Wikimedia Commons

Commons:Village pump/Archive/2019/09

< Commons:Village pump‎ | Archive

Category:Blood bikes

I created a category Category:Blood bikes to bring together a number of images in support of the corresponding wikipedia article en:Blood bike. The category I created is plural, the wikipedia article is singular. I suspect if I had called them the same then the wikidata would match, but I'm out of my depth with that stuff. Can someone who knows better please take a look and if appropriate sort out my mess? --2A00:23C6:FA02:EC00:4DF9:3ED3:6ECD:F339 20:34, 1 September 2019 (UTC)

Oxyman linked them for you on Wikidata. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 20:41, 1 September 2019 (UTC)

Stats of WLM 2019

Hi everyone! Where are can find the statistics of uploaded photos from Wiki Love Monuments 2019? I can only find here the stats from earlier editions... Thanks for your help and have a nice day :) Tournasol7 (talk) 12:23, 2 September 2019 (UTC)

@Tournasol7: I suggest you ask at Commons:Wiki Loves Monuments/Help desk.   — Jeff G. ツ please ping or talk to me 12:29, 2 September 2019 (UTC)
Thanks, Tournasol7 (talk) 13:31, 2 September 2019 (UTC)

Own photographs display

Do we have a tool which let me filter files I have uploaded and display it in big thumbnails? Juandev (talk) 12:16, 3 September 2019 (UTC)

Create campus categories for old Chinese unis

Many Chinese unis today are actually located on the sites of (kind of) unrelated unis. This usually happens because 1. some unis had been disbanded but the campuses were taken over by others, e.g. all christian unis. 2. 1952 communist shuffling zh:中國高等院校院系調整 3. some schools were split from unis and later upgraded to unis by themselves.

I was looking at the cases in Canton. Category:Lingnan University disappeared in 1952. Its campus was taken over by SYSU. SYSU's campus was at 石牌, but its Schools of Agriculture and of Technology were split. Now Category:South China Agricultural University and Category:South China University of Technology are two separate unis and share the 石牌 campus.

To straighten out the category structure, I come up with the following proposals:

  1. Create campus categories.
  2. Do not embed a separate university in another, even if one was split from the other. Instead, write a short description to explain.

The proposed structure would be

Unis
Uni A
X Campus
buildings, events, etc.
Uni B
X Campus

If necessary, X Campus could be split into

Geographical location
X Campus
X Campus of Uni A
buildings that exist in Uni A's era...
X Campus of Uni B
buildings that exist in Uni B's era...

This would be necessary if the latter uni has renovated the campus and built many new buildings for example, because the new buildings are technically not related to the former uni at all.

What do you think?--Roy17 (talk) 22:29, 3 September 2019 (UTC)

Merry-go-rounds

 
I'd sure call this a merry-go-round and it is certainly not a carousel. - Jmabel ! talk 21:20, 6 September 2019 (UTC)

Category:Merry-go-rounds is a redirect to Category:Carousels, so what do we call the playground equipment shown in the accompanying picture? - Jmabel ! talk 21:20, 6 September 2019 (UTC)

Well, as kids in the UK, we used to call those "roundabouts". -- Begoon 22:20, 6 September 2019 (UTC)
We have Category:Playground roundabouts, which is as good a name as any for an object that goes by different names around the world. "Merry-go-round" has more than one definition, and this is also a carousel. Category:Playground carousels and Category:Playground merry-go-rounds might be added as likely redirects. --Animalparty (talk) 00:07, 7 September 2019 (UTC)
Seems to me that Category:Merry-go-rounds should be a disambiguation, not a redirect. - Jmabel ! talk 15:47, 7 September 2019 (UTC)
I Confirm that the hand-pushed playground equipment like shown in the photo were called "Merry go rounds" in Louisiana. (In New Orleans, the mechanical carousels were often called "the flying horses", but that may just be a regionalism). Hm, are they they actually varieties of the same thing, an amusement ride spinning on an axis, differing only by size/elaborateness/motive force? [1], [2]? -- Infrogmation of New Orleans (talk)
Since we seem to have an established populated category at the UK English term, I suggest a redirect to there from common US English term. Category:Playground roundabouts already has a hat note for the US English term. It is already within Category:Human_powered_carousels, a subcategory of Category:Carousels, so I think that should cover it? -- Infrogmation of New Orleans (talk) 02:17, 8 September 2019 (UTC)
No, I don't think that covers it. Someone looking up Category:Merry-go-rounds is equally likely to mean either of these two things, so I think it should be a disambiguation. - Jmabel ! talk 04:53, 9 September 2019 (UTC)
OK. I suggest you propose this on Category talk:Merry-go-rounds, and if there are no serious objections, make it so. Cheers, -- Infrogmation of New Orleans (talk) 21:31, 9 September 2019 (UTC)
Done. Jmabel ! talk 00:41, 10 September 2019 (UTC)
This section was archived on a request by: Jmabel ! talk 00:41, 10 September 2019 (UTC)

File:Sahara.png

Should it be {{PD-self}} or {{PD-NASA}}? It is derived from something created by NASA. Nigos (t@lk • contributions • Uploads) 12:34, 1 September 2019 (UTC)

Pinging @Emilfarb as uploader.   — Jeff G. ツ please ping or talk to me 12:59, 1 September 2019 (UTC)
It should be {{PD-NASA}}, but there is no source. Regards, Yann (talk) 17:31, 4 September 2019 (UTC)

Link to Commons at a Wikipedia page

The Wikipedia page Het Hogeland does not have a link to Commons. The link to wikidata gives "Het Hogeland (Q27959423)". However there exists a category Hogeland with the link to Wikidata "Categorie:Het Hogeland (Q60310625)". How do I get the right link to Commons on the Wikipedia page? Wouter (talk) 18:49, 3 September 2019 (UTC)

@Wouterhagens: {{Commonscat|Het Hogeland}} should work. Rodhullandemu (talk) 19:00, 3 September 2019 (UTC)
On enwp, just use {{Commonscat}} with no locally-defined text to fetch the value using the Commons sitelink on Wikidata (the inverse of the link that lets the Wikidata Infobox work here). Hopefully other wikis can adopt that same system at some point. Thanks. Mike Peel (talk) 19:24, 3 September 2019 (UTC)
@Wouterhagens: Duplicate Wikidata items on the same object can be merged. This may be such a case. Ask at the community portal of wikidata. --C.Suthorn (talk) 20:00, 3 September 2019 (UTC)
There's no duplicate here - Category:Het Hogeland (Q60310625) is the category item, Het Hogeland (Q27959423) is the topic item, and they link to each other via category's main topic (P301) and topic's main category (P910). Both Template:Wikidata Infobox and en:Template:Commonscat follow those links to display the correct info in the category / category link in the article respectively, versions on other wikis might not do that yet. Thanks. Mike Peel (talk) 04:56, 4 September 2019 (UTC)

Category:3 ft 6 in gauge trams

Japanese use metrics for example (Google 軌間 1067). The parent cat is being discussed Commons:Categories_for_discussion/2017/10/Category:Track_gauge_by_size. Absolutely no reason to move to Category:3 ft 6 in gauge trams without a consensus against common usage. All similar moves about 1067mm tracks by User:Andy Dingley should be reversed.--Roy17 (talk) 16:55, 4 September 2019 (UTC)

Per the long discussions years ago. The canon measurement was in Imperial. That's why it's a "round number" in Imperial, obscure and arbitrary in metric. It arose because Japanese railways were built (or at least designed) by foreign engineers, and these were largely Scottish and using Imperial. Also Japan didn't adopt metric until after the gauges were established.
And apart from repeatedly trolling me again AN/U, why raise this here? Andy Dingley (talk) 17:13, 4 September 2019 (UTC)

Splitting བོད་ (Tibet) into Tibet (region) and Tibet Autonomous Region (for region of the PRC)

@Gryffindor: Re: བོད་ I had moved it before to 西藏 བོད་ as it seems to be used to mean the Tibet Autonomous Region (specifically the PRC region) but it has been moved back to "བོད".

I'm OK with keeping བོད་ where it is and keeping the title only in Tibetan (as it represents a particular region in multiple countries), but making a second gallery specifically about the Tibet AR which would be titled in both Mandarin and Tibetan, with Mandarin being first (as it's a region of the PRC). WhisperToMe (talk) 16:57, 4 September 2019 (UTC)

FM X-XX-XX on the back of photos

Notes on File:Gen. Lei Fuk Lam and Dr. Sidney K. Wei.jpg end with FM 6-11-25. Does anyone know what FM stands for? I presume 6-11-25 is 11 June 1925.--Roy17 (talk) 16:33, 3 September 2019 (UTC)

Most likely the initials of the editor that wrote the caption for the image at P & A Photo. RAN 16:01, 5 September 2019 (UTC)

Ponzi

I stumbled on this image when looking at the Charles Ponzi article on English Wikipedia. It does not look right; it looks like the head was photoshopped onto the rest of the image. I'm not sure what should be done about this, and would appreciate learning from the expertise of the contributors here. Risker (talk) 06:28, 5 September 2019 (UTC)

It does look odd, though it hasn't been modified from the source. Pi.1415926535 (talk) 06:47, 5 September 2019 (UTC)
I also agree it looks odd, but his head also looks slightly oversized in this photo and a little odd in this photo too. So it's probably just a combination of a strange camera angle and mild macrocephaly. =) - Themightyquill (talk) 08:14, 5 September 2019 (UTC)
  • It's indoors in the 1920s. So lighting is limited and photographic emulsions are slow. These combine to give a restricted depth of focus. I don't think the head is larger, more that it's in a clearly sharper focus and so appears more prominent. It's not an uncommon effect in many photos of that era. Andy Dingley (talk) 13:58, 5 September 2019 (UTC)
I uploaded the image, I agree, the aperture of the lens is wide open to let in more light, so the depth of field is narrow, so his body is in focus and the background is out of focus. There are many copies of the photo printed in papers, this was the highest resolution scan of the original I could find. There as a better copy on eBay a few years ago. RAN 15:31, 5 September 2019 (UTC)

Heap of stones

 

This is not a landmark. People just enjoy collecting lose stones and making a heap of them. The same as they enjoy building dams in small streams and building sandcastles. Can the classification of Cairns be broadened? Or is this trivial?Smiley.toerist (talk) 10:25, 6 September 2019 (UTC)

I'd say it is a cairn, but what is its educational purpose? I think we have better images of "proper" cairns and could live without this. Rodhullandemu (talk) 10:41, 6 September 2019 (UTC)
No doubt we could "live without" most photos on Commons. In addition to landmarks, more mundane subjects are also covered in Commons. This photo seems unobjectionable. -- Infrogmation of New Orleans (talk) 12:56, 6 September 2019 (UTC)
My penny. I think putting it in Sea cairns is pushing it as it's not a landmark or ancient en:daymark, or even Balanced Rocks which are to my mind natural phenomena. However it's certainly OK in Piles of Rocks. Broichmore (talk) 14:19, 6 September 2019 (UTC)

Hong Kong: Lennon Wall Flag

Would the upload of Hong Kong's Lennon Wall Flag be accepted as an SVG here? Syced (talk) 07:12, 10 September 2019 (UTC)

I contacted the artist, and he has put it into public domain: https://www.instagram.com/p/B2N_O-5AsOW/ (see where it says "license ： public domain"). Syced (talk) 08:59, 10 September 2019 (UTC)
license: public domain is vague by Commons standard. Personally I think it's OK and you can upload it and tag it as pd-self, but some other people might disagree.--Roy17 (talk) 22:32, 10 September 2019 (UTC)
  This section is resolved and can be archived. If you disagree, replace this template with your comment. --Roy17 (talk) 22:12, 12 September 2019 (UTC)

Wiki Loves Science

Will there be a competition this year like in 2017? Basile Morin (talk) 12:32, 8 September 2019 (UTC)

According to this website, it will take place in late 2019, but I don't know any details. De728631 (talk) 14:35, 8 September 2019 (UTC)

Tool for Changing File Category in Batch

Hello all,

I noticed that there are many stereo cards under Category:Boston Navy Yard. I plan to consolidate them under Category:Stereo cards of Boston Navy Yard. Do we have any tool to automate this process, i.e. If a file is under certain categories, then remove those categories and add a new category? -Mys_721tx (talk) 13:15, 8 September 2019 (UTC)

Hi there. There's in fact a handy tool called Cat-a-lot. You can activate it in Preferences - Gadgets: Tools for categories. See the manual here: Help:Gadget-Cat-a-lot. De728631 (talk) 14:10, 8 September 2019 (UTC)
Thanks! All moved to the new category. -Mys_721tx (talk) 14:42, 8 September 2019 (UTC)

Brazil flag

Which one is the official version of the Brazil flag? A user uploaded "A" and claimed that it is the official version as defined by Law No. 5,700 of September 1, 1971. Can someone check? 大诺史 (Talk/留言/토론/Discussion) 17:28, 8 September 2019 (UTC)

@大诺史: "A" does not appear to be accurately constructed under law 5700, or under the most recent iteration of the law, law 8421 of 11 May 1992. The distance from the vertices of the yellow parallelogram to the edge of the flag are not the same and not in the correct proportion on either the horizontal or the vertical axis. I didn't check any further than that. CC: @JpBrazil1: – BMacZero (🗩) 21:56, 8 September 2019 (UTC)
Also, the stars on "A" are horribly misplaced when compared to the template in law 8421. – BMacZero (🗩) 21:58, 8 September 2019 (UTC)
@BMacZero: I believe "B" is more commonly used? 大诺史 (Talk/留言/토론/Discussion) 10:40, 9 September 2019 (UTC)

Photos of stenocara gracilipes

Hi, we have several photos of Stenocara gracilipes, but it would be great to source at least one showing the creature colleting water from condensing fog, as seen here. What are the channels to source such an image? Grateful for any help. Rui Gabriel Correia (talk) 11:24, 9 September 2019 (UTC)

Statistics of deleted files

Hi, Are there statistics of deleted files vs. uploaded files available over the last 10 years? I would like to know if the ratio is stable, has increased, or decreased. Regards, Yann (talk) 17:23, 4 September 2019 (UTC)

I've looked for it and was unable to find it, I wonder if it's possible for us to try to ask the Wikimedia Foundation (WMF) to keep more statistics of Wikimedia Commons, many other websites do it and such statistics are very handy for people interested in the project to look at. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 14:50, 5 September 2019 (UTC)
You might want to take a look at Wikistats. While I haven't found a dedicated "uploads vs deletions" table, there are some interesting statistics including the total number of "articles". De728631 (talk) 20:49, 5 September 2019 (UTC)
This data should be in the WikiReplicas. So it can be accessed by quarry- https://quarry.wmflabs.org/query/38827 Bawolff (talk) 16:23, 6 September 2019 (UTC)
@Bawolff: Thanks a lot! So I made a Wiki table: Commons:Statistics of uploads vs deletions. Regards, Yann (talk) 14:51, 8 September 2019 (UTC)
That's useful. Thanks for creating it. --Túrelio (talk) 15:43, 8 September 2019 (UTC)

Wow, that page looks great, can a similar page be created for number of active users by year (and possibly by month), and maybe even for users by user groups, or files by file types. In fact we still can't see in which months or on which days (of the week) most deletions or uploads are done, also statistics on how many active users Wikimedia Commons has at any given time plus how many average edits they perform. Just some ideas. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 20:23, 10 September 2019 (UTC)

User:Fæ/Userlist

Hi at all. User:Fæ/Userlist doesn't work since march 18th. Can anyone fix the problem? Thank you very much. Best regards. --DenghiùComm (talk) 08:30, 9 September 2019 (UTC)

@DenghiùComm: The page works, but Fæ's bot hasn't been updating it due to the issues presented at m:Hardware donation program/Fæ.   — Jeff G. ツ please ping or talk to me 12:13, 9 September 2019 (UTC)
What's wrong with Fae just using toolforge for this? In fact, how is that list even being generated unless being done via toolforge db replicas (I mean yes, there are other ways, but they're much much more complicated) Bawolff (talk) 22:35, 10 September 2019 (UTC)
The number of people contributing to wikipedia/commons is very limited. The number of people writing software voluntarily even more so. What Fæ is (not) doing, could be done by anyone else with toolforge and more efficiently. Only: Noone else does. For the time being it might be wise, to let Fæ do this work in the fashion that suits Fæ best (I take it as given that Fæ knows about toolforge and made an informed decision against using it.). --C.Suthorn (talk) 16:29, 11 September 2019 (UTC)

Modifying pibot

Can we modify pibot to not add in wikidata infoboxes for categories like Category:Photographs by Carl Brandt, the Category:Carl_Brandt already had the correct infobox and the "Photographs_by" only needs the Creator infobox. Once tagged with a biographical infobox people add in other biographical categories which confuses Google into thinking the pictures taken BY the photographer are pictures OF the photographer. "Photographs_by" was supposed to help Google Image figure out these are not images of the photographer. Thanks! RAN 21:26, 11 September 2019 (UTC)

P.S. Anyone know why my ~~~~ renders into a signature and date stamp minus the link? — Preceding unsigned comment added by Richard Arthur Norton (1958- ) (talk • contribs) 21:29, 11 September 2019‎ (UTC)
@Richard Arthur Norton (1958- ): It appears that you kept the "Treat signature as wikitext (without an automatic link)" checkbox checked on Special:Preferences#mw-prefsection-personal, but removed any links from the "Signature" field there. This edit worked just fine.   — Jeff G. ツ please ping or talk to me 23:54, 11 September 2019 (UTC)
@Richard Arthur Norton (1958- ): You probably want to create Category:Carl Brandt (which has an image in it but doesn't exist) and update the Commons sitelink at Carl Vilhelm August Brandt (Q63437107) to point to that. Thanks. Mike Peel (talk) 04:40, 12 September 2019 (UTC)
Yes! An elegant solution, add in the supra-category and link to that only. RAN (talk) 23:44, 12 September 2019 (UTC)

First images from the UNESCO archives now on Commons

Dear all

Category:Media files produced by UNESCO: 2019-09

I’m very pleased to say that the first batch of 100 very high resolution photos (plus descriptions) from the UNESCO archives has been uploaded to Commons by our friends at Wikimedia Sweden. There are some amazing photos in this first batch including Italian National Archives material being washed and dried in a train station after a flood, and the installation of a Henry Moore sculpture. We would really appreciate it if you could take 5 minutes to add some to Wikipedia articles, Wikidata items etc so that UNESCO will be encouraged to make more images available.

Category:Media files produced by UNESCO: 2019-09

Thanks very much

John Cummings (talk) 18:57, 12 September 2019 (UTC)

That's awesome. Thanks for helping to make this happen John. GMGtalk 19:23, 12 September 2019 (UTC)
Thanks @GreenMeansGo: , more to come :) Please do add them to articles. John Cummings (talk) 19:27, 12 September 2019 (UTC)
Great, I added three images to three articles (example). 4nn1l2 (talk) 06:46, 13 September 2019 (UTC)
Thanks very much @4nn1l2: , really nice to have them used in many languages :) John Cummings (talk) 11:40, 13 September 2019 (UTC)
@John Cummings: Wonderful news! I've added three of the above images (I couldn't find a home for the TV studio pic) to articles across several projects. Some feedback, which I trust is useful: a) I found multiple instances French-language text marked up as English-language descriptions. b) "Category:Monochrome photographs" could have been applied during upload. c) Categories like "Category:1970 photographs" had been applied, whereas the metadata included month-specific dates, so say, "Category:November 1970 photographs" could be used. d) They're very large tiffs (which is great); perhaps it would be worth running a bot to add medium-size PNG or JPG equivalents? e) Please include structured data! Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:15, 13 September 2019 (UTC)
Thanks @Pigsonthewing: , very glad for the feedback on metadata structure, that's why we did a small batch first :) John Cummings (talk) 11:40, 13 September 2019 (UTC)
Also thanks to User:Alicia Fagerving (WMSE) for moving the reverse sides to their own category, which makes sorting through them a great deal easier. GMGtalk 13:45, 13 September 2019 (UTC)

Fatherland front flag (Austria)

With regards to the colours used in the flag and according to the source given for en.wiki upload. The civil flag was used from 1919-1938. Fatherland Front was founded in 1933 and later banned in 1938. So the colours will most probably follow the civil flag that was used in the time and not the current the Austrian flag. 大诺史 (Talk/留言/토론/Discussion) 07:51, 13 September 2019 (UTC)

Which files is this related to? I see a few different ones in Category:Fatherland Front (Austria).--Roy17 (talk) 22:41, 13 September 2019 (UTC)
@Roy17: Forgot to link the file, File:Flag of the Fatherland Front of Austria.svg. 大诺史 (Talk/留言/토론/Discussion) 10:24, 14 September 2019 (UTC)

Oversight email issue

As suggested on Commons:Oversight I sent an email to oversight-commons@lists.wikimedia.org - I have received an automated response saying:

Your mail to 'Oversight-commons' with the subject

[redacted]

Is being held until the list moderator can review it for approval.

The reason it is being held:

Post by non-member to a members-only list

Either the message will get posted to the list, or you will receive notification of the moderator's decision.

This is not an appropriate response, especially as we advertise that service to members of the public. What can we do to make it better? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:41, 14 September 2019 (UTC)

@Andy Mabbett: Would you prefer an automated response from OTRS or no response at all?   — Jeff G. ツ please ping or talk to me 15:43, 14 September 2019 (UTC)
OTRS would seem a better approach than this. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:20, 14 September 2019 (UTC)

Turn 1924 into a calculation

The text reads "Associated Press articles before 1924 are in the public domain." in the category for the articles. How can I change "1924" into a calculation so it increments by a year on January 1 of each new year? RAN (talk) 16:00, 17 September 2019 (UTC)

In this case, you would write {{#expr:{{CURRENTYEAR}} - 95}} which currently comes out as 1929. --LA2 (talk) 18:33, 17 September 2019 (UTC)
Just what I needed ... Thanks!
  This section is resolved and can be archived. If you disagree, replace this template with your comment. --Roy17 (talk) 23:22, 21 September 2019 (UTC)

How do we stop Corruption in Africa?

We are used to the sufferings and inadequate attention being paid to our daily existence. The climes of Mama Africa is fast becoming the den for thieves and unrepentant political con men. There is no doubt that if things keeps falling out of order, we will wake up one day to the news that War and bigots' fallacies have been given the required space to bring the world to a standstill.

A lot of foreigners are aware that thousands do try to cross into Europe through unsolicited means. Yet this is negligible to the pain the concerned citizens are subjected to at home. The world needs to see to the underground war in Africa, instead of spending the millions building temporary and permanent migrants' deportation camps.

The relevant world leaders need to talk to the present presidents of African countries and give them the magic initiatives that will end the ignited madness that is spreading to the world. Without doubt, there is no darkness but ignorance. Corruption is now rampant in Africa that nobody sees it as a scourge and backwardness to the common good of the ever increasing population of the poor. No matter how inconvenient it is, we must stand in solidarity and speak the truth. That is the only way we can say the missing value chains can be discovered and put to work.

It is clear the developed world has its own inequalities as well, but Africa is worse. New governments get into power: They resolve to increase tax, lower the standard of living, get foreign loans at the expense of the local stock market, steal more funds and take it abroad to increase political tension. This encourage nepotism, ruling-party hypocrisy and payment of millions of dollars to ensure that no party wins at their expense.

Sadly this trend has been with us since time immemorial, and it will never go away. As long as the level of poverty is excruciating and condemnable. The conscience of rights and wrongs has been degraded to the point that as long as money is rolling in, you will have a lot of fans on social media. May be I was born yesterday because I can't say I know how we got here, but we need higher help from the clouds somewhere. Africa used to be a continent where moral views reign supreme, but now our psyche has been eaten into by a capitalistic virus.

There is an underground war in Africa, many are on the way to Europe, America and Asia. Illegal or otherwise, when you see them be very assured that they are refugees of tribal sentiments, and as such they desrve some peace of mind, good job offers and glorious beginnings. Help them and cater to the needs. But above all this, have it in mind that more are on the way. So I am using this humble medium to call on all African leaders looting their countries' treasuries to have a rethink in order to make things better and give hope a chance in their homeland. Africans are not daft as being painted by foreign media outlets. Let us fight ignorance together and create awareness to stop the unstoppable multi-headed monster called corruption. — Preceding unsigned comment added by PoetryVacancy (talk • contribs) 13:13, 18 September 2019 (UTC)

Hey PoetryVacancy. This is a space for discussing and coordinating contributions to Wikimedia Commons, a repository of free media for use by our sister projects and the general public. While political corruption may be a serious problem that many are working to solve, I'm afraid it does not pertain to the mission of this project, and you will need to find somewhere else to discuss strategies for preventing it. GMGtalk 13:28, 18 September 2019 (UTC)
  This section is resolved and can be archived. If you disagree, replace this template with your comment. --Roy17 (talk) 23:22, 21 September 2019 (UTC)

Notification of DMCA takedown demand - Artistic view of how the world feels like with schizophrenia

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

Affected file(s):

To discuss this DMCA takedown, please go to COM:DMCA#Artistic view of how the world feels like with schizophrenia. Thank you! Joe Sutherland (WMF) (talk) 19:52, 16 September 2019 (UTC)

Structured Data - computer-aided tagging tool

Note: this is a copy and paste of a message I put on the SDC talk page and sent out to the SDC focus group. It will also be going to the Commons mailing list.

The development team is starting work on one of the last planned features for SDC v1.0, a lightweight tool to suggest depicts tags for images. I've published a project page for it, please have a look. The tool has been carefully designed to try to not increase any workload on Commons volunteers; for starters, it will be opt-in for auto-confirmed users only and will not generate any sort of backlog here on Commons. Additionally, the tool is highly privacy-minded for the contributors and publicly-minded for the third party being used, in this case Google. The implementation and usage notes contain more information about these and other potential concerns as a starting place. It's really important that the tool is implemented properly from the start, so feedback is welcome. Questions, comments, concerns are welcome on the talk page and I will get answers as quickly as possible as things come up. The first desktop designs will be ready in a few weeks, the team would like to hear from you first. Keegan (WMF) (talk) 21:04, 17 September 2019 (UTC)

Bad signature

Why is my signature not correct here? It only says undefined --193.171.152.104 06:39, 17 September 2019 (UTC)

Your IP Address is not a registered user. Please see COM:QIC#Evaluating images, and login if you want to review QICs.   — Jeff G. ツ please ping or talk to me 16:07, 17 September 2019 (UTC)
Your link didn't work for me, Jeff, so I took the liberty of attempting to fix it. – BMacZero (🗩) 15:30, 18 September 2019 (UTC)

Renewal notice and image match as an example

Can someone provide a link for a renewal notice for an individual photograph that I can add to some boilerplate text. Generally we certify that a renewal notice is absent, but I would like to include a renewal example to show what it looks like. We also need the photograph hosted somewhere online, maybe one hosted by Corbis or Getty. I know it is hard to match a prose description of a photograph to an actual photo, especially when a dozen photographers were present from a dozen news organizations. A Good example if you can match them would be the one discussed by the Library of Congress: "It was discovered that copyrights were placed and renewed on the photographs of the Dionne Quintuplets taken by the [Newspaper Enterprise Association]." Renewal notices are here: http://onlinebooks.library.upenn.edu/cce/ RAN (talk) 18:39, 17 September 2019 (UTC)

I found a copy of the image. Now we just need the renewal notice. RAN (talk) 12:51, 18 September 2019 (UTC)

Sheriffs (American county level ones)

To my surprise, Commons doesnt have a category of sheriffs yet? Should/Can there be one? I have the impression that sheriffs are quite important figures in local politics, innit?--Roy17 (talk) 22:32, 10 September 2019 (UTC)

I saw Category:Sheriff elections in the United States by chance today. Are there American sheriffs that are not county sheriffs?--Roy17 (talk) 22:12, 12 September 2019 (UTC)
Louisiana "parishes", Alaska "boroughs", though both are really counties by other names.
I believe (though I'm not sure) that in the 19th century the title was sometimes used at a more local level, especially before counties were properly established in the Western territories. - Jmabel ! talk 04:09, 13 September 2019 (UTC)
@Roy17 and Jmabel: Seems like our main category tree for this is under Category:Sheriff's departments of the United States. The sheriffs (individual persons) themselves, if they have individual categories, are put into the local XY-county sheriff's office or XY-county sheriff's department categories. Example: Category:Dave Reichert ← Category:King County Sheriff's Office ← Category:Sheriff's departments of Washington (state) ← Category:Sheriff's departments of the United States by state ← Category:Sheriff's departments of the United States. --El Grafo (talk) 08:17, 19 September 2019 (UTC)
We should probably single out a category (or category tree) for the sheriffs themselves, though. This is as if we had categories for city governments but not for mayors. - Jmabel ! talk 15:22, 19 September 2019 (UTC)

Ilkinm files deletion.

There was photos of my grandfather-hero of 2nd WW. I put them to his page Yaqub Qasımov, but you deleted all of them. Is it vandalism, or German nazism? I dont understan. I demand to repair my files, listed below. He defended your future against nazi Germany. Have respect for your hero.

İlkinm (talk) 18:22, 17 September 2019 (UTC)

Obviously you marked the files as own work and created today. As your Grandfather is probably dead, this is likely wrong. You have to provide the true information on the photographer or prove that the photographer died more than 70 years ago (the time may vary with the place where the photo was took). It would also be useful to give a description in english on the file description page, add catagories and add the real date of the photo. --C.Suthorn (talk) 20:17, 17 September 2019 (UTC)

See: Commons:Deletion requests/Files uploaded by İlkinm. --Túrelio (talk) 13:44, 19 September 2019 (UTC)

Category:School strikes for climate in Atlanta

Category:School strikes for climate in Atlanta was created today with 15 files imported from flickr. The 15 files are identical to the 15 files in Category:Global Climate Strike in Atlanta, 17 March 2019. Does flickr-import not make duplicate checks? --C.Suthorn (talk) 16:22, 18 September 2019 (UTC)

Template:Mbox changed to float left?

I just noticed that {{Mbox}} and the bajillion warning and info message templates which transclude it, had its box shifted to the left (or to the right, in rtl scripts). However, its history seems unchanged. What’s going on? -- Tuválkin ✉ ✇ 21:42, 18 September 2019 (UTC)

Structured data captions

I noticed some of this user's edits in my watchlist. Are these additions acceptable? Most of the added captions incorrectly use underscores instead of spaces, there are multiple spelling errors, and most of them seem to be direct transcriptions of the filenames. Jc86035 (talk) 23:57, 19 September 2019 (UTC)

No, these are not correct. I left the user a message and undid the edits. Commons:File captions could use a bit of attention. Multichill (talk) 17:53, 20 September 2019 (UTC)

File talk:Schmittsohnlogo.gif

Can we remove the April Fools’ Day joke from this talk page? --92.211.147.115 19:29, 20 September 2019 (UTC)

Question about the deletion of my photograph of a historical marker

Here is the deletion request discussion: Commons:Deletion_requests/File:Alta_Loma_Historical_Marker.jpg.

The best I can understand, this photograph of a marker was deleted because it contains text. And text of a sentence or more on a photographed object was assumed to be copywrited and constitutes a derivative work. There are more than thousands of other historical marker pictures on Commons. Nearly all of these are the same as mine in regard to the reason for the deletion. Historical markers are erected to mark the location of something of historical consequence, so almost all bear text describing the person, place, event, structure, etc. commemorated. If mine violates the law, the others are also violations. Surely the administration is not going to single mine out while allowing the thousands of other violations to remain.

I was able to find each of the following example marker pictures in 3 seconds or less. They were the first four Texas markers I came to. That is, they were not *picked out* to support my belief.

Some places require an explicit question so: Are the thousands of equivalent pictures going to be deleted? Wiki name (talk) 16:59, 6 September 2019 (UTC)

Hi,
Each case has to be evaluated, but potentially yes. Old boards without a copyright notice would be OK. Recent ones are not. You can nominate them for deletion, specially if you have information about them. Regards, Yann (talk) 17:16, 6 September 2019 (UTC)
As I think this decision is absurd I'm not likely to abrade others. I guess I was asking if commons cares enough about the large number of copyright violations to take an initiative. I hear you saying "No, we don't care about the others, just yours."
I don't understand Old and New. How are these defined? Seems like all markers with text would fall under this ruling -- unless maybe they were erected in the 1920s. Wiki name (talk) 18:42, 6 September 2019 (UTC)
@Wiki name: In this case, "old enough" is 1977 or earlier. Anything published (or erected) before that time is not copyrighted unless it includes a properly-formatted copyright notice on the work itself. For more detailed reference, see Commons:Hirtle chart. Unfortunately Commons can be pretty spotty at noticing non-obvious copyright violations like these, so our enforcement of them is probably really inconsistent. I'm sure there are other pictures like yours that we should delete, but nobody with the right knowledge has noticed them yet. – BMacZero (🗩) 19:00, 6 September 2019 (UTC)
I really appreciate your answering my questions. Yes, there are definitely others, like 3 of my 4 examples.
The markers I'm primarily interested in are Texas markers so I can avoid uploading ones Commons thinks are illegal. I only uploaded that one because the Alta Loma Wikipedia page had no content.
Texas markers are not intended to be copyrighted so as I read the Hirtle chart this provision would apply. (I don't know how to format this)"1978 through 28 February 1989 Published without notice, and without subsequent registration within 5 years None. In the public domain due to failure to comply with required formalities {{PD-US-1978-89}}" Wiki name (talk) 02:07, 7 September 2019 (UTC)
"Texas markers are not intended to be copyrighted" - If you can find such a statement in a publication or official webpage of the agency that put up the markers, that could be very helpful. If that's just your guess, sorry - for new works at least since 1989, US Federal law has made copyright automatic. -- Infrogmation of New Orleans (talk) 18:18, 7 September 2019 (UTC)
As far as making an initiative to delete these, remember that Commons isn't really an organized effort - it's just a loose collection of volunteers. We should delete any other files like yours, but for that to actually happen requires someone with the time and inclination to trawl through all those and check the dates the signs were erected. Personally, I'll admit it's not a task I'd be excited to take up. – BMacZero (🗩) 19:09, 6 September 2019 (UTC)
That definition of "old enough" is specific to the United States. - Jmabel ! talk 21:18, 6 September 2019 (UTC)
I'm only interested in the U.S., primarily Texas Wiki name (talk) 02:07, 7 September 2019 (UTC)
As mentioned, 1977 seems the last date it is safe to just assume such are ok. Up to 1989 seem likely, to be ok, but some research would be needed to confirm that the State of Texas either did or didn't file registration for such markers; if not {{PD-US-1978-89}} would apply. (Checking Texas State Historical Committee / Texas History Commission webpages may produce some useful info.) Post 1989, US Federal law makes everything (that is not de minimus or too simple to copyright, eg short simple text stating fact like "This house was built in 1992") copyrighted automatically. For those more markers/plaques that are recent than 1989 we'd need to find some explicit statement from the State of Texas, or whatever agency put up the marker, that they are releasing the text to public domain or free license. Yes, US copyright law can be convoluted, but to have legally free licensed media Wikimedia Commons needs to comply with it as best we can. -- Infrogmation of New Orleans (talk) 18:12, 7 September 2019 (UTC)
Here we go wrapping around this axial again. The person who started all this, Sphilbrick, wrote to the State of Texas office responsible for Historical Markers asking about copyright. He received a reply from the office and the person responsible for Historical Markers in Austin, Texas saying "We have no problems with folks sharing photos of our marker. I am not aware of any copyright or other limitation . . ." Surely the primary person responsible for Historical Markers for the State would know if their office had filed or been notified of any copyrights. Wiki name (talk) 00:33, 8 September 2019 (UTC)
Also, see my reply to RAN below Wiki name (talk) 00:39, 8 September 2019 (UTC)
  • You can try and contact the entity that erects the signs and ask them to release the text under a creative commons with attribution license. Or you can ask them when the sign was erected, or if the text is older than 1989. Some jurisdictions add a date to the sign of when it was posted. RAN 18:52, 7 September 2019 (UTC)
This saga started in late July. When Sphilbrick with no warning deleted my picture. I contacted the Texas State Historical commission to ask if they had copyright restrictions on markers. In a reply they confirmed the following statements:
<begin quote>The State of Texas claims no copyright on Texas Historical Markers or pictures of them taken by individuals nor the publication of these pictures.
In addition there is no copyright on the text contained on the markers.
Physical access to a small number of markers in order to photograph them is restricted, though the photos and their text are not restricted.<end quote>
I sent him a copy but he seems to have trouble finding his email. So he wrote the same people and got the reply I quoted above to Infrogmation's last. Wiki name (talk) 00:33, 8 September 2019 (UTC)
@Wiki name: Please forward the e-mail to Commons:OTRS WhisperToMe (talk) 18:19, 21 September 2019 (UTC)

As for Texas, basically the full text of all the state historical markers in the whole state has been collected into the book Why Stop?: A Guide to Texas Historical Roadside Markers which has been published in several editions. That book would be far more infringing than any Wikimedia Commons photograph, if there were anything to infringe... AnonMoos (talk) 00:50, 8 September 2019 (UTC)

When you say 'unsucessful attempt' above, if you mean Sphilbrick's comment. He received a prompt reply, but as I mentioned, he has trouble finding his email. He was unable to find either of the two emails I sent him. He eventually found the one from Texas which I quoted from above. — Preceding unsigned comment added by Wiki name (talk • contribs) 14:14, 8 September 2019 (UTC)
Email sent Wiki name (talk) 17:33, 8 September 2019 (UTC)
I'm wondering the same thing, but Jeff G. says this on his page "I do not accept any messages sent via PM, IM, or Email." Earlier I found some sort of internal way to send him a message but I can't find it anymore. i_am_jim (talk) 01:28, 13 September 2019 (UTC)
@i_am_jim: Sorry, I have added a disclaimer to that (thanks for reading so deeply). You are specifically authorized to use Special:EmailUser/Jeff G., whether or not the "Email this user" tool shows. And @Infrogmation of New Orleans no, I still haven't heard from the Texas Historical Commission.   — Jeff G. ツ please ping or talk to me 13:35, 13 September 2019 (UTC)

The Speaker Explains

https://www.youtube.com/playlist?list=PLj3mInRJqIenu11GhIbyxFmif0SqYJVTO is interesting. Before I import all of them, I'd like to ask,

Are these five videos recordings of parliamentary proceedings?--Roy17 (talk) 22:41, 13 September 2019 (UTC)

However, User:Taivo and User:Ankry think they are. Taivo deleted File:The Speaker Explains - Bobbing in the House of Commons Chamber.webm and File:The Speaker Explains - Keeping Order in the House of Commons.webm. Ankry closed the undeletion request twice: Commons:Undeletion_requests/Archive/2019-09#File:The_Speaker_Explains_-_Keeping_Order_in_the_House_of_Commons.webm.--Roy17 (talk) 23:22, 21 September 2019 (UTC)

Fixing incorrect files names

Hi! Doing maintenance works on the frwiki I’m sometime facing issues with some Commons files name not respecting the “Manual of Styles”.

I first had a chat with Zhuyifei1999 on this topic and he redirected me here.

The problem is that files names shouldn’t have:

  • a space after the “(“ - intitle:/"(" +/ - 1,342
  • a space before the “)” - intitle:/ +")"/ - 1,239
  • a space before the “.” - intitle:/" ."/ - 10,835 (mostly redirects)
  • a space before the “,” - intitle:/" ,"/ - 7,046
  • a “,” followed by a “.” - intitle:/"," *"."/ - 694

A bot has been requested in 2014 but only for the removals of spaces before file extensions. And it was a one-time run.

I think it is necessary to have a permanent bot for the issues I’m talking about and perhaps others you could add.

Do you think I can send this request to bots?

Regards. --FDo64 (talk) 20:16, 16 September 2019 (UTC)

No, I don't think so. Filenames are supposed to be stable and not changed just because it looks better. See also Commons:File renaming. Multichill (talk) 20:36, 16 September 2019 (UTC)
Thanks. I already read this page but it is not because it looks better. It's because it is wrong. --FDo64 (talk) 21:35, 16 September 2019 (UTC)
Many Commons files are imported from external sources, eg. various archives, Flickr etc. Such files can't be expected to follow style guides, and I don't think it is necessary to change them. MKFI (talk) 13:16, 17 September 2019 (UTC)
@FDo64: Can you give some examples of files that you think are wrongly-named? Commons has (deliberately) quite low standards for file names, and we wouldn't usually bother renaming a file just to remove a space. --bjh21 (talk) 13:33, 17 September 2019 (UTC)
Hi! As requested, one example: File:Albero Alto ( Huesca ) . - panoramio.jpg
I also added in the error list two items, statistics and the way to have a full list.
I’m not judging at all but it sounds strange to me knowing that all Wikipedias have strict rules for page naming and nothing in Commons?
I’m just talking about basic typographic conventions, nothing else. And part of my request has been accepted in 2014.
--FDo64 (talk) 18:01, 17 September 2019 (UTC)
Not a well-chosen name, but nothing there so wrong as to merit changing it once it's there.
Yes, this is very different from other WMF wikis. We are as concerned as they are on category and (gallery) page names, but for naming image files stability is very valuable, so there is a higher bar to change the name once it's there. - Jmabel ! talk 23:56, 17 September 2019 (UTC)
If you are using a script to check style issues, then perhaps it should be tweaked to ignore file names.--Roy17 (talk) 23:30, 21 September 2019 (UTC)

Question about source info.

I am making an .svg using 4 free-to-remix pictures from this site as a base. I am only using the pictures as a guide for contour outlines. When it comes to providing a source to the image, how should I go about referencing each image?

I was thinking about the {{Extracted from}} template, but it only allows for 1 image, and 4 separate boxes are quite bulky.

Another thing to consider is that since my version of these 4 images can be considered a remix, and nose of the licences on the images even require credit for remixed works. If I can leave out the source entirely and just say "own work" that would be prefered, but I am unsure if this is possible.

That being said, providing a source for the image might be an interesting case. The picture is not uploaded, or complete for that matter, so I can't show you what I mean at this time. I just wanted to know in advance. --Diriector Doc (talk) 05:15, 19 September 2019 (UTC)

{{Derived from}}, perhaps? Pi.1415926535 (talk) 05:19, 19 September 2019 (UTC)
+1 - Jmabel ! talk 15:24, 19 September 2019 (UTC)

Second iteration of draft recommendations published

Hey there. The second (September) iteration of draft recommendations (m:Strategy/Wikimedia movement/2018-20/Recommendations) are published at Meta-wiki. What we reviewed and discussed in the past month was the first (August) iteration. Now is your time to review the current (second) iteration of the recommendations. --George Ho (talk) 11:12, 21 September 2019 (UTC)

COM:INUSE, bots, and templates

COM:INUSE and its friend COM:NPOV say that if a file is used on another Wikimedia project then Commons should treat it as being "realistically useful for an educational purpose", and so not delete it just because it's of poor quality or not accurate. The idea is that if another project finds a file useful then Commons shouldn't overrule it by deleting the file. Recently, I've found a few cases where I'm not sure if this principle should apply with full force, and I'd appreciate some other opinions.

The situation is where a file has been added to a page solely based on its name, usually by a bot or by a template that uses {{#ifexist}}. An example is File:Hy-թագավոր.ogg. This is used on wikt:en:թագավոր because DerbethBot added it, and on wikt:hy:թագավոր and wikt:mg:թագավոր because each of those pages has a template that effectively does {{#ifexist:File:Hy-{{PAGENAME}}.ogg|[[File:Hy-{{PAGENAME}}.ogg]]}} dating long before this file was uploaded. By a strict reading of COM:INUSE, since this file is in use it can't be deleted, but that seems a little silly. How should we handle this? One approach would be to declare that this kind of use doesn't count for COM:INUSE purposes. Another would be to say that the file must be renamed first (and the redirect deleted), but that feels like unnecessary bureaucracy to me. Thoughts? --bjh21 (talk) 15:29, 21 September 2019 (UTC)

Can't a new version of the file simply be uploaded with the correct pronunciation? Files can be overridden by other files and "COM:INUSE" is quite useful, but Wikimedia Commons should host realistically educational content so if it's useful it should stay. As this concerns an incorrect pronunciation, can't another soundbite be uploaded on the existing file? Or maybe it could be renamed to illustrate how a certain accent would pronounce it. I've seen people willingly remove files from other Wikimedia websites and then nominate files here as out of scope (sometimes rightfully, sometimes not) so I am not sure if making "COM:INUSE" even weaker than it currently is is a good thing to do.
Deletion should always be a last resort and if the file can be improved in other ways it should. But as the pronunciation simply is wrong it should be replaced by another file on those local Wikimedia websites thus no longer making it eligible for the "COM:INUSE" exception in the first place. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 15:58, 21 September 2019 (UTC)
What if no other recording is available? Not everybody speaks good Armenian and is willing and able to make voice recordings. Commons:Overwriting existing files would prohibit uploading it with the same name, since it would be "Different files on the same topic". --ghouston (talk) 03:32, 22 September 2019 (UTC)
If the file is misleading due to its filename, maybe the solution is just to rename it? (Eg. to File:Hy-թագավոր (incorrect pronounciation).ogg or something similar? I think that our rules concerning COM:INUSE should not be changed due to possible abuse of the changed rule. Ankry (talk) 07:48, 22 September 2019 (UTC)

Two de-adminship requests are open for community feedback

Desysop requests are thankfully a rare event for Wikimedia Commons. There are two open desysop requests which are due to close in 3 days. Community members are welcome to add votes to keep or remove sysop access, or to ask questions.

The discussion related to alleged misconduct which resulted in these requests is available here, and the procedure which governs desysop requests is Commons:Administrators/De-adminship.

This neutral village pump notification is to encourage wide feedback from the community, rather than just those that follow admin related noticeboards. Thanks --Fæ (talk) 11:16, 23 September 2019 (UTC)

Request to extend Wiki Loves Monuments 2019 deadline due to multiple attacks on servers and consequent misfunctioning

Today again the upload wizard is out of use (at least for Europe). Looking fo support here. Thank you. --Sailko (talk) 17:40, 23 September 2019 (UTC)

Structured Data - blogs posted in Wikimedia Space

There are two separate blog entries for Structured Data on Commons posted to Wikimedia Space that are of interest:

  • Working with Structured Data on Commons: A Status Report, by Lucas Werkmeister, discusses some ways that editors can work with structured data. Topics include tools that have been written or modified for structured data, in addition to future plans for tools and querying services.
  • Structured Data on Commons - A Blog Series, written by me, is a five-part posting that covers the basics of the software and features that were built to make structured data happen. The series is meant to be friendly to those who may have some knowledge of Commons, but may not know much about the structured data project.
I hope these are informative and useful, comments and questions are welcome. All the blogs offer a comment feature, and you can log in with your Wikimedia account using oAuth. I look forward to seeing some posts over there. -- Keegan (WMF) (talk) 21:33, 23 September 2019 (UTC)

Classic Manager

"Classic Manager" is a South Korean website and app that streams out-of-copyright music. Though information on the site is scant, press articles say it uses "an algorithm which automatically filters music freshly freed from copyrights and adds them to its free archives" ([3], [4]). Can anyone confirm? Does it have material that we don't, but should?

Here, for example, is an album:

https://classicmanager.com/albums/ATMCD1244

I note that this score:

https://media.classicmanager.com/score/Musette+in+D+major%2C+BWV+Anh.126+%28Bach%2C+Johann+Sebastian%29.pdf

contains a "transcription copyright R.S.B. 2012" footnote, but perhaps that's copyfraud? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:47, 23 September 2019 (UTC)

I can't tell what copyright rules they're using, but in the US, all recordings not explicitly released are copyrighted; even the pre-1923 works will take a few more years to enter the public domain. South Korea has a 70 year copyright on recording, if they're being careful, but while many countries have 50 year copyrights on recording, COM:CRT doesn't mention audio recordings for many languages. As for the album under question, if it is the Vanguard Classics version, which would be consistent with the data available, it wouldn't be PD in South Korea (Vanguard Classics wasn't founded until 1950) and the first publication would seem to be in the US, where it would be in copyright for many, many years. If it wasn't published in the US, or we're worried about Austria as well, COM:AUSTRIA has nothing on copyright on audio recordings, so IMO we would have to assume life+70 applied, with a death date of 1987 for the conductor. (If someone objects, well, find the right rules for Austria and add them to that page.)
Strictly following COM:L, they're not PD in the US, and thus aren't suitable for Commons. If we're ignoring the US when it's not the source nation, we still need to be clear about when and where the album was published and what the rules are in that country. I would hesitate to download any works from that site, given their lack of clear copyright details.--Prosfilaes (talk) 09:28, 24 September 2019 (UTC)

Category grouping

 
  1. i do not think grouping of categories makes any sense at commons at all.
  2. is there a way to opt out of category grouping?

--C.Suthorn (talk) 05:18, 25 September 2019 (UTC)

Category grouping is very convenient when editing wikitext. The bot is only triggered when category changes are made to a file, which isn't that often, and thus should not be adding any additional items to your watchlist. You can also choose on your watchlist to ignore all bot edits; since Commons doesn't have many bots, you're unlikely to miss anything else. You can also avoid this problem by adding templates like {{Wiki Loves Pride 2019}} during the initial upload, rather than later with VFC. Side note: you did also choose to upload over 900 images of a single event; any change that affects all those images is going to generate 900+ watchlist items. Pi.1415926535 (talk) 05:47, 25 September 2019 (UTC)
editing wikitext is mostly limited to the info template on commons, categories will seldom be inside the template. I do not want to ignore all bot edits, not even the edits of one bot, only edits, that actually do not change anything (category grouping). If I had known, I was to add this images to wlp, I would have done so at upload. But there will always be edits, that trigger bot actions like this grouping. Captions and depict statements are on their way. With 55M files and about 300 languages there will be many edits, that will flood watchlists of users, if there is no way to intelligently surpress mass edits of a specific nature, or have these joined into a single watchlist entry. --C.Suthorn (talk) 11:49, 25 September 2019 (UTC)
Tracked in Phabricator
Task T167224
There is already a task for this. --GPSLeo (talk) 12:08, 25 September 2019 (UTC)

Exception for file size restriction?

Tracked in Phabricator
Task T233721

I am trying to upload a PDF of a translation of Rycaut from the 1700s. I just made it from high quality raw scans, but the system stated:

  • "You can only upload files with a size of up to 4.00 GB. You tried to upload a file that is 4.01 GB."

Is there a way to get an exception to this or override this? I would really like to upload the high quality Rycaut scan.

Thanks WhisperToMe (talk) 00:39, 24 September 2019 (UTC)

@WhisperToMe: See User:Alexis Jazz/common.js. Note: I never actually tried most of the crap in there. I don't know what restrictions exist server side. - Alexis Jazz ping plz 00:55, 24 September 2019 (UTC)
Thanks for the tip! I'm not sure how to load the script (either into the browser or into my Commons account) - I checked "Interface: Editing and uploads" but I am not sure which one reflects common.js WhisperToMe (talk) 01:07, 24 September 2019 (UTC)
@WhisperToMe: User:WhisperToMe/common.js. You only need the one line. (and the bluray comment, if you wish) - Alexis Jazz ping plz 01:08, 24 September 2019 (UTC)
Got it! Thanks! WhisperToMe (talk) 01:18, 24 September 2019 (UTC)
This will not work. The 4GB limit is fixed due to limitations of the Mediawiki software. --GPSLeo (talk) 10:17, 24 September 2019 (UTC)
The file size restriction is not 4GB, but 4GiB. = 4,294,967,296 Bytes. That is the limit of an actual file system (FAT32, FFS, ISO9660), that is probably used by WMF for file storage. If this is the case, an exception is not possible, only a migration of all WMF file servers to a different file system (ext4, NTFS, ...). You can split your file into files with upto 4GiB. --C.Suthorn (talk) 10:18, 24 September 2019 (UTC)
If the WMF file servers use FAT32 I will eat my hat. - Alexis Jazz ping plz 11:42, 24 September 2019 (UTC)
Is your hat eating limited to FAT32, or do other limitations to 4GiB also apply? Politparade.webm and File:Genderwahn.webm are both 4GiB and both do not come with a UHD or QHD trnscoding, as the transcoding would be larger than 4GiB. --C.Suthorn (talk) 12:45, 24 September 2019 (UTC)
Alexis Jazz is right; the Wikimedia Foundation is running on Debian GNU/Linux, thus not using FAT32. ISO9660 is a write-once filesystem, and I don't even know what FFS is; Wikipedia turns up any number of options. The default option, assuming no change ever in twenty years, would be ext2, which can do 16 GB per file; nowadays ext3 or ext4 would be more likely. If the WMF managed to chose a filesystem that doesn't support more than 4GB, I'm not sure which one it was; I can't find a reasonable choice without that support. I suspect it's a bug in the Wikimedia code, not a filesystem choice.--Prosfilaes (talk) 12:54, 24 September 2019 (UTC)
Why are you trying to do this? Very few people will have the capability or patience to download such files. Why not upload each page as a separate TIFF if you are determined to achieve max quality with no lossy compression. -- Colin (talk) 13:00, 24 September 2019 (UTC)
I think you underestimate how many people can download a 4 GB file in modern times. I would say there are fewer people who would have the patience to download each and every page of even a 100 page book, which is way more tedious and just as much data transfer as pointing a download manager/wget at a file and letting it take its time in the background.--Prosfilaes (talk) 13:38, 24 September 2019 (UTC)
No, I stand by "few people will have the capability or patience". A PDF with scanned images that is > 4GB.... something very wrong here. Like uploading uncompressed BMP! -- Colin (talk) 15:25, 25 September 2019 (UTC)
You've never edited on Wikisource, and you've never uploaded any PDF to Commons. I submit you've never wrestled with the problems of taking a large set of images of a work and leaving embedded images suitable for cropping and rotation, to produce an optimal transcription with highest quality images at the end of the day.--Prosfilaes (talk) 04:24, 26 September 2019 (UTC)
It is in the Phabricator task. The problem is the used OpenStack Object Storage(Swift). Is seem that there are solutions for this in OpenStack. But this needs changes in MediaWiki and all other components. --GPSLeo (talk) 14:52, 24 September 2019 (UTC)
w:FFS can refer to several file systems, and the Wikipedia article w:Unix File System misleadingly doesn't mention the filesize limitation of the earlier versions. Still, using the Berkeley Fast File System or any version of it would be unsupported on Linux and I can't imagine why professional sysadmins would use it with Linux.
Welcome to the 21st century? Programs are deep and huge, and you're talking, not to people running the system, not to people who support the system, but to people who run a project on the system. Would you ask the people on an arbitrary GitHub project what filesystem GitHub is running on?--Prosfilaes (talk) 15:06, 24 September 2019 (UTC)

Hello, the issue is not really in file system - Swift, which we use for storing files, can support files up to 5 GiB without any issues. The issue that MediaWiki stores filesize into a field that can be up to 2^32, which is 4 GiB. Because of that, it can't accept bigger files, since it can't store how big the file actually is anywhere - the field is too small to accept that information :). As was mentioned above, there are tasks about this. Best, --Martin Urbanec (talk) 16:13, 24 September 2019 (UTC)

5GiB? That looks like an odd untouchable number. --C.Suthorn (talk) 17:19, 24 September 2019 (UTC)
@WhisperToMe: Have you tried using lossless PDF compression? Kaldari (talk) 21:07, 24 September 2019 (UTC)
@Kaldari: I haven't yet tried it. What steps should I do? WhisperToMe (talk) 23:08, 24 September 2019 (UTC)
@WhisperToMe: Depends on what OS you're using and what programs you have. For a PDF like the one you're working with, the image compression is going to make the most difference. If the images embedded in your PDF are color or greyscale (which I imagine they are), you can use JPEG 2000 compression at lossless quality. If the images are monochrome you can use JBIG2 compression at lossless quality. Most other compressions are going to be lossy, but at high quality levels it's virtually impossible to tell the difference. Hope that helps. Kaldari (talk) 15:03, 25 September 2019 (UTC)
Thanks! I use Windows 10 (not the best OS by any means) WhisperToMe (talk) 15:06, 25 September 2019 (UTC)
What program on Windows 10 are you using to create these PDFs? How many pages and what resolution is each page. There's something very bad going on if your PDF is 4GB. -- Colin (talk) 15:25, 25 September 2019 (UTC)
@Kaldari: JPEG 2000 is possibly still protected by patent, so not allowed on Commons. JBIG2 may also still be protected. - Alexis Jazz ping plz 16:21, 25 September 2019 (UTC)
In that case, you'll want to use JBIG1 for monochrome images (as it is definitely no longer patented), and maybe PNG for color and greyscale (if your PDF program supports it). And like I said, high quality JPEG compression, although lossy, should be considered as well. Kaldari (talk) 16:36, 25 September 2019 (UTC)
I think you guys are over complicating this. You are over-optimising by picking obscure compression formats. It is quite likely zero compression is being used here, which is just not sensible. JPG at 11/12 or 90% level compression (depending on your software) is quite adequate. We'd need some convincing reason why lossless compression is essential here. All us photographers make do with JPG and that's the reason these other formats didn't catch on. Also, not all software supports them. Also what is the reason this is being uploaded as a PDF of scans? Is the actual manuscript beautiful? Or do you simply want folk to read the text, in which case let's consider OCR and we're looking at kilobytes rather than gigabytes. -- Colin (talk) 17:13, 25 September 2019 (UTC)
@Colin: The book has illustrations and I would like for those to be the highest quality. That's also why there's a book about Armenia that I am fixing to upload in full PDF format too: the illustrations. Anyhow now that I implemented the workaround for 4 GIB, I'll try to upload these two when I'm in Hong Kong... the VPN situation in Mainland China makes it hard for me to do large uploads. I'll stay in Hong Kong and use that time to do my uploading. WhisperToMe (talk) 18:34, 25 September 2019 (UTC)
Historically, there's a lot of images out there that are frustrating, because they're tiny or overcompressed; scans of books have often lost periods due to aggressive despeckling for compression, or OCR has been applied and text replaced with the OCRed version, whether or not that OCRed version is correct. We do not uploaded OCR to Wikisource, period. If you have to, you can transcribe, with or without the use of OCR, but a final human comparison either way, but the scans are vastly more valuable because people can go back to the original text and compare.--Prosfilaes (talk) 03:52, 26 September 2019 (UTC)
My point is we should have asked for the requirements and thought about the best solution, rather than assume what you were trying to do was the best option and needed a fix. I still think there is merit in uploading TIFF for each page, especially for illustrations. There are people who restore or appreciate old illustrations and the only real way to show them off to people is to produce a JPG from the TIFF original. Expecting folk to download a 4GB PDF and extract images from that is too much, so in the end nobody would see it. Also you can describe and categorise each page if helpful, to indicate what the illustration is of. Then people can find them by search. -- Colin (talk) 07:05, 26 September 2019 (UTC)
the use case is wikisource. that is the way people work at wikisource: create a pdf of images in highest quality, upload it, then create a transcription, and others can check the quality from the pdf or cooperate in the transcription. they work with pdfs, they have tools for that. you can surely create a tiff with multiple images in it, but it would probabky not be smallet and you would need other tools to work with. In this case, i would like to know, how many pages are in the pdf, and how big it is. if it was 5GB and consisted of 50 images, a workaround would be to have two pdfs with 25 pages each. but large files are needed for video. File:Politparade.webm has 4GiB, but my original webm is 4.3GiB. The uploaded file has no uhd and qhd transcodes, as they would be larger than 4GiB. stores have started to sell 8K TV sets, there are already 8K videos at commons. --C.Suthorn (talk) 09:30, 26 September 2019 (UTC)
@Colin: JPG is really, really bad for text. If there is no other way, I'd go for at least 95% quality. - Alexis Jazz ping plz 10:48, 26 September 2019 (UTC)
I know. But for illustrations it might be fine. The folk who restore and nominate historical works at FPC often use TIFF or PNG as a master copy but offer a JPG as the one for use on Wikipedia, etc. I'm not aware that huge PDFs are generally the medium of choice for scanned works. -- Colin (talk) 16:24, 26 September 2019 (UTC)
If you want a copy to use on Wikipedia, you might want to upload a JPEG. The primary use is for Wikisource, where PDFs or DJVUs are the medium of choice for scanned works; anything else is a complete pain in the ass. Illustrations should be pulled out and formatted as needed, probably JPEG. But the book as a whole should stored as a unified work.--Prosfilaes (talk) 02:55, 27 September 2019 (UTC)

Bad geocoords

In Category:Problem tags I don't see anything in particular to indicate incorrect location coordinates. Is there an appropriate template to use for this? - Jmabel ! talk 21:32, 25 September 2019 (UTC)

@Jmabel: I've got a bot task running at the moment that's tagging a specific uploader's images as somewhat inaccurate (per their request, due to technical issues). There isn't a great template for "The location in the exif is wrong, don't add it" at the moment. We're using {{Location estimated}} because it's the closest existing template. {{Location rounded}} and {{Location withheld}} may be useful as well. These templates are in Category:Geocoding templates. --AntiCompositeNumber (talk) 23:45, 25 September 2019 (UTC)

Still, there ought to be some way to say outright "the coordinates here are clearly wrong." - Jmabel ! talk 04:29, 26 September 2019 (UTC)

The new Metadata "depicts"

Does that information end up in the exif data when I download? RAN (talk) 03:09, 26 September 2019 (UTC)

No. Structured data is unconnected to file metadata or the page wikitext. For these reasons I don't touch it, the practical use case was never made for the system change from my perspective as a content creator. --Fæ (talk) 03:56, 26 September 2019 (UTC)

Starting a project to digitalize Aiurri magazine

Hello! The Basque Wikimedians User Group is about to start a project with Aiurri magazine to upload part of their archive to the Commons. This magazine is published under cc-by-sa-3.0 license, but the historical image archive and its corresponding magazine information is not online. We will measure during one month how much we can digitalize and upload in three different tasks: image scanning and description, tiding it to magazine news maybe in Wikisource and uploading full pdfs of more recent magazines.

This project will last one month, and we will measure after that what we have achieved, in order to make this project with other similar magazines.

You can read more about the project here: Commons:WikiProject_Tokikom -Theklan (talk) 21:48, 26 September 2019 (UTC)

This is the first test upload. If you have anything to make the information better, please, let it know: File:Laborde_erakusketaren_prentsaurrekoa.jpg -Theklan (talk) 10:15, 1 October 2019 (UTC)

The consultation on partial and temporary Foundation bans just started

Hello,

In a recent statement, the Wikimedia Foundation Board of Trustees requested that staff hold a consultation to "re-evaluat[e] or add community input to the two new office action policy tools (temporary and partial Foundation bans)".

Accordingly, the Foundation's Trust & Safety team invites all Wikimedians to join this consultation and give their feedback from 30 September to 30 October.

How can you help?

  • Suggest how partial and temporary Foundation bans should be used, if they should (eg: On all projects, or only on a subset);
  • Give ideas about how partial and temporary Foundation bans should ideally implemented, if they should be; and/or
  • Propose changes to the existing Office Actions policy on partial and temporary bans.

We offer our thanks in advance for your contributions, and we hope to get as much input as possible from community members during this consultation!

-- Kbrown (WMF) 17:13, 30 September 2019 (UTC)

Anyone have any idea as to why Commons would want WMF T&S issuing partial or temporary bans, rather than the community relying on the normal Commons administrators and global stewards to review cases? I have followed the English Wikipedia disputes, but have no firm ideas about why this would be relevant here. Naturally the global banhammer that WMF T&S uses to ban accounts permanently is unaffected. --Fæ (talk) 17:35, 30 September 2019 (UTC)
  • The wording used is «how partial and temporary Foundation bans should ideally implemented, if they should be; and/or». Does this mean that answering just no/none is an acceptable and effective course of action? Or would it just be discarded, as usual in these WMF consultations? -- Tuválkin ✉ ✇ 20:11, 30 September 2019 (UTC)
This is about w:WP:FRAM. And WMF didn't read m:Requests for comment/WMF-community trust person. Hence, I am disappoint. - Alexis Jazz ping plz 13:58, 1 October 2019 (UTC)
But let's be honest, if they can't ban people from one platform they'll ban them from them all, Classiccardinal was a sysop on one project before the WMF globally banned him. Do we really need another Russavia case? --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 10:40, 2 October 2019 (UTC)

Mirrors

 

Is there an appropiate classification for objects en facades made up of small mirrors?Smiley.toerist (talk) 09:51, 30 September 2019 (UTC)

More images in Category:Pasaż Róży. Smiley.toerist (talk) 10:39, 1 October 2019 (UTC)
I think Category:Mosaic murals is pretty good. As far as the mirror aspect goes, we also have File:Pillared hall with glass and mirror mosaic decorations, in the Jalnavas or Fountain Palace, City Palace, Udaipur.jpg, File:Najaf, the Gate of Wisdom.pdf, File:Institut français de Budapest. Mirror effect and mosaic. - 17 Fő Street, Budapest District I.JPG, File:2015 Woodstock 024.jpg and probably some others. But no base category. Category:Disco balls ? =) - Themightyquill (talk) 11:16, 3 October 2019 (UTC)

Policies around images created by Artificial Intelligence applications (such as GANs)?

Hello all, I've just uploaded a series of images generated by StyleGAN, an AI model that generates completely new faces from nothing. The model was created by NVIDIA but it runs on my own machine. I've shared these images on the commons and saw that there was no category for these types of images, so I created Category:Generative Photography, which is where you will find the images. Are there any policies or issues with sharing these images on Commons? What kind of images should be included or excluded? As I see it, Commons has a benefit of access to faces without identities for a variety of purposes, but at the very least as examples for illustrating Wikipedia's articles relating to the technology. Thanks! Owlsmcgee (talk) 03:50, 23 September 2019 (UTC)

Please, explain what you mean by "sharing the images". Ruslik (talk) 05:05, 23 September 2019 (UTC)
By "shared" I mean I have uploaded images I created using the GAN to Commons. Owlsmcgee (talk) 06:17, 23 September 2019 (UTC)
@Owlsmcgee: Impressive, though not perfect. The perspective isn't right on some. But the copyright for this is somewhat uncharted territory. Images purely generated by an algorithm can't be copyrighted because you can't copyright an algorithm. But here, the question arises: to what degree are these images identical to the source images? Is the software copy-pasting eyes and mouths and so on? It could be derivative work. And to what degree did you put any work into these? If you just clicked "generate" or entered some basic parameters and clicked "generate", you can't claim own work and Creative Commons. We may need a variant of {{PD-animal}} for this. @Clindberg: I have no idea about this.. - Alexis Jazz ping plz 05:31, 23 September 2019 (UTC)
Thanks for the question, Alexis Jazz. The images are generated, not copy/pasted or reassembled. The algorithm looks at millions of portraits of images from Flicker and figures out, based on a kind of "average", what colors the pixels need to be to fit in with the Flicker archive. As a result of coloring these pixels, and comparing them to the images in the Flicker archive, the machine passes on to us an image we see as a human face. There's no resemblance to any original images aside from, generally, the size of the images the machine was trained on (1024 x 1024). The algorithm was created by NVIDIA, run on a piece of software called RunwayML. The AI is creating a new image at every iteration, so there's no repeat faces between any two users of the software. I'm intrigued by the idea that clicking "generate" isn't enough to claim own work and Creative Commons, and I've read up on the "monkey selfie" case. If that's the standard applied here, should the images be reclassified as CC0/public domain? Owlsmcgee (talk) 06:17, 23 September 2019 (UTC)
@Owlsmcgee: Copyright exists to protect creative expressions. Clicking "generate" does not require any creative input. If you take a photo, you decide what you want to take a picture of, pick an angle, lighting, maybe change camera settings and shoot. When you draw or paint something, every stroke is your decision. When you compose a melody, you pick the notes. When you write a text, you pick the words and build the sentences. But tapping a button may as well be done by a drinking bird.Homer figured that out long ago Drinking birds can't paint, write or compose, but they can click buttons. And they're not entitled to copyright. The images would be classified as public domain in this case. Not even CC0, because CC0 is a tool to release your rights. You can't release rights you don't have. See also m:Wikilegal/Copyright for Google Translations. - Alexis Jazz ping plz 06:47, 23 September 2019 (UTC)
I get all that, and I'll see to it that they are changed. Thanks! 216.38.129.234 16:39, 23 September 2019 (UTC)
Thank you everyone, I've updated the images to pd-reasons: because "as the work of an artificial intelligence, it has no human author in whom copyright is vested," based on the pd-animals template. Owlsmcgee (talk) 04:04, 24 September 2019 (UTC)

Images are now all in Category:AI generated faces. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:56, 23 September 2019 (UTC)

While the issue of copyright (or lack thereof) has been addresses, other relevant policy includes COM:EDUSE: Images must be realistically useful for an educational purpose. A small number of faces to demonstrate the technology is probably acceptable, but large number of images of people who have never existed are in essence the same as snapshots of my friends at a party (see COM:PARTYPICS): very unlikely to be used for any other purposes. Issues of project scope exist in a spectrum, and involve some amount of subjective opinion: I'm not saying the images you've uploaded are beyond scope, but it's an issue that may conceivably be raised. --Animalparty (talk) 21:15, 24 September 2019 (UTC)

I think that a "{{PD-algorithm}}" template could be created, this could then be used for any algorithm created files. --Donald Trung 『徵國單』 (No Fake News 💬) (WikiProject Numismatics 💴) (Articles 📚) 07:44, 27 September 2019 (UTC)

I've added the following copyright notice to each of the file based on PD-animals: "This file is in the public domain, because as the work of an artificial intelligence, it has no human author in whom copyright is vested." Owlsmcgee (talk) 03:58, 4 October 2019 (UTC)
Is Category:Generative photography still needed? It's now sitting empty. - Themightyquill (talk) 11:06, 3 October 2019 (UTC)
I think it's useful, but I believe Andy Mabbett removed them from that category when they created the "AI-generated faces" category. Not sure if this was a deliberate filing. Owlsmcgee (talk) 03:58, 4 October 2019 (UTC)
I didn't create the latter. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 07:52, 4 October 2019 (UTC)
Yes - and you can see that for yourself, as it's still used. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 07:52, 4 October 2019 (UTC)

freepd.com

Hello.

Can we use musics from this site : https://freepd.com/

I think no.

Thanks a lot.

--ComputerHotline (talk) 08:00, 28 September 2019 (UTC)

That is all CC0. But as they write these platforms do not want that you have the permission which is given by the license. These platforms want that you are the copyright holder, what you are not. --GPSLeo (talk) 21:57, 28 September 2019 (UTC)
Ok. Thanks a lot. --ComputerHotline (talk) 08:08, 30 September 2019 (UTC)

A word of caution. This is an aggregator website, as are many others, like Europeana. Our batch uploads should be from the original source, not from aggregation sites. They often have weak metadata (especially losing original EXIF data), lower resolution images, or may have automatically tampered with the saturation or other original features of the file. They may also be making invalid presumptions about the original source, and harvesting images from other sites which themselves have failed to give a correct attribution, or to choose the most appropriate license. For example "CC0" is not the same thing as saying that a file is public domain by age. --Fæ (talk) 13:34, 2 October 2019 (UTC)

The site and the content is made by the artists themselves and so seems to be the original source. And I think audio files does not have any EXIF data. --GPSLeo (talk) 10:21, 4 October 2019 (UTC)

Bad signature

Why is my signature not correct here? It only says undefined --193.171.152.104 06:39, 17 September 2019 (UTC)

Your IP Address is not a registered user. Please see COM:QIC#Evaluating images, and login if you want to review QICs.   — Jeff G. ツ please ping or talk to me 16:07, 17 September 2019 (UTC)
Your link didn't work for me, Jeff, so I took the liberty of attempting to fix it. – BMacZero (🗩) 15:30, 18 September 2019 (UTC)
I reopen this topic, because I wasn't able to answer in time. The issue is that the signature is incorrect for nominations. Since I do not want to vote there shouldn't be a need for an account. --193.171.152.104 10:29, 30 September 2019 (UTC)

Can anyone help here? The signature works fine on regular pages (like this one). Maybe it's the template that causes the problem? --193.171.152.104 12:09, 7 October 2019 (UTC)

Retrieved from "https://commons.wikimedia.org/w/index.php?title=Commons:Village_pump/Archive/2019/09&oldid=370600534"