. Many folks are alarmed by this strategy and the lack of any detail about how it is being implemented by the WMF, it's not a fringe concern, it's core to the relationship between Commons and Wikidata, and why diverting or by design forcing our Commons volunteers into editing on Wikidata to display or add basic information about Commons content is highly problematic, and long term those volunteers are likely to feel misled. --Fæ (talk) 10:48, 4 December 2020 (UTC)
My point was solely that you should have started a separate thread for this if you wanted, rather than going off topic. I think you're wrong/developing conspiracy theories, though. Mike Peel (talk) 15:46, 4 December 2020 (UTC)
@Fæ: As respectfully and non-confrontationally as possible, I and I'm sure many others already know how you feel about Wikidata and everything related to it and are a bit tired of seeing you frequently jump to attack it in threads that are just looking for solutions to specific technical problems. – BMacZero (🗩) 20:36, 4 December 2020 (UTC)
Facts about Commons on the Commons VP are not "attacks" on Wikidata.
Interesting point you make, but I don't recall mentioning the commercial access service on village pump at all in 2020, so why you feel that providing some facts and links to a WMF board member writing about it are "frequent" seems like shutting down me personally and fooling the reader in to thinking I'm a fringe obsessive. Putting 'non-confrontationally' in the same sentence given this context of a subject being mentioned once, speaks for itself. --Fæ (talk) 20:56, 4 December 2020 (UTC)
@Fæ: You may choose not believe me, but I have no personal ill will against you or desire to "shut you down" and the primary reader I am addressing is you (though this is by no means unique to you). I feel you have critically misinterpreted parts of my comment (such as what it is I feel you are attacking) in an effort to defend yourself from it. What I intended here is not to elicit defense but to encourage and suggest a more collegial tone in these discussion (I know this is an important goal for you and I hope you can agree it has been lacking in this discussion). I know many aspects of Wikidata etc. have not been handled well, but posts like your original reply here hurt your efforts to, for example, fix the problems with proposals by making it somewhat harder for "Wikidata moderates" like me to assume you are acting in good faith (probably a failing on my part, but there it is). – BMacZero (🗩) 18:26, 5 December 2020 (UTC)
I guess as I cannot even edit Wikidata, my IP being blocked on that project, I guess it's pretty irrelevant for me to continue commenting about it, I may as well be blocked on Wikidata. I'll take your tip and focus just on the damaging impact that we see on Commons, like forcing Commons contributors to edit Wikidata entries when they click on the edit link within a Commons displayed template, with all the consequent error conditions that can then arise, and for which there never has been a Commons consensus for. --Fæ (talk) 20:52, 5 December 2020 (UTC)
@Fæ: I'd be happy to help you get that IP unblocked on Wikidata (and can think of several colleagues who would do likewise). The project would benefit greatly from your input; and welcomes bot operators. To use the vernacular, I'm sure we'd all like you to be inside the tent, er, facing outwards... Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:26, 11 December 2020 (UTC)
@Somatochlora: Next time, try Help:Purge-ing the page cache on the non-mobile page. The appearance of pages on Commons is only updated when they are changed locally, so you may have to do this when you change something elsewhere that affects the page appearance. It's possible that the mobile and non-mobile versions have distinct caches, and someone may have done this after seeing your message. – BMacZero (🗩) 17:24, 3 December 2020 (UTC)
That must be it, thanks! Somatochlora (talk) 17:31, 3 December 2020 (UTC)
Normally when this happens, I do a null edit (click edit then save without changes) and then purge the page, and that clears the cache. It can either be due to normal caching, or it can be an occurrence of phab:T233520, but the null edit + purge fixes both of those possibilities in the individual case. Interesting to hear that the mobile version updates quickly! (and any feedback on the choice of data to show in the infobox on mobile would be useful, BTW). Thanks. Mike Peel (talk) 21:24, 3 December 2020 (UTC)
ISTM cache-updating has been particularly slow recently. I overwrote an image a couple of weeks ago, and on some pages its thumbnails continued to show (a distorted version of) the previous picture for several days, despite a few purging attempts on my part. And in the past couple of days I’ve seen a handful of Talk and project pages not showing recent posts (except as diffs)—something I don’t recall encountering before. (Manual purging did work in these cases.)—Odysseus1479 (talk) 21:53, 3 December 2020 (UTC)
@Odysseus1479: I usually find that out-of-date thumbnails (especially if distorted) are caused by your browser's caching of the image, which is separate from Mediawiki's server caching. You could try clearing your browser's cache next time for that (Ctrl+F5 or Shift+F5 on most browsers). – BMacZero (🗩) 22:01, 3 December 2020 (UTC)
@BMacZero: I doubt it could be my local cache because I’d never visited a couple of the category pages in question before. Unless the browser caches individual images, recognizing them regardless of the page they appear on? Even MediaViewer would briefly show an enlarged version of the previous thumbnail, before pulling up the current image in full-window resolution.—Odysseus1479 (talk) 22:45, 3 December 2020 (UTC)
@Odysseus1479: That's actually exactly how it works, to my understanding. That type of caching is generally based on the image's URL rather than the page's URL because the main point is to save the browser from downloading images it's already downloaded. That also explains why you might get an old thumb but an up-to-date full-sized image because those are different image URLs (e.g. https://upload.wikimedia.org/wikipedia/commons/thumb/4/48/Little_Gibraltar_Catalina_Island_Panoramic.jpg/180px-Little_Gibraltar_Catalina_Island_Panoramic.jpg for the category thumb and https://upload.wikimedia.org/wikipedia/commons/thumb/4/48/Little_Gibraltar_Catalina_Island_Panoramic.jpg/1280px-Little_Gibraltar_Catalina_Island_Panoramic.jpg for MediaViewer). – BMacZero (🗩) 03:06, 4 December 2020 (UTC)

Community Wishlist Survey 2021

We invite all registered users to vote on the 2021 Community Wishlist Survey. You can vote from now until 21 December for as many different wishes as you want.

In the Survey, wishes for new and improved tools for experienced editors are collected. After the voting, we will do our best to grant your wishes. We will start with the most popular ones.

We, the Community Tech, are one of the Wikimedia Foundation teams. We create and improve editing and wiki moderation tools. What we work on is decided based on results of the Community Wishlist Survey. Once a year, you can submit wishes. After two weeks, you can vote on the ones that you're most interested in. Next, we choose wishes from the survey to work on. Some of the wishes may be granted by volunteer developers or other teams.

We are waiting for your votes. Thank you!

SGrabarczuk (WMF)

15:03, 11 December 2020 (UTC)

Two tram stops categories

Category:Tramway de Bordeaux stops (38) and Category:Tram stops in Bordeaux (23) Smiley.toerist (talk) 17:48, 10 December 2020 (UTC)

Well, the user who recently split the two categories seems to hold the opinion that "Category:Tramway de Bordeaux stops" relates to "Category:Tram stops in France by system" whereas "Category:Tram stops in Bordeaux" relates to "Category:Tram stops in France by city". Maybe you could begin by talking with him. -- Asclepias (talk) 04:28, 12 December 2020 (UTC)
    • The answer to your questions "Does anyone other than Tramway de Bordeaux operate trams in Bordeaux? Does Tramway de Bordeaux operate only within the limits of Bordeaux?" are respectively no and no. Like many French cities, the boundaries of the city of Bordeaux are quite tightly drawn, and the tramway serves quite a number of places that are not within the city. There is thus an ambiguity here: does 'tram stops in Bordeaux' mean 'tram stops on the Bordeaux tramway' or 'tram stops in the city of Bordeaux'. And indeed if you look at the category Tram stops in France by city you can see that different editors have taken different and contradictory views on that, with a number of 'cities' (actually communes, but there isn't a lot of difference in French usage of the terms) listed that are places served by trams but which don't have their own tram systems. It seemed (and seems) to me that the best way to address this is to acknowledge the two different possible meanings, and have separate category trees for system and city. I'm open to other possibilities, but my past experience elsewhere is that any attempt to coerce stops that are actually in one commune/municipality/whatever into another on the grounds that the stop is on the latter's tramway is likely to meet resistance. Hence, for example, the two categories Tram stops in Zürich and Tram stops served by Zürich trams‎. -- Chris j wood (talk) 02:21, 13 December 2020 (UTC)

Commons links from enwp categories

Hi all, just to note that I've started an RfC at enwp's Village Pump (proposals) on the future of links from enwp categories to Commons categories, which may be of interest to people here. Thanks. Mike Peel (talk) 20:10, 12 December 2020 (UTC)

Bots for switching uses of PNG to SVG?

I've noticed that quite a few instances of PNG files, such as File:Science-symbol-2.png, are still used on a whole bunch of projects, despite the existence of an SVG replacement. Has there been discussion of having a bot automate the task of swapping them out? (The bot would be making edits on the projects themselves, so I'm not 100% sure this is the right forum, but it seems the most centralized place.) {{u|Sdkb}} talk 19:35, 7 December 2020 (UTC)

AFAIK, globally replacing images with an SVG replacement is frowned upon, and CommonsDelinker wouldn't process any requests whose replacement is an SVG. I do wonder if this is still what the Commons and global community think. pandakekok9 13:26, 12 December 2020 (UTC)
CommonsDelinker instructions did remove the remark about "In order not to start World War III," though. - Jmabel ! talk 09:39, 13 December 2020 (UTC)

Deletion request log pages with wrong numerals?

Sometimes users create deletion request log pages with wrong numerals:

The dates should use the numerals 0-9 but these pages use other numerals. What's causing this? Is there a script which produces numerals in the user's local language instead of the numerals we use in the page titles? The edit summaries suggest that these were created using the normal tool for nominating files for deletion. --Stefan2 (talk) 22:16, 11 December 2020 (UTC)

i suspect they are all related to the android app. feedback can be filed at https://github.com/commons-app/apps-android-commons/issues .--RZuo (talk) 12:40, 13 December 2020 (UTC)

File:Sajjad Ali.jpg

Can anyone this file to Wikimedia commons?.Thank You.192.142.198.103 13:51, 13 December 2020 (UTC)

File:Flag of Germany.svg – input needed

Please join the discussion at File talk:Flag of Germany.svg#edit request. —⁠andrybak (talk) 09:16, 14 December 2020 (UTC)

Commons:Uploading from the Seattle Municipal Archives Flickr stream

For a decade or so, I've been uploading from the Seattle Municipal Archives Flickr stream and dealing with various metadata issues, etc. There is some really good content; see Category:Seattle Municipal Archives via Flickr. The task is not as trivial as it sounds; please follow the link in the subject line for this section if you want details.

For at least the next several months I believe I will be too busy to do this, or at least to do it thoroughly. I'm backlogged probably a month, and could use someone else either to take over or at least to help. I'd be glad to mentor: that would be such good return on my time that it would be worth putting in a bit of "the time I don't have."

Any takers? - Jmabel ! talk 14:37, 15 December 2020 (UTC)

Reproduction prohibited?

I came across the following file with four times the watermark "reproducción prohibida". The user has uploaded many more images. In the Spanish article Marquesado de San Lorenzo de Valle Umbroso is also the same image without watermark. In the article is given the reference "Jose C. Gil. Documentos Originales de concesión del Título. Madrid, España.". I expect that these images come from that publication. Should the "reproducción prohibida" be considered as a watermark or does it mean more?

  • with watermark
    with watermark
  • without watermark
    without watermark

Wouter (talk) 13:53, 15 December 2020 (UTC)

As the original is from 1687, it is certainly in the public domain. Attempt by an institution to claim some exclusive rights to it is Copyfraud. The version without the watermark should be preferred; when there a good quality unwatermarked image I see no reason to have a watermarked version as well. -- Infrogmation of New Orleans (talk) 20:04, 15 December 2020 (UTC)
Thanks for the explanation. I have added {{Watermark}} to all the images with a watermark. Wouter (talk) 10:27, 16 December 2020 (UTC)

Historical events

Decision to make? Example: Category:Battle of Navarino, 1827.

The images of record for this battle were made by Category:George Philip Reinagle. We have sketches, studies, lithographs, watercolours and oil paintings of his eye witness records, based on a core of 12 views, extrapolated many times. We have other works of art that are inspired or copied from his (very well known) work by others.

To complicate matters the event was in the days before photography, art was the medium of record.

So we have cats for "Battle of Navarino" and "Battle of Navarino in art". In this context art for me is made up fantasy images. Whereas Sketches of record by participants in the battle, are something more than that; not art as such. We have History cats by date out there, before the American Civil War it was all art. That war itself, was a hybrid of art and photos.

What are the community's views on this? How best to categorize pre-photographic events such as this? How should Reinagle's work be represented? category: "Battle of Navarino as seen by by Reinagle"? or left in the main cat, which would be my preference given that we're not likely to exceed 80-90 images in total (80 of them art?)... Broichmore (talk) 21:30, 4 December 2020 (UTC)

-- Tuválkin ✉ ✇ 12:18, 9 December 2020 (UTC)
What! Reinagle's record is the "definitive" documentary eye witness source of what happened there. He's a sailor sketching it in real time, not some cartoonist imagining it. The most accurate art presentations of the battle are copies or derivations of Reinagle, the only exception is Lee. --Broichmore (talk) 16:30, 10 December 2020 (UTC)
  • I do understand where your coming from.
How are we going to give Reinagle's images their proper value, in comparison to others others. If anything it should be under Category:George Philip Reinagle and Category:Battle of Navarino.
I suspect this strategy is too complicated to implement, beyond making a sourced note on the main cat page that images from Reingold and Lee in particular are accurate images.
Going on from that, we only have 34 art images. I really feel, that when there are under 100 images in a category there is no justification for diffusing them out of sight into sub categories. The human eye knows the difference between a coin, a monument, an image of the battle.
Perhaps it would be better to sponsor the notion that 100 or more images are required before splitting into two categories? Broichmore (talk) 16:32, 17 December 2020 (UTC)

Uploading from Flickr by tag

Here for example, is a set of 57 Flickr images sharing a unique tag. Is it possible to upload all images with that tag, in one go? If I use Flickr2Commons, specifying the Flickr account name and the tag, it finds no results. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 09:39, 16 December 2020 (UTC)

Not an answer to your question, but it looks like in 2015 User:Fæ tested uploading one page from each edition of this book in Category:Bird-life of the borders. You could also have a look at the 56 Flickr images from the 1907 edition to see if there are any differences with the 1889 edition. -- Asclepias (talk) 14:19, 16 December 2020 (UTC)
Tags are entirely get-able via the FlickrAPI. It's a pity that the Upload Wizard does not accept a suitable url based on tags, rather than only on an album.
I'm not currently working on the FlickrAPI, any mass upload from Flickr comes with a lot of baggage. I have no idea what could be done with F2C, or which parts are reliable. --Fæ (talk) 15:30, 16 December 2020 (UTC)
  • For what it's worth, I filed a request to add tag support into the UploadWizard in February (phab:T245063), but actually making a patch for it is a little beyond my abilities. Looking at the code, it seems UW supports a few other things as well as albums - there's code in there for "galleries" and "favourites", both of which you could potentially set yourself - create an account, favourite everything you want to upload, and then run that through UW. A bit clunky, and I haven't checked if it actually works, but it might be an interesting experiment. Andrew Gray (talk) 17:34, 17 December 2020 (UTC)
Update: favorites doesn't currently work, but it looks like a pretty quick fix - outlined at phab:T270432 if anyone wants to give it a shot. Andrew Gray (talk) 20:43, 17 December 2020 (UTC)

Licensing/authorship scam

Being discussed elsewhere — Rhododendrites talk |  17:57, 17 December 2020 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

I saw this come up on Twitter: Beware This Wikipedia Photo Copyright Scam (PetaPixel). Basically someone was changing authorship of an image then contacting reusers to scam them. It made me wonder if we might use an edit filter or some other tool to keep track of changes to author fields or license templates (which should be pretty rare). I'm still curious about that, but it looks like the user in question actually changed captions to indicate authorship, so that's a trickier problem... :/ — Rhododendrites talk |  17:30, 17 December 2020 (UTC)

@Rhododendrites: FYI, we have three sections on reacting to this issue already: COM:ANV#"A. Sturdivant" is back, COM:ANU#Suzane.Rob2020 altering attributions to enable copyfraud, and COM:AN#False authorship and subsequent copyright claims. The latter section and COM:FILTERT would be appropriate places to suggest new filters.   — Jeff G. ツ please ping or talk to me 17:46, 17 December 2020 (UTC)
Thanks, Jeff. Let's call this thread closed, then. :) — Rhododendrites talk |  17:57, 17 December 2020 (UTC)
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Update to MediaSearch and a small survey on search preference

Greetings,

The Structured Data team has released licensing filters for Special:MediaSearch. When looking for files you can now narrow or expand your choices by deciding what free license permissions you would like to work with, or use to share with particular reuse parameters.

  • All licenses
  • Use with attribution
  • Use with attribution and same license
  • No restrictions
  • Other

Please give it a try, and leave feedback on the feature or anything else about the new search in general to let the team know how you like it or how it can be further improved.

Additionally, there is a quick survey available on the MediaSearch project page about search preference. The survey is available to logged-in and anonymous users, but is only available to be taken once–it should disappear after filling it out. However, the survey is cookie-based so it will reappear again if you use another browser. Anonymized survey results will be published on-wiki soon after the survey closes. The survey is hosted on-wiki, and the privacy statement is available.

The Structured Data team looks forward to what people think about the new feature for MediaSearch, and their preferences for search in general. This information will help to continue to improve the experience of trying to find the media that is being curated here. Thanks for your time, I'll be posting some short followup reminders about the survey over the next few weeks. Keegan (WMF) (talk) 21:05, 17 December 2020 (UTC)

what ever i am searching , i always get masses of bookcovers

Could that be fixed? --Itu (talk) 18:27, 12 December 2020 (UTC)

For bookcovers specifically I don't know, but if you mean .PDFs of books, you can try the simple trick of adding -pdf to your search terms. There may be some more official or sophisticated way to do it, but simple ways usually work well enough for me. -- Asclepias (talk) 20:52, 12 December 2020 (UTC)
For anyone interested in this problem, please subscribe and add tokens to Phab:T269929 Add a Commons user search preference to exclude documents. Thanks --Fæ (talk) 18:37, 13 December 2020 (UTC)
User:Itu Try using Special:MediaSearch.--Snaevar (talk) 06:07, 18 December 2020 (UTC)
Thank you. But now i'm running into further problems: Searching "suitcases packed" i get a lot of insects, because there is a long text with plenty of insect photos from a flickr source.
So search engine should weight words less relevant in big descriptions .... --Itu (talk) 06:21, 18 December 2020 (UTC)
At the right upper corner of the search result page (search, not mediasearch) is an (i) icon linking to the cirrussearch help page. Search allows a number of parameters described on that help page, for example exluding pdf files, which will remove most book covers. However it should really be much easier to exclude pdfs or to do other things like search for a specific period (images of Obama from before his presidency, but not limited to one specific year for an example). But as this is supposed to be covered by SDC, it is unlikely that search will be improved (outside of improvements through SDC). --C.Suthorn (talk) 07:40, 18 December 2020 (UTC)
@Itu: You must help the search a little. You can gradually refine your search as needed. In your search suitcases packed, the excellent photos of insects are from one flickr user, so they can be filtered out. Try to search suitcase packed -33398884. -- Asclepias (talk) 15:25, 18 December 2020 (UTC)

Christmas categorizations

I wish to suggest that Wikimedians who celebrate the Christmas holidays, and others who find it of interest, take on a project to improve access to Christmas related media on Wikimedia Commons. We have lots of media related to the holiday that has minimal or not categorization. Firstly, we have over a thousand files just in Category:Christmas. Note that we have many useful subcategories; such as Christmas by year (eg, Category:Christmas 1952, etc) and Christmas by country (eg, Category:Christmas in Mexico, etc), and intersection categories for the two (eg Category:Christmas 2016 in Germany etc). There are categories for Christmas trees, Christmas cookies, Christmas cards, etc etc. Sorting images in "Category:Christmas" into one or more of the subcategories would help users find what they might be looking for. Additionally, a quick search for "Christmas uncategorized" shows over 800 files. (More are seen at some of the terms for the holiday in other languages.) Season's Greetings and Sortings! -- Infrogmation of New Orleans (talk) 20:18, 15 December 2020 (UTC)

I'll do a bit of work on this. GeorgHH • talk   13:11, 16 December 2020 (UTC)
In principle it would make more sense to addd SDC tags (depicts) to these files. --C.Suthorn (talk) 14:23, 16 December 2020 (UTC)
I don't know why you think it would make "more sense", but do both if you like. Cheers! -- Infrogmation of New Orleans (talk) 14:10, 18 December 2020 (UTC)

A guideline on disruptive editing

I realized that we don't have a guideline about disruptive editing, so I thought to create one so that not only our definition and interpretation of "disruptive editing" is clear, but to also send a message that we don't appreciate disruptive users here in this project. I created a draft at Commons:Disruptive editing. As you can see, it's mostly derived from the enwiki equivalent. Please do leave some feedback on its talk page on what is to be improved before it can become an official guideline. Thanks! pandakekok9 13:30, 17 December 2020 (UTC)

It reads as having a focus on edits, rather than uploads and copyright issues. We have had folks who are uninterested in engaging in consensus but their contributions on uploads are highly worthwhile. There is much less of a need for "repeatedly disregards other editors' questions or requests for explanations" to be considered a problem on Commons unless a contributor is unambiguously ignoring specific policy requirements.
Indeed we have also seen editors being hounded for years, and griefing with repeated demands for replies to questions can easily create a hostile environment, which COM:BP includes, but can be hard to request action for when the griefer is experienced and knows how to game the system. --Fæ (talk) 13:55, 17 December 2020 (UTC)
Is there also a guideline about 'disruptive' deletion nominations at COM:DR? Case in point being deletions that seem to have more to do with moral (or copyright) paranoia than carefully thought through legal or policy considerations. ShakespeareFan00 (talk) 18:00, 17 December 2020 (UTC)
Hi! I recently opened a disruptive deletion request here trying my best to convey my thought. Sorry for being disruptive though. Please close the DR if necessary. Good luck with the guidelines! sB talk 19:00, 17 December 2020 (UTC)
@ShakespeareFan00: There doesn't seem any. I addressed this at the draft I made. Please give feedback. Thanks, pandakekok9 03:56, 18 December 2020 (UTC)

Some renaming help please?

If you have renaming rights maybe you could help with this backlog?

As part of the ongoing large COM:IA books project, some very old books were uploaded with bad titles by mistake. There's only ~135, see search, but it's the sort of naming problem that is probably best done by hand. The IA "title" is in the description, but it takes a bit of thoughtful selection to decide how much of the text to use in the filename, plus some are in Latin.

The bug is fixed, but these were uploaded before the glitch was noticed.

We'll probably put up an announcement about IA books in January, but right now the total uploads are over 920,000 documents, and naturally this makes for some large backlogs of all sorts. Thanks! --Fæ (talk) 08:28, 18 December 2020 (UTC)

Weird rendering of Gif-file

At the Wikidata-infobox here the foto looks very strange. On the same page a smaller version of the same Gif-file looks OK. Is that caused by a bug? - Robotje (talk) 11:43, 18 December 2020 (UTC)

@Robotje: It appears our systems are very bad at displaying downscaled versions of GIF photos. The GIF format can only display 256 distinct colors, and is not recommended for storing photos on Commons per COM:FT.   — Jeff G. ツ please ping or talk to me 15:08, 18 December 2020 (UTC)
Would it be an option to concvert the .gif file in a .jpg file? - Robotje (talk) 15:13, 18 December 2020 (UTC)
@Robotje: ✓ Done, see File:Piet Adema.jpg, changed in Wikidata, you can change it in Dutch Wikipedia.   — Jeff G. ツ please ping or talk to me 15:35, 18 December 2020 (UTC)
Thanks. - Robotje (talk) 18:34, 18 December 2020 (UTC)
@Robotje: You're welcome.   — Jeff G. ツ please ping or talk to me 08:47, 19 December 2020 (UTC)

Commons:Do not disrupt Commons to illustrate a point as a guideline

See Commons talk:Do not disrupt Commons to illustrate a point#Guideline. Note that this is not yet a proposal. I'm just soliciting some opinions on whether there is anything that need to be added or removed before this can become a guideline. pandakekok9 14:30, 19 December 2020 (UTC)

Please review

Please review the photo and files in this category: Category:Farsnews review needed. Wishing health and happiness for all users. 90 TV (talk) 19:08, 19 December 2020 (UTC)

Categories of dates before 1582-10-15 should not be formatted in ISO 8601

As ISO 8601 disallows the Julian calendar, may I propose renaming all categories before 1582-10-15 to English months like 18 March 1582? Categories of dates since 1582-10-15 should be strictly defined per the Gregorian calendar.--Jusjih (talk) 22:17, 14 December 2020 (UTC)

Who says that such categories use Julian calendar dates? ISO 8601 allows dates before 1582-10-15, using the Proleptic Gregorian calendar. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:43, 15 December 2020 (UTC)
Per "prior agreement", there's nothing to suggest a change is necessary to whatever the "norm" is right now. --Fæ (talk) 11:37, 15 December 2020 (UTC)
There's no demonstrated need to do this, and such action would almost certainly break vast amounts of formatting templates and other emplaced methods. — Huntster (t @ c) 13:11, 15 December 2020 (UTC)
 Oppose per Andy, Fæ and Huntster. ISO 8601 style dates fit here just fine.   — Jeff G. ツ please ping or talk to me 13:19, 15 December 2020 (UTC)
Then is Category:1492-10-12 in the Julian calendar, or the proleptic Gregorian calendar?--Jusjih (talk) 05:32, 20 December 2020 (UTC)

Poker vandalism

I would like to make other users aware of a vandalism problem with images related to poker. This has been occurring for several months and is still an ongoing issue. New accounts are being created and used to add spam links for gambling websites to the descriptions of these images. Most of these accounts are only used to make one edit. Every edit is listed as a minor edit and all the edit summaries are identical. If you view my contributions, you will see that I have reverted over 60 spam edits. Johnj1995 (talk) 17:27, 19 December 2020 (UTC)

Nice find and well done. For two of the links, I found them also being spammed on English Wikipedia and have requested global blacklisting on Meta. ~Kevin Payravi (talk) 02:50, 20 December 2020 (UTC)
I'd like to say nice catch too. But please be careful with your edit summaries, because I was confused about this revert you did. The revert was correct (it's not the Amazon rainforest), but the edit summary was wrong. Btw, you might want to apply for patroller and rollbacker. pandakekok9 12:36, 20 December 2020 (UTC)

Are the Iraqi most-wanted playing cards "PD-USgov"

I recently came to know about the most-wanted Iraqi playing cards issued by the government of the United States of America. Usually works by the American government are licensed "{{PD-USgov}}", but as the images on these cards are sometimes high quality I am not sure if the American government actually authored all photographs on them.

Now generally United States government works are accepted at face value, but in this case I can't be fully certain that the American government actually made these, nor can I find definitive information that confirms that they didn't. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 09:48, 20 December 2020 (UTC)

The Wikipedia article says that the pictures came from "numerous intelligence agencies and 'open sources'", which doesn't reveal enough info. You could try contacting the Defense Intelligence Agency under the Freedom of Information Act (though I'm not sure if that applies as well to non-US citizens). pandakekok9 12:42, 20 December 2020 (UTC)

File page cleanup bot?

Is there any bot that does some generic cleanup on file pages, like adding a "Summary" section header above the {{Information}} template? Media description pages seem to be quite messy and ununified. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 16:17, 12 December 2020 (UTC)

There is no general consensus for forcing a specific format on image pages, nor specific templates. There are plenty of image pages that have no information templates, just the description and source typed in. They don't need fixing. --Fæ (talk) 18:15, 12 December 2020 (UTC)
@Fæ: Actually, we have Category:Media missing infobox template for the image pages that have no information templates, so I guess that suggests they do need fixing. And I was talking specifically about the pages that have an information template but not the heading. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 18:20, 12 December 2020 (UTC)
Yes, we have that category, people write bots and reports for whatever they wish to try. However, there is no consensus for mass changes to standardize having subsections on image pages, nor to limit sections to a specific set. --Fæ (talk) 18:27, 12 December 2020 (UTC)
Is Magog the Ogre's tool against consensus? 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 18:38, 12 December 2020 (UTC)
No consensus exists. --Fæ (talk) 22:03, 12 December 2020 (UTC)
SchlurcherBot (talk · contribs) does some.--RZuo (talk) 12:40, 13 December 2020 (UTC)
@RZuo: with all due respect, your last remark makes no sense. I assume you are a non-native speaker trying to be terse in English (my apologies if that's wrong), but you were terse to the point of incomprehensibilty. Maybe "some does" with "consensus" from the previous remark from the antecedent? Could you repeat using a few more words? - Jmabel ! talk 19:21, 13 December 2020 (UTC)
@1234qwer1234qwer4: my bot went through the official approvals and does this sort of work for new uploads, yes. Magog the Ogre (talk) (contribs) 15:38, 14 December 2020 (UTC)
@Magog the Ogre: Why not for old ones, too? 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 15:45, 14 December 2020 (UTC)
@Jmabel: I don't get your point. I asked if there was a bot doing any cleanup, and RZuo replied SchlurcherBot does some. Nothing unclear to me. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 15:45, 14 December 2020 (UTC)
@1234qwer1234qwer4: I see, the mention of SchlurcherBot was part of the sentence. I read that as a ping, and thought I couldn't see the antecedent for "does some"! - Jmabel ! talk 17:12, 14 December 2020 (UTC)
@RZuo: {{U}} would have been slightly clearer.   — Jeff G. ツ please ping or talk to me 17:17, 14 December 2020 (UTC)
i am well aware of the different templates. and i intentionally included a link to contribs so ppl can see how well the bot performs rightaway.--RZuo (talk) 04:35, 21 December 2020 (UTC)

Error in download : ask for rapid delete

Hello. I have just download this file File:Tarente statère 320 avJC.jpg but I have not look enough the licence CC BY-NC-SA 3.0 - I don't notice the NC ! So the picture will be deleted. Thanks. - Siren-Com (talk) 00:14, 21 December 2020 (UTC)

@Siren-Com: I marked the file to be deleted. In the future, you can just add a template such as {{speedy|G7}} to the file (see {{Speedy}}. – BMacZero (🗩) 22:08, 21 December 2020 (UTC)

Musics under Filmmusic website

Hello.

Since December 17th 2020, all musics on Filmmusic website are under "Filmmusic standard license". Before this date, there was under CC-BY license.

Before this date, I've uploaded these musics here : [4].

The website : [5].

Is this "license" is compatible with Wikimedia Commons ?

Thanks a lot.

--ComputerHotline (talk) 22:02, 21 December 2020 (UTC)

I looked at the "What is not allowed?" section and that looks too restrictive to be allowed on Commons. Too bad. Multichill (talk) 22:31, 21 December 2020 (UTC)

Category of rotten food?

i was looking for photos but there isn't actually a category for all kinds of rotten/spoiled food? Category:Plant rot seems to be for rotten plants. i cant find one for meat.

is it a good idea to build this cat tree? should their names be Category:Rotten XXXX?--RZuo (talk) 06:13, 21 December 2020 (UTC)

If there is content for such cats their creation would seem a good idea to me Oxyman (talk) 12:42, 21 December 2020 (UTC)
Agreed, and "spoiled" is more general than rotten (e.g. canned food where the canning failed) so Category:Spoiled food should probably be a parent of Category:Rotten food. - Jmabel ! talk 13:53, 21 December 2020 (UTC)
Agree with Jmabel; spoiled is more general, and rot is a more specific subcategory. Both potentially useful. -- Infrogmation of New Orleans (talk) 15:00, 21 December 2020 (UTC)
according to oxford dictionary:
spoiled is "(of food) having become unfit for eating"
rotten is "Suffering from decay"
so they are actually different concepts. spoiled is about whether it's edible. rotten is about whether it's decomposing.
but the two concepts are highly overlapped. very few things could be rotten but not spoiled.
so we will build this tree as Category:Spoiled XXXX.
but should we redirect Rotten XXXX to Spoiled XXXX? or make them subcats?--RZuo (talk) 15:16, 22 December 2020 (UTC)
Category:Spoiled food could mean a lot of things. Bugs, mold, rot, or even just getting dirty or drying out could spoil food. Rot is technically synonymous with Category:Fermentation isn't it, just with a negative connotation? We have many examples of edible Category:Fermented food. I'm surprised we don't have more images of mold on food. Aside from intentional ones like blue cheese, I see Category:Moldy roast chicken, Category:Bread mold, Category:Fruit mold, as well as (the mis-named?) Category:Rotting tomatoes and Category:Spoiled rice? - Themightyquill (talk) 15:46, 22 December 2020 (UTC)

move photo

move photo

File:Four for McGovern poster with venue information and list of ushers. Pictured is Carole King, Barbra Streisand and James Taylor.jpg
to
File:Three for McGovern poster, Carole King, Barbra Streisand and James Taylor, with venue information and list of ushers.jpg
24.7.56.99 17:58, 22 December 2020 (UTC)

Why no "View user in global rights log" on Commons?

compare en:Special:UserRights/Jimbo_Wales with Special:UserRights/Jimbo_Wales, how come we dont have this convenient link here? how do we add it?--RZuo (talk) 04:35, 21 December 2020 (UTC)

Perhaps this could better be suggested in the Phabricator, as the people actually capable of changing that frequent there. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 21:59, 22 December 2020 (UTC)
@RZuo and Donald Trung: because nobody every bothered to update MediaWiki:Viewinguserrights? ✓ Done now.
If in the future you're puzzled about what message needs to be updated exactly, use the magic language qqx. That will show what messages are in use on a page. Multichill (talk) 12:52, 23 December 2020 (UTC)

Is this pdf upload allowed ?

Hello. Is THIS PDF UPLOAD allowed ? --Havang(nl) (talk) 14:37, 24 December 2020 (UTC) Same question for THIS DJVU UPLOAD which is not up[loaded by me, but which I like to correct on wikisource. --Havang(nl) (talk) 14:42, 24 December 2020 (UTC)

@Havang(nl): The second certainly is as the author died in 1941, which is at least 70 years ago as specified by French copyright law. I replaced the license with {{PD-old-auto-expired|deathyear=1941}} so a US license is also shown as required by Commons policy. I was not able to quickly find out the death year of the author of the first - if that is unknowable, {{PD-old-assumed}} will work, but it would be better if the death year could be confirmed and the same license used as the other. – BMacZero (🗩) 18:59, 24 December 2020 (UTC)
Merci : Sylvin n'est pas le problème, c'est la requête de google ; ne pas utiliser commercialement. Mais je viens de lire, que cela n'engage pas le copyyright , qui est domaine publique. --Havang(nl) (talk) 19:18, 24 December 2020 (UTC)
That Google cover page is usually removed. Many users, especially wikisource users who use djvu files frequently agree that the limitations on that google cover page are invalid. This is after all an PD file, so the only rights Google has is trough rules on derivative works. While it would be nice to get an confirmation from an lawyer, their claims are doubtful. (I understand basic french)--Snaevar (talk) 17:05, 25 December 2020 (UTC)

Illegal occupancy of Jammu and Kashmir

Hello everyone, I have seen many flags of Pakistan occupied Jammu and Kashmir here as named "Azad Jammu and Kashmir". Azad means free in Urdu/Hindi. I object this as an Indian because that part of J&K was illegally occupied by Pakistan in 1948. Why should not we rename all the flags of that region available here from the name "Azad Jammu & Kashmir" to "Pakistan occupied Jammu & Kashmir"?--Junior Jumper (formerly Tæ) 14:55, 24 December 2020 (UTC)

While I agree with your perspective, the reason why Pakistani Kashmir is called "Azad (Free) Jammu and Kashmir" is because it's the official name of the region, while Indo Kashmir is simply called "Jammu and Kashmir", compare this with "the Democratic People's Republic of Korea" (North Korea) Vs. "the Republic of Korea" (South Korea) with the latter being an actual democracy, names are often propaganda tools and Wikimedia Commons tends to prefer official names like this. Changing this to "Pakistan occupied Jammu & Kashmir" wouldn't be neutral. Alternatively it could be titled "Jammu & Kashmir (Pakistan)" to contrast it with "Jammu & Kashmir (India)", but Pak Kashmir is already officially called "Free Jammu & Kashmir" so the current naming scheme is the most technically correct by going by official names. Happy ☧-mas --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 23:24, 24 December 2020 (UTC)

Commons:Uploading from the Seattle Municipal Archives Flickr stream

I'm reviving this from the archive because it got no response at all, not even a suggestion of how I might proceed. - Jmabel ! talk 20:55, 24 December 2020 (UTC)

For a decade or so, I've been uploading from the Seattle Municipal Archives Flickr stream and dealing with various metadata issues, etc. There is some really good content; see Category:Seattle Municipal Archives via Flickr. The task is not as trivial as it sounds; please follow the link in the subject line for this section if you want details.

For at least the next several months I believe I will be too busy to do this, or at least to do it thoroughly. I'm backlogged probably a month, and could use someone else either to take over or at least to help. I'd be glad to mentor: that would be such good return on my time that it would be worth putting in a bit of "the time I don't have."

Any takers? - Originally posted Jmabel ! talk 14:37, 15 December 2020 (UTC), reposted added Jmabel ! talk 20:54, 24 December 2020 (UTC)

Again: I've been doing this for years, it is worth doing, but I am not having time to do this right. I can even continue to help out, but cannot carry it all myself. - Jmabel ! talk 20:55, 24 December 2020 (UTC)
Is this something a bot can handle? Otherwise, I would post the Commons page into Category:Requested files at least. This feels like one of those times where WikiProjects would be helpful here. Perhaps post at the Seattle one on English? -- Ricky81682 (talk) 21:11, 24 December 2020 (UTC)
@Ricky81682: Not a hope for a bot: one of the steps is to actually research the photos and provide better descriptions.
Seattle group in the 'pedia is a good idea, I'll follow up, though probably not today. - Jmabel ! talk 16:03, 25 December 2020 (UTC)
@Jmabel: I'd offer to help but I don't know anything at all about the Seattle area. -- Ricky81682 (talk) 19:16, 25 December 2020 (UTC)
@Ricky81682: Not surprising. It's not like it was NYC, but the city government here has one hell of an archivist's office, I think 6-7 FTEs. Pretty much any materials the City ever had are still there, and they know how to find it. - Jmabel ! talk 20:31, 25 December 2020 (UTC)

Holiday reads

Good news everyone!

Our huge book upload project COM:IA books, has not stopped for the holidays, in fact minutes ago, Category:Scans from the Internet Archive went over 1,000,000!

We now have an incredible backlog of public domain books, magazines and reports of all types. Here you can find favorite classics to Nobel prize winners from Anderson, Hans Christian to Zola, Émile.

These are easy to download to a tablet or phone and use your own PDF reader for a shelf of books to browse, even when the internet goes down, or use the link on the Commons page to surf to the Internet Archive online reader.

Illus. Search suggestions
Christmas returns poetry, prose, photos and production stats for Christmas trees. Enjoy the list, find something to save and read, or dip in and add some easy categories, like publication years and location, author, subject matter. Don't forget to bookmark any treasures, I assure you with this large library it's easy to lose them!
Christmas Carol, oh my, this was a popular meme in the early 20th century. Enjoy reading through some variations of Dickens original, or perhaps help curate the many spin off works into better Commons categories.
National Geographic always has been classic coffee table reading, and provides a brilliant insight in to how the world was viewed a century ago, at least from an American perspective. The magazine series and associated publications are crying out for good curation, and putting in perspective from the modern understanding of "empire" and debunked theories of "human races".
If you get tired of Youtube over the holiday, maybe it's time to add Shakespeare to your tablet Christmas reading? His writing was influenced by outbreaks of terrifying bubonic plague, with 25% of Londoners dead and all theaters closed for a year and a half, he responded by writing King Lear and Romeo and Juliet.

Time to start preparing Christmas lunch, and like many of you during the pandemic, with lockdowns and disruption, for the first time spending the day alone, thank goodness we live a time of videoconferencing! Happy holidays, stay safe, stay mellow. :-) --Fæ (talk) 10:02, 25 December 2020 (UTC)

How can I rotate a PDF image or get someone else to?

I posted a PDF (File:Wagener Terrace.pdf), but the image needs to be rotated 90 degrees clockwise. I tried inserting the note recommended by Wikimedia, but the reply was added that PDFs can't be rotated. How can I get rotate the image or get someone to do it? I don't even need to get the original image spun, just to rotate the image as used in a Wikipedia article, if that is easier.

(PLEASE do not reply here with any Wiki-jargon.) --ProfReader (talk) 14:28, 26 December 2020 (UTC)

@ProfReader: Hi, and welcome. I'm sorry to inform you that we don't have an automatic way to rotate a PDF image. Looking at File:Wagener Terrace.pdf, I found that XenonX3 had already rotated it for you.   — Jeff G. ツ please ping or talk to me 14:46, 26 December 2020 (UTC)
I also uploaded a png version as File:Wagener Terrace.png for the possibility of future edits of the image. XenonX3 (talk) 14:48, 26 December 2020 (UTC)
@XenonX3: The png version will look fuzzy when scaled down (due to design decisions discussed in phab:T192744), so you may want to upload a jpg version which will look sharper, too.   — Jeff G. ツ please ping or talk to me 14:50, 26 December 2020 (UTC)
@Jeff G.: Thanks, I didn't know that. I just uploaded File:Wagener Terrace.jpg then. XenonX3 (talk) 14:55, 26 December 2020 (UTC)
@XenonX3: Thanks, and you're welcome.   — Jeff G. ツ please ping or talk to me 15:00, 26 December 2020 (UTC)

Croptool

Any chance that Croptool will be expanded with an oval crop and a circular crop? --RAN (talk) 19:08, 11 December 2020 (UTC)

croptool has recently been changed: You no longer choose between lossless and precise, but the tool tells you if a crop was lossless or precise. This is a dramatic change of usage and it still does not handle copyright (adding information about the person doing the crop, required by cc-by-sa). Guess from this, how likely it is, that the author of the tool works on requests from the users of the tool. --C.Suthorn (talk) 06:43, 12 December 2020 (UTC)
The change to lossless vs. precise is odd, but I don't think the derivative made from a crop violates any attribution requirements: Whether overwriting or creating a new file the tool describes the changes in the file history upon upload (e.g. "cropped 50 % horizontally, 42 % vertically using CropTool with lossless mode"), and for new file adds {{Extracted from}}. I think either one of these would satisfy the condition of "indicate if changes were made," done so "in any reasonable manner". Having both is more than sufficient. I also don't think that mere cropping is sufficient threshold of originality to grant the cropper any ownership rights or attributions requirements (cutting a rectangle out of an existing image is not really a new creative work), but the user performing the cropping/modification is clearly indicated in the file history, which is visible on every file page. I don't see any CC-license requirements that the person making the changes must be indicated, only that changes were made. --Animalparty (talk) 21:25, 12 December 2020 (UTC)
As has been pointed out before a Crop can make an image illegal, if a part of the image, that was DeMinimis in the orginal becomes a main subject in the crop. Also many images have been composed by the photograph/author to include the content it contains, and often also to a specific aspect ratio. The Change is certainly a creative act and has threshold of originality. Some have argued in the past, that the CropTool needs to be deactified until this issue is solved. --C.Suthorn (talk) 16:51, 16 December 2020 (UTC)
Is there any evidence that the threshold of originality in the US can be exceeded by a simple crop? Since it can't be exceeded by colorizing a black and white picture, and the legal difference between a pan-and-scan and uncropped movie (not just a single frame) was debated just like the difference between a black & white movie and the colorized version, I would be stunned if a simple rectangular crop could exceed the threshold of originality. While "certainly a creative act" is not a factually answerable question, this is.--Prosfilaes (talk) 09:19, 18 December 2020 (UTC)
See COM:Forum: it violates cc-by-sa-4.0 --C.Suthorn (talk) 17:58, 21 December 2020 (UTC)
That's not how you link to something, and it's absolutely not responsive to what I said, nor is it responsive to what Animalparty said. Do you want to try that post again?--Prosfilaes (talk) 07:30, 23 December 2020 (UTC)
Nein, das werde ich nicht versuchen. Ich habe hier und im Forum auf einen Mißstand hingewiesen, auf den auch andere schon in der Vergangenheit hingewiesen haben. Leider hat das nicht zu einer Behebung geführt, aber immerhin muss ich mir später, wenn das Problem akut wird, nicht vorwerfen lassen, ich hätte nichts gesagt. Im Gegensatz zu einem spanischen Landedelmann werde ich aber nicht gegen Windmühlen kämpfen. Und das schöne an Wikipedia ist ja, das nahezu alles öffentlich ist und nahezu alles auch in öffentlichen Archiven landet. Fröhliche Weihnachten. --C.Suthorn (talk) 08:58, 23 December 2020 (UTC)

Removing coordinates from categories when they are in the infobox

Hi all. I'm planning on starting to remove uses of {{Object location}} from Commons categories where the coordinate is on Wikidata and {{Wikidata Infobox}} is present (e.g., [6], [7]), as part of d:Wikidata:Requests for permissions/Bot/Pi bot 18. I think this falls under the existing Pi bot tasks here (deploying infobox/removing IDs in the infobox), but if anyone disagrees then I can submit a new bot request. Any comments? Thanks. Mike Peel (talk) 19:33, 25 December 2020 (UTC)

I think it's reasonable extension of task, but bot should remove only matching (or very closely matching) coordinates. Other cases should be handled manually. --EugeneZelenko (talk) 15:38, 26 December 2020 (UTC)
@EugeneZelenko: I check that they are within 100 metres, is that sufficient? Thanks. Mike Peel (talk) 17:13, 26 December 2020 (UTC)
I think 100 meters are too much. 10 should be better threshold. --EugeneZelenko (talk) 17:27, 26 December 2020 (UTC)
OK, changed to 10 metres. Thanks. Mike Peel (talk) 17:47, 26 December 2020 (UTC)
Please don't. You can't get a scale on Wikidata, which you need when curating categories. You don't want to be zooming in all the time from some impossible large area. Keep them but at least make them consistent, and don't use en:Wikipedia, Historic England or Geograph coords as they are often laughably wrong. I've only been doing this for 13 years, but YMMV. Rodhullandemu (talk) 15:46, 26 December 2020 (UTC)
Pinging @Mike Peel about "You can't get a scale on Wikidata".   — Jeff G. ツ please ping or talk to me 17:09, 26 December 2020 (UTC)
@Rodhullandemu: The infobox auto-zooms based on the area of the item, that seems to work reasonably well? Discrepant ones would have to be manually checked after the bot run - one of the advantages of having them on Wikidata is that you only need to fix them once rather than here+elsewhere. Thanks @Jeff G.: for the ping. Thanks. Mike Peel (talk) 17:13, 26 December 2020 (UTC)
@Mike Peel: You're welcome.   — Jeff G. ツ please ping or talk to me 17:24, 26 December 2020 (UTC)
 Oppose No consensus for this mass change. A solution can and should be found that enabled editing coordinates on Commons, not forcing Commons editors to log in to Wikidata (if they can) and edit a completely different project just to make corrections or amendments to Commons pages. Commons should remain the "master record" about media on Commons. --Fæ (talk) 10:05, 27 December 2020 (UTC)

About the license of These Logos

If anyone can assisst about the license of File:KurulusOsmanATVtitlePoster.png and File:KurulusOsmanTitleImage.png of both logos because it seems that they can fall under PD-textlogo. Anyhow, please assist if they are compitable with this license and change their license with Trademark template because they are uploaded under wrong license. If they are compitable please anyone also submit for conversion or convert these into SVG.Thank You.59.103.122.87 09:23, 27 December 2020 (UTC)

Babel boxes

Do we have anywhere a multilingual template that can be put on a user's talk pages to suggest that they add Babel boxes? - Jmabel ! talk 14:27, 28 December 2020 (UTC)

I am only aware about {{Welcome}}, which contains such a suggestion. Ruslik0 (talk) 13:10, 29 December 2020 (UTC)
In passing, in with other things. I'm thinking about something to use when you are trying to talk to someone in particular on their user talk page and are unsure if there is a common language. - Jmabel ! talk 13:18, 29 December 2020 (UTC)
@Jmabel: Beyond babel boxes, I look at the wikis to which they contributed the most via CentralAuth, sorted on Edit count twice (ascending then descending) if there are many.   — Jeff G. ツ please ping or talk to me 07:25, 30 December 2020 (UTC)
@Jeff G.: Yes. If I recall correctly a few days later, the contributor in question had only French-language contributions, but I didn't want to presume that was their only language. My French is pretty weak. I was trying to see if there might be a better choice. - Jmabel ! talk 01:46, 31 December 2020 (UTC)

Leiden University Libraries Digital Collections uploads

L.s
After a lot of nice support from some wikipedians, I want to slowly start uploading again.
I have a request to upload immages from the Leiden University Libraries Digital Collections. These are mostly Creative Commons CC BY Licenses. The program i wrote will check for this. Right now i am doing some test runs to see if it all works accordingly. If by any chance something goes wrong, please let me know and i will try to fix it. If someone has any other ideas on how to improve the uploads, please let me know as well and again i will see if i am capable of rewriting the program in such a way as to make it work. In the past i had some suggestions as to use photo-, and art templates. However, i could not find any way on how to distinguish these at the source. The catagorization of the source is copied at the moment, but we can change those later to more appropriate categories if necessary.
Regards,
Mr.Nostalgic (talk) 08:18, 29 December 2020 (UTC)

You may wish to examine the EXIF data on the files. It may be that the artwork photos have been consistently taken with different cameras, or have EXIF data comments, that distinguish them from the photo collections. It is possible to filter these differences as post-upload housekeeping. It's the sort of thing that can be requested at COM:Bots/Work requests. --Fæ (talk) 11:50, 30 December 2020 (UTC)

Account deletion

I am sometimes approached by Commons contributors who want me to delete their account. Some want to have their their uploads deleted, some just want their account to be deleted or blanked.

Do we have a FAQ about this? I could not find any guide on Commons. All I have found is on other sites and it is not clear how it applies to Commons: https://meta.wikimedia.org/wiki/Right_to_vanish https://en.wikipedia.org/wiki/Wikipedia:Courtesy_vanishing

Here is my current understanding for Commons:

  • Having one's uploads deleted is almost impossible, you would have to go through a deletion request for each file. Some cases can be slightly easier:
  • Deleting one's account is impossible, but:
  • You can request for your user to be renamed.
  • You can blank your user page and your discussion comments (but not other people's comments on your user talk page).

Is that correct? How about creating a FAQ if one does not exist already?

Thanks! Syced (talk) 03:28, 30 December 2020 (UTC)

@Syced: Have you reviewed COM:FAQ#How do I close my account?   — Jeff G. ツ please ping or talk to me 07:18, 30 December 2020 (UTC)
Not sure how I missed that, thanks! Syced (talk) 00:46, 31 December 2020 (UTC)
@Syced: You're welcome!   — Jeff G. ツ please ping or talk to me 00:56, 31 December 2020 (UTC)

"COVID-19 pandemic in XXX" categories

The new year is upon us and the SARS-Coronavirus-2 pandemic doesn't appear to be going anywhere, but most categories like "Category:COVID-19 pandemic in Bonn" are subcategories of things like "Category:COVID-19 pandemic in North Rhine-Westphalia" and "Category:2020 events in Bonn" which clearly indicate that the pandemic happened in 2020. I actually wanted to bring this up last month, but due to personal circumstances couldn't. What would be the best course of action here? As it's likely that the SARS-CoV-2 pandemic isn't going to be stopping anytime soon.

I have options like creating a "Category:COVID-19 pandemic in Bonn in 2020" and then a "Category:COVID-19 pandemic in Bonn in 2021" and move all current content into the former, or simply add "Category:2021 events in Bonn" already (though I am less a fan of the latter). Any more suggestions? --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 13:46, 31 December 2020 (UTC)

I don't think there is a need to split the pandemic categories by year. We don't have news articles so it's not like there's a ton in these categories that needs to be be identified by year. The time of the category long-term will not be that important. I would rather we keep it as COVID-19 pandemic in Bonn and have it in both 2020 and 2021 categories. The individual images can then be split into more well-defined subcategories. If the category is big, then it could be split but I don't see the need. Besides I could see COVID-19 related artwork or other things continuing past 2021. -- Ricky81682 (talk) 19:58, 31 December 2020 (UTC)
Creating by-year-categories is unnecessary for the covid-19 pandemic media. GeorgHH • talk   20:25, 31 December 2020 (UTC)
Should it even be treated as an "event"? Other pandemics such as Category:Spanish flu seem not to be. --ghouston (talk) 04:12, 1 January 2021 (UTC)

Month in country by city template coding

I have a disagreement with @Kleeblatt187: about how to organize and code various month in city category template. For an example, Template:Berlinmonthyear used at Category:March 1902 in Berlin mandates the creation of Category:March 1902 in Germany by city which is then within Category:March 1902 by city (or a red link). This could be useful if you wish to navigate Germany by city for every month but gets silly very quickly as in each month at Category:1900 in Germany by city. I set it as "if March 1902 in Germany by city exists, use it or use March 1902 by city." Older categories can be directly put into the "by city" category while newer ones can be put into a cities in Germany one if someone just creates it. Either way, I don't see how removing the ifexists function does anything other than make these categories more hidden unless every single month and year that has even a single German city is created. I'm avoiding the actual question of whether categories like Category:March 1902 in Berlin shouldn't just be upmerged to the parent categories and dealing with what is here now. -- Ricky81682 (talk) 21:08, 31 December 2020 (UTC)

We have had a talk at User talk:Kleeblatt187#Template:Leipzigmonthyear, in my opinion we are still talking. --Kleeblatt187 (talk) 21:12, 31 December 2020 (UTC)
Yeah, we can get more opinions here. As to how long the autocategorization takes, it takes a second to edit each category manually and save the change which will populate the category if it exists. Otherwise, I don't know how many redlinks have been created since we were only taking about a single template right now and I don't care to manually check every one of the individual templates to see what is missed. It's simple enough to aggregate all that and figure out. I question whether a single category like in the year 1900 is helpful but I'm not stopping you from creating all those. -- Ricky81682 (talk) 05:14, 1 January 2021 (UTC)
I' ve answered even to this at the original place. Best wishes, --Kleeblatt187 (talk) 09:46, 1 January 2021 (UTC)

Can't find DEFAULTSORT conflict

I added a DEFAULTSORT template to https://commons.wikimedia.org/wiki/Category:Jakob_Br%C3%BCllmann and when I saved it, I got an ugly red message that said, 'Warning: Default sort key "Brüllmann, Jakob" overrides earlier default sort key "Brullmann, Jakob".' I understand the problem,as my entry has an umlaut-U while the conflict has only a U. I have looked everywhere, including the Wikidata page, and can't find an "earlier default sort key". Can someone help me resolve this? Thanks in advance. Seauton (talk) 22:06, 31 December 2020 (UTC)

The category is proper categorized by wikidata without the manually added defaultsort key. Why do you think a manually sort key is necessary for this category? Second, so far as i know, umlaut-u (and the other) are not allowed/useful, but i cant tell you the appropriate help page for that, cant find it now. At last, probably is the description of Category:Pages with DEFAULTSORT conflicts helpful. Happy new year GeorgHH • talk   00:18, 1 January 2021 (UTC)
According to MediaWiki help, Sorting occurs based on the Unicode values of the characters. That explains why we avoid special characters, because they would sort after everything, even Z. – BMacZero (🗩) 02:49, 1 January 2021 (UTC)
There are cases where a DEFAUTLSORT template is needed and better than a sort scheme generated solely from Wikidata (e.g. artists with prominent pseudonyms that differ from their legal names). Wikidata infobox loses its damn mind and defecates all over Commons in bright red whenever some robot gets confused. It sucks, but ignore it and a bot will ostensibly soon remedy the situation. Template:Wikidata Infobox is the source of almost every display problem in Commons categories. --Animalparty (talk) 02:57, 1 January 2021 (UTC)
Sure, when there are multiple names a sort order can't be reliably generated. But it's easy enough to add a defaultsort=no option to the infobox. --ghouston (talk) 04:08, 1 January 2021 (UTC)

Very large number of troubling edits with SuggestedEdits

Tracked in Phabricator
Task T244672

A couple months ago, I opened a discussion at AN about SuggestedEdits, a feature of the Wikipedia app that makes changes to Commons. I noticed that there were a lot of people making changes with that feature and that more than half of the edits were vandalism, incorrect, spam, nonsense, etc. That thread didn't get much of a response. Today I looked again at the change log for that feature and still see many, many problematic changes being made. See the other thread for my evidence there. Just from the last couple hours there are incorrect depicts statements (that Q# points to a work of creative writing), spam?, confusion, nonsense, addresses?, who knows why, customer service phone numbers, partial unhelpful translation of the English description to Arabic for the English caption, and, less problematic, changes that are so vague as to be barely useful...

I will say that my spot check this time had better results than it did last time. More than half of the edits this time look at least a little bit helpful, but still a lot of problems (and more edits that I cannot read [perhaps it was rolled out in other languages since then? -- access to the feature in one's own language may account for the decrease in problems]).

I'd like to open a discussion about this, because I haven't seen much of substance (maybe I've just missed it). I like the idea of bringing new people in, and of crowdsourcing our data, but the question is whether the good information being added is worth misinformation and worse? I don't know the answer. — Rhododendrites talk |  17:11, 26 December 2020 (UTC)

More examples at Commons talk:Suggested Edits#Thread at AN. I've also noticed this, and reverted some instances of it, but checking through a huge log every day would currently be too much for me. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 17:40, 26 December 2020 (UTC)
To add, it may be helpful to look at the CentralAuth pages for the users in question. Check this out, for example. Importing mw.loader.load('//en.wikipedia.org/w/index.php?title=User:The Voidwalker/centralAuthLink.js&action=raw&ctype=text/javascript'); makes the process quicker. 𝟙𝟤𝟯𝟺𝐪𝑤𝒆𝓇𝟷𝟮𝟥𝟜𝓺𝔴𝕖𝖗𝟰 (𝗍𝗮𝘭𝙠) 17:54, 26 December 2020 (UTC)
This was introduced as a "trial". How exactly do we get it removed?
The trial failed, there's no ambiguity about this being a catastrophically bad change for this project. --Fæ (talk) 13:44, 27 December 2020 (UTC)
Actually the trail was a success: It showed very clearly a number of shortcomings of the software. A responsible Company would now revert the software to a prior version (or switch of editing using the feature) and work on fixing the known shortcomings, than make another try. --C.Suthorn (talk) 09:27, 28 December 2020 (UTC)
Also known as the Brexit is a success post-hoc rationale. --Fæ (talk) 10:22, 28 December 2020 (UTC)
@Fæ: Not quite. C.Suthorn is saying, perhaps tongue in cheek, that the trial was a success, in that it proved that the feature was a failure. Sort of like if I'm called in to estimate how long a software project will take and I say "not a hope that this can deliver, certainly not in the year they promised, maybe never," then two years later when they haven't delivered, my estimate was clearly a success. - Jmabel ! talk 14:16, 28 December 2020 (UTC)
Congratulations to everyone! Now, how exactly do we get it removed? --Fæ (talk) 14:20, 28 December 2020 (UTC)

Gamification and consensus

It seems to me that if we really want to glean useful "depicts" statements from unregistered or inexperienced users, we would need something more like a game where we would need effectively to find out that there is a consensus (or near-consensus) among several different people as to what the image depicts before actually editing the wiki; in theory, interim results could go in an HTML comment in the wikitext, if there is no more convenient place to store them. @Fæ: , I believe that last is what you do with equivalently untested suggestions you scrape from third-party metadata, isn't it? - Jmabel ! talk 14:24, 28 December 2020 (UTC)

The upload projects I work on literally use 1st party metadata, made by humans so subject to error like anything else. Second party would be something like presuming Flickr tags have meaning, third/Nth party would be something like scraping Wikidata (which itself almost always breaks the provenance chain, and ignores copyright issues by presuming everything is CC0 without any consistent verification to prove it one way or the other).
No, we have had several gamification projects. These proved that there are not enough interested casual volunteers for it to help systematically. We're not Galaxy Zoo. --Fæ (talk) 14:36, 28 December 2020 (UTC)
I understand, that there was (before SDC went live) a community decision to not have the developers fill in depicts on the server side via batch jobs, but that the community wants to do this themselves with individual edits. IMHO a decision driven by ignorance on maths. The task is much to big as >65M files need depicts. This decision needs to be undone, depicts prefilled by database admins (based on categories) after that the community can start to work on depicts using tools, that by then will hopefully become available. --C.Suthorn (talk) 22:31, 28 December 2020 (UTC)
Also if we would like to keep the category->property linkage in sync (which we would like to do) then it would need somekind of trigger system. When category will change it would update the category derived property values too. In anycase this would need the decision from the community that this would be ok and wanted. (i will support the proposal)--Zache (talk) 14:32, 29 December 2020 (UTC)
@Fæ: You say the "several gamification projects … proved that there are not enough interested casual volunteers for it to help systematically" but clearly we are getting "interested casual [but often not very competent] volunteers" who are either trying to help out with Depicts or messing around in a game-like way and providing metadata that isn't very useful. What I'm suggesting is that we use gamification techniques (which does not necessarily mean that the user sees it as, "Look, I'm playing a game!" although it can include that) to actually extract potentially useful information from this very noisy data. A good approach could also include knowing, over time, which IP addresses (or whatever) seem to be correlated with proposing tags/depicts that "stick," and giving tags coming from those addresses more weight.
Largely unrelated to the above and mainly addressed to Fæ: there's a lot I'm not thrilled with about the way the structured date (doesn't) related to the wikitext. Three years ago, I proposed what I still think is a perfectly good scheme for serialization/deserialization and while no one seems to have picked up on my ideas to implement them (and I have much too much I'd rather be doing to do that work, especially for free and with no cooperation), no one has really poked any holes in them either. I don't love the overall approach they've taken, but I think we should nonetheless try to make it succeed, and I really hope this part is negotiable, where if some of us can see how to make it give Commons a moderate amount of rather good data instead of, not to put too fine a point on it, a shit-ton of mostly bad date, we should work with them. - Jmabel ! talk 15:26, 29 December 2020 (UTC)
This makes no sense. Every single case study of this untested trial system that a couple of developers thought was a fun use of WMF funding has demonstrated that it has suffered from:
  1. Terrible user interface
  2. Breaks the upload Wizard experience, driving newbies to waste time filling in fields with no definition of what they are for, and they then have to duplicate in other fields
  3. Squanders volunteer good will and time
My own time is far, far, better spent creating verifiable public domain content that completely by-passes structured data nonsense (literally non-sense). Please reconsider how you invest your own time, this "debate" and the fanaticism around it that has operated in a vacuum of fact, testing, or demonstrable usage is a ghastly time sink. The WMF has this month been pushing idea that copyright statements in structured data has been a great investment, but as it would have been trivially easy to create a part of the API that used the wikidatabase to hold the same information, again it's nonsense to rewrite the facts to fit the myth that for some reason or other, structured data makes sense to be badly retrofitted to this project. --Fæ (talk) 15:36, 29 December 2020 (UTC)
  • @Fæ: I agree that having it in the upload sequence is ridiculous, but I believe SuggestedEdits has nothing to do with the upload sequence. Am I wrong there? I also have not invested a ton of time in structured data on Commons, as such. - Jmabel ! talk 02:38, 30 December 2020 (UTC)
@Fæ: We have API's (in example: https://magnus-toolserver.toolforge.org/commonsapi.php ), but it is not the point if we can create the API on top of the categories, but transform the actual data as more refined form so we can do things which arent possible currently. In example querying the data or making cross-database queries. --Zache (talk) 06:49, 30 December 2020 (UTC)
This is not the whole story, one can for example use the API to interrogate the list of all files with a specific copyright template on their page, or which may have an inherited category from that template. The database the API relies on can, of course, be queried much faster than through http requests. The biggest potential claimed benefits of structured data, none of which has been realized so far, can be done using other methods which would have no impact at all on newbies, or require Commons regulars to type in lots of extra data in extra fields/tabs/pages/spaces that behave entirely differently from "pre-structured data" user interface changes, which were never properly proposed or tested.
The trial of "suggested-edits" failed. How do we get it removed rather than allowing it to damage this project indefinitely and squander volunteer time correcting the firehose of negative-value changes? --Fæ (talk) 10:36, 30 December 2020 (UTC)
How to get it removed: Get a community decision via RfC, then create a phab ticket and if it doesn't help then block the edits from suggested-edits using abuse filter? --Zache (talk) 13:41, 30 December 2020 (UTC)
A "trial" that had no proposal and no RfC, now requires this level of bureaucracy to get it removed when it blatantly fails and damages Commons' mission?
That's kakistocracy. --Fæ (talk) 16:13, 31 December 2020 (UTC)
@Fæ: Is there a target for that last remark? Do you mean the Foundation in general? I guess that would be OK to say. But if it is aimed at some individual whom you are accusing of being "the worst" and "ruling us," that's at least pretty close to a personal attack.
It's clear to me you are angry about some of this (so am I), and it's clear that a lot of your anger is justified, but in my view at least several of your recent remarks have been really unconstructive, they come off as if you are aching more for a fight than for a solution. I like most of what I've seen of you here, you certainly do a lot of good work, and it's not as if I never flare up myself, I do. - Jmabel ! talk 18:20, 31 December 2020 (UTC)
propsing an abuse filter to remove a feature that the foundation obviously wants at all cost, seems unconstructive, it is like asking for the return of super protection. --C.Suthorn (talk) 18:48, 31 December 2020 (UTC)
There was couple steps before that (ie. RfC and phab) and if there would be no solution with those the situation would be already unconstructive. However, my point was more like that there is methods what community could do if they want to get rid of the feature all the way to the point where WMF needs to do something. Complaining doesn't afaik move things forward and I don't think that the suggested-edits is bad on purpose, but most likely the reason is that they don't have resources to change things on fly so we should least try to define the problems (in written form) so they can do something to them. Problem from common editor point of view is that this likely takes time. --Zache (talk) 19:03, 31 December 2020 (UTC)
Could someone sketch out, say, the necessary polite and constructive American English 30 words to request removal that can go in a Phabricator request? This discussion is already a consensus. Thanks --Fæ (talk) 19:23, 31 December 2020 (UTC)

@Fæ: I think the right way to go to phabricator is to be able to say something like (completely made-up data here) "Of the 198 edits made this way December 16-18, 94 were clearly nonsense or vandalism, 23 appear to be well-intentioned but wrong, 122 were correct but low-value, and only 42 were of any significant positive value. The work people are expending cleaning up the bad edits equals or exceeds what little value we are gaining from this," and then link the present discussion, which by then will probably be in an archive.

It would help if we had a representative sample that would show just how bad these edits are. Do we have that? If not, I also think someone has to put together something more rigorous than just our informal but near-unanimous impressions. I have a sneaking suspicion that if anyone has already done that, it's someone from who for some reason would like to keep this (mis-)feature, and they are probably not sharing the result of an objective study where those results were not to their taste. If I'm right that we don't have such a study available to us, then the next step here is probably to see if we can interest one of the now many people who study Wikipedia and related projects and enlist them into structuring an experiment and writing a paper. @Benjamin Mako Hill: I imagine this would not be exactly up your alley, but I bet you know who might be interested in doing a paper like that. Of course, it may turn out there already is one. - Jmabel ! talk 04:38, 1 January 2021 (UTC)

Suggested edits (Android) reverts (2.12.2020 - 1.1.2021)
Category Total edits Reverted edits % Quarry Examples
Depicts 4417 69 1.6 50943 50946
Captions 4991 184 3.7 50944 50947
Some analytics. Based on m:Wikimedia_Apps/Team/Android/AppEditorTasks tools vandal protection works like this:Access to the Suggested Edits feature will now be paused for a user for one week the first time their edit quality drops below a certain threshold. They will be given information on how to improve their edits so that they can productively contribute once the feature is re-enabled. The Suggested Edits feature will be permanently disabled for a user in case of consistently low edit quality. The user's account is left unchanged - only the Suggested Edits feature becomes disabled. If the user's account is banned by an admin on either the Commons or Wikidata projects, the SE feature will also be disabled for the user.
So based on the number of the reverts and revert rate the problem is not either huge OR the problematic edits arent reverted in commons. Later is a problem as it also means that the "low edit quality" detection wont work as it requires that community is tracking the edits and reverting the problematic ones. There is also somekind of bug as the system should have been permanently disabled the Suggested Edits feature for the user Cubaces4ever2 by now based on the edit/revert history. However, banning the user should be do the trick too. Also if we look the captions examples some of the problematic edits could be legitly captured by abuse filter without being unconstructive or controversial. However, doing so would require again some communication between community and developers (= phab ticket) as it would also require adaptations to the the user interface so it would be able to show nice error message. Also I think that it should also disable the feature if user is hitting to the abuse filter like the reverts will do. --Zache (talk) 12:52, 1 January 2021 (UTC)
So it sounds like in order to address the issue we'll need to take the time to go through them all and do the reverts. I've done some that I've seen, but not all, and I've only looked a couple times (when the error rate was between 35-55% among those in English). — Rhododendrites talk |  15:50, 1 January 2021 (UTC)

The table can be read as "about 5000 new captions per month via the Android app". That is about 60000 captions per year. With about 60000000 files that do need captions and about 10 languages for the captions (english, spanish, chinese, turkish, korean, arabic, russian, hindi, malaian, bengal) this means it will only take about 10000 years for the captions to be added (assuming no more files are uploaded in the meantime). In only 1000 years every file will have one depict statement. --C.Suthorn (talk) 18:34, 1 January 2021 (UTC)

It's ~150000 captions per month if we count all and not just the android app, but even that doesn't change the conlusion that we need all the automation what we can get to populate those. --Zache (talk) 18:47, 1 January 2021 (UTC)
What counts as a "revert" here? Only if the person who changes it is an admin, so has the "revert" tool and chooses to use it? If the caption is simply removed again but not with that tool, does that count as a "revert"? If, as so often, once a bad caption is there, I overwrite it with a good one, does that count as a "revert"? - Jmabel ! talk 01:07, 2 January 2021 (UTC)
Edits which are tagged with mw-reverted by the mediawiki. If the caption is overwrited with a good one then system doesn't count it as a revert. With wikicode there could be also edits flagged by mw-manual-revert but it seems that it is not used for structured data edits and removals are flagged using mw-reverted only.--Zache (talk) 06:03, 2 January 2021 (UTC)
As I thought. So the few percent that are "reverted" as such wildly underestimates the number of bad edits and, as I said above in the remark that mentions Mako, this calls for proper study. - Jmabel ! talk 16:16, 2 January 2021 (UTC)
It seems no. Even if we count any edits which are done to captions with same lang after the suggested edits the number is just 184 (mw-reverted) + 86 (any edits to same lang caption) = 270 (5.4 %) (Quarry: 50999). Even if this is missing something then I am still pretty sure that there isn't any double digit number of reverted caption edits (with any method) from Suggested edits feature. --Zache (talk)
And this was answer to the bolded sentence. The proper study would be interesting if somebody wants to do the leg work for checking the edits. --Zache (talk) 04:00, 3 January 2021 (UTC)
An example: An IP-user added a caption with this edit: https://commons.wikimedia.org/w/index.php?title=File:Venus_Berlin_2019_787.jpg&oldid=523150277 the caption reads "fetish medusa venus 2019". It is a good faith edit. But should it be reverted never the less? Should it not be "Fetish Medusa Venus 2019" or "Fetish Medusa Venus Berlin 2019" or better "Woman wearing latex fetish clothing at Venus Berlin 2019" and if the last, then there are 17 more files, that should get this caption (for consistency). And the caption is not really single language. It is the name of a website, the name of an event and a year. This could be flagged "all languages", only the captions do not offer such a flag. --C.Suthorn (talk) 07:41, 3 January 2021 (UTC)
Retrieved from "https://commons.wikimedia.org/w/index.php?title=Commons:Village_pump/Archive/2020/12&oldid=525256973"