Commons:Village pump/Archive/2017/02

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

Rescuing old version from file history

The history of, for example File:Adam D'Angelo and Steven Levy in 2011.jpg incudes a cropped version of the original. Is there a tool that will "rescue" that, and create a new page with that version, with the history, description etc as it was at the time of its upload? Andy Mabbett (talk) 21:08, 31 January 2017 (UTC)

By far the easiest way you @Pigsonthewing: can use to separate out a version of the cropped image from the original upload is, if you have the "CropTool" activated in your preferences (found under Gadgets), you can just crop the existing image, to approximately the same size as the previous crop, and CropTool will give you the option of make a new file, which is what you want, or overwriting the file. When making a new file it copies the existing details and tag, and places an {{Extracted from}} and {{Extracted}} template in each file as appropriate linking the two images. The history will not be copied. Alternately you can tag the image with a {{Split}} template but I don't think that copies the history either, so in that case you may need to make a request at Commons:History merging and splitting/Requests. Ww2censor (talk) 23:18, 31 January 2017 (UTC)
The extracted-from (or other derivative-of) link should be all that’s needed for attribution or provenance; why else would one want to copy the history? (Not that there’s much to speak of in the above example.)—Odysseus1479 (talk) 03:03, 1 February 2017 (UTC)

February 01

Redirect loop when saving Flickr photos using Wayback Machine

I'm going to save some photos from Flickr for license evidencing. When I go to I'm unable to save any page from Flickr and I get error about redirect loop. Do you experience the same problem? --Rezonansowy (talk) 21:41, 20 January 2017 (UTC)

This may be caused because the photographs are for Friends only or restricted. If you are uploading to Commons, using either standard upload tools or adding {{Flickrreview}} will ensure the license is checked automatically. The automatic bot checks are sufficient evidence against any future challenge. -- (talk) 21:54, 20 January 2017 (UTC)
@: Thanks for hint! But in this case photos are public and in addition I noticed that Flickrreview bot is also unable to check licenses of new photos right now. I tried with this example:
--Rezonansowy (talk) 22:41, 20 January 2017 (UTC)
@: Could you or someone help me with this please? The problem is still persistent and Flickrreview bot is probably affected. Try this example I provided above. --Rezonansowy (talk) 10:49, 23 January 2017 (UTC)
Using your example, I see, which displays fine for me using Chrome. The issue may be your ad-blocker or similar. -- (talk) 10:54, 23 January 2017 (UTC)
@: Does creating new snapshots of Flickr photos (on work for you as well? --Rezonansowy (talk) 11:38, 23 January 2017 (UTC)
No, I get the same error you've had; "ERR_TOO_MANY_REDIRECTS". I suggest emailing or checking their forum for comments. -- (talk) 20:21, 23 January 2017 (UTC)
OK, I'll mail them soon. --Rezonansowy (talk) 22:16, 24 January 2017 (UTC)
For saving a Flickr photo's details in the Internet Archive, something that may work and which may be better than nothing is to follow the "Download this photo" option (the icon is a downwards arrow with a line underneath) to reach the photo's download page and to then archive the download page. For example, in the case of, the download page would be instead. The download page has links for downloading the photo in various sizes and also has information about the photo's author and license. --Gazebo (talk) 05:26, 28 January 2017 (UTC)
@Gazebo: Thanks! --Rezonansowy (talk) 16:30, 1 February 2017 (UTC)

January 21

Questions about file types

Commons:File types I'm confused as to why certain types of files are not allowed on Commons. From a technical perspective, it is not terribly hard to allow new kinds of files. In terms of community support or fitting the general mission of the WMF in education, I don't think anyone would argue that (e.g.) FLAC audio or 3-D renderings are inappropriate--they have broad community support for several years and they are free formats. Since Commons Archive accepts not only these but any type of file with mammoth file sizes, I don't understand what is holding back the community here from accepting the many file types that users would like to have available. —Justin (koavf)TCM 07:25, 31 January 2017 (UTC)

The main requirements on file types here have been: 1) Open file specification not subject to royalties to implement. 2) Represents some type of concrete visual, audio, or audio-visual rendering (i.e. not merely abstract data which could be rendered in many different ways). 3) Formats representing visual data should have visual thumbnails available in Wikimedia software that can be seen in web browsers. I don't think there will be an upload-any-format policy on Commons anytime soon.
Adoption of 3D here has been held back due to lack of a main predominant standard file format. Don't know about FLAC... AnonMoos (talk) 08:25, 31 January 2017 (UTC)
FLAC files are supported.
Please note that some of the documentation does not always correctly differentiate between: allowed, resource-wise feasible and technically implemented. And having a file format supported for upload/download (the only thing that Commons Archive allows) is of course different than it being supported for usage inside wiki pages with thumbnails, players etc.
It's indeed curious that the site bypasses an assortment of community rules, but it's not unique in that regard. Video2Commons does that as well and the community doesn't really seem to mind (which I find a bit hypocritical, but whatever). —TheDJ (talkcontribs) 12:58, 31 January 2017 (UTC)
@AnonMoos: I was in no way suggesting that Commons could/would/should whitelist every file extension or MIME type--that's madness. In the case of 3-D renderings for instance, we have .blend and we know that the community wants these. Are you saying that the only thing holding it back is generating thumbnails? Also, we do have at least a few "redundant" file types--e.g. DJVU/PDF. Is it a problem to have multiple file types that perform the same/similar function? @TheDJ: can you explain what you mean about hypocrisy? I'm not following. —Justin (koavf)TCM 04:15, 1 February 2017 (UTC)
See also phabricator:T3790. --ghouston (talk) 21:43, 1 February 2017 (UTC)
Koavf -- Being "duplicative" is not much of a problem in itself, if there are other strong reasons why we should support a format. TIFF duplicates JPEG and PNG, and has well-known problems (such as being it being practically almost impossible for a software program to understand/process every theoretically-valid TIFF file), but institutional archivists wanted TIFF, so it was added. PDF and DVJU are multi-page whole-document formats which are in practical use. As far as I know, .blend is the internal file format of Blender software, not a general-purpose standardized 3D file format. If it were added to Commons, it would be for the same reason that .xcf (GIMP) files were added... AnonMoos (talk) 03:49, 2 February 2017 (UTC)

rotation bug?

Hi... I just uploaded a batch of four photos. The originals were all in landscape (desired layout) but the wizard rotated all of them to portrait. I have requested rotation from the bot. Why were they rotated in the upload? This has never happened to me before. Is there a bug? I don't want to be uploading anymore until it's fixed (as rotating anew is not lossless). Thanks! --Saintfevrier (talk) 14:19, 31 January 2017 (UTC)

  • I gather we recently started paying more attention to rotation info in EXIF data. I'm guessing that info ws wrong. That's not definitive, just a guess. Does anyone have something more solid? - Jmabel ! talk 16:27, 31 January 2017 (UTC)
  • And if it can be fixed in the EXIF data, that is lossless. - Jmabel ! talk 16:28, 31 January 2017 (UTC)
    • @Saintfevrier: Hi, You have to check the exif orientation which was set by your cam. Looks like your cam has set a wrong rotation (or at least not in a common way). When you take photos with your cam, then your cam is writing orientation datas in the exif files, mediawiki thumbs will be generated on that basis (it is not a bug). The rotations are lossless. Best --Steinsplitter (talk) 17:42, 31 January 2017 (UTC)
      • Thanks for the replies! I have uploaded with the same camera before (it's my S5 smartphone), and I haven't changed the settings in the meantime. I will check though. Nonetheless it's good to know that rotations are lossless. Thanks again!--Saintfevrier (talk) 23:22, 1 February 2017 (UTC)


Hello, I find it very hard to know how to get in touch with 'Biccie', the person that has uploaded a picture of my sculpture. A picture that I would like to replace with another picture because the one that is now visible is a picture of the sculpture after it was vandalised. Thank you for some help. Jacqravelli — Preceding unsigned comment added by Jacqravelli (talk • contribs) 09:10, 1 February 2017 (UTC)

You don't need to get in touch with the uploader. Just upload a new picture yourself (you need to create an account for that) and replace it in the articles where the current picture is used. --Magnus (talk) 10:39, 1 February 2017 (UTC)
@Jacqravelli: Just go to User talk:Biccie, click on "Add topic" at the top of the page and leave a message. Please do not edit the page "manually" when you are not sure what you are doing. However Biccie has been inactive since November 2014, I am quite sure that his new user name is Bic. Anyway, please upload the new photo under a new name, as per COM:OVERWRITE. --jdx Re: 10:53, 1 February 2017 (UTC)

Thank you for your help. I tried to upload a new picture but I keep getting: 'We could not determine whether this file is suitable for Wikimedia Commons. Please only upload photos that you took yourself with your camera, or see what else is acceptable. See the guide to make sure the file is acceptable and learn how to upload it on Wikimedia Commons'. The picture I use is a jpeg file. I don't know what to write in the space called categories? Really difficult. Thank you, Jacqueline — Preceding unsigned comment added by Jacqravelli (talk • contribs) 19:09, 1 February 2017 (UTC) Sorry!--Jacqravelli (talk) 19:14, 1 February 2017 (UTC)

  • It looks like the categories on File:Rapenburg_65_-_Bronzen_plaquette.JPG are fine, so you can just duplicate those.
  • That warning sounds like you might be trying to upload a low-resolution JPEG without EXIF information. Is that the case? (Only new users get that warning: usually when they are trying to upload something like that, they "just snagged it" off of a different site.) - Jmabel ! talk 21:20, 1 February 2017 (UTC)

Kan misschien iemand een nieuwe foto maken van de bronzen plaquette op Rapenburg 65 in Leiden? De huidige foto zou volgens de maakster niet langer representatief zijn. Regards, Bic (talk) 22:55, 1 February 2017 (UTC)

Valued images - why not video or audio files??

In the section Commons:Valued_images#Eligibility there is a point about not allow neither videos or audio. Have a reason for that? This is powerful medias when we are talking about educational value, and this is our away to highlight educational value. We can't remove this restriction? And I already so some videos passing as VI, and makes sense. -- Rodrigo Tetsuo Argenton m 23:32, 1 February 2017 (UTC)

February 02

Clean up of ""

Daniel's old WikiSense tool has links all through templates and many other places within Commons. It would seem opportune for us to look to clean out the prominent uses of the link, especially help:, commons: and template: namespaces where it is directional to utilise.  — billinghurst sDrewth 06:33, 23 January 2017 (UTC)

What does "directional to utilise" mean? Thanks -- (talk) 10:40, 23 January 2017 (UTC)
Fae directional (we direct) utilise to use.  — billinghurst sDrewth 09:48, 2 February 2017 (UTC)
Most likely, "directional to utilise" refers to situations where a page talks about using the WikiSense tool or suggests or directs the user to use the WikiSense tool. --Gazebo (talk) 11:03, 26 January 2017 (UTC)
Probably, certainly it looks like tech jargon rather than English. -- (talk) 11:09, 26 January 2017 (UTC)
For anyone who wants to replace or remove references to WikiSense, the following list is not complete but may well be useful, particularly with regard to templates, categories, and help information:
Commons pages with references to ""
--Gazebo (talk) 06:33, 28 January 2017 (UTC)
Pinging @Billinghurst: --Gazebo (talk) 05:58, 2 February 2017 (UTC)
Thanks. Not something that I have the time or incentive to work my way through, seemingly the same for others. There needs to be a decision whether there is a replacement, or we are just removing text.  — billinghurst sDrewth 09:48, 2 February 2017 (UTC)

Enormously large SVGs

Some files uploaded as SVG files contain only photografical data and no vector data, for example File:WLM-Calendar 2016 TOP-101 Germany A1.svg, File:WLM-Calendar_2016_TOP-101_Germany_A0.svg, File:TTC_dome_and_LANYARDome.svg, and take much much more space than if they were uploaded as vector files. Maybe we should reupload them as JPG files and delete original SVG files? --Tohaomg (talk) 20:41, 27 January 2017 (UTC)

  • "than if they were uploaded as vector files": I assume you mean "than if they were uploaded as raster files". I would think this is entirely appropriate to do; anyone (including you) can do it exactly as easily as anyone else. By the way, though: while it will make them quicker to download, it won't solve any file space issue. Once a file is on our servers, it is never physically deleted. - Jmabel ! talk 21:33, 27 January 2017 (UTC)
I thought if a file is deleted it is deleted for good. If no space will be freed, than really, no reason to delete it. I will reupload and replace all inclusion by myself then. --Tohaomg (talk) 22:02, 27 January 2017 (UTC)
With some extremely rare exceptions, Admins can look at all "deleted" files. - Jmabel ! talk 00:39, 28 January 2017 (UTC)
Tohaomg and SVG and JPG have different proposals, not equivalent type of file, SVG is better for impressions, modifications, magnifications... -- Rodrigo Tetsuo Argenton m 12:02, 2 February 2017 (UTC)

January 28

question about pictures supplied by subject

I edit (under the same name) and whilst I regularly challenge and remove COI or promo content in articles, and fully subscribe to the belief that a subject or connected parties should not contribute to their article other than at arms length via requests, with full declarations, I do feel that many of our profiles on a range of subjects lack photos. I know many people would happily contribute to their own article, and actually supplying a photo is a contribution of great benefit that in no way impinges on any of our core principles of neutrality/POV etc. I also remember hearing andy mabbett talk at wikimania 2014 about a similar project he spearheads for subjects' voices and wonder if this is something that has been explored in the past for photos? Would we encourage people to add photos of themselves? Would that just result in a lot of professional or inappropriate shots or something else undesirable? Sometimes people do add their own pics whilst adding their own content (e.g.). I ask this because anecdotally, I feel there's a sincere lack of photos in many articles (e.g. this British MP, by all accounts an important person; no photo en:Thangam_Debbonaire) - perhaps the single issue where I feel wikipedia perhaps is not an all-encompassing source of information and knowledge. I'm not sure of the stats across the project (or across the other projects) of articles/profiles without pictures where one would be expected or desired, but based on my browsing I feel we could do with many more. Based on my experience of people not understanding that wiki is editable by anyone, let alone the intricacies of the licensing and upload process, there are probably a LOT more photos out there not covered by commercial licenses that people would happily provide that they/we can add and this is something I would support. I would be willing to burden all/most/some of the work in getting something off the ground, however I need to ask this because my experience on the media project is very limited and perhaps this is an old discussion dismissed for reasons that eventually became apparent. Rayman60 (talk) 01:34, 2 February 2017 (UTC)

I don't know of any guide for contributing such photos. They are obviously in scope for Commons, but licensing is an issue; if it's a selfie, verifying that the uploader is actually the person in the photo; if somebody else took the photo, verifying that it has a license from the copyright holder. In both cases OTRS can be used. --ghouston (talk) 03:52, 2 February 2017 (UTC)
I wrote en:Wikipedia:A picture of you. It's on Wikipedia deliberately, as that's the "brand name" the general public best recognise. It also avoids (I hope) any Wikimedia jargon. Please feel free to reproduce it here and/ or other projects, and to translate it. Andy Mabbett (talk) 13:42, 2 February 2017 (UTC)
@Pigsonthewing: I edited the article slightly (this is, of course, what happens in a wiki) but I figured that I would do the courtesy since it's a personal essay. I realize that you want to keep it very simple and friendly but I think a small caveat is okay. —Justin (koavf)TCM 19:05, 2 February 2017 (UTC)

How can I serve Commons by YouTube

Hello.I'm recording short reflections and uload them to YouTube (With timedtext).can I serve Commons by this method?Thank you --ديفيد عادل وهبة خليل 2 (talk) 13:09, 2 February 2017 (UTC)

Hi, Please see COM:SCOPE. I think that these are out of scope. Regards, Yann (talk) 13:31, 2 February 2017 (UTC)
@Yann: Is it not possible to find any benefit from this method?Thank you --ديفيد عادل وهبة خليل 2 (talk) 13:37, 2 February 2017 (UTC)
@ديفيد عادل وهبة خليل 2:; Thank you for your offer, but Commons is not the place for personal content such as this. Please read COM:SCOPE. Ubcule (talk) 18:54, 2 February 2017 (UTC)

How to split my User:Piotrus/gallery?

For years, Commonist has been dumping all my uploads there, and it is now to big for me to open. I wonder if there is any bot or script or such that could split it into subpages by year? --Piotr Konieczny aka Prokonsul Piotrus Talk 19:44, 2 February 2017 (UTC)

@Piotrus: Here is a link to directly edit if you are having browser problems: I would recommend maybe copy and pasting all of that into a text editor and saving the page as blank. Then off-line, figure out how many items you want per page and come back to Commons to edit them. I'm 100% certain a tool could help with this but I simply am too ignorant about that. This is a quick fix in the moment if you want to try it. —Justin (koavf)TCM 20:01, 2 February 2017 (UTC)
You have yourself to move the content year by year. Open the edit link provided by Justin. For me, it took some time, but in the end it loaded. Create subpages per year (service links: User:Piotrus/gallery/2013, User:Piotrus/gallery/2014, User:Piotrus/gallery/2015, User:Piotrus/gallery/2016). In the edit pane of your main gallery page first search for the first section entry with 2013 = (the equal sign is needed to distinguish it from files with this mark), and copy this section and all below into the opened empty page for 2013, save it, then delete in the main pane. Do it in the same way for the further years. Do this in future once in every year, or change the according setting in commonist year by year (I think I would prefer to do this in Commons.). — Speravir – 01:10, 3 February 2017 (UTC)

February 03

Need help concerning {{#switch}}

I've been wanting Template:NASA Photojournal and Template:NASA Photojournal with coordinates to have a "mission" parameter that both changes the image displayed on the right column of the infobox to a mission insignia, and have that insignia link to a corresponding mission page. For this, I had been using {{#switch}}. I had created a subtemplate, Template:NASA Photojournal/missions, in which I had copy+pasted the switch template from Template:Infobox album/link on the English Wikipedia, and simply swapped out the names for mission names, and have them result in file names.

Originally I had the mission names result in a complete file template in this particular revision,

|Curiosity|Curiosity rover|MSL|Mars Science Laboratory=[[File:Mars Science Laboratory mission logo.png|center|120px|link=Wikipedia:Curiosity (rover)]]

and then later, I had them instead result in just the filename, as of this revision, and have a second mission parameter placed in the "link" paramater of the file template,

|Curiosity|Curiosity rover|MSL|Mars Science Laboratory=Mars Science Laboratory mission logo.png

Both of these times did not work. What exactly am I doing wrong, do you guys think? Essentially, the end goal of this is that I want this particular line in the NASA Photojournal templates,

| rowspan="4" style="background-color:#f7f7f7; width:10%;" | [[File:Jet Propulsion Laboratory logo.svg|center|120px]]

To turn into this,

| rowspan="4" style="background-color:#f7f7f7; width:10%;" | [[File:{{NASA Photojournal/missions|{{{mission|}}} }}|center|120px|link={{{mission|}}}]]

This is what I was aiming for in my second attempt. I want the image on the right of the infobox to reflect the mission the particular file was produced by, displaying their insignia and having it link to the corresponding Wikipedia page. Something like this. I want to be able to have a single "mission" parameter, where I can simply type in the mission name to invoke this, rather than have two parameters for a file name and a mission name. Philip Terry Graham 04:06, 3 February 2017 (UTC)

The "lc:" applied to the argument converts it to lower case, which you are comparing with names like "Cassini", so it will never match. I removed the lc: and now it produces some output. If you instead want the parameter to be case-insensitive, put the lc: back but compare with lower case strings only. --ghouston (talk) 05:50, 3 February 2017 (UTC)
@Ghouston: Wow, I should've realised that.... haha! Thanks for pointing that out to me. Everything is working smoothly as intended now; thanks for the help! It's really appreciated! :) Philip Terry Graham 06:04, 3 February 2017 (UTC)

Template for media needing translation checked?

Yes check.svg Resolved

I recently created a German version of an SVG file I made some time ago, and ended up being forced by an upload error to render the text as paths rather than as editable text. I am not a German speaker, and am not certain my translations were correct, but since the text is no longer editable, a German speaker will now have a very hard time correcting what I wrote. I am willing to make corrections to the German text if/ once a German speaker comes along and has recommendations for changes, but I wanted to flag this file as one needing to have its translation checked— and I could find no template for doing this. Do we have such a template?? Please advise. Thanks! KDS4444 (talk) 20:42, 30 January 2017 (UTC)

@KDS4444: This is a crappy fix but you could use {{Update}}. —Justin (koavf)TCM 21:12, 30 January 2017 (UTC)
TC Yeah, that is a dubious solution, but no one seems to have any other ideas for me, so I went ahead and used it— thank you! (Maybe someone (maybe me) should put together a template that will suit the job better?) KDS4444 (talk) 20:30, 3 February 2017 (UTC)

January 31

Suggestions for improvements for Wiki Loves Monuments


The Wiki Loves Monuments international team is working on preparing the next edition. After many years of organising the contest, we believe there are still a lot of ways to make the project run even smoother, for everyone − organizers, participants, and of course the Wikimedia Commons community.

We welcome your suggestions regarding WLM. What specific improvements would you suggest? What do you love most about WLM that we should keep doing?

Please share your suggestions on Commons:Wiki Loves Monuments 2016/Feedback.

Cheers, Jean-Fred (talk) 21:52, 3 February 2017 (UTC)

Copyrighted or public domain

On Flickr I came across this abum. These images would be a valuable addition if they are indeed public domain. It appears the owner of the images was the one who published them to Flickr liscensed to allow commercial use and modification. However, on the images themselves he included a watermark indicating a copyright. What is your judgement? Is the liscense on Flickr insufficient evidence of public domain status due to the copyright mark on the images? Or do you believe the liscense with which it was uploaded onto Flickr (uploaded by what, for all extents and purposes, appears to be the account of the copyright holder as indicated by the watermarks) appears to negate the copyright on the watermark?
SecretName101 (talk) 23:07, 3 February 2017 (UTC)

These photos are copyrighted, but licensed with cc-by-2.0, so they should be fine to upload to Commons. --ghouston (talk) 01:31, 4 February 2017 (UTC)
Thank you.
SecretName101 (talk) 03:22, 4 February 2017 (UTC)
I was utilizing the media in that album as an example. I was uncertain what the general protocol would be in similar instances. I appreciate your chiming-in.
SecretName101 (talk) 03:27, 4 February 2017 (UTC)
  • Watermarks are discouraged, but permitted. Do put {{Watermark}} on such images when uploading. - Jmabel ! talk 05:54, 4 February 2017 (UTC)

February 04

Wiki Loves Music

An alto flute in G by Yamaha (model YFL-482H).

Hi! Together with some other German Wikipedians, I have created a project called Wiki Loves Music to improve the quality of musical instrument images on Wikipedia. I would like to ask two questions:

  • What is the best place on Commons where I can reach out to people interested in creating high-quality photographs of objects like this one? I would like to get some feedback because I have no real previous experience with creating images for Wikipedia.
  • Do you think it would be a good idea to nominate this or other images from the category Wiki Loves Music as featured image? A particularly good one seems to be this picture of a clarinet.

Thank you, --Gnom (talk) 14:48, 3 February 2017 (UTC)

One thing people involved in Wiki Loves Music can do is to support and latter help organize Commons_talk:Photo_challenge/themes#Musical_instruments. --Jarekt (talk) 21:22, 3 February 2017 (UTC)
That's a good start, thank you Jarekt! --Gnom (talk) 11:17, 4 February 2017 (UTC)
Hi Gnom thanks for your ping there and thank to Jarekt for informing you. I looks that you just "rebranded" a standard project about musical instrument with attention to media file (which is fine). About the "like" aspect, if this also has some "social undertones", I was asking around and apparently some users were thinking about creating a m:Wikimedia user groups about "wiki"-artists. We have m: Art+Feminism User Group but is very specific but the idea was to create a group about all performing arts, inclusing music. In any case also an user group of wikimedian interested in music and musical instruments could be interesting and have enough critical mass. It could be a way to organize plays and similar events at wikimedia gathering (so far the collaboration on this issue has always been fragmented), but also to organize international competition to get media file similar to "WIki loves X" (think about not just image of instrument, but whole concerts). An in the end I think that for a manufacturer or the organizer of an event could be a much better partner to interact than a national chapter. Just telling you, so you know.
About the challenge, tell us what could be the best moment for example and we schedule it. I inform "local" projects as much as I can if I have the time, so we welcome synergy with them. We especially need expert users for better categorization and translation of descriptions and rapid reuse of files, many are uploaded by newbies with few expertise of commons.
FInally, take a look at Commons:Structured_data/Overview. When the images on commons will be organized in a more "efficient" way it will be much more easy to target missing items and create standard descriptions in many language. I hope that at least one of the de-N user expert in musical instrument could take a look in the future and act as a bridge with his/her local community. Bye!--Alexmar983 (talk) 11:56, 4 February 2017 (UTC)
Hi Alexmar983, thank you for your great thoughts!
First, you are correct that the name of the project is a little too broad for the relatively small focus. For example, there is another project on German Wikipedia aimed at taking pictures of photos at rock concerts called de:Wikipedia:Festivalsommer.
Next, there is definitely a "social" aspect to this project. At the end of June, we are planning a musical instrument editathon at a Museum with a collection of historic instruments in Hamburg, see de:Wikipedia:Wiki Loves Music/Hamburg 2017, and more than 25 Wikipedians have already showed interest in attending.
Accordingly, do you think we could schedule the photo contest maybe for the first week of July of 2017?
I very much hope that we will be doing some chamber music at the weekend in Hamburg, and maybe we can showcase our project with some live music at de:Wikipedia:WikiCon (maybe even at Wikimania?).
Finally, we have reached out to the Structured Data folks and they told us that for now, all we should care about at this point is creating good categories for our pictures, which I have been doing a lot recently (e.g. Category:Boehm system clarinets).
Again, thank you for sharing your thoughts! --Gnom (talk) 12:50, 4 February 2017 (UTC)
Gnom the challenge is always one month, or more (see later), which can be July. than you have all the next weeks to decide with more calm which pictures to upload. It's max. 4 per person, but something between 25 and 100 is a decent number as a core group of uploaded media files. It's good for a challenge to have a solid base, when it is very specific. So, there is no specific reason not to do it, no problem for me. And I suppose also no problem for all the other "organizers" (we are three or four names) or voters to the ideas (ca a dozen names). If the number of upload is limited, it is possible that the competition will be prolonged to August but that's no a problem too. So yes, we can do it. I will leave a message in the theme talk pages and if noone opposes at the end of June we create the page and than it's official.--Alexmar983 (talk) 13:03, 4 February 2017 (UTC)
Oh wow, so it would be all of July? That would be really cool. --Gnom (talk) 16:39, 4 February 2017 (UTC)

CQ Roll Call Photograph Collection from the Library of Congress

I recently uploaded this photo to the Commons. It comes from here at the United States Library of Congress. The photo was taken by Maureen Keating of either the Congressional Quarterly (CQ) or Roll Call publications from The Economist Group. Per the deed of gift, The Economist Group dedicated to the public all rights it held for the photographs from the CQ Roll Call Photograph Collection (of which this photo is a part) to the public domain (see here). Unfortunately, the Commons does not appear to have a specific copyright tag suitable for this collection. In lieu of one I have added a CC0 1.0 tag, because it seemed most appropriate. Is this enough? Should I use a different tag? -Indy beetle (talk) 01:12, 4 February 2017 (UTC)

CC0 is only valid if applied by a copyright holder. I think Commons is lacking a tag that says that something has been declared to be public domain, but without a fallback license; I don't know if there's any obstacle to creating one. It would be like {{PD-Author}} but without the statement "{{{1}}} grants anyone the right to use this work for any purpose, without any conditions, unless such conditions are required by law." If the deed of gift is anything other than a simple dedication to the public domain, then a custom template would be useful. --ghouston (talk) 01:38, 4 February 2017 (UTC)
we should not be deleting items to force a custom license creation. rather we should be making it easier to reflect good faith attempts to release items into the public domain. this was the practice before CC licenses. the anti-"public domain" jihad tends to undermine the credibility of commons. Slowking4 § Sander.v.Ginkel's revenge 11:36, 4 February 2017 (UTC)
I've added a new LoC license template for the CQ Roll Call Photograph Collection at {{PD-CQ Roll Call}} and updated the license section of File:Ted Kennedy 1999.jpg to use the new license. —RP88 (talk) 01:57, 4 February 2017 (UTC)
Thank you for the template! -Indy beetle (talk) 02:26, 4 February 2017 (UTC)

File:Writings of Henry David Thoreau (1906) v7.djvu

@Taivo: Re: Commons:Deletion requests/File:Writings of Henry David Thoreau (1906) v7.djvu I believe this file was recently deleted erringly (I may have made a mistake with tagging files). It is a working file on Wikisource: s:Index:Writings of Henry David Thoreau (1906) v7.djvu, and should not have been deleted here. A similar file was marked as a duplicate, but this file should have remained. Apologies for the confusion. Thank you! Londonjackbooks (talk) 12:33, 4 February 2017 (UTC)

✓ Done Londonjackbooks (talk) 14:02, 4 February 2017 (UTC)

Why we don't have "permission" in Upload Wizard

I never realise that we do not have the permission label in Wizard, someone know why? Or if it's hidden... -- Rodrigo Tetsuo Argenton m 02:36, 25 January 2017 (UTC)

  • Not sure I follow you. It asks whether it's your own work, and gives the alternative "Use a different license". Or are you just remarking that it doesn't give you the opportunity to explicitly filly out the "permission" field of {{Information}}? I'm pretty sure the latter is deliberate: an assumption that the few percent of people who need to do something more complex than just provide a license can either use Special:Upload rather than the Wizard, or can go in and edit after the fact, rather than confuse the bulk of uploaders with a rare issue. - Jmabel ! talk 16:06, 25 January 2017 (UTC)
    • UploadWizard does allow you to fill in custom licenses. You have to select (on the "Release rights" screen) "This file is not my own work." → "Another reason not mentioned above". Matma Rex (talk) 12:26, 26 January 2017 (UTC)
      • Matma Rex Even in "Another reason not mentioned above", we do not have "permission", that's the problem.-- Rodrigo Tetsuo Argenton m 23:14, 1 February 2017 (UTC)
        • @Rodrigo.Argenton: I'm sorry, I don't understand what you mean. Matma Rex (talk) 10:13, 2 February 2017 (UTC)
          • Matma Rex if we want to upload something like File:Chladni plate 14.jpg via Wizard, we need to do another edition to add the "permission" label as this. For one file, I prefer to use the Upload basic than Wizard, however we can't do a massive upload using the basic, so we use Wizard and then we need to perform some batch task... My questions is: there is a way to skip the batch, and add this line directly via Wizard, or just third party uploaders? And why we don't have this option in Wizard, if not possible. -- Rodrigo Tetsuo Argenton m 11:57, 2 February 2017 (UTC)
  • The Upload Wizard is designed only for the most common use cases. More specialized cases, such as batch uploads for GLAM partnerships, usually require 3rd party tools or custom uploading scripts. Kaldari (talk) 02:40, 6 February 2017 (UTC)

Names Differing

There is a difference in the names of this car by market.Is it better to merge categories:

?Thank you --ديفيد عادل وهبة خليل 2 (talk) 09:58, 4 February 2017 (UTC)

  • I'd say keep both and make a parent category, something like Category:Nissan Navara or Frontier; add some text to all three categories explaining the situation. But someone else may have a better idea. - Jmabel ! talk 17:05, 4 February 2017 (UTC)
I have created the following categories:
Category:Nissan D40 platform
Category:Nissan D23 platform
That one was already existing:
Category:Nissan D22 platform
I think it's the best solution.
--Carnby (talk) 10:42, 5 February 2017 (UTC)

Category tree

I can see there is a different system of creating categories for polish villages and czech villages. Is there any policy or help to describe, how to create category tree?--Juandev (talk) 16:47, 4 February 2017 (UTC)

  • Probably not. If you can, follow the pattern of the most similar categories (i.e. use one from the same country as your model). I suspect this would be a hard one to get consistent: lots of existing categories, lots of creators, not a lot of people working on both countries. But I suppose you could try to drive something if you really care. - Jmabel ! talk 17:20, 4 February 2017 (UTC)
    • Well its hard for me to study with a new country new system. But anyway thanks for your help.--Juandev (talk) 09:59, 5 February 2017 (UTC)


These two images

are almost the same: I think one of them could be deleted but I'm not sure about. Another problem is that Symbol code ww 26.svg belongs to a complete series of symbols made by -xfi-, while [[:|Symbol Shower13.svg]] was created first, and so probably should be kept.--Carnby (talk) 10:27, 5 February 2017 (UTC)

Just start a deletion request if you think an image should be deleted. The merits can be discussed there. In this case, I'd suggest to delete the older file if it is not part of a series. Also, the older file is not in use as opposed to the younger one. Sebari – aka Srittau (talk) 10:53, 5 February 2017 (UTC)

February 06

Upload wizard error

Screenshot of upload wizard error.jpeg

As can be seen in the above screenshot, the upload wizard is failing for me. It is doing so consistently, and has been for several I am using Firefox (50.0.1 - the up--to-date version) on a high-spec Windows10 machine, The problem persists, even after I restart my machine. Even if I try to upload more than one image at once, it reports "1 of 1 file uploaded" and does not show the "continue" button. Andy Mabbett (talk) 16:05, 4 February 2017 (UTC)

This is not the best place to report this. I've filed a Phabricator ticket on your behalf. Ijon (talk) 00:47, 5 February 2017 (UTC)
Thank you. I'll watch for updates there, but I often find replies here are useful in resolving technical issues. I also note that the head of this page says "This page is used for discussions of the operations, technical issues, and policies of Wikimedia Commons (emphasis added), with no mention of Phabricator. Perhaps that should be changed? Andy Mabbett (talk) 12:44, 6 February 2017 (UTC)
Well, yes you can report issues here, but the developers are less likely to notice them. Other non-developer community members will have to report them to Phabricator anyways to get attention from the developers. Even if a developer see an issue reported here they will likely copy it to Phabricator anyways, as it is usually how development work is tracked nowadays. --Zhuyifei1999 (talk) 13:15, 6 February 2017 (UTC)
@Pigsonthewing: Thanks for reporting this. (Thank you for taking the screenshot especially, this is always very helpful.) Are you using Firefox? This sounds similar to an issue recently reported by Schwede66 at Commons:Help desk#Upload Wizard stopped working. Can you look at my instructions there, and confirm if you see the same error message in the console? Matma Rex (talk) 13:27, 6 February 2017 (UTC)


@Matma Rex: Yes, Firefox - as noted above ;-)

Here you go:

Extended content
Gadget "popups" styles loaded twice. Migrate to type=general. See <>.  Special:UploadWizard:1850:45
Gadget "Cat-a-lot" styles loaded twice. Migrate to type=general. See <>.  Special:UploadWizard:1850:182
This page is using the deprecated ResourceLoader module "jquery.ui.position".  load.php:77:291
This page is using the deprecated ResourceLoader module "jquery.ui.widget".  load.php:121:887
This page is using the deprecated ResourceLoader module "jquery.ui.core".
Please use "mediawiki.ui.button" or "oojs-ui" instead.  load.php:20:84
Use of Mutation Events is deprecated. Use MutationObserver instead.  RemoteAddonsParent.jsm:645:4
NS_ERROR_UNEXPECTED:   load.php:47
	bound  self-hosted

Thank you. Andy Mabbett (talk) 14:24, 6 February 2017 (UTC)

@Pigsonthewing: Right, this looks like the same issue. Try disabling Firefogg. I'm afraid no one is maintaining it anymore, and Firefox is about to stop supporting it ( and I just filed phab:T157319 to investigate replacing it with something else. Matma Rex (talk) 15:20, 6 February 2017 (UTC)
@Matma Rex: That's fixed it. Thank you. Andy Mabbett (talk) 16:04, 6 February 2017 (UTC)

Way to search uncategorized images only, by keyword?

Through keyword searches I've found several uncat images to add to Category:Cannabis, and I'd like to round up more. Is there any way I can search only uncategorized images for keywords like "cannabis/hashish/marijauana/marihuana" so I can find images applicable to my WikiProject to rope into the proper cats? Goonsquad LCpl Mulvaney (talk) 04:18, 6 February 2017 (UTC)

Include "media needing categories" (with quotes) or "media needing category review" to your search query. For more detailed searches incategory: search operator can be used (you can use also "-" not operator). MKFI (talk) 07:32, 6 February 2017 (UTC)
@MKFI: thanks that helps already! I don't understand "search operator" though, what do you mean by that? I'll continue filing uncat images for my topic, so I'm already on a roll. Goonsquad LCpl Mulvaney (talk) 03:39, 7 February 2017 (UTC)
@Goonsquad LCpl Mulvaney: see Help:Searching. For example you could use search operator "-incategory:Cannabis" to not include images which have already been categororized (note that subcategories are still searched). MKFI (talk) 07:19, 7 February 2017 (UTC)

Suggestions for Deletion Requests

I made two suggestions on how to improve wording/handling of deletion requests:

Please discuss on those pages. Thank you! Sebari – aka Srittau (talk) 10:35, 6 February 2017 (UTC)

Wikimedia Foundation is hiring community members as strategy coordinators

Hello all! At the moment, the Wikimedia Foundation is hiring 20 contractors - 17 strategy coordinators for specialized languages and 3 Metawiki coordinators. I was am posting this on your noticeboard to reach out to any community members who would be both interested in being a part time contractor with us for three months and a good fit for any of the movement strategy facilitation roles. Even if you are not personally interested in the position, we would appreciate your assistance in encouraging community members to apply, either individually or with local wiki announcements. You can find the Job Description for the position at this page. There is a less-formal description of the tasks they would be working on here on Meta. Kbrown (WMF) (talk) 19:07, 6 February 2017 (UTC)

Tech News: 2017-06

19:45, 6 February 2017 (UTC)

I particularly like the "Script error" caption in Good change, though. Sebari – aka Srittau (talk) 21:43, 6 February 2017 (UTC)

February 08

Why did these deletion requests end without waiting 7 days?

Wikicology said "No valid reason for deletion". However, I think that there is still room for discussion. The text on these description plates may have copyright.--G I Chandor (talk) 03:39, 6 February 2017 (UTC)

You are right, these should not have been closed. I reopened them and warned the closing user. Sebari – aka Srittau (talk) 10:19, 6 February 2017 (UTC)
  • Pictogram voting comment.svg Comment - Thanks for the heads up Srittau. I apologize for hastily closing the discussion but it seems pretty obvious that the text on the plate is below COM:TOO. I am not going to comment on the discussion since I already closed it earlier. All the best. Wikicology (talk) 11:48, 6 February 2017 (UTC)
Sebari, thank you.--G I Chandor (talk) 03:07, 7 February 2017 (UTC)

Commons:Deletion requests/File:天秤櫓(重要文化財)の説明 (31737146871).jpg is not correctly linked to Commons:Deletion requests/2017/02/02.--G I Chandor (talk) 03:09, 7 February 2017 (UTC)

Fixed. Sebari – aka Srittau (talk) 09:38, 7 February 2017 (UTC)

RFC for videos of speeches

Some high profile videos of speeches have been recently deleted from Commons, putting many others under threat of deletion, and with implications for Commons to host presentations or speeches in the future. In order to assess consesus on what "significant doubt" means in these circumstances, I have posted a RFC at:

Feel free to chip in with views, or to support one of the alternatives.

Thanks -- (talk) 18:55, 8 February 2017 (UTC)

February 09

New aide-memoire for the cartography of international treaties

I'm updating the maps illustrating the participation of the countries to international treaties for a while. I noticed some problems in the maintenance : the sources of data are not as constant as they could be, new users are editing the maps with graphic editors and are breaking the CSS instructions, etc. Moreover, numerous subnational entities are requiring further research and verifications, and I suspect that no map currently uploaded is perfectly reflecting the real status of application of a treaty in all these entities.

So here is what is aimed to be a comprehensive guide to illustrate the treaties : Commons:Map resources/Aide-memoire to illustrate status of treaties. Please help me propagating the guide and completing the list of instructions for each entity based on this list. Louperivois Ψ @ 05:16, 1 February 2017 (UTC)

CC User:Abjiklam, User:IdiotSavant, User:Popo le Chien, User:Allstar86, User:ZaidRock11, User:Крокозябрус, User:Mimid, User:Turnless, User:L.tak, User:Anastan

Louperivois Ψ @ 05:16, 1 February 2017 (UTC)

Thanks, good idea! I have added some things (and think it still should be a bit easier): main point is that most treaties we cover are not UN treaties... (but in time, we may have to make list on reliable pages of often used depositories (UN, ILO, WIPO, EU, Netherlands, Switzerland...)... I wouldn't focus on the "real status", because virtually on all states "it's more complicated" then a dependent territory status, and -reasonable, but sometimes arbitrary choices- were made that we can use based on the specific treaty. What I am missing is the abiity in the worldmap to colour subnational entities that may become party to treaties (the territories/provinces of Canada in many cases; and in one case even Schotland!). Who would be able to fix that? L.tak (talk) 21:59, 1 February 2017 (UTC)
Thank you! There is this map with all secondary divisions, but the code is not clean at all (no ISO for subnational items, full of Inkscape crap) and the size of the file is not worth it for using it everywhere. Living in Canada, I don't see a lot of cases (eg. Francophonie whose Quebec is a full member) where it is useful. In these cases the mapmaker could replace the SVG block of the entire country concerned by the subdivised one of the other map (when the code will be cleaned and will use ISO 3166-2:CA). Louperivois Ψ @ 22:38, 1 February 2017 (UTC)
Nice initiative. I can try to help cleaning up the code for national subdivisions. Do you know which countries to prioritise? Also, besides a clean code and ISO tags, is there something else that is needed for a subdivision map? Abjiklɐm (tɐlk) 23:44, 7 February 2017 (UTC)
@Louperivois, L.tak: You can use File:BlankMap-World6-Subdivisions.svg. For now, only Canadian and British subdivisions are included since those are the only countries you mentioned, but let me know which others are needed. For the UK, both countries/provinces and nations are included, as mentioned in en:ISO 3166-2:GB. Abjiklɐm (tɐlk) 01:46, 8 February 2017 (UTC)
I forgot to mention, the subdivisions must be called by id in the css. That means you must use #ca-qc, not .ca-qc. Abjiklɐm (tɐlk) 01:50, 8 February 2017 (UTC)
@Abjiklam:, thanks, that helps a lot!!! I asked for this a couple of times at map workshop wiki/commons pages, but to no avail... Your chosen subdivisions seem the most relevant to me, as those have relatively much autonomy (also for age of consent-svg's and driver's licence minimum age svg's this would need those territories mostly)I have made two maps now: File:ProtectionOfAdultsConvention.svg where Scotland is involved, and File:HagueTrustConvention.svg applying to 8 provinces of Canada). Others that may be relevant are Brazil and Mexico with regards to same sex marriage etc. My priority would be the possibility to make the internal borders invisible: either for all subdivisions, or (preferably but complicated) to show subdivision-borders only if subdivisions are called in the colouring... For the w:Hague Trust Convention for example, the US signed as a single entity (so no borders desirable), while canada extended it to 8 provinces only (no internal borders needed, but nice to have). L.tak (talk) 21:00, 8 February 2017 (UTC)
@L.tak: Actually I meant for this map to be just a repository of subdivisions. Basically, like Louperivois said, if you need a map with only Canadian provinces, you could take the BlankMap-World6 file, and replace the Canadian code block with the equivalent from BlankMap-World6-Subdivisions. Not only does that ensure you only have the subdivisions you need, but it also keeps the file size only as big as it needs to be. Maybe there is a way to make a tool to automatize the process, but for now it would have to be done manually. Abjiklɐm (tɐlk) 15:57, 9 February 2017 (UTC)

Excessive taxonomic information in categories

Screenshot of Category:Myrmarachne

If I visit a category like Category:Myrmarachne (permalink), I have to scroll down 3 pages before I actually get to see any images. This amount of taxonomic information seems excessive to me, especially because it seems unlikely that anyone is actually going to maintain it. Do we really want Commons to have more detailed taxonomic information than Wikipedia, Wikispecies, and Wikidata? It seems out of scope to me and likely to rot after a few years. This isn't an isolated case, either. There are dozens (maybe hundreds) of categories that have been set up like this. Should this be discouraged? Should the species lists be removed? Kaldari (talk) 01:39, 6 February 2017 (UTC)

I don't have any comment on whether or not something should be changed about this, but you (or any other editor) can have all those lists automatically display collapsed by going to Preferences/Gadgets/Improved navigation/CollapsibleTemplates, and enabling that option. Daphne Lantier 02:38, 6 February 2017 (UTC)
Kaldari (and let me ping @Liné1: here) I think that moving forward we should concentrate on moving content of our taxonomic pages to Wikidata, and pulling it from wikidata for display purposes with something resembling d:Module:Taxobox. That is the only viable option for long term maintenance. --Jarekt (talk) 19:32, 6 February 2017 (UTC)
@Kaldari, Jarekt:
Interesting that, a contributor with a scrolling-ache.
About the importance of the data: With those species list any contributor can know if his species Myrmarachne tagalica exists. In case of M. tagalica, you can see in Category:Myrmarachne that it is recognized by 4 of the 5 sources, which seems a good confirmation. Before the introduction of those templates, contributors could create categories without prooving their name! References were only for wikipedias, not for wikicommons.
About the format, {{Species2}} and {{Genera2}} are supposed to be collapsing. It works on firefox and IE, not on certain version of chrome. Sorry about that.
Regards Liné1 (talk) 19:52, 6 February 2017 (UTC)
This way lies madness. What should be done is that templates should call the lists of species (or whatever) from Wikidata. The data then only needs to be maintained on that project. Andy Mabbett (talk) 21:35, 6 February 2017 (UTC)
maybe you should talk to user:Josve05a, he's the one who added them. [6]
i agree needs to be on wikidata, but the not playing nice with other projects is not confined to english wikipedia. Slowking4 § Sander.v.Ginkel's revenge 03:19, 8 February 2017 (UTC)
Currently WIkidata does not store lists of species under a specific taxa - you can only query a list of subtaxa which have items already created on Wikidata, which is greatly lacking. As long as there is no good way t link such things, I'll keep adding long lists and templates to the cats, since it does not hurt, and can easially be removed once a better tool/template exists.(tJosve05a (c) 03:26, 8 February 2017 (UTC)
why don't you take up your critique of wikidata there? as we know categories are broken here, and this "custom snowflake" is one symptom. you can hack commons to do a wikidata function, since you have more control on commons, but the harm from the lack of collaboration is clear. Slowking4 § Sander.v.Ginkel's revenge 19:32, 9 February 2017 (UTC)

Babels, assessing differences?

Hello! I'm a bit puzzled about the options for babel templates. What are actually the differences in the abilities of using a language between a native English speaker and those who assessed themselves as "en-4" and "en-5"? And is there a reason to have the level five only for English (I didn't see it for German, French, Russian or Spanish, for instance)? Hopefully, somebody has some clues about this subject. Regards, Grand-Duc (talk) 14:26, 9 February 2017 (UTC)

  • de-5 definitely exists (I tried it). I presume the others do, too.
  • My take on the rest:
    • en-5 would be distinguished from en-N only in that one is not actually native. For example, I have a Romanian friend who has lived in the U.S. over 30 years and whose written English is as strong as virtually any native speaker; I have another U.S.-born friend who Japanese is good enough that on the phone no one would know he was not Japanese; and a U.S.-born cousin of mine has worked as an editor in Norway. They would be at the 5 level in those languages.
    • en-4 would be very strong but short of native level. Possibly not native-level on figures of speech, might occasionally not come up with the most felicitous word, etc. They probably read at a native level, but might fall a bit short on writing. (Speaking is basically irrelevant for our purposes, since we are dealing almost entirely in writing.)
  • I hope that helps. - Jmabel ! talk 15:52, 9 February 2017 (UTC)
  • Pictogram voting comment.svg Comment en-5 indicates someone with a professional command of English, such as a writer, college graduate, teacher, translator, etc. en-4 indicates a non-native speaker with advanced understanding of the language. Daphne Lantier 22:22, 9 February 2017 (UTC)

Change to audio/video transcoding queue

Audio/video transcoding (to produce appropriate output files for video at various resolutions, or converting from WAV/FLAC/Opus to Ogg Vorbis for audio output) now uses two background job queues: a new high priority queue runs audio and SD & low-resolution videos, while the old queue continues to serve any old items plus HD video and all transcodes of long files (over 15 minutes). This should make new uploads of audio and video files much more responsive in terms of being available for online viewing sooner after upload.

It'll take a few days for the old items to run through the queue, and we will likely need to adjust the balance of the queues in response to actual load data. Thanks for your patience! --brion (talk) 20:23, 9 February 2017 (UTC)

Thanks a lot! See Special:TimedMediaHandler. Yann (talk) 22:02, 9 February 2017 (UTC)

February 10


What is the maximum mp of an image to be a featured image. 24 mp Am I right?--Shriheeran (talk) 09:28, 10 February 2017 (UTC)

There's no maximum threshold of size for candidates to featured or quality status, only a lower resolution limit of 2 megapixel. If somebody out there shoots with a digital medium format camera, eg. a Hasselblad boasting 50+ MPx, or has a large rendering output of a CGI software or some large panoramic image, then 24 MPx is not that large... Regards, Grand-Duc (talk) 10:01, 10 February 2017 (UTC)

Manila Light Rail Transit

I created two new categories: Category:Manila Light Rail Transit BN vehicles and Category:Manila Light Rail Transit 1200 series. However I now realise that the distinction between serie 1000 (BN) and 1100 is not easy, as they are painted the same way.Smiley.toerist (talk) 11:26, 10 February 2017 (UTC)

Separate "school" and "school building" categories?

Hi, guys! I notice many Hong Kong articles have separate "Schools in..." (...學校), example Category:Schools in Southern District, Hong Kong; and "School buildings in ..." (...校舍), example Category:School buildings in Southern District, Hong Kong (which instructs users to also add photos to "Schools in Southern District") categories while most cities have it all in the same category. Do you guys think such categories should remain separate or should they be merged together? I'd like to ask the Hong Kong Commons users their opinions on the matter. @Ghouston:@Cyueomchem:@Eddia Liu Sun-Ming: WhisperToMe (talk) 20:57, 10 February 2017 (UTC)

That's a common problem: Should a building and the institution using that building use one or two categories? Often both are closely related, i.e. the building was built for the institution and the institution only ever used that building (common with schools). In those cases, one category might work, especially if none of the files in the category are not specific to the building. In cases where the building was used by multiple institutions or where the institution used multiple building, I prefer multiple categories. Sebari – aka Srittau (talk) 21:30, 10 February 2017 (UTC)
"if none of the files in the category are not specific to the building" or, more clearly, applying DeMorgan's law, "if all of the files in the category are specific to the building". - Jmabel ! talk 22:51, 10 February 2017 (UTC)
My view: if a school has one building (or even if it has one cluster of buildings, none of them of any particular architectural notability) there is no reason to create a separate category for the building. - Jmabel ! talk 22:53, 10 February 2017 (UTC)
Note: We have a similar issue with banks vs bank buildings. See Commons:Categories for discussion/2014/06/Category:ANZ banks in Australia and Commons:Categories for discussion/2013/07/Category:Bank buildings in Portugal. I think the differentiation is more legitimate in the case of banks (or universities) which are more likely to have multiple buildings than schools, but I'm fairly neutral on the issue. - Themightyquill (talk) 23:29, 10 February 2017 (UTC)
It was discussed previously at Commons:Categories for discussion/2011/09/Category:Education buildings. There wasn't any consensus to get rid of "buildings" categories. --ghouston (talk) 00:24, 11 February 2017 (UTC)

Generally, the to category can be "Schools" and if needed, some special types of school buildings (school canteens, school drill-halls, school dormitories etc.) can have their specific subcategories. To create specific subcategories for main school buildings is IMHO counter-productive. As regards individual schools (or banks, museums, parliaments etc. etc.), they can have an unified category, but as soon as we need to distinguish the building-related files from the organization-related files, we can create a subcategory for the main building or all buildings of the institution. --ŠJů (talk) 18:08, 11 February 2017 (UTC)

Somebody decided at some point that since schools are organisations, not buildings, that should be reflected in the category system. Likewise with Category:Housing vs Category:Residential buildings (although I notice now that Housing has found it's way back into Buildings, which defeats that idea). The logic is sound, but in practice for schools it would take a lot of work to actually carry through that scheme right through the category system, and since that hasn't been done, it's inconsistent in different sub-trees. It's also inconsistent with other trees like Category:Hospitals or Category:Hotels, which have the same issue, but in Commons have been retained as subcategories of Buildings. --ghouston (talk) 00:10, 12 February 2017 (UTC)

February 11

Cut-off date for Public Domain files

Since this has been discussed before here, pointer to a new discussion: Commons:Village pump/Copyright#Cut-off date for .7B.7BPD-old.7D.7D. Hopefully this time we can come to a conclusion. Sebari – aka Srittau (talk) 17:27, 11 February 2017 (UTC)

400,000 photographs of archaeological objects found by members of the public in England and Wales

Starting with a suggestion from Pigsonthewing, over the last couple of weeks we have uploaded to Commons the entire collection of photographs released by the Portable Antiquities Scheme (PAS). This is an amazing collection of images to illustrate British archaeology, with finds ranging from the Neolithic axes through to lost oddities of the 18th century. The collection represents a tricky challenge for Commons volunteers, as the quality of both artefacts and photographs varies hugely. Consequently a search for ancient Roman period coins will provide you with both excellent condition examples of gold coins with profiles of different emperors, alongside unidentified "dog biscuit" quality coins which are 'probably' Roman!

So, this both a notice that the collection is now on Commons, and a great resource to reuse and illustrate Wikipedia articles, and a curation challenge to identify the best examples for reuse and ensure these are easily found in useful categories. This is also an opportunity for our amateur archaeologists and numismatists to see if they can do better than the experts, and work out what any of the large number of unidentified objects in the PAS database might be. :-)

Some interesting example searches are:

The Project page includes other example searches, further explanation about the upload process, initial categorization, and links to GLAM dashboard reports. If you help out with categorizing the collection, your name will be immortalized at the project list of volunteers :-) -- (talk) 12:46, 10 February 2017 (UTC)

Thank you, , for the acknowlwdgment, and, much more importantly, for all your good work on this collection. Andy Mabbett (talk) 21:03, 11 February 2017 (UTC)
I'll leave some messages here and there.--Alexmar983 (talk) 11:44, 12 February 2017 (UTC)

Not updated file map

In relation to Commons:Bots/Work requests#Nonsense coordinates added by videoconvert, I found that the map [7] displays this file although the false coordinates were removed at 2015-06-16T17:00:47‎ from the file page.

Regrettably the {{Location}} template and the target map contain no link to authors or admins of the tool or a to a feedback discussion about it. In previous discussions about similar problems, nobody claimed his responsibility for the tool. --ŠJů (talk) 19:08, 11 February 2017 (UTC)

toollabs:contact/search/wiwosm points to @Kolossos, Simon04: & Retsam, except for the last which does not have an SUL linked, both of the other two are still active on de.wp. --Zhuyifei1999 (talk) 19:22, 11 February 2017 (UTC)
The commons files are loaded via @Para: why does [8] contain File:Tres_deu_entrevista_Obrint_Pas.webm although it does not contain any coordinates? – Simon04 (talk) 20:19, 11 February 2017 (UTC)
FWIW, this tool does not seem to have its source code published (afaict) and is very likely without a license. TLSC may look into this tool soon. --Zhuyifei1999 (talk) 09:42, 12 February 2017 (UTC)

February 12

Vandalism or not? How to deal with it?

Please look at Special:Contributions/2600:8800:3080:2C50:194E:E87A:2510:F53C. It took over 70 edits to achieve this: Special:Diff/207519196/233251428. He/she changed the surname of the person a few times (both surnames are OK according to enwiki), changed order of the same words a few times, etc. It could easily be done in 7 (seven) edits and for me it is hard to believe that someone could be so undecided (or simply stupid) and needs 70+ edits. IMO these are deliberate, disruptive actions. It is also worth to mention that I have already seen such behaviour on Commons a few times. So how to deal with it? --jdx Re: 08:07, 12 February 2017 (UTC)

Just ignore it. Ruslik (talk) 08:28, 12 February 2017 (UTC)
I agree with Ruslik. In the end the page is in better shape than at the beginning. --Jarekt (talk) 20:17, 12 February 2017 (UTC)
Yeah, but in the meantime someone has to patrol the changes. BTW. Are you sure that e.g. after these changes [9], [10], [11], [12], [13], [14], [15] made during two minutes, the page actually is in better shape? Face-smile.svg --jdx Re: 20:42, 12 February 2017 (UTC)

Location maps

Does anyone can help me to create a location map? I want only a part of File:Cyprus adm location map.svg. Xaris333 (talk) 16:12, 12 February 2017 (UTC)

have you checked out Commons:Image map resources ? a screenshot of open street map might serve your needs. Slowking4 § Sander.v.Ginkel's revenge 17:59, 12 February 2017 (UTC)

Bugged filename blacklist blocking bugs?

Not bugs (Hemiptera), but weevils (Coleoptera > Corculionidae): I’ve been trying to upload three simple photos from Flickr (my own account) as:

  • Rhynchophorus ferrugineus‎ (estrago geral).jpg
  • Rhynchophorus ferrugineus‎ (estrago no fuste 1de2).jpg
  • Rhynchophorus ferrugineus‎ (estrago no fuste 2de2).jpg

(and similar filenames) and have been getting these rejected. What’s wrong with these filenames? -- Tuválkin 22:45, 12 February 2017 (UTC)

Alas, for some reason the following were accepted for upload, although I’m sure most of us would prefer my 1st choice:

-- Tuválkin 22:54, 12 February 2017 (UTC)

What specific error message did you see when attempting this? And which tool for Flickr uploads were you using? If I view, I see nothing that would prevent me from creating the page (compare to e.g., which shows a clear error message immediately). Matma Rex (talk) 20:41, 13 February 2017 (UTC)

February 13

Works not in the public domain (special extensions)


I see that three works of Jean Moulin are present in Commons and considered as public domain because he died more than 70 years ago. But there is a special extension of copyright for authors who "died for France". Alas, he is so registered in the French defense Ministry database. Should the pictures be deleted?

Thank you for your attention,--Canaricolbleu (talk) 15:57, 13 February 2017 (UTC)

De-Recognition of Wikimedia Hong Kong

This is an update from the Wikimedia Affiliations Committee. Translations are available.

Recognition as a Wikimedia movement affiliate — a chapter, thematic organization, or user group — is a privilege that allows an independent group to officially use the Wikimedia trademarks to further the Wikimedia mission.

The principal Wikimedia movement affiliate in the Hong Kong region is Wikimedia Hong Kong, a Wikimedia chapter recognized in 2008. As a result of Wikimedia Hong Kong’s long-standing non-compliance with reporting requirements, the Wikimedia Foundation and the Affiliations Committee have determined that Wikimedia Hong Kong’s status as a Wikimedia chapter will not be renewed after February 1, 2017.

If you have questions about what this means for the community members in your region or language areas, we have put together a basic FAQ. We also invite you to visit the main Wikimedia movement affiliates page for more information on currently active movement affiliates and more information on the Wikimedia movement affiliates system.

Posted by MediaWiki message delivery on behalf of the Affiliations Committee, 16:25, 13 February 2017 (UTC) • Please help translate to your languageGet help

This noticeboard is for the Wikimedia Commons project. Why was this message, obviously intended for Affiliates, posted here? -- (talk) 16:32, 13 February 2017 (UTC)
because they are mass mail spamming all the projects. leave you feedback here [16]. Slowking4 § Sander.v.Ginkel's revenge 03:33, 14 February 2017 (UTC)

Tech News: 2017-07

18:06, 13 February 2017 (UTC)

California Highway Patrol images

Can someone please confirm that images posted on social media by California Highway Patrol officers, such as those on:

meet the requirements to be tagged with {{PD-CAGov}}? Andy Mabbett (talk) 19:22, 13 February 2017 (UTC)

Ghost category link

I created Creator:Pindar using the automated Creator generator, and added it to Category:Pindar following the model of other similar categories. But I am seeing a [[Category:Creator template home categories|]] displayed as unlinked text below the Creator template, and can't figure out why it is there or how to make it go away. Please help. --EncycloPetey (talk) 21:05, 13 February 2017 (UTC)

@EncycloPetey: Gotcha. Let me know if you need more help. —Justin (koavf)TCM 05:58, 14 February 2017 (UTC)

February 14

from flickr user Simon YANG photos

from flickr user Simon YANG photos [22] maybe copyvio, see Commons:Deletion requests/Files uploaded by KOKUYO, plese help check--shizhao (talk) 06:29, 14 February 2017 (UTC)

see Commons:Deletion requests/undefinedSimon YANG--shizhao (talk) 06:33, 14 February 2017 (UTC)

Template for generic public domain dedications

The question of files that have been released to the public domain by non-Commons users keeps coming up. In these cases the authors haven't accepted specific "fallback licenses" such as on {{Cc-zero}} and {{PD-author}} so it seems that these templates can't be used. I'm thinking about creating {{PD-dedication}}, with the wording "The copyright holder of this work (usually the creator) has released it into the public domain. This may not be legally recognized in all countries. In particular, moral rights may still apply." It has a problem though, that Commons requires a work to be free in its source country, and I don't think there are many countries where the validity of such statements has been legally tested. Can we assume it's OK until proven otherwise? Legal challenges to public domain dedications seem to be rare or non-existent. I suppose they could arise if heirs wanted to enforce copyright or if somebody changed their mind, but perhaps it's not worth worrying about. --ghouston (talk) 23:37, 4 February 2017 (UTC)

Isn't that what exactly {{PD-author}} is? Or am I missing anything? --Zhuyifei1999 (talk) 08:18, 5 February 2017 (UTC)
PD-author includes "In some countries this may not be legally possible; if so: {{{1}}} grants anyone the right to use this work for any purpose, without any conditions, unless such conditions are required by law." That's fine if it's a Commons user and they agreed to that license, but I don't see how you can apply it to files found elsewhere. --ghouston (talk) 11:20, 5 February 2017 (UTC)
We have {{PD-self}} for our users who grant a PD. So I think we can strip-off the "fallback licenses" from {{PD-author}} which is intended to use where author and uploader are different. Jee 11:32, 5 February 2017 (UTC)
I don't think that would be a good idea, because you'd need to check every file using the tag first and change to PD-self if appropriate, and that would be difficult because in some cases it's not clear if the name given to PD-author is the real name of the uploader. It also matches the same template in Wikipedia, so would mess up file transfer. --ghouston (talk) 21:09, 5 February 2017 (UTC)
I think we would have to very carefully consider the ramifications before introducing that template. I'm not really a fan of introducing such a generic tag, given how fact specific public domain dedications can be and the problem that the necessary legal language to successfully effect a dedication might vary from jurisdiction to jurisdiction (exactly the problem {{Cc-zero}} was intended to solve). Just take a look a couple of topics upwards at Commons:Village_pump#CQ_Roll_Call_Photograph_Collection_from_the_Library_of_Congress which resulted in {{PD-CQ Roll Call}}. I trust The Library of Congress, as the agency responsible for overseeing the U.S. Copyright Office, to draft a public domain dedication that is legally binding in the United States. That dedication ended up with a bunch of caveats which we had to be careful to reflect in our template. Despite their reliability, I am not sure I'd extend a similar trust to the Library of Congress that they can draft pubic domain dedications acceptable to non-U.S. jurisdictions for the non-U.S. works in their collection.

Do we even know if just saying "this work of mine is public domain" is legally sufficient in the United States? I've seen some of the dedications that the Library of Congress has accepted, and most, if not all, are more comprehensive. If we think that just saying "this work of mine is public domain" is sufficient in the United States, I might be willing to entertain the idea of introducing a U.S. specific public domain dedication template for U.S. authors who say their work is in the public domain, something like {{PD-US-dedication}}. If we do this I think we'd need to include a warning in the template that the work might be still protected in countries other than the United States. Adding PD dedication templates for other countries has the problem that we require works to be free in both the U.S. and the country of origin, so we'd have to untangle what kinds of public domain dedications we think are simultaneously effective in the non-U.S. country of origin as well as the United States. —RP88 (talk) 23:08, 5 February 2017 (UTC)

I'm wary too, that's why I started a discussion instead of just creating the template. CC0 is far better. The template should have a footer advising to use CC0 instead if it's your own work. This probably would have been a good discussion to have 10 years ago, before the CC licenses became really popular and all kinds of vague free use statements on websites were converted into "PD author". There are plenty of files in Category:PD-author where the source website is dead or changed, the license is unverified and even doesn't help. I think {{PD-CQ Roll Call}} is a simple public domain release with some fine print that it doesn't apply to all files in their collection, and such sources are accepted on Commons. There's no Commons requirement for a US sourced file to be free anywhere outside the US. I already mentioned the problem of files from outside the US, and restricting it to US sourced files would be an option. --ghouston (talk) 06:55, 6 February 2017 (UTC)
"I trust The Library of Congress, as the agency responsible for overseeing the U.S. Copyright Office, to draft a public domain dedication that is legally binding in the United States." don't bogart that joint, pass it over. those are separate divisions. LOC sets the standard for collection copyright disclosure. but what does legally binding mean? there is no case law. rather, it is all in the minds of the wikilawyers. only on commons does anyone care about the PD tag. everyone else thinks it is PD. Slowking4 § Sander.v.Ginkel's revenge 03:25, 8 February 2017 (UTC)
I don't think there's any legal risk to Wikimedia for hosting files with a simple PD release, since there's no reason to think they are copyright violations in the US and therefore there's no risk to the DMCA safe harbor. Image reusers can make their own decisions, although it would be nice to give them the most accurate information possible. I could make a {{PD-dedication}} template and leave it to hypothetical future deletion requests whether a simple public domain release from outside the US should be accepted or not. --ghouston (talk) 04:14, 8 February 2017 (UTC)
i look forward to your keeping PDM from flickr. the idea that the origination from LOC is a lower risk is without evidence. having no case law, you cannot distinguish between flickr PDM and LOC "PD no known copyright". Slowking4 § Sander.v.Ginkel's revenge 19:36, 9 February 2017 (UTC)
On Flickr, anyone who really wants to waive copyright can use CC0, and it has a good chance of being valid outside the US. If they use PDM on their own work, I have no idea what their intentions are, or if they understand the options at all. It doesn't give much confidence for reusing their work. --ghouston (talk) 01:39, 10 February 2017 (UTC)
anyone who is an expert on the deletionist tendency on commons, if they cared. i think i will upload all my items as PDM on flickr, so they will be public domain everywhere in the world except commons. my intent is clear, right? Slowking4 § Sander.v.Ginkel's revenge 13:40, 14 February 2017 (UTC)

Commons Categorizer Meetup 2017 in Montréal

Dear fellow Commoners,

I just submitted a proposal to organize a second Commons Categorizer Meetup at Wikimania 2017. If you are interested in attending this meeting, please put your name on the preliminary attendees list and name discussion topics, you would like to talk about.

Thank you,

--MB-one (talk) 17:46, 13 February 2017 (UTC)

  • @MB-one: The link you give here for Wikimania 2017 goes nowhere. - Jmabel ! talk 00:38, 14 February 2017 (UTC)
@Jmabel: Thanks for the info. It seems to work for me. If someone knows how to fix this, please do so. --MB-one (talk) 09:02, 14 February 2017 (UTC)
So you are saying w:Wikipedia:Wikimania 2017 works for you? Can you give the URL of the page it leads to, because it sure doesn't work for me! - Jmabel ! talk 16:17, 14 February 2017 (UTC)
works for me, but here you go Slowking4 § Sander.v.Ginkel's revenge 02:14, 15 February 2017 (UTC)


I have create File:PaphosDistrict.svg with the help of File:Cyprus adm location map.svg. The coordinates I have add are wrong. Anyideas how can I find the correct coordinates? Xaris333 (talk) 21:05, 13 February 2017 (UTC)

Input needed on category discussions

All interested people are invited and requested to comment on current category discussions. Discussions include proposed deletion, renaming, or reorganizing of categories, among other things. You can look at Category:Categories for discussion to see many of the categories being discussed.

There are some current discussions from January and December that each address multiple categories, in some cases dozens or more: if these proposals are accepted, some would result in changes to frequently-used categories, so why not let your thoughts be known?

There are also many discussions from the past several years that didn't get enough participation to determine consensus, so they're still open. Please consider lookng at these discussions regularly and participating on the ones where you have any interest. Thanks. --Auntof6 (talk) 16:43, 6 February 2017 (UTC)

As I've said before, the CfD process is fundamentally broken, since you can have a lot of images in the categories on your watchlist, but unless you have the category itself on your watchlist, you will never receive a notification that any discussions that could affect categorization of those images are going on. This is what led to the "adolescent girls" category fiasco (which took many years to clean up) and other problematic outcomes... AnonMoos (talk) 18:45, 11 February 2017 (UTC)
@Xxxx: This is why I watch both the discussion pages and Category:Categories for discussion. Doing that shows me both new discussions and categories that are tagged for discussion (those aren't completely redundant because some discussions are for multiple categories). I do have to add each new month's discussion page to my watchlist. Maybe those options would help you. --Auntof6 (talk) 04:43, 14 February 2017 (UTC)
However good it would theoretically be if people on Commons undertook to follow your suggested course of action, it's quite clear empirically that under the current set-up, the great majority of them won't. Therefore the current CfD process is broken in a basic way. AnonMoos (talk) 01:36, 16 February 2017 (UTC)

February 10

Links between Commons and Wikidata (and vice-versa)

Here's an update of results of queries into linking patterns between Commons categories and galleries and Wikidata. (Data from a couple of weeks ago).

For comparison, a previous version was posted here at VP in December 2015; there are also some earlier historical versions, going back to September 2014.

Commons categories
Commons galleries
total linked
Wikidata articles
(~ 22,165,947)
~ 1,268,063 100,042 ~ 1,299,996
~ 1,209,119
~ 1,235,579
Wikidata categories
396,087 558 396,094
total linked 1,426,002 100,086 ~ 1,696,090 items / 1,523,993 pages
props: ~ 1,590,788 items /
1,419,074 pages

The numbers are slightly crude -- "Wikidata articles" includes everything that isn't an item for a category, as does "Commons galleries", and in some cases I wasn't quite able to run the query I wanted (these are marked with the tildes ~). But I think they are not far off.

Compared to 2015, perhaps the most notable thing is that the number of categories that can be identified in some way with an article-like Wikidata item continues to rise -- now up to 1,268,063 from 992,272 (ie up 275,791 since December 2015).

From Wikipedia articles, these Commons categories are now all accessible via sidebar links (regardless of whether the two are connected by sitelinks or the Wikidata P373 property).

In the other direction, only sitelink connections cause Wikipedia articles to show up in the sidebars of Commons category pages. As a workaround, a Commons cat -> Wikidata lookup user-script exists ("wdcat.js"), which can automatically show a to link to the Reasonator tool, which summarises the information about an item on Wikidata including Wikipedia links, if a P373 pointing to the category exists on a Wikidata item. This can be enabled by adding the line importScript('User:Jheald/wdcat.js'); to your common.js on Commons. There is also the template {{On Wikidata}}, which can be added to the top of a category.

A rough breakdown of the P373 links includes 270,445 for people; 223,303 for "administrative territorial entities"; 112,725 for buildings; 48,159 for physical geographic features... etc.

I wrote in 2015, "Getting more of these Wikidata links into place should make it significantly easier for external programs to find the right Commons categories to put images into. The identifications should also become particularly valuable for Commons templates ... [once] Commons templates become able to systematically draw information from all Wikidata items."

This is now starting to become a reality, with template {{Authority control}} now drawing live data automatically from Wikidata, and instances of {{Creator}} templates now able to be generated automatically from a Wikidata item. Jheald (talk) 11:17, 13 February 2017 (UTC)

Discussion (wikidata links)

Jheald thank you for the update. I actually find those numbers quite interesting and 1.2M P373 properties is a great news. It would be great to see this tables with some links to the actual queries that produced the numbers. For example 16 sitelinks of category items pointing to commons galleries are clearly wrong and should be looked at, an already provided query would be helpful. I colorcoded part of your table to indicate: good, potentially problematic, and bad links, following two principles
  1. that in general category items should not have P373/P935 properties, which should be kept in article items and
  2. that there was never consensus about cross-namespace sitelinks between commons categories and wikidata article items, so at the moment it is very unpredictable if a sitelink from wikidata will lead to category or gallery on Commons. One potential issue I see is that as more templates on Commons rely on wikidata, some follow the sitelinks and some look for explicit q-code. The category pages relying on a sitelink might break when someone moves the category sitelink to newly created category item.
One sideeffect of 1.2M P373 properties is that some fraction of them is wrong and should be corrected either on Wikidata or on Commons. d:Wikidata:Database reports/Constraint violations/P373 lists many pages trip some validity test. It would be great if some members of Commons community are actively engaged in fixing reported errors, especially since many of them need to be fixed on Commons. --Jarekt (talk) 15:27, 14 February 2017 (UTC)
Commons categories
Commons galleries
Wikidata articles
(~ 22,165,947)
~ 1,209,119
Wikidata categories

Hi User:Jarekt. The queries can be found here. I couldn't get Autolist to work any more (WDQ retirement ?), so this time the table was all put together with WDQS SPARQL queries. (And in some cases, a calculator). Jheald (talk) 15:45, 14 February 2017 (UTC)

Here's a query specifically for the category item -> gallery sitelinks :
If you do remove a sitelink, please add a P935 (at least for the moment, even though you've coloured that box in orange), so the information is not lost.
It's possible that for some of these category item <-> category sitelinks are blocked by existing article item <-> category sitelinks. I don't necessarily see the latter as a bad thing, so let's hold off from removing any, at least for the moment. Jheald (talk) 16:06, 14 February 2017 (UTC)
Jheald, Thanks for, I cleared the 16 by moving them to properties (of category or article items) and replacing the sitelinks with links to categories. Although I colored all properties if category items as orange, They are OK (in my book) if category item does not have a matching article item. The 545 P935 properties in Wikidata category items should be checked if they are duplicated in article items and possibly removed. that would clear up many "Unique value" violations in d:Wikidata:Database_reports/Constraint_violations/P935. --Jarekt (talk) 17:21, 14 February 2017 (UTC)
Okay, so first a correction: I think I wrote above that if both a sitelink and a P373 were present, then a Wikipedia would show a sidebar link to both the category and the gallery. It seems this is not generally correct, and also different wikis do different things. As far as I can see (except perhaps when specifically over-ridden)
I thought there were at least some projects that tried to link to both, but maybe not. If only Dutch wiki is linking to them, is anybody visiting Commons gallery pages at all? Jheald (talk) 18:12, 14 February 2017 (UTC)
Turning to P935 ("Commons gallery"), I had thought the original point of that was for Wikipedia category pages to be able to show a link to a Commons gallery as well as a Commons category. But (i) they don't seem to; and (ii) if only 545 have ever been added, then maybe nobody cares.
I don't really understand why P935 (and P373) shouldn't be allowed to exist on both category- and article-type items. It used to be that for P373, there was a separate uniqueness report for use on article items and use on category items. I don't see why that couldn't be resumed. It is certainly a useful report -- for example here's a report for P373 non-uniqueness for category items as a WDQS query: Unfortunately I can't get the corresponding query for non-cats to complete in the time Jheald (talk) 18:41, 14 February 2017 (UTC)
But here's a query that captures most of the article items that are the worst offenders: (I can't get the central SELECT below '4' without it timing out). Jheald (talk) 18:54, 14 February 2017 (UTC)
And some of the classes corresponding to the above query: where 4 or more article-items are linking to the same Commonscat, here is the class of at least one of them: Jheald (talk) 19:09, 14 February 2017 (UTC)
There's clearly a thing about multiple sub-taxons all linking to the same Commons category. (Also multiple bits of German law). Removing these non-unique P373s will make a visible difference on Wikipedia pages, which will no longer show this link in the sidebar. But is that an issue? Jheald (talk) 20:15, 14 February 2017 (UTC)
The most controversial thing, of course, in your colouring is the question of Commons categories sitelinking to Wikipedia articles -- which seems to always have been what a majority on Commons have preferred, because that way the Wikipedia articles appear as direct sidebar links. At least 20,000 of these block Category <-> Category links ( -- it can find 20,000; any more and it times out), which is a bit more than shown in the table, though there may be some double-counting where multiple category items have P373s to the same category on Commons.
The question is, does this matter? Compared to Dec 2015, Commons category <-> Wiki article sitelinks are overwhelmingly the dominant sort of sitelinks that people have been bothered to add, up almost 200,000 while cat <-> cat sitelinks are only up about 40,000.
Indeed the number of new Commons cat <-> Wiki article sitelinks is a substantial chunk of the total increase in the number of Commons cats we can identify with Wiki articles in any way, because they include 60,000 new links that only have this sort of sitelink and no P373. As for the other 140,000 I am not able to tell whether these were links that a P373 already existed for, and somebody has added a sitelink; or whether the sitelink and the P373 were added at the same time; or whether a new sitelink was added first, and then the P373 was added later by somebody else scanning for new sitelinks. Jheald (talk) 20:38, 14 February 2017 (UTC)
Jheald, I wrote this reply yesterday but somehow did not save it:
  • That is interesting about "In other projects" links from Wikipedias to Commons. You are right only Dutch wiki shows link to a gallery and everybody else links to Category. That means that P935 properties might be mostly unused by most wikipedias. In a way, in the most of the cases that is a better option.
  • Your point about "only 545 [P935s that] have ever been added", is incorrect. 545 were added to category items where they should go only if there is no article item linked through category's main topic (P301). However there are still 92,865 P935s added to article items. Apparently someone cares deeply about them.
  • I do not have a strong opinion about duplicate copies of P935 / P373 and other properties at article and category items, except that I prefer to have them in a single place so that if someone corrects it in one place than it is correct as oppose to having the same data stored in multiple locations which slowly gets out of synch. For example, I lately deleted some 2 photo gallery which was not updated since 2006, the sitelink was automatically removed but P935 have to be removed by hand. I am afraid that if I remove one by hand but miss the other one than someone might restore the deleted one years latter based on the other copy. So to summarize, having a single copy of data on Wikidata helps with maintenance.
  • Wow 501 items linked to Category:Melolonthinae through P373, like d:Q14869209 for example. They were added recently by User:Jeblad (bot). I am clearing those up.
  • As for "the most controversial thing" or Commons cat <-> Wiki article sitelinks. In my opinion is, as long as they are accompanied by P373 they can and should be ignored. I am afraid that if your pages rely on them then they might break if or when someone moves the sitelink to category item. They are inherently unstable as they might be removed and they are unpredictable as you have 4:1 chance of hitting category as oppose to gallery page on commons if you follow them. With P935 / P373 properties you know exactly where the link is pointing. But if you do not use them than they can not hurt you.
--Jarekt (talk) 12:44, 15 February 2017 (UTC)
Hi Jarekt! I think it may actually have been me that added most of those P935s to article items :-) Compare 6 December 2015 with 29 November 2015 immediately below. But I agree, I do think that they are useful: a P935 is a bit easier to check than grabbing the sitelink then filtering the string to exclude categories.
Regarding the Commons cat <-> Wiki article sitelinks: I think there may have originally been a notion (if we think about things like eg Multichill's 2013 roadmap) that eventually every Commons category would have an item on Wikidata, and therefore it was going to be important to make sure that sitelinks were kept strictly aligned. But I think that is no longer the case, and nobody is seriously contemplating an item on Wikidata for every Commons cat. I do, on the other hand, think that there is a need to make structured data possible for categories -- for example, to allow internationalisation of category names in the same way that Wikidata items have multilingual labels (something that it seems was highlighted in the application for the recent grant); also perhaps to record what categories aim to cover, eg to move the contents of the "category contains" template I suggest below into structured data. And there may be all sorts of other things that might turn out to be useful for categories -- eg what kinds of ways to sort the images may make sense. So I expect that there will be structured data for categories. But rather than storing that on Wikidata, I'd expect the structured data to be in Commons's own wikibase, more directly attached to the page. This also helps because the roles of categories can be slightly different on Commons vs Wikipedias, so a category page here should not necessarily be presumed to have the same name as the category on Wikipedia in that language (or the label on Wikidata). Having the data page for it stored here (rather than on Wikidata) allows it to keep its own name; and also avoids any difficulty of preserving synchronisation with Wikidata -- if categories are renamed, merged or redirected the data page on Commons will automatically follow. So I think that (potential) rationale for requiring strict cat <-> cat linking should now be considered dead.
Which leaves the question of templates, and the question of how much we lose by always having to explicitly specify a Q-number in a template to draw data from, rather than being able to draw it implicitly from the linked page. I can see the argument; but in reality I think we lose rather little -- principally because there is very little information on the Wikidata category items to draw from. Instead, as per Multichill's diagram, one would usually be relying on the category to have a P301 "category's main topic" set -- but at the moment there are only 200,000 with P373s that do (query: ). And the more steps you put into the chain, the more risk of things not quite being right. (Some of those P301s may never have been checked by humans; and others may rely on how different languages interpret the category and the item, because category approximation between different wikipedias can be even more rough-and-ready that some of the article approximations in a single item -- which might mean that the P301 ended up being more true in some languages than others). It's hard to tell exactly how good all the approximations are, but one interesting point of data is that there are currently 7000 category items where their "main topic" has a P373 pointing to a different Commons cat that the category item itself. (count: list:; to be compared with 180,000 where the round-trip does work: Now in some cases that is because one of these three links is clearly wrong. But in many cases it is because the linking properties have identified something, the relevance of which one can understand, but which is a bit more or a bit less specific than the original, and along the way precision has been lost.
So perhaps it may not be such a bad thing that, the way things are at the moment, templates on categories have to identify a specific wiki item to draw from, rather than trusting in sitelinks and category -> topic matching.
And perhaps the costs aren't so great, if it is links of the Commonscat <-> article item type that are the ones that Commons users seem to prefer to make. Jheald (talk) 19:37, 15 February 2017 (UTC)
Structured data on Commons is being discussed at Commons:Structured data. I agree that the current system of using q-codes by Commons templates works mostly fine. I have however a few concerns mostly related to long term maintenance. I have seen cases of people deleting wikidata items without knowing that they are used, breaking our templates. The fact that the same data (Commonscat <-> article item link) is stored at two locations creates problems when 2 copies are out of synch. Just like 7000 category items whose P373 are out of synch with P373s of the matching article items. Q-codes actually work in cases where sitelinks do not, for example if the page sitelinked is a template than the template can't access its properties when it is being transcluded, but can when a Q-code is used. So at the moment I create category items only if there are other categories sitelinked in the item, and I try not to rely on sitelinks when accessing properties from commons categories. I also agree with an issue that relying on too many correctly set inter-item links is often not rebust. For example in taxonomy items you rely on long chain of properly set parent taxon (P171) properties to create taxonomy template. But than edits like this can break the chain and leave all lower rank organisms with a broken taxonomy template. --Jarekt (talk) 21:27, 15 February 2017 (UTC)

Other categories: "Category contains"

The increase in Wikidata <-> Commons category links with P373 is something really good.

On the other hand, while the number of categories we can identify with article-like Wikidata items is up 275,000 that compares with an increase in the total number of Commons categories from 4,587,397 to 5,499,772 over the same period -- up 912,375. So even now only 23% of Commons categories can be identified with Wikidata article item.

Are we doing well or not? It's hard to tell. One issue is the (unknown) number of complex intersection categories here -- for which for the most part there aren't, and probably it never makes sense for there to be, a corresponding item on Wikidata. At the moment it's hard to identify these categories, or describe them in a machine interpretable way.

So I wonder if it would make sense to start rolling out something that would try to store a description of what such an intersection category is trying to capture. As a proof of concept, I've created an experimental template {{Category contains}}, that can store a specification for the category in terms of Wikidata items and properties represented in their SPARQL query form. The template also produces a strapline, that could be shown at the top of the category:

The above strapline was generated by the specification wdt:P31?/wdt:P279* wd:Q146, which translates as "cat" (Q146); all subclasses of cat, and sub-sub-classes etc (P279*); and all instances of any such subclasses (P31). It's safe to say that Wikidata doesn't (yet) know as much about cats as Commons, but this is the kind of thing that might be placed on the Commons category c:Category:Cats c:Category:Felis silvestris catus.

Four test examples of the template, as it might be used, can be found on:

If such templates started to be placed systematically on Commons categories, I can see at least three possible uses becoming possible through these machine-interpretable specifications becoming more available. One is querying, to show what Wikidata knows about items it thinks would be related to the category or its subcategories; another is auto-categorisation, which I think this could help; and another is better bot-identification of anomalies and/or apparent missing information between what's in Wikidata and the contents of the category structure. Taking each of these in turn:


As introduced already with the query link in the strapline above, given a machine-interpretable specification we can use the Wikidata SPARQL query service (WDQS) to see what items in Wikidata appear to match that specification, and some of what Wikidata knows about them. In particular, do they have principal images associated with them (Wikidata property P18), and Commons categories (Wikidata property P373). The WDQS service also lets us see the images in an image-grid format (click on the "Edit on" link at the very bottom left of the screen; then the blue "Run" button that will appear about half-way down on the left; and then the "Image grid" in the drop-down menu released by the "Display" icon that appears on the right at the same level as the "Run" button.

For geographical items, like the hits for the queries in the Bedfordshire categories above, one can also see all the items (that Wikidata knows coordinates for) plotted on a map -- re-run the query as above, but this time choose the "Map" option in the "Display" drop-down list. Each dot on the map can be clicked on to show a summary of what was in the row of the query (including the preferred image, if there is one); the query has also been written to plot the dots in different colours, according to whether the items have images or not. (Each set can be turned on or off via a drop-down menu released by the "layer" icon at the top right of the map).

Yes, there are still some things that are a bit clunky. It would be nice to be able to switch views without having to go back into the query editor. It would be nice if the image links went to eg Media Viewer, from where one could get to the File information page, rather than going to the image at complete full size from which there is no way back. It would be nice if the names of the Commons categories were links, that could be clicked. The last two issues could to some extent be worked around by adjustments to the query. But the WDQS output display has been improving so quickly, by such leaps and bounds, that with luck before too long all these nit-picks may have been fixed at source. At any rate, this current query should give a taste of what is possible.

Categorisation help

One of the difficulties with the category system is knowing just what categories it might make sense to add for images of a particular thing. (Or to know what categories one might place a new category for that thing into).

Having machine-interpretable specifications for the categories could help here. If an item's properties were quite well-populated on Wikidata, one might be able to ask a tool to start at quite a high-level category, and then see what sub-categories in that category's category tree the item could fit into.

Recursively, at each stage the tool could retrieve the sub-categories of categories currently considered viable (ignoring categories it had already seen, to avoid loops), retrieve their machine-readable specifications, then for each one ask the WDQS service whether the item would meet that specification. (SPARQL has a special query command for this, ASK { ... } , which is usually very very quick, because it doesn't have to create a list, just come up with a single true / false response.)

The tool could then return as suggestions all sub-categories for which the item complied with their specification, that had no deeper sub-categories that the item could be percolated further down into.

Anomalies and missing data

It's quite hard for tools to extract information from the category tree at the moment -- eg what Commons thinks are all the interesting locations in a particular county -- because as you down a category tree, the sorts of things you're seeing can suddenly swerve off in quite different directions. For example a category descent that appears to be about more and more specific locations can suddenly swerve off into information about residents of a particular place, then works they have created, then things somewhat tangentially related to some of those works, etc, etc.

Having the machine-interpretable specification on categories should help to much better identify what is likely to be in them; and so how far they are likely eg still to be about the locations, rather than something else -- potentially making it much easier to see e.g. if Wikidata can match the suggestion conveyed by the fact of a preferred picture of an object appearing in a particular category here.

Wikidata limitations

At the moment, Wikidata searches can produce a lot of good hits. But they also typically miss a lot of good matches too, because the extent to which particular properties have been populated can be quite patchy.

So for example, the 'cat' search above misses a lot of what is in the Commons subcategories (even discounting the ones that take a left swerve), because either relevant items don't exist, or (perhaps more often the case) because one of the links in the chain that would be needed to make it satisfy the search is missing, because one of the properties on one of the items in that chain hasn't yet been filled in.

To see an example of that, here is a recent search I ran to see which U.S. civil war battles Wikidata knew we had maps for on Commons: One can see that data coverage for the geographic coordinates is pretty good; for dates about 50/50; probably a lot have commons categories that are missing; while as for maps, only one has been linked. One also doesn't know how many battles may be missing completely, because they haven't been marked as "part of" (P361) "the U.S. civil war" (Q8676).

Similarly, the Grade I houses in Bedfordshire search is a pretty good match for the contents of the Commons category (give or take some items that one could argue about whether or not they were houses or not). But the Grade I churches search is a lot less complete compared to the category -- in this case, because the churches ideally ought to have a statement to say which parishes they are in, the parishes which counties they are in, and that way they would be identifiable as churches in Bedfordshire. But for most of the churches that is not yet in place. (And for those that are returned, they mostly just take a short-cut saying they are in Bedfordshire, rather than which parish).

So what Wikidata can return at the moment for a search related to a category here may well be pretty incomplete. But it's a wiki -- it can be edited, facts can be added, it can be improved. And even if results corresponding to one of our 'complex categories' are still quite patchy at the moment, I hope that what does get returned can still be interesting enough, that it may be worth putting a template like that at the top of categories, to machine-describe what's primarily in them, and see what Wikidata can say so far about those things. Jheald (talk) 11:17, 13 February 2017 (UTC)

Files not deleted following a delete decision

I was thinking of nominating File:Howe.jpg for deletion as a probable copyright violation. When I click on the link I get the message "Consider reading the deletion debate –Commons:Deletion requests/Files uploaded by Peterandfrankie– that links to this page." That discussion concluded with Deleted, yet this file and two of the others discussed have not been deleted. Was this an oversight or a error? Should I renominate this page? Verbcatcher (talk) 05:50, 15 February 2017 (UTC)

Thank you for pointing this out. I deleted the file, because it was reuploaded after it was deleted in the deletion request. Sebari – aka Srittau (talk) 05:56, 15 February 2017 (UTC)
Thanks. The same issue appears to also apply to File:Spcstaff1957.jpg and File:Spcreport1931.jpg, from the same discussion. Verbcatcher (talk) 06:00, 15 February 2017 (UTC)
There were more images re-uploaded by a suspected sockpuppet of the original uploader under different names. I deleted them all. Sebari – aka Srittau (talk) 06:10, 15 February 2017 (UTC)
  • Besides sock reuploads, files with simple names like File:Howe.jpg are often uploaded by good editors months or years after deletions of unrelated versions. I've seen some really common file names with half a dozen different uploaders and images over a period of several years. Face-wink.svg lNeverCry 08:19, 15 February 2017 (UTC)

Merging categories

Hello.I suggest merging Category:Framed images and Category:Images with intentional borders.Thank you --ديفيد عادل وهبة خليل 2 (talk) 12:19, 15 February 2017 (UTC)

Those aren't the same thing though. Framed images is for pictures of things with frames - paintings, etc. Intentional borders means the image itself has a border. -mattbuck (Talk) 13:35, 15 February 2017 (UTC)
@Mattbuck: You're saying Category:Framed images could alternatively be caled "Images with physical frames" ? - Themightyquill (talk) 16:29, 15 February 2017 (UTC)
Those seem to me different kinds of category, The latter (which appears to have a lot of content that doesn’t belong) is more like a maintenance category, pretty much saying “Do not crop these images.” As I understand its purpose, “borders“ in this name refers to margins, and the “intentional” implies that they could be perceived as accidental, as if someone had merely neglected to trim them. OTOH the former category is simply descriptive and yes, “physical frames” is apt, while we have Graphic frames for the kind that are part of the artwork. That’s my take on them anyway; I have no knowledge of their history or background.—Odysseus1479 (talk) 21:02, 15 February 2017 (UTC)

Images incorrectly marked as "No Source"?

I notice that @ShinePhantom: has tagged a number of images as having {{No source since}}. In some cases, I note that a source *has* been explicitly specified beyond the default "own work":-

It may legitimately be argued that the source is vague, incorrect or insufficient, but I'm not sure that these should be tagged as "no source". Are these correctly tagged or not? Ubcule (talk) 13:10, 12 February 2017 (UTC)

I think the taggings are fine, this is indeed insufficient source information, so that the copyright situation of both files cannot be verified. We learnt from earlier discussions that older uploads (e.g. more than 3 years ago) should typically have a regular DR, but these two uploads are recent enough to use the tag. Jcb (talk) 13:18, 12 February 2017 (UTC)
It seems odd that there isn't an "insufficient source", "disputed source" or similar to distinguish such cases from "no source", though. "No source" says to me that there isn't any proper source given, period. Ubcule (talk) 19:56, 13 February 2017 (UTC)
You can legitimately convert them to deletion requests, then put the logical case as to why they are public domain. I agree that where logically the date and source are obvious (just not a hyperlink), the tag is unhelpful and literally incorrect. There is very little assessment required before using these tags, it is a generally good thing to challenge any you notice where there may be a case. -- (talk) 13:35, 12 February 2017 (UTC)
you should not be saying "missing essential source information" when you mean "contested source information". that is what DR is for. circumventing the deletion process tends to undermine the credibility of commons. Slowking4 § Sander.v.Ginkel's revenge 00:46, 13 February 2017 (UTC)
@:; The option to convert them to DRs isn't necessarily appropriate . As you imply, this places the onus on the nominator to (a) know that the image is free *and* (b) to demonstrate this.
My point *wasn't* that I was sure these images were free nor even that the sources above were necessarily sufficient. It was simply concern whether tagging them as "no source" when this wasn't the case was a mis-characterisation that would see them subject to a lower standard of deletion- and chance of response- than they should have been.
(To some extent, this is much what @Slowking4: said above).
I agree there are cases we can reasonably treat as having no source, such as "own work" when it's obvious it isn't and that was likely chosen purely as the default option, but this isn't the case here. Ubcule (talk) 19:48, 13 February 2017 (UTC)
the problem being that habitual misuse of the tag tends to undermine the credibility of the "tag and delete 7 days later" process. there is not a "disputed source" tag because that be be honest. when you do not have an honest broker, then all deletion becomes biased. Slowking4 § Sander.v.Ginkel's revenge 13:38, 14 February 2017 (UTC)
@Slowking4:; Just because someone hasn't given a good enough source doesn't mean they weren't making a good faith attempt to follow the rules. In at least one case above, I'm pretty sure that they were. Ubcule (talk) 13:50, 14 February 2017 (UTC)
makes no difference to the deletionist admin; AGF is for admins on this site; all others need to jump through hoops within seven days, ten years after upload. or have their items deleted. see also Commons:Village_pump/Archive/2016/10#Proposal_to_ban_Jcb_from_using_the_.27no_source.27_template_for_six_months Slowking4 § Sander.v.Ginkel's revenge 13:53, 14 February 2017 (UTC)
  • Pictogram voting comment.svg Comment - I am concerned about Jcb's comment above. Afaik, "No source" is obviously not the same as "Insufficient source information", thus using the tag when there is an insufficient source information is in fact wrong and I tend to describe such tagging as recklessness or cluelessness. However, I do want to agree with Fae that where an insufficient source information is provided, the image should be converted to DR. Our policy allow such nomination. Of course, a discussion with the uploader may be necessary to resolve the issue prior to nomination. Wikicology (talk) 06:56, 17 February 2017 (UTC)

Recategorising categories.

I need to recategorise a number of categories. Is there a tool, like HotCat, that can assist with that? Andy Mabbett (talk) 23:56, 14 February 2017 (UTC)

HotCat works on categories. Do you mean like Cat-a-lot? It has an option to allow moving subcategories as well as files (see preferences in the Cat-a-lot window). --ghouston (talk) 00:23, 15 February 2017 (UTC)
Cat-a-lot should make quick work of it as Ghouston says. Help:Gadget-Cat-a-lot is the documentation page for the gadget. lNeverCry 04:47, 15 February 2017 (UTC)

Thank you, both. Yes, I meant cat-a-Lot, and I wasn't aware of that option under preferences. It's done the job. Andy Mabbett (talk) 13:11, 16 February 2017 (UTC)

February 15

Updating large videos

This video File:Pneumonia.webm

Was updated with this version File:Pneumonia-2.webm

Because the file is 216 MB it was not possible from what I understand to upload the second file as a new version of the first.

Wondering if we should delete the first and move the second to its place? We still want to save the translated subtitles as those are not changed. And we want the second version to replace all the uses of the first version without a bot going around and removing those. Doc James (talk · contribs · email) 07:33, 15 February 2017 (UTC)

@Doc James: This is confusing to me. File:Pneumonia.webm is in slightly smaller dimensions (1,812 × 1,018 pixels versus 1,919 × 1,080 pixels) but is also three seconds longer and actually a slightly larger file. These two videos are not identical in terms of content. Why do both of these exist in the first place? Neither is above 216MB by a sizeable enough margin. —Justin (koavf)TCM 07:48, 15 February 2017 (UTC)
@Doc James: Also, in case you weren't aware, the subtitles are saved in SubRip format as a completely separate file in the TimedText namespace:,,,, etc. Overwriting or deleting File:Pneumonia.webm will still leave those subtitles intact. I actually don't know what happens if you move the original file though--does anyone know if MediaWiki is smart enough to move all associated subtitles as well? —Justin (koavf)TCM 07:51, 15 February 2017 (UTC)
Thanks. I have asked the creators to comment. Doc James (talk · contribs · email) 07:57, 15 February 2017 (UTC)
@Koavf: I know. It's not T122038TheDJ (talkcontribs) 12:04, 15 February 2017 (UTC)
Hello @Doc James: and @Koavf:! A few things have changed in this video. Most importantly, Pneumonia.webm has a factual inaccuracy in it, specifically, we were not clear that mycoplasma are technically bacteria, so we have changed the audio and visuals to reflect this. We also have changed our encoder, switching to VP9 instead of VP8, which has drastically reduced the file size of the video from 216 MB to 162 MB. We've also made minor adjustments to the video resolution as you pointed out. Normally, I would update the video by uploading a new copy of the file directly to the original file. However, Wikimedia has a 100 MB file size limit. You can overcome this limit when uploading new videos, but I have found no way to work around this limit for existing videos. Finally, I can easily migrate old subtitles to the Pneumonia-2.web video manually once we've resolved the issue of removing Pneumonia.webm. OsmoseIt (talk) 16:36, 16 February 2017 (UTC)

Date nav box on Category:1999-01-15

There is a problem with the date nav box on this page:

Try going to the previous day and see where it takes you.

Jasonanaggie (talk) 22:16, 16 February 2017 (UTC)

Looks like Category:1999-01-14 incorrectly claimed that it's June. I think this should fix it: [23]. Matma Rex (talk) 22:34, 16 February 2017 (UTC)
Nice work; I tried to hunt down the error but I figured it was a problem with the nav box template. Thanks! Jasonanaggie (talk) 02:43, 17 February 2017 (UTC)

February 17

.kml pages

Someone could explain to me that they serve pages with .kml extension, such as File:India Jammu and Kashmir locator map.svg/overlay.kml. Thank you. --Metrónomo's truth of the day: "That was also done by the president" not an excuse. 17:12, 10 February 2017 (UTC)

And File:Geniesserpfad-karlsruher-grat.png/GPX? --Metrónomo's truth of the day: "That was also done by the president" not an excuse. 17:15, 10 February 2017 (UTC)
In the old days templates {{Map}} linked to /overlay.kml pages so they can be displayed in GoogleEarth. Two years ago User:André Costa (WMSE) removed those links from {{Map}} and it was broken since. Maybe someone will have time to fix it again. --Jarekt (talk) 18:08, 10 February 2017 (UTC)
the Kml overlay bit of {{Map}} was broken since it hooked into a deprecated Google service (per discussion page). /André Costa (WMSE) (talk) 18:17, 10 February 2017 (UTC)
No Google Map link was removed from {{Location}}, but that is unrelated to KML handling. See for example File:Warsaw Ghetto Map - 1940-10-15a.png and go to the template below {{Map}} and click KML opens up in Google Earth (if you have it installed on your computer). That is what the removed link in {{Map}} was doing. --Jarekt (talk) 04:29, 11 February 2017 (UTC)
The {{Overlay}} template has a link for viewing an image as an overlay. At the same time, the "Google Maps" and "Bing" links in the template don't appear to work properly; those links cause Google Maps or Bing to display a map based on the user's current location (perhaps using geolocation of the user's IP address) instead of displaying a map based on the KML page. --Gazebo (talk) 09:52, 15 February 2017 (UTC)
When the link was introduced to {{Map}} it was possible to view the kml overlay on (at least Google Maps). Since they now require a user to have a specialist piece of software (Google Earth) installed I would classify it as broken. I forgot the discussion link which should be Template_talk:Map#Kml.2Foverlay. /André Costa (WMSE) (talk) 09:27, 17 February 2017 (UTC)

Images from Fukushima

Hi, I came accross File:223.6.2 13偵:瓦礫等の除去・調整(福島県浪江町)② 東日本大震災における災害派遣活動 1.jpg in Category:Type 00 Personal protective equipment, but

  1. these do not seem to be military, but civilians;
  2. I don't see any evidence of a free license, but I don't read Japanese. Anyway a license review is needed for the whole set uploaded on Commons.
What do you think? Regards, Yann (talk) 19:45, 13 February 2017 (UTC)
This is the entire folder[24]. Appears to be partly military related.
User:Takot what is the copyright of materials generated by the Japanese military? Doc James (talk · contribs · email) 07:41, 15 February 2017 (UTC)
The account belongs to the Japan_Ground_Self-Defense_Force. So User:Yann appears to be okay to me. Doc James (talk · contribs · email) 07:45, 15 February 2017 (UTC)
Thanks for raising a question and inviting me to this discussion. It strongly suggests it is a work of Japanese Ground Self Defence Force, however, this Picasa Web Album account is not verified as an official account of them. Plus, while the Japanese Government is paying efforts to deploy a CC-compatible license (Government Standard Terms of Service), I cannot say the government works are immediately PD or any free license, as of today. --Takot (talk) 08:37, 15 February 2017 (UTC)
User:Ks aka 98 told me that this Picasa account is linked from the JGSDF official website and thus the photo mentioned is available under CC-BY equivalent condition. I'll give you more details later. --Takot (talk) 08:51, 15 February 2017 (UTC)
According to this JGSDF official website (the domain itself is verified as domains are exclusive to the Japanese government), the account in question is indeed their account. The contents are available at CC-BY equivalent, noting that 「それらの写真は全て、出典元(例:陸上自衛隊HPより引用)を明記していただければ、商用・非商用を問わず、インターネット・出版物等に使用していただくことができます。」("those photos are avilable both for commercial and non-commercial purposes on the internet or publications as long as the proper credit is clearly stated (e.g. cited from JGSDF website)") --Takot (talk) 10:54, 15 February 2017 (UTC)
Some remarks: the source URL was broken, but I fixed it. This should be done for the whole set. We need {{Licensereview}}, one way or another. Could we use the Web archive to do that? At the very least, we need to add more information that there is now. May be a template explaining all this. Regards, Yann (talk) 22:43, 15 February 2017 (UTC)
Agreed. Creating template to explain it would be a solution. You mean "Web archive" as --Takot (talk) 02:07, 18 February 2017 (UTC)

Loading percentage

Hello.Why is there no "uploading percentage" such as YouTube?Thank you --ديفيد عادل وهبة خليل 2 (talk) 08:00, 16 February 2017 (UTC)

Where exactly (=link) is your "there"? Asking as there are many ways to upload stuff. --Malyacko (talk) 12:02, 16 February 2017 (UTC)
@Malyacko:During uploading to YouTube, loading and processing percentages show.I'm talking about the normal way --ديفيد عادل وهبة خليل 2 (talk) 12:13, 16 February 2017 (UTC)
There is when you upload with Google Chrome, but not when you use Firefox, at least that's my experience with the "experienced" upload interface. Blue Elf (talk) 13:13, 16 February 2017 (UTC)
The "normal way" (linked from the sidebar) is Special:UploadWizard, not Special:Upload. --Malyacko (talk) 13:28, 17 February 2017 (UTC)
There is a progress bar when using the default interface at Special:UploadWizard. Matma Rex (talk) 17:42, 16 February 2017 (UTC)

Cut-off date for {{PD-old}}

At COM:VPC we have had a discussion to try to reach a consensus about how old an image has to be before we assume PD if the PMA+70 rule applies and we don't know when the author has died. After several users stated their opinions and arguments, I have extracted a proposed compromise from the discussion, to see if we can find each other somewhere in the middle, see Commons:Village_pump/Copyright#Proposed_conclusion. Not many users have responded to that proposed compromise. Of the users who responded, some find the compromise acceptable, some don't. Currently there is not yet a consensus. More input from different users would be appreciated. Jcb (talk) 12:10, 17 February 2017 (UTC)

Other eyes needed


I was ready to put a {{NoUploads}} on Category:Donald De Lue and sending all files to DR when I had a doubt, can maybe some of these file be indeed free? At least for the works in France (like Spirit of American Youth Rising from the Waves in Colleville-sur-Mer) where there is no FoP, the deletion is almost certain but what about works in USA? There is FoP there but not sure if it applied for such artworks (and not sure if it's true for both 3D statue and 2D-like relief). As there is maybe a lot of different cases, I'd like some other point of view before starting a RfD.

Cdlt, VIGNERON (talk) 09:56, 15 February 2017 (UTC)

According to Commons:Freedom_of_panorama#United_States, FoP in the US is for architecture only - artworks like statues are not covered by this. --El Grafo (talk) 14:52, 15 February 2017 (UTC)
What happens when sculpture and architecture are combined, like at Category:United States Court House and Post Office Building, Philadelphia? Kaldari (talk) 07:24, 16 February 2017 (UTC)
Those particular sculptures almost certainly are not copyrighted, but if they were the issue would be whether the photograph emphasizes the copyrighted sculpture (in which case it would not be OK) or the inclusion of the sculpture is de minimis (in which case it is OK). For example, you can certainly photograph a building that happens to have some small bas reliefs. - Jmabel ! talk 16:23, 16 February 2017 (UTC)
for US public art, you should be checking SIRIS, i.e. [25]; public buildings from the 30s & 40s tend to be PD WPA, or PD US no notice. need a second license for the sculpture. also user:smallbones can speak to philly. Slowking4 § Sander.v.Ginkel's revenge 00:43, 19 February 2017 (UTC)
See Commons:Public art and copyrights in the US about the copyright of publicly displayed sculptures in the US. Before 1978 if they were displayed in a public setting (not a closed gallery) and they don't have a visible copyright mark they are in the public domain. Freedom of panorama then is irrelevant. The SIRIS website mentioned above notes the writing on and around the sculpture, so should be checked to see if there is a copyright mark and also for the date of display. All my photos of public sculpture involve a close examination for a copyright mark and checking the date of display. Any blanket ban of De Lue's American work is totally uncalled for. Smallbones (talk) 03:05, 19 February 2017 (UTC)

George Eastman House

Looks like George Eastman House has a new website: It is way classier but it aint yuge, because try as you might neither button «browse our collection» nor «advanced search» yields anything remotely close to the wealth of material the previous version of the site, clunky and fugly as it was, did. All links to previous resources are now, “obviously”, lost and pointing redirected to a limp — sad!

A couple years ago I uploaded from GEH all photos by/from Charles Chusseau-Flaviens that pertain Portugal (resulting that Category:Photographs of Portugal by Charles Chusseau-Flaviens‎ includes 845 files and its parent Category:Charles Chusseau-Flaviens‎ has only 28…) and I did add 2 source urls for each image. They look like this:

  • number)_ful.html
  • number).jpg

It is possible to mend this case of linkrot by adding*/ right before each of these urls. Even for those which were not archived the result isn’t any worse. Anyone against? Any better ideas? -- Tuválkin 02:31, 16 February 2017 (UTC)

For linking to archived versions of pages at the Internet Archive, it could be useful to use*/ which uses HTTPS and as such would provide more privacy and security for users than an HTTP link.
It appears that a number of the photos have a 12-digit number (identified as "(some number)" above) in the URL. For a photo that has such a number, it may be possible to link to a "search results" page with just an entry for the photo by dividing the 12-digit number into three groups of four digits separated by periods, and to append the number to the URL. For example, in the case of File:1er_de_l’an_au_Palais_de_Belém.jpg, there is the source URL, and for linking to the "search results" page, appears to work. For File:BarbarosHayreddin-class1910s.jpg, which has the source URL the URL for the "search results" page would be --Gazebo (talk) 09:39, 16 February 2017 (UTC)
For the photos in the Category:Photographs of Portugal by Charles Chusseau-Flaviens‎, it might be possible to do a search and replace operation (such as with VisualFileChange) by using regexps and replacement strings. For the first variety of source URL, the regexp would be /http://www\.geh\.org/ar/chus/([a-z])+/m([0-9]{4})([0-9]{4})([0-9]{4})_ful\.html(#[a-z]+)?/ and the replacement string would be$2.$3.$4. For the second variety of source URL, the regexp would be http://www\.geh\.org/ar/strip[0-9]{2}/m([0-9]{4})([0-9]{4})([0-9]{4})\.jpg and the replacement string would be$1.$2.$3. --Gazebo (talk) 10:41, 18 February 2017 (UTC)

Is just that all my drawings were deleted withoug any concensus

Hello I'm Vicond.In the past an User posted to me a deletion request of basically all my uploads of images See:. Subsequently, an Administrator deleted withoug any consensus, of the community all my work.For me It's was very absurd.Speaking about the matter of the images I'ts represent a lot of drawings MADE BY MOUSE,that I'm absolutely sure that could be posted in the category with the same name.

In other side I draw a lot of drawings that are intimate linked with the customers and tippical culture of my comunity.I wonder: I'ts not represent any condition of educative way in any case??

I know It's was very time ago but on that time the possibilities to communicate with the commons admin was very limited. So, I wonder if any human person here can help me to resolv this inconvenient. Can I resque my drawings??Meaby making a kind of selection?

Thank you very mutch. — Preceding unsigned comment added by Vicond (talk • contribs) 16:03, 17 February 2017 (UTC)

  • As an admin, I can still see these. I looked at four, more or less at random. I don't see any reason these would be any more in scope than any other art by a non-notable artist. There's nothing to stop you from posting these elsewhere on the Internet, and if you've lost the originals I'd be glad to help with a temporary undelete so you can recover them, but I don't see any reason for these to be on Commons. - Jmabel ! talk 16:38, 17 February 2017 (UTC)

Frist of all:Thak you very mutch by your amability.Sadly and in adiction I have to say that I loose the works...Even the four that you see is corresponding to others uploads made by others users made more recently.In fact if you go to the pages of the files, you can note that the names of the pages is not correspondig with the names of the primary files, like one kind of redirecting. In any way I will be gratefull to recover this works or at least the most part of them, if you believe is possible that. I promisess to be more careful when I made my uploads in the future.Honestly:I thinked that could served in an educative way like expression of the culture of my country. Thank you very mutch aggain. — Preceding unsigned comment added by Vicond (talk • contribs) 16:03, 17 February 2017 (UTC) — Preceding unsigned comment added by Vicond (talk • contribs) 17:12, 17 February 2017 (UTC)

I'm also an administrator and have reviewed the deleted materials which were not educational. In fact, the remaining images in your gallery also seem to have no educational purpose. Commons is not here to be your filing cabinet or scrap book, sorry! Ellin Beltz (talk) 18:10, 17 February 2017 (UTC)
it's true, commons is not flickr, and artists and institutions should be using the latter. Slowking4 § Sander.v.Ginkel's revenge 23:35, 18 February 2017 (UTC)

Improving PD-art templates

Instead of {{PD-art-70}} and {{PD-art-100}}, we could have a template that looks like, say, {{PD-art-year|1920}}, and which works out the necessary text to display, changing it as the years roll on.

In other words, {{PD-art-year|1917}} would display like {{PD-art-70}} this year, but from 1 January 2018, would display like {{PD-art-100}}. Andy Mabbett (talk) 20:21, 18 February 2017 (UTC)

We allready have {{PD-old-auto}}, like {{PD-old-auto|deathyear=1935}}. Maybe a good idea to create the same or similar for PD-art. --JuTa 20:51, 18 February 2017 (UTC)
you could just do {{PD-art|PD-old-auto|deathyear=1935}} Slowking4 § Sander.v.Ginkel's revenge 23:31, 18 February 2017 (UTC)

Czechia vs. Czech Republic

Our categories related to the Czech Republic seem to be a hodgepodge with respect to whether they use Czechia vs. Czech Republic in their names. I would normally think this is a no-brainer: we normally call things by their common English-language names if they have one, and as far as I know Czech Republic is by far the normal usage in English. But I'm aware that or its equivalents might not be by any means the majority usage in other languages, and this is a multilingual project. Still, it seems to me that this is a place where any uniform decision would be better than no uniform decision. As it is, no one can ever correctly guess any category names for this country without looking up the individual category. ~~- Jmabel ! talk 00:36, 10 February 2017 (UTC)

I would say that in Polish Czechy is far more common than the official Republika Czeska. Anyway, en:Czech Republic#Etymology may give some hint – since May/July 2016 Czechia is the official short name of the state in English. Regarding category names on Commons, IMO Czech Republic should be used everywhere. --jdx Re: 01:58, 10 February 2017 (UTC)
Czech Republic seems best, except there's an open discussion about historical categories at Commons:Categories for discussion/2016/12/Category:History of Czechia. --ghouston (talk) 02:04, 10 February 2017 (UTC)
Regrettably, "Czech Republic" is quite unusable (or even absurd and ridiculous) for pre-1969 context (1848 in the Czech Republic, Comenius among "Educators from the Czech Republic" etc.). The word "Czechia" is maybe not very known among uneducated people but English language has no better word for this meaning – timeless and geographical name of the Czech country. "Czech lands" or "Czech Lands" can be an equivalent (especially in historical context) but generally we don't prefer such country names which emphasize some level of division of the country (except for United States, United Kingdom and Soviet Union which have no simple geographical names). The history-related categories were renamed urgently. However, as it seems, unification of the remained categories requires some time to wait until the irrational resistance weakens. --ŠJů (talk) 17:55, 11 February 2017 (UTC)
I'm not sure it is any less problematic to say that all of the Germans who lived in Bohemia were in "Czechia" than in the "Czech Republic".
There are many cases where we use anachronistic names. For example, Category:Old maps of Israel includes many maps that predate Zionism, let alone Israel. Category:Washington (state) in the 1860s refers to a geographical area that was not a state at that time. - Jmabel ! talk 20:33, 11 February 2017 (UTC)
We also have categories for former entities, such as Category:Maps of the British Mandate of Palestine. Wouldn't those be better than anachronistic names? Washington in the 1860s was Washington Territory. --ghouston (talk) 00:29, 12 February 2017 (UTC)
@Jmabel: The first thing is meaning of the terms, a second and different thing is how well-known is which subject or which term.
* Bohemia is only one part of Czechia, i.e. everybody who lived in Bohemia lived in Czechia but not everybody who lived in Czechia lived in Bohemia.
* Obviously no Czech Republic existed before 1969 (uneducated people didn't notice it even before 1993), but Czechia existed through centuries. E.g. who lived in Bohemia or in Moravia in 1880s, 1920s or 1950s, that lived in Czechia but surely not in the Czech Republic. Although the word Czechia is relatively modern (its Latin form appears in baroque period), Czechia as the area of Czech lands and Czech language is relatively immemorial subject, even though German language prevailed in some periods and areas. Formation of the Czech nation (including Moravians) and the unified state of two lands (under Premyslids) is dated to the 11th century, i.e. since that times, it is not an anachronism to apply retrospectively the term Czechia. German colonization of Sudeten areas of Bohemia and Moravia began in 12th century but it is considered as an internal colonization (supported by the Bohemian King and not disrupting the national identity of the country). Even though Czechia was at least bilingual for centuries, its determination as "primary Czech lands" is unquestionable. (The one-time secondary Czech lands as Lusatias and most of Silesia are commonly not count as parts of Czechia.)
Germany has not so definite meaning in relation to the period before 1871 but "Germany" is obviously more timeless and more universal name of the whole country than "Bundesrepublik Deutschland" or "Federal Republic of Germany". Israel is a bit different case, their state was not created mostly by any previous inhabitant nation of any previously existing country, there was any historic discontinuity. European national countries are relatively stable since medieval times, although empires, states, kingdoms and republics arise, perish, split or join. Czechoslovakia was comprised by two previously existing countries of Czechia and Slovakia. The federalization didn't create the countries, just as the dissolution of the federal "peel" of the two countries. Even the countries of Bohemia and Moravia (constituting parts of Czechia) are existing continuously, even though they have not their own self-governments currently. --ŠJů (talk) 20:31, 13 February 2017 (UTC)
My only strongly held conviction about this particular case is that any uniform decision would be better than no uniform decision. The current state of the categories is a mess. - Jmabel ! talk 00:35, 14 February 2017 (UTC)
Generally, everybody can agree with it, but such an opinion doesn't solve the problem. "Czech Republic" is quite unsuitable and absurd for pre-1969 contexts. I mean that a complete shift to "Czechia" is the only possible solution, but the word has still some adversaries. Maybe, the situation need to grow mature for such a decision. --ŠJů (talk) 23:59, 19 February 2017 (UTC)

Google Cultural Institute sync?

Back in 2012 a big batch of Google Art images were uploaded. Google Art was renamed to Google Cultural Institute and more content has been added. I noticed some more recent pictures, but I don't think anyone recently did a batch upload to add the missing images here. Am I right or did I miss a big upload? Can someone sync the public domain paintings to Commons? Multichill (talk) 17:14, 17 February 2017 (UTC)

@Multichill: Downloading from Google is a complex task, with all those obfuscation and encryption Google uses. Searching the web about it, almost all the solutions I see are:
  • Solutions that once worked before Google obfuscated the images, but no longer works after that.
  • Use some programs to resize the browser to some insanely large size then take a screenshot.
  • Search on Commons (maybe we should be proud of this)
  • Use a site (which I already forgot). The method was mainly, AFAICT, reverse-engineered image url and de-obfuscation, with the image concatenated in user's browser.
A year ago I had a script that could download this, using headless browsers to fetch the images and imagemagick to concatenate them. However, with Google redesigning some parts of the site, and Ubuntu Trusty being deprecated on Tool Labs, the script no longer works and needs a complete rewrite/overhaul. I'll try to get it done during March/April due to the headless browser requiring Firefox 53 (currently alpha). --Zhuyifei1999 (talk) 01:30, 20 February 2017 (UTC)

Do not crop template

The scale is important on en:Portable Antiquities Scheme

Do - or could - we have a template, say {{Do not crop}}, to warn when an image like the above is used on an article where cropping (say, to remove the scale) would be detrimental to its purpose?

If so, the crop tool should not allow an image marked with the template to be overwritten. Andy Mabbett (talk) 17:25, 17 February 2017 (UTC)

It may be good idea but someone will need to make a change to MediaWiki:Gadget-CropTool.js. Ruslik (talk) 19:29, 17 February 2017 (UTC)
Preempt by creating a cropped version as a new file. -- (talk) 10:05, 19 February 2017 (UTC)
I don't think that would stop well meaning but uninformed individuals; and in some cases there are multiple possible ways to crop. Andy Mabbett (talk) 19:24, 19 February 2017 (UTC)
See the “Merging categories“ thread above (under Feb. 15); I think that’s what Images with intentional borders is supposed to do when populated by {{Border is intentional}}, which last I take to mean more or less the same thing as {{Do not crop}}. It might be easier to exclude the category from CropTool than to make it look for a template.—Odysseus1479 (talk) 17:50, 19 February 2017 (UTC)
No, it is not the same as (though it is comparable to) 'Border is intentional'. Andy Mabbett (talk) 19:24, 19 February 2017 (UTC)

February 18

All geolocated images which I have uploaded displayed on a map

Hi there, I have made tons of pics all over the world and uploaded them here. I would like to see an wiki-integrated map with a layer of my locations. What do I have to do? Thanks and bye --Mattes (talk) 12:55, 18 February 2017 (UTC)

Firstly, put all your images in one category (mine is Category:Images by Andy Mabbett). Andy Mabbett (talk) 20:23, 18 February 2017 (UTC)
✓ Done Category:Work by Mattes --Mattes (talk) 08:52, 19 February 2017 (UTC)

Category problem

I recently came across Category:CL, meant for images of a Korean K-pop singer. However the category is mostly filled with images of cathedrals, castles, canal locks and such, uploaded from Panoramio. Only 7 out of 354 actually belong there. Could this be fixed, and if so how? --Auric (talk) 16:53, 18 February 2017 (UTC)

  • I think most of that miscategorizations comes from CL (or rather CyL) standing for Castille-Leon, a region of Spain. Before deleting, better move them to Category:Castilla y León. -- Tuválkin 01:07, 19 February 2017 (UTC)
✓ Done. -- Tuválkin 10:14, 19 February 2017 (UTC)

IAEA images

Hi, I came accross IAEA Flickr stream, which contains many useful free images. Could someone with a bot upload them? Independently some files were imported, but are not in Category:Files from IAEA Flickr stream. Could someone with a bot add them there? Thanks, Yann (talk) 22:10, 18 February 2017 (UTC)

Those are NC, so you might want to have a word with the UNESCO Wikimedian in Residence, User:John Cummings. Slowking4 § Sander.v.Ginkel's revenge 23:41, 18 February 2017 (UTC)
Hmmm, well I have no idea how to do it with a bot... You might want to try using Flickr2Commons to find and upload any Wikipedia compatible license images. It looks like the Flickr account uses a mix of different licenses and Flickr2Commons will filter out the ones that aren't compatible with Wikimedia Commons.... --John Cummings (talk) 21:10, 19 February 2017 (UTC)
Flickr2Commons detects about 12,000 images with a free license among a total of 18,000+ images. But then it crashes on Chrome+Win7 on a 8 GB RAM machine... :/ Regards, Yann (talk) 21:50, 19 February 2017 (UTC)

February 19

Tech News: 2017-08

19:25, 20 February 2017 (UTC)

European Space Agency release its content under a CC BY-SA 3.0 IGO license


If someone could import them en masse with a bot or something, that would be great.

Regards. --Thibaut120094 (talk) 14:54, 20 February 2017 (UTC)

Wow! We should also review deleted files from ESA. Regards, Yann (talk) 15:35, 20 February 2017 (UTC)
@: FYI. Andy Mabbett (talk) 16:53, 20 February 2017 (UTC)
This is incredible news!! It seems from the announcement, however, that we can't automatically regard their content as CC BY-SA as we can automatically regard NASA content as PD. It's just a statement of intent to apply the license a lot more broadly. Am I reading that right? If so, hopefully they will retrospectively apply it to most of their online archives soon, if they haven't already.
Addendum: I found their use policy, which confirms that the CC BY-SA license only applies "where explicitly so stated". So it looks like this is an announcement that they plan to explicitly state it in a lot more places in the near future. I can't wait! A2soup (talk) 04:07, 21 February 2017 (UTC)
Absolutely fantastic, but as has been suggested, this should only be taken as moving forward under specific circumstances. It only applies, for now, to material produced in whole by ESA, and not to anything created under partnership with industry or other agencies. So, we'll still have to exercise caution when uploading, but things are about to get a whole lot easier. Very exciting, as I especially hope this opens up even more of the Rosetta archive. Huntster (t @ c) 08:41, 21 February 2017 (UTC)

Recruiting photographers

How do we recruit photographers online and/or offline? So far, Wikipedians demand more free photos as they are challenging the use of copyrighted photos. If more photographers are recruited, they can let us share some of their own photos of things and people, living and deceased. Thoughts? --George Ho (talk) 09:52, 19 February 2017 (UTC)

I searched for an image of w:Jill Saward at Flickr but found none except photos of the singer with the same name. I don't live in the UK. Nevertheless, I found Twitter webpage and official website of her husband (widower), Gavin Drake. I don't know what to appropriately say to him, given that Saward died last month. I don't feel comfortable taking advantage of a grieving widower just for an image of Saward. If contacting him is not a good choice, how else can I convince any photographer to contribute to Commons? --George Ho (talk) 19:07, 19 February 2017 (UTC)

One possibility is to look at the history page of the Wikipedia article and look for the people that have contributed to the article. In particular when they contribute also to Commons, you can ask them. Possibly they have ideas how to obtain the desired photos. Success. Wouter (talk) 19:30, 19 February 2017 (UTC)
@Espresso Addict, Philip Cross, Aircorn, Stephen, Marchjuly: Your thoughts about this? --George Ho (talk) 20:01, 19 February 2017 (UTC)
Also pinging Jayron32. --George Ho (talk) 20:03, 19 February 2017 (UTC)
Firstly, please don't use "free" and "copyrighted" as if they were opposites. They are not. We have lots of free content here that is protected by copyright.
Secondly, we already have File:Jill 080627b.jpg, for which the uploader claims to have sent in permission evidence via e-mail.
As for the subject heading, if all you want are free media, then you don't necessarily need to recruit photographers to Commons – you "only" need to convince them to publish their content under a free license. Creative Commons does a lot of work on that. LX (talk, contribs) 23:56, 19 February 2017 (UTC)
@LX: Oh... the Saward photo is still pending. Hmm... I didn't know that. That case aside, I have another case of obtaining a permission to use a photo of w:A. A. Gill. I contacted some photographers about the Commons. I provided them links to contribute to Wikimedia Commons, like a Welcome page and COM:Flickr files. I did imply permission to use their photos, but I also said that they can upload their photos to Commons themselves. I've not yet received a response from them. Well, I did receive one email saying that this photo is not his work but someone else's. I replied and still am awaiting another response. --George Ho (talk) 01:32, 20 February 2017 (UTC)
we have a desperate need for photos. why don't you submit an idea lab for small grants for photo stringers of living people at premieres and book festivals? Slowking4 § Sander.v.Ginkel's revenge 04:15, 20 February 2017 (UTC)
@Slowking4: I am not sure which "idea lab" you refer to. Can you clarify? Commons:Picture requests looks messy and requires a lot of cleanup. --George Ho (talk) 04:36, 20 February 2017 (UTC)
it is the on ramp to grants - here you go meta:Grants:IdeaLab/Inspire - Slowking4 § Sander.v.Ginkel's revenge 12:04, 20 February 2017 (UTC)
@Slowking4: Reading IdeaLab, that requires a lot of resources and effort. What happens if I just bring up the idea and then not contribute afterwards? --George Ho (talk) 20:09, 20 February 2017 (UTC)
idealab is where you pitch ideas - the probability of implementation goes up as the quality of the planning goes up. - it will take resources to stipend photographers. Slowking4 § Sander.v.Ginkel's revenge 20:33, 20 February 2017 (UTC)

@Slowking4: I created meta:Grants:IdeaLab/In memoria and commemorations, but it's more about images of people who became recently deceased and ones who are now deceased. When shall I create an idea about recruiting photographers in general? --George Ho (talk) 22:26, 20 February 2017 (UTC)

I refer people to en:WP:A picture of you (if I'm asking for a picture of them) or en:Wikipedia:Images from social media (if I'm asking them to donate pictures they've taken themselves, and then posted to sites like Twitter or Facebook. Note that the pages are on Wikipedia, because that's the brand that most people recognise, and that they are written in plain language, with (I hope) no Wikipedia jargon. Andy Mabbett (talk) 15:15, 22 February 2017 (UTC)
i have also started m:Grants:IdeaLab/images of living people. we will see if we can pry loose some expense money. Slowking4 § Sander.v.Ginkel's revenge 17:16, 22 February 2017 (UTC)

February 20

WikiProject Women

Hi, I created this project. Your contributions, opinions, suggestions, and critics are welcome. Regards, Yann (talk) 13:35, 21 February 2017 (UTC)

You left out Annie Besant's name in the caption. Not too sure why you chose her, since her main life's work was Theosophy, and she held a number of extremely bizarre (and sometimes repulsive) beliefs in connection with it... AnonMoos (talk) 14:19, 22 February 2017 (UTC)
Hi, I don't understand what you want to say by "Annie Besant's name in the caption". The objective is to help the creation and promotion of images of women, by women and for women, irrespective of their origin, opinions, etc. And she also was a women's rights activist, although she is better known for her work on theosophy. Regards, Yann (talk) 14:41, 22 February 2017 (UTC)
Dude, her name ain't in the caption. It says "British women's rights activist, writer and orator", but does not include her name. To see some of the dark side of the swirling vortex of strange occultism that was Theosophy, start with en:Root race. Would you really include a photo of en:Ilse Koch on your page? (Not that I'm comparing Besant and Koch...) AnonMoos (talk) 15:11, 22 February 2017 (UTC)
dude, go change the caption. not sure why you want to sea lion a project with anti-theosophy ideology. Slowking4 § Sander.v.Ginkel's revenge 17:10, 22 February 2017 (UTC)
If I were to edit the caption, it would basically be impossible for me not to change it to be a lot less evasively blandly flattering than it is now. And if she were just a follower or believer in theosophy, it might not be so relevant -- but she was in fact its main founder, and so basically a cult leader... AnonMoos (talk) 22:27, 22 February 2017 (UTC)

The first video uploaded to YouTube

Hello.Is there a benefit of being here?See Commons:Deletion requests/File:Me at the zoo.webm.Thank you --ديفيد عادل وهبة خليل 2 (talk) 13:16, 22 February 2017 (UTC)

Cropping an image to remove possible non-de minimis image of a fictional character?

Please see this deletion request discussion: Commons:Deletion requests/File:ShimajiroDisplayChina.jpg - A user is arguing that the image should be deleted because there is a cardboard cutout of a copyrighted character but I proposed cropping it out so what is left is de minimis (it is a display of Shimajiro-themed merchandise in China) - the character originates from Japan WhisperToMe (talk) 15:33, 22 February 2017 (UTC)

Babel box is too wide


I have noticed that the box displaying the language abilities of users is too wide, as a result of which there is a distinct white edge (empty space) on the right.

The width of the box is 250px, but I think this should be 242. The examples on the right will hopefully show you the difference.

The width appears to be defined by class="mw-babel-wrapper".

I don't know if this issue can be resolved locally, but perhaps a developer or somebody with editing rights can look into it. ErikvanB (talk) 20:36, 22 February 2017 (UTC)

I've often noticed little things like this. The only answer is usually to design your own, as you've done above, or as I've done on my user page. I've also seen a lot of messy pages where the user doesn't seem to worry about it. These are the people whose pictures and posters on their walls at home are all crooked and uneven... Face-wink.svg lNeverCry 21:39, 22 February 2017 (UTC)
Ha! Ha! Thanks. ;-) ErikvanB (talk) 22:29, 22 February 2017 (UTC)
Also please note that there is deprecated template {{Babel}} and MadiaWiki's parser function {{#babel}}. There are differences between them. --jdx Re: 04:38, 23 February 2017 (UTC)
  • You also have the option of adding the language categories to your user page manually. Daphne Lantier 08:25, 23 February 2017 (UTC)

February 23

Redirected categories, inmovable interwikis

Is there any possibility that interwikis migrate with redirecting category, see for example: Otherwise, it is very dedious to fix these manually--Estopedist1 (talk) 10:13, 23 February 2017 (UTC)

I am not sure that they should migrate. Ruslik (talk) 18:51, 23 February 2017 (UTC)

Intersection categories

I'm working on some categories; see, for example, Commons:Categories for discussion/2017/02/Category:Taken with Canon EOS 60D and Canon EF 100mm F2.8L Macro IS USM. It looks like it's a straightforward intersection of Category:Taken with Canon EOS 60D and Category:Taken with Canon EF 100mm F2.8L Macro IS USM. This should be doable with the Help:FastCCI tool, but it just hangs. Trying it on the command line, I think it's broken?

 $ wscat -c 'wss://'
 connected (press CTRL+C to quit)

There are, for Canon alone, about fifty camera models and sixty lens models. The combinatorial blowup here is potentially over three thousand, and that doesn't even include filters. Is FastCCI broken? Should there be three thousand intersecting categories instead of two sets of fifty and sixty? I don't think intersections are warranted here, but I feel like I'm missing something. COM:CAT#Principles says that intersection categories are sometimes reasonable, but I don't think they make sense here, especially if we have tools which can perform the intersection. --grendel|khan 21:45, 18 February 2017 (UTC)

I don't know about FastCCI, but a search query can give the intersection: incategory:"Taken with Canon EOS 60D" incategory:"Taken with Canon EF 100mm F2.8L Macro IS USM". Creating the thousands of intersection categories doesn't sound like a great idea. Next somebody will want Category:Taken with Canon EOS 60D and Canon EF 100mm F2.8L Macro IS USM on 2017-02-19. --ghouston (talk) 03:59, 19 February 2017 (UTC)
Also, once you choose a particular set of intersections, you make it harder to do different intersections. E.g., incategory:"Taken with Canon EOS 60D" incategory:"CC-Zero". There's also a good chance that a lot of this intersection work will be undone when Structured data is implemented. --ghouston (talk) 04:11, 19 February 2017 (UTC)
Thank you; that search query is extremely helpful! (It's even worse than you think; there's intersection categories for all manner of equipment; see Category:Taken with Canon EOS 60D and Canon EF 100mm F2.8 Macro USM + Raynox DCR-250, for example.) One thing that worries me is that policy (COM:CAT) is mostly silent on this. ("A category can combine two (or more) different criteria; such categories are called "compound categories" or "intersection categories"."--there's no guidelines for when it is or isn't appropriate to do this sort of thing.) Is this the right place to seek a firmer policy, or at least to push for a flatter categorization for the camera-equipment categories? --grendel|khan 08:37, 19 February 2017 (UTC)
  • As I mentioned at the deletion discussion in question, when it comes to camera + lens combinations, I don't find the intersection to be too problematic. Yes, there are numerous possible categories. However, a simple camera + lens combination is useful for showing the potential performance of a certain body with a certain lens, as camera performance can vary considerably based on lens used (and vice versa). I don't think it will reach the point where we have Camera + lens + date, and if it does such intersections would be deletable under the simplicity principal at COM:Categories.
@--ghouston: I thought the more advanced search tools allowed us to select how deep of a search to go (i.e. up to 3 subcategories). In such a case, if you tell the tool to search down three levels wouldn't searching "Taken with Canon EOS 60D" automatically bring up images in "Taken with Canon EOS 60 and foo lens"? — Chris Woodrich (talk) 08:08, 23 February 2017 (UTC)
Probably, but I expect it would be slow when there are a lot of categories to be scanned. --ghouston (talk) 08:49, 24 February 2017 (UTC)

What does it mean?

This message is followed by a table with metadata in every file page.

"This file contains additional information, probably added from the digital camera or scanner used to create or digitize it.

If the file has been modified from its original state, some details may not fully reflect the modified file."

What does the second sentence mean? Please explain it.--維基小霸王 (talk) 16:57, 21 February 2017 (UTC)

This refers to en:Exif data. Image editing programs can remove or modify that data. For instance, en:Exif#Problems has an example where the embedded thumbnail may not be updated when an image is modified, so the thumbnail will show an outdated version of the image. clpo13(talk) 17:26, 21 February 2017 (UTC)

Would it be possible to insert a link to w:EXIF, into the text mentioned above? Andy Mabbett (talk) 20:17, 24 February 2017 (UTC)

February 22

Is there a way to export all vector graphics from a pdf automatically?

Hi all

I'm working with some open license publications that have 100s of open license vector graphics graphs in them. Currently I have worked out that I can extract them one page at a time using Illustrator or Inkscape as svg files, however it is taking far to long. Can anyone think of a way to extract the images in some way more automatically? Is there a way to automated making svg files from all pages in a pdf? There are going to be 1000s of pages in total, maybe even 10,000s.


--John Cummings (talk) 09:17, 24 February 2017 (UTC)

I have never used it, but you might give a try to PDF2SVG. --jdx Re: 10:10, 24 February 2017 (UTC)
Thanks Jdx, unfortunately I can't get it to run for some reason.... I'll keep trying. --John Cummings (talk) 16:10, 24 February 2017 (UTC)
OMG Jdx, I got it working and it is amazing. It works perfectly :) :) :), I will write a guide. --John Cummings (talk) 16:21, 24 February 2017 (UTC)

Removal of out of scope mass announcements

I would like to propose that mass announcements and posts which are posted across many projects but have no obvious connection for Wikimedia Commons, or raise no issues for the Wikimedia Commons project, can be removed from the Village pump, and any contributor that takes action to do this may do so with a presumption of good faith from the wider Community. As stated at the top of this noticeboard, it "is used for discussions of the operations, technical issues, and policies of Wikimedia Commons", so posts which are simple blanket mass announcements and raise nothing of obvious interest about the operation, technical issues or policies of Wikimedia Commons should not be here, while those that meet the scope should be kept for discussion. For notices of wide general interest to the entire Wikimedia Community, CentralNotice has the best impact, and using that procedure for mass communication helps to keep this noticeboard in scope. Thanks -- (talk) 20:02, 20 February 2017 (UTC)

If you're referring to "Tech News: 2017-08" directly above, then I would prefer that they keep posting it here. There are some other announcements which are less relevant to Commons, but also less frequent... AnonMoos (talk) 01:04, 21 February 2017 (UTC)
How about replacing post like that one, which are at best tangential, with a link to the master post on meta? By the way you can subscribe and have copies on your talk page, if you really want transcluded copies. -- (talk) 02:18, 21 February 2017 (UTC)
Edit summaries or the Hovercards/Page Previews beta feature are available on Wikimedia Commons. How did you get to "no obvious connection"? --Malyacko (talk) 13:35, 21 February 2017 (UTC)

If there's a specific regular post that is cluttering this page, then we can debate removing it. Otherwise this just seems like rule creep. The scope quoted above has remained unchanged since the page was created and I doubt that a whole lot of thought went into it. It used to also say "Other discussions are welcome here until pages are created to hold them" and now we have some other pages which are listed. There is more variety here than that "scope" implies. Including, for example, discussions about new websites containing free images, or noting legal and political threats/opening of free content. Even notices of wide community impact (such as WMF) may warrant discussion among Commons regulars, something that the central notice does not provide. Surely this should be a page where people can post whatever they, in good faith, feel the community might like to discuss (that doesn't have a dedicated forum) without having to read 101 rules or run the risk of someone rudely dismissing their edit. I'd rather this page supported good-faith-inclusionism than some kind of unilateral "good-faith" removalism. -- Colin (talk) 13:53, 21 February 2017 (UTC)

+1. Yann (talk) 14:10, 21 February 2017 (UTC)
Symbol keep vote.svg Agree with Colin --Jarekt (talk) 17:38, 21 February 2017 (UTC)
+1 --El Grafo (talk) 10:40, 22 February 2017 (UTC)

really dislikes the tech news has advocated several times to get rid of tech news on this page. See also discussions at Commons:Village_pump/Archive/2016/02#Tech_News:_2016-07 and Commons:Village_pump/Archive/2016/11#Tech_News:_2016-44. I'll note that this page can be unsubscribed from the delivery list by simply removing it from m:Global message delivery/Targets/Tech ambassadors. Matma Rex (talk) 16:50, 22 February 2017 (UTC)

Please don't make statements about what is in my head, unless you have mind reading powers. I do not dislike tech news, and I find it interesting to check over, which I can do by following it on meta and anything tech and more urgent I tend to find out about from announcements on Wikitech-l. Movement wide newsletters and announcements posted by bots are not a good fit, as frustratingly to ask questions you have to go to yet another forum off Wikimedia Commons. If someone who contributed to the tech newsletter were to customize the post, or even just introduce it to point out anything of interest for Commons contributors, that would be great and would indicate some human interest rather than automatons relentlessly delivering box-ticking of the Comms plan. -- (talk) 17:06, 22 February 2017 (UTC)
yeah, some projects have such self regard, that they like broadcasting on noticeboards. some others narrow cast on user talk after opt in. i prefer the latter; it is more considerate. Slowking4 § Sander.v.Ginkel's revenge 17:13, 22 February 2017 (UTC)
Symbol oppose vote.svg Oppose this proposal. Seems to be a solution to a non-existent problem to me. Sure, there are other methods for mass announcements, but removing these announcements from the village pump don't do any good for the community. And I think some conservative users here prefer announcements on the village pump over CentralNotice. -- Poké95 06:59, 26 February 2017 (UTC)

February 21

Sorting character for unidentified subjects

Please could we discuss it in Category talk:unidentified subjects#Sorting character thanks--Pierpao.lo (listening) 10:53, 25 February 2017 (UTC)

Multiple copyvio uploads 23:12, 25 February 2017 (UTC)

✓ Done Files deleted, user warned. Yann (talk) 23:33, 25 February 2017 (UTC)

February 26

Reuploading a large number of files

What's the easiest way of reuploading lots of files? I've tried using VicuñaUploader, but it refuses to reupload files. Jc86035 (talk) Use {{re|Jc86035}}
to reply to me
10:31, 25 February 2017 (UTC)

  • By "reupload" do you mean to write over an existing file? (I don't have an answer, just trying to clarify the question.) - Jmabel ! talk 17:48, 25 February 2017 (UTC)
In case the answer to Jmabel’s question is yes, then it is a feature: In Commons:VicuñaUploader you can read “It checks file-name conflicts before sending files to Commons”. — Speravir – 18:59, 25 February 2017 (UTC)
  • Sure, but the lack of a manual override is a bit of a misfeature. - Jmabel ! talk 21:51, 25 February 2017 (UTC)
@Jmabel: It depends a bit from point of view, I think. At least it should not be made too easy to overwrite files to hinder (?) vandalism. — Speravir – 19:05, 26 February 2017 (UTC)
Clearly it shouldn't be something absolutely anyone can do, but equally clearly there should be some way to give someone believed to be reliable a means to do this. - Jmabel ! talk 22:17, 26 February 2017 (UTC)
@Speravir: Is there another program or script that allows reuploading? Jc86035 (talk) Use {{re|Jc86035}}
to reply to me
05:49, 26 February 2017 (UTC)
@Jc86035: Just before the cited sentence there is a reference to Commonist, see Commons:Commonist. The VicuñaUploader is more or less a fork of it. (Update: More less. “Fork” is very probably the wrong word here. — Speravir – 21:56, 26 February 2017 (UTC)) And then there is Pattypan, like VicuñaUploader programmed by Yarl, but I do only know, that it exists. — Speravir – 19:05, 26 February 2017 (UTC)
@Speravir: I've tried Commonist and it does work, but it annoyingly overwrites all the description pages. Jc86035 (talk) Use {{re|Jc86035}}
to reply to me
12:23, 27 February 2017 (UTC)
Though it may not help: There is also Commons:Upload tools. I didn’t myself know of this until today. — Speravir – 00:37, 28 February 2017 (UTC)

Tech News: 2017-09

19:55, 27 February 2017 (UTC)

February 28

سلام، چگونه می توانم به یک مقاله ویکی پدیا یک عکس اضافه کنم؟

اینجا را ببین.. Maybe a Persian speaker can give you more help.—Odysseus1479 (talk) 09:11, 28 February 2017 (UTC)

ships menus

Menu Transatlantique French line Liberte.jpg

I have a few questions:

  • Wat is the date of the menu?
  • Who is the painter/illustrator of the French frigate (a ship within a ship) on the menu?
  • I have made a detail scan and removed stains. Is it posible to use the detail picture to correct the main picture?Smiley.toerist (talk) 11:48, 28 February 2017 (UTC)
@Smiley.toerist:, maybe adding the Category:1856 in France could be useful. Lotje (talk) 12:54, 28 February 2017 (UTC)
It seems to be the same image as this one by "F. Roux". Oddly, the menu says the ship was launched 26 April 1856, contradicting the info at en:French frigate Audacieuse (1856). They seem to have used the same cover image at least as early as 1950, but this ebay sale of a very similar menu dates it to 1961. I guess you are planning to change the source from "own work"? - Themightyquill (talk) 12:57, 28 February 2017 (UTC)
Oeps, I forgot the own work part after uploading many of my own pictures. The double licences can be confusing (the painting and the menu itself). In the detail picture of the painting I removed the Liberty ship categories to avoid confusion. The painting is old enough no matter wich Roux painter it is (fr:Famille Roux (peintres de marine)). The abbreviation ´Cie´ I know but wat is ´Gle´?Smiley.toerist (talk) 13:44, 28 February 2017 (UTC)
Would it make sense to link the Category:Scale model of Audacieuse-MnM 23 MG 10 to the Category:French frigate Audacieuse? Lotje (talk) 14:04, 28 February 2017 (UTC)
✓ DoneSmiley.toerist (talk) 20:33, 28 February 2017 (UTC)
@Smiley.toerist: As per en:Compagnie Générale Transatlantique, it's CompagnIE GénéraLE. - Themightyquill (talk) 19:23, 28 February 2017 (UTC)